(en)A vehicle detection system ( 1 ) comprising at least one detection unit ( 2 ). Each detection unit corresponds to a detection volume (S n ) and each detection unit comprises at least two spaced-apart magnetic field sensors including first and second magnetic field sensors ( 6 a, 6 b ; FIG. 3 ). Each magnetic field sensor is configured to provide a series of instantaneous magnetic field measurements spaced apart in time. A processing unit ( 3 ) is configured to receive the time series of instantaneous magnetic field measurements. The processing unit is configured to store the time series of instantaneous magnetic field measurements in a table ( 8 ; FIG. 5 ). The processing unit is configured to determine the presence of a vehicle ( 7 ) in each detection volume in dependence upon the magnetic field measurements in the table.

1.ApplicationNumber: US-201615146657-A
1.PublishNumber: US-2016328961-A1
2.Date Publish: 20161110
3.Inventor: GARCÉS CADENAS FRANCISCO JAVIER
Rovirosa Xavier Marti
JUNGWIRTH TOMÁS

4.Inventor Harmonized: GARCÉS CADENAS FRANCISCO JAVIER(ES)
Rovirosa Xavier Marti(ES)
JUNGWIRTH TOMÁS(ES)

5.Country: US
6.Claims:
(en)A vehicle detection system ( 1 ) comprising at least one detection unit ( 2 ). Each detection unit corresponds to a detection volume (S n ) and each detection unit comprises at least two spaced-apart magnetic field sensors including first and second magnetic field sensors ( 6 a, 6 b ; FIG. 3 ). Each magnetic field sensor is configured to provide a series of instantaneous magnetic field measurements spaced apart in time. A processing unit ( 3 ) is configured to receive the time series of instantaneous magnetic field measurements. The processing unit is configured to store the time series of instantaneous magnetic field measurements in a table ( 8 ; FIG. 5 ). The processing unit is configured to determine the presence of a vehicle ( 7 ) in each detection volume in dependence upon the magnetic field measurements in the table.

7.Description:
(en)CROSS REFERENCE TO RELATED APPLICATIONS
This application claims the benefit of European Application No. 15380017.2 filed on May 5, 2015.
FIELD OF THE INVENTION
The present invention relates to a system for detecting vehicles, in particular motor vehicles, such as motorcycles, automobiles, minibuses, buses, trucks or lorries.
BACKGROUND
Vehicle detection systems can be used (particularly in urban areas) in many different ways. For example, they can be used to monitor traffic flow, to locate car parking spaces and to control traffic control signals (often referred to simply as “traffic lights”).
Some vehicle detection systems use magnetic field sensors to sense local disturbances in the magnetic field of the Earth. For example, some vehicle detection systems employ a number of discrete magnetic field sensors, each sensor measuring the magnetic field in a respective volume of space. Each sensor records a baseline level of magnetic field when no vehicle is present and can detect the presence of a vehicle when the measured magnetic field deviates from the baseline level by more than a given amount. U.S. Pat. No. 6,546,344 B1 describes an example of such a system.
However, when magnetic field sensors are positioned close together, for example in adjacent car parking spaces or traffic lanes, their measurements may be spurious or inaccurate. This may arise due to perturbations to the Earth's magnetic field resulting from the proximity of other vehicles in adjacent volumes of space. A magnetic field perturbation within an unoccupied volume of space may be large enough to trigger a false positive or to prevent detect of a vehicle.
SUMMARY OF THE INVENTION
The present invention seeks to provide an improved vehicle detection system based on measurements of magnetic fields.
According to a first aspect of the present invention there is provided a vehicle detection system comprising at least one detector. Each detector is configured to measure a magnetic field in a respective volume of space. Each detector comprises at least two spaced-apart magnetic field sensors, including first and second magnetic field sensors. Each magnetic field sensor is configured to provide a time series of instantaneous magnetic field measurements. The vehicle detection system also comprises a processing unit configured to receive the time series of instantaneous magnetic field measurements. The processing unit is also configured to store the time series of instantaneous magnetic field measurements in a table. The processing unit is also configured to determine presence of a vehicle in each volume in dependence upon the magnetic field measurements in the table.
This can be used to reduce or even avoid sensor blind spots due to uneven variations in magnetic field.
A vehicle detection system may comprise a third magnetic field sensor.
Magnetic field sensors may be spaced apart by at least 15 cm, at least 30 cm, at least 50 cm or by at least 1 m. Magnetic field sensors may be spaced apart by a distance between 30 cm and 40 cm.
The processing unit may be configured to determine a speed of a vehicle in a volume. The processing unit may be configured to determine that a vehicle enters a volume. The processing unit may be configured to determine that a vehicle exits a volume. The processing unit may be configured to determine the direction of travel of a vehicle through a volume.
Each detector may comprise first and second units which are spaced apart, and a link connecting the first and second units. The first and second magnetic field sensors may be disposed in the first and second units respectively. The link may be flexible. Each detector may comprise at least one further unit. Each further unit may contain a magnetic field sensor and be connected to the other units by one or more links.
The magnetic field sensors may be anisotropic magnetoresistance sensors. Each magnetic field sensor may be a three axis sensor. Each detection unit may be sealed against particles of dirt and/or dust. Each detection unit may be waterproof. Each detection unit may be encapsulated to the IP-68 certification standard.
The vehicle detection system may comprise a cable comprising two or more detectors spaced along the cable. The detectors may be connected to the cable in a bus configuration. The cable may comprise one or more energy lines and one or more data lines. The detectors 2 may be connected to the data line(s) and power line(s) in a bus configuration. The cable may comprise cladding or armour surrounding the energy line(s) and data line(s).
Thus, a failure of, or damage to a detector will not affect other detectors connected in parallel to the same cable.
The processing unit may be connected to two or more cables, each cable comprising two or more detectors positioned along the length of the cable.
The vehicle detection system may further comprise a short-range wireless transceiver configured to exchange signals with a corresponding short-range wireless transceiver of a vehicle so as to capture the identity of vehicle.
The processing unit may be configured to determine the presence of a vehicle in each volume in dependence upon the magnetic field measurements corresponding to two or more volumes.
Thus, the processing unit may be able to compensate (partially or fully) for magnetic field variations caused by a vehicle which is present in an adjacent volume of space. This can improve the accuracy and reliability of detecting whether or not a vehicle is present in a volume of a space.
The processing unit may be configured to determine the presence of a vehicle in each volume in dependence upon the magnetic field measurements corresponding to two or more consecutive points in the time series.
Thus, the accuracy and reliability of detecting a vehicle in a volume of a space may be improved, because magnetic field variations during entry or exit of a vehicle may be larger than static changes in magnetic field before or after the entry or exit of the vehicle.
The processing unit may be configured to detect a start of an event and an end of an event in a volume in dependence upon a variance of the respective magnetic field measurements calculated based on a number of consecutive points in the time series. The processing unit may determine the presence of a vehicle in each detection volume in dependence upon a change in the respective magnetic field measurements between the start and end of an event.
The processing unit may be configured to store an occupancy flag corresponding to a detection volume in a second table in response to determining the presence of a vehicle in that detection volume. The processing unit may be configured to update the occupancy flag for a detection volume after the end of a detected event in that detection volume.
The processing unit may be configured to determine the presence of a vehicle in each detection volume in dependence upon the magnetic field measurements in the respective detection volume and also the magnetic field measurements in each adjacent detection volume which is flagged as occupied in the second table.
The processing unit may be configured such that when an event is not detected, the processor will execute a policing control check such that the occupancy flag of each detection volume is reversed if a threshold number of magnetic field measurements are consecutively stored in the table, each of the threshold number of magnetic field measurements being inconsistent with the respective occupancy flag stored in the second table.
The vehicle detection system may be for monitoring availability of car parking spaces. The vehicle detection system may be for monitoring the volume and speed of traffic on a road. The vehicle detection system may be for detecting the numbers and distribution of vehicles around a junction controlled by a traffic signalling system.
The processing unit may be configured to store a data structure defining, for each detection volume, a number of coupled detection volumes which at least partially overlap the detection volume. The processing unit may be configured to receive the time series of instantaneous magnetic field measurements. The processing unit may be configured to store the time series of instantaneous magnetic field measurements in a first table. The processing unit may be configured to store occupancy flags corresponding to each detection volume in a second table, the occupancy flags indicating presence of a vehicle in that detection volume. The processing unit may be configured to store baseline magnetic field values corresponding to each detection volume in a third table, the baseline magnetic field values being instantaneous magnetic field measurements obtained in the absence of vehicles. The processing unit may be configured to store a fourth table comprising, for each detection volume, correction values corresponding to each coupled detection volume. The processing unit may be configured to calculate, for each detection volume, difference values between the corresponding magnetic field measurements stored in the first table and a corrected baseline obtained by summing the baseline magnetic field value for the detection volume and the correction values corresponding to each coupled detection volume. The processing unit may be configured to determine presence of a vehicle in each detection volume in dependence upon the corresponding difference values. The processing unit may be configured, in response to determining the presence of a vehicle in a detection volume, to update the corresponding occupancy flag in the second table and, for each coupled detection volume, to store the difference values corresponding to the coupled detection volume as correction values corresponding to the detection volume.
According to a second aspect of the invention there is provided a method comprising receiving a time series of instantaneous magnetic field measurements from at least two spaced apart magnetic field sensors including first and second magnetic field sensors. The method also comprises storing the time series of instantaneous magnetic field measurements in a table. The method also comprises determining presence of a vehicle in each volume in dependence upon the magnetic field measurements in the table.
The method may further comprise determining a speed of a vehicle in a detection volume.
The method may include capturing the identity of a vehicle by exchanging signals with a vehicle using a short range wireless communications protocol such as ZigBee®.
The presence of a vehicle in each detection volume may be determined in dependence upon the magnetic field measurements corresponding to two or more detection volumes. The presence of a vehicle in each detection volume may be determined in dependence upon the magnetic field measurements corresponding to two or more consecutive points in the time series.
The method may include detecting a start of an event and an end of an event in a detection volume in dependence upon a variance of the respective magnetic field measurements calculated based on a number of consecutive points in the time series. The presence of a vehicle in each detection volume may be determined in dependence upon a change in the respective magnetic field measurements between the start and end of an event.
The method may include storing an occupancy flag corresponding to a detection volume in a second table in response to determining the presence of a vehicle in that detection volume.
The occupancy flag for a detection volume may be updated after the end of a detected event in the respective detection volume. The presence of a vehicle in each detection volume may be determined in dependence upon the magnetic field measurements in the respective detection volume and in each adjacent detection volume which is flagged as occupied in the second table.
When an event is not detected, the method may include executing a policing control check such that the occupancy flag of each detection volume is reversed if a threshold number of magnetic field measurements are consecutively stored in the table, each of the threshold number of magnetic field measurements being inconsistent with the respective occupancy flag stored in the second table.
According to a third aspect of the invention there is provided a computer program comprising instructions which, when executed by at least one processing unit, cause the processing unit to perform a method according to the second aspect.
According to a fourth aspect of the invention there is provided a computer readable medium (which may be non-transitory) storing the computer program according to the third aspect.
According to a fifth aspect of the invention there is provided a method of authenticating a vehicle detected by a vehicle detection system which includes one or more detectors, a transceiver device corresponding to each detector and a processing unit. Each detector is configured to measure magnetic fields in a volume of space at least partially overlapping a corresponding space for a vehicle. Each transceiver device is arranged proximate to the respective space for a vehicle and includes a wireless personal area network module. The processing unit is configured to communicate with an external device and to determine presence of a vehicle in each space in dependence upon measurements from one or more detectors. The method includes, in response to detecting a vehicle entering a space corresponding to a detector, one or more transceiver devices broadcast a plurality of challenge messages according to a schedule. Each challenge message has a corresponding correct response. The one or more transceiver devices broadcast each challenge message before a subsequent challenge message is broadcast. The method also includes, for each challenge message broadcast, in response to one or more transceiver devices receiving at least one corresponding response message broadcast by a wireless personal area network transceiver in or on the vehicle, incrementing a number of received responses and checking whether the response message corresponds to the correct response. The method also includes, in response to the number of received responses exceeds a first threshold and a fraction of correct responses exceeds a second threshold, authenticating the vehicle by transmitting a message to the external device to indicate that an authorised vehicle has entered the space.
Thus, because transceiver devices are proximate to, rather than within, the corresponding spaces for vehicles, a parked vehicle is less likely to block or interfere with transmissions between the transceiver device and a wireless personal network area transceiver in or on the vehicle. The condition of receiving a minimum number of response messages, a minimum fraction of which must be correct, before authenticating a vehicle further improves the reliability of authenticating an arriving vehicle.
The plurality of challenge messages may be broadcast according to a predetermined schedule. The plurality of challenge messages may be broadcast according to a dynamically determined schedule. Any second and subsequent response messages received in reply to a specific challenge message may be ignored. The processing unit may include a list of challenge messages and corresponding correct responses. The challenge messages which are broadcast may be selected at random from a list stored in the processing unit.
The vehicle detection system may include two or more detectors and two or more corresponding transceiver devices. Challenge messages may be broadcast by any transceiver devices corresponding to spaces which are adjacent to the space into which a vehicle has been detected entering. Two or more detectors may be connected into a branch. Challenge messages may be broadcast by any transceiver devices corresponding to detectors connected to a single branch.
Thus, the reliability of authenticating an arriving vehicle further improved because two or more transceiver devices broadcast the same challenge messages. Even when broadcasts from a transceiver device are blocked or interfered with by parked or moving vehicles, transmission by two or more transceiver devices increases the probability that at least one transceiver device has a favourable transmission path.
The method may also include, in dependence upon the number of received responses exceeds the first threshold and a fraction of incorrect responses exceeds a third threshold, transmitting a message to the external device indicating that an unauthorised vehicle has entered the space.
The method may also include, in dependence upon a predetermined duration has elapsed and the number of received responses does not exceed the first threshold, transmitting a message to the external device indicating that an unauthorised vehicle has entered the space.
The method may also include, in dependence upon the vehicle is authenticated, one or more transceiver devices broadcast an identity request message. The method may also include, in response to one or more transceiver devices receiving a message including an identity of the vehicle, transmitting the identity of the vehicle to the external device.
Each transceiver device may further include at least one signalling device, each signalling device having two or more output states. The method may further include, in response to authenticating the vehicle, controlling the signalling device(s) of the transceiver device corresponding to the space to switch from a first output state to a second output state.
The first output state may correspond to a first colour. The second output state may correspond to a second colour. The signalling device may emit light. The first output state may correspond to steady illumination of the first colour. The second output state may correspond to steady illumination of the second colour. The second output state may correspond to intermittent illumination of the first or second colours.
The method may also include, in response to transmitting a message to the external device indicating that an unauthorised vehicle has entered the space, controlling the signalling device(s) of the transceiver device corresponding to the space to switch from the first output state to a third output state.
The third output state may correspond to a third colour. The third output state may correspond to steady illumination of the third colour. The third output state may correspond to intermittent illumination of the first, second or third colours.
Each signalling device may include at least one light emitting diode. Each signalling device may be a multi-coloured light emitting diode. Each signalling device may be an electrochromic device or a mechanical device.
Each detector may include at least two spaced-apart magnetic field sensors including first and second magnetic field sensors. Each magnetic field sensor may be configured to provide a time series of instantaneous magnetic field measurements to the processing unit. Detecting a vehicle entering a vehicle parking bay may include receiving a time series of instantaneous magnetic field measurements from each detector. Detecting a vehicle entering a vehicle parking bay may also include storing the time series of instantaneous magnetic field measurements in a table. Detecting a vehicle entering a vehicle parking bay may also include determining presence of a vehicle in each parking bay in dependence upon the magnetic field measurements in the table.
The method may also include storing an occupancy flag corresponding to a space for a vehicle in a second table in response to determining the presence of a vehicle in that space for a vehicle. The presence of a vehicle in each space for a vehicle may be determined in dependence upon the magnetic field measurements corresponding to the respective parking bay and also the magnetic field measurements in each adjacent space for a vehicle which is flagged as occupied in the second table.
According to a sixth aspect of the invention there is provided a non-transitory computer readable storage medium storing a computer program including instructions which, when executed by at least one processing unit, cause the processing unit to perform the method of authenticating a vehicle.
According to a seventh aspect of the invention there is provided a vehicle detection system including one or more detectors, each detector configured to measure magnetic fields in a volume of space at least partially overlapping a corresponding space for a vehicle. The system also includes one or more transceiver devices, each transceiver device corresponding to a detector and arranged proximate to the respective space for a vehicle, each transceiver device including a wireless personal area network module. The system also includes a processing unit configured to communicate with an external device and to determine presence of a vehicle in each space in dependence upon measurements from one or more detectors. The processing unit is also configured, in response to detecting a vehicle entering a space corresponding to a detector, to control one or more transceiver devices to broadcast a plurality of challenge messages according to a schedule. Each challenge message has a corresponding correct response. The processing unit is also configured to control the one or more transceiver devices to broadcast each challenge message before a subsequent challenge message is broadcast. The processing unit is also configured, for each challenge message broadcast, in response to one or more transceiver devices receiving at least one corresponding response message broadcast by a wireless personal area network transceiver in or on the vehicle, to increment a number of received responses and to check whether the response message corresponds to the correct response. The processing unit is also configured, in response to the number of received responses exceeds a first threshold and a fraction of correct responses exceeds a second threshold, to authenticate the vehicle by transmitting a message to the external device indicating that an authorised vehicle has entered the space.
The processing unit may also be configured, in dependence upon a number of received responses exceeds the first threshold and a fraction of incorrect responses exceeds a third threshold, to transmit a message to the external device indicating that an unauthorised vehicle has entered the space.
The processing unit may also be configured, in dependence upon a predetermined duration has elapsed and the number of received responses does not exceed the first threshold, to transmit a message to the external device indicating that an unauthorised vehicle has entered the space.
The processing unit may also be configured, in dependence upon the vehicle is authenticated, to control one or more transceiver devices to broadcast an identity request. The processing unit may also be configured, in response to one or more transceiver devices receiving a message including an identity of the vehicle, to transmit the identity of the vehicle to the external device.
Each transceiver device may also include at least one signalling device, each signalling device having two or more output states. The processing unit may also be configured, in response to authenticating the vehicle, to control the signalling device(s) of the transceiver device corresponding to the space to switch from a first output state to a second output state.



BRIEF DESCRIPTION OF THE DRAWINGS
Certain embodiments of the present invention will now be described, by way of example, with reference to the accompanying drawings in which:
FIG. 1 is a schematic plan view of a vehicle detection system used to monitor vehicle parking bays which includes a plurality of detectors and a processing unit;
FIG. 2 is a schematic perspective view of a detector shown in FIG. 1 ;
FIG. 3 is a schematic side view of a detector shown in FIG. 1 ;
FIG. 4 is a schematic block diagram of a detector shown in FIG. 1 ;
FIG. 5 is a schematic block diagram of a processing unit shown in FIG. 1 ;
FIG. 6 is a schematic view of a data structure storing magnetic field measurements;
FIG. 7 is a schematic view of the content of a data object shown in FIG. 6 ;
FIGS. 8A to 8C schematically illustrate updating a data structure shown in FIG. 6 ;
FIG. 9A illustrates an example of a data structure used in a method of detecting the presence of a vehicle in a volume of space;
FIG. 9B illustrates an example of a data structure used in a method of detecting the presence of a vehicle in a volume of space;
FIG. 10A schematically shows a plot of magnetic field with position when two coupled vehicle parking bays are unoccupied by vehicles;
FIG. 10B schematically shows a plot of magnetic field with position when one of two coupled vehicle parking bays is occupied by a vehicle;
FIG. 11 is a process flow diagram of a first method of detecting the presence of a vehicle in a volume of space;
FIG. 12 is a process flow diagram of a second method of detecting the presence of a vehicle in a volume of space;
FIG. 13 is a schematic plan view of a vehicle detection system used to monitor vehicle numbers and speeds on a road;
FIG. 14 is a schematic plan view of a vehicle detection system used to help control an automated traffic signal system as a junction;
FIG. 15 is a schematic plan view of a vehicle detection system used to monitor vehicle parking bays which includes a plurality of detectors, a plurality of transceiver devices and a processing unit;
FIG. 16 is a schematic block diagram of a transceiver device shown in FIG. 15 ;
FIG. 17 is a process flow diagram of a method of identifying a vehicle;
FIG. 18 in an exploded projection view of a first transceiver device;
FIG. 19A is a plan view of a first transceiver device;
FIG. 19B is a cross-sectional view along the line labelled A-A′ in FIG. 19A ;
FIG. 19C is a side view of the first transceiver device;
FIG. 20A is a plan view of a second transceiver device;
FIG. 20B illustrates a Bluetooth® chip incorporated into the second transceiver device;
FIG. 20C shows a cross-sectional view along the line labelled B-B′ in FIG. 20A ; and
FIG. 20D shows a cross-sectional view along the line labelled C-C′ in FIG. 20A .



DETAILED DESCRIPTION OF CERTAIN EMBODIMENTS
Referring to FIG. 1 , a vehicle detection system 1 is shown.
The vehicle detection system 1 is used for detecting the occupancy of bays P 1 , P 2 , . . . , P 16 in a vehicle parking environment. The vehicle detection system 1 comprises at least one detector 2 and a processing unit 3 . Each bay P 1 , P 2 , . . . , P 16 contains a corresponding detector 2 . The detector(s) 2 are connected to the processing unit 3 by a wired network in the form of one or more bus cables 4 . The detector(s) 2 connected to a single bus cable 4 constitute a branch 5 . Detectors 2 which are connected in a branch 5 are connected in a daisy chain configuration by the bus cable 4 . The bus cable 4 may comprise one or more data lines (not shown) and one or more power lines (not shown). The data line(s) and power line(s) are contained within cladding or armour (not shown). The detectors 2 may be connected to the data line(s) and power line(s) in a bus configuration. In this way, if a detector 2 becomes inoperable, other detectors 2 in the same branch will not be affected. The detectors 2 need not be equally spaced along the length of the bus cable 4 .
The detectors 2 are disposed underneath the corresponding bays P. For example, a detector 2 may be emplaced under a bay P, or a detector 2 may be placed in a slot or trench (not shown). A detector may be sealed in a pipe (not shown) which is emplaced in a slot or trench. A slot or trench containing detectors may be filled in or covered. The detectors 2 may be placed on the surface of the corresponding bays P and covered by a protective cover (not shown). There are sixteen bays P 1 , P 2 , . . . , P 16 for parking vehicles. The number of detectors 2 and bays P is not limited to sixteen, and the vehicle detection system 1 may be extended to include a general number, N, of detectors 2 , each contained by a respective bay P 1 , P 2 , . . . , P n , . . . , P N , wherein P n denotes the n th of N vehicle parking bays.
Each detector 2 is sensitive to a corresponding volume of space S, which is to say that each detector 2 measures magnetic field(s) which may be influenced by the presence of magnetic materials within the corresponding volume of space S. For example, the n th detector 2 is sensitive to a corresponding n th volume of space S n which is focussed on the respective n th bay P n . The n th detector 2 is sensitive to a corresponding n th volume of space S n which encompasses at least part of, for example all of, the volume above the respective n th bay P n .
Each detector 2 includes first and second magnetic field sensors 6 ( FIG. 3 ). The first and second magnetic field sensors are spaced apart by a distance L ( FIG. 3 ). Each of the magnetic field sensors 6 is configured to provide a time series of instantaneous measurements of magnetic field B (t), in which t is the time at which the magnetic field was measured. Each magnetic field sensor provides magnetic field measurements B x , B y , B z corresponding to three orthogonal components of the vector magnetic field B (t).
When a vehicle 7 does not occupy a bay P n , the instantaneous magnetic field B (t) measured by the corresponding detector 2 , which is sensitive to the respective volume of space S n , represents a baseline magnetic field B 0 arising from the Earths magnetic field and the ambient environment. The baseline magnetic field B 0 may be substantially invariant across a monitored environment, or the baseline magnetic field B 0 may vary with position, i.e. B 0 ( x ) in which x is a position vector, depending on the amount and distribution of magnetic material in the ambient environment. The baseline magnetic field B 0 may also vary in time.
When a vehicle 7 occupies a parking bay P n , elements of the vehicle 7 that have high relative magnetic permeability μ/μ 0 will produce localised distortions of the baseline magnetic field B 0 . As a result, the instantaneous magnetic field B (t) measured by the n th detector 2 , which is sensitive to the respective volume of space S n focussed on the parking bay P n , may deviate from the baseline B 0 values. Depending on the spatial extent of the magnetic field distortions produced by the vehicle 7 , the instantaneous magnetic fields B (t) measured by detectors 2 corresponding to nearby/adjacent bays P p , which are nearby/adjacent to the occupied bay P n , may also deviate from the baseline B 0 values.
The processing unit 3 receives the time series of instantaneous magnetic field B (t) measurements from the pair of magnetic field sensors 6 in each of the detectors 2 . The processing unit 3 stores the time series of instantaneous magnetic field measurements B (t) in a magnetic field history table 8 ( FIG. 5 ). The processing unit determines the presence or absence of a vehicle 7 in each bay P n in dependence upon the magnetic field measurements B (t) stored in the history table 8 . The presence or absence of a vehicle 7 in the n th bay P n is determined in dependence upon magnetic field measurements B (t) from the n th detector and also from nearby/adjacent detectors 2 .
The detectors 2 may additionally support wireless personal area network communications (also referred to herein as “WPAN”) protocols such as ZigBee®, Bluetooth® or Z-Wave®, and vehicles 7 may be fitted with WPAN transceivers (not shown). The processing unit 3 may include information to enable determination of whether a detected vehicle 7 includes a recognised WPAN transceiver.
The vehicle detection system 1 may detect the occupancy of parking bays P n in a vehicle parking environment which is an open air vehicle park or lot, a vehicle parking structure provided across at least one floor of a building and/or at least one basement level, or in an on-street vehicle parking environment. The vehicle detection system 1 may detect motor vehicles, such as, for example, motorcycles, automobiles, minibuses, buses, trucks, vans or lorries.
Referring to FIGS. 2 and 3 , the detector is generally elongate and each detector includes first and second units 9 , 10 . The first and second units 9 , to are spaced apart and connected by a link member 11 . The first magnetic field sensor 6 a is disposed in the first unit 9 and the second magnetic field sensor 6 b is disposed in the second unit 10 . The link member 11 is flexible. Alternatively, the link member 11 need not be flexible and may be rigid.
The first and second magnetic field sensors 6 a , 6 b are disposed within the first and second units 9 , to such that the first and second magnetic field sensors 6 a , 6 b are separated by a centre-to-centre distance L. The distance L may be at least 15 cm, at least 30 cm, at least 50 cm or at least 1 m. The distance L may be up to 3 m. Preferably, the distance is between about 30 cm and 40 cm, for example, about 35 cm. The first and second units 9 , to have a diameter d 1 of approximately 22 mm. The link member 11 has a diameter d 2 of approximately to mm. The bus cable 4 may have the same or similar diameter to the link member 11 . However, other dimensions d 1 , d 2 may be used for the first and second units 9 , 10 , the link member 11 and the bus cable 4 , depending on the dimensions of the components used to manufacture the detector 2 .
The detector(s) 2 are sealed to prevent the ingress of particles such as particles of dirt and/or dust. The detector(s) 2 are sealed so as to be waterproof. The detectors are encapsulated to the IP-68 certification standard, in which IP stands for Ingress Protection. The IP-68 standard refers to dust tight protection against particle ingress and waterproof to immersion beyond 1 m depth of water.
Referring also to FIG. 4 , the detector 2 includes the first and second magnetic field sensors 6 a , 6 b , a controller 12 and a network interface 13 connecting the detector to a bus 14 . The detector may optionally include a WPAN module 15 .
The second unit 10 is longer than the first unit 9 and the second unit 10 accommodates the controller 12 , network interface 13 and optionally the WPAN module 15 . However, the controller 12 , network interface 13 and optionally the WPAN module 15 may instead be provided in the first unit or distributed between the first and second units 9 , 10 .
The controller 12 receives a first set of instantaneous magnetic field measurements B a (t) having orthogonal components B x a (t), B y a (t), B z a (t) from the first magnetic field sensor 6 a . The controller receives a second set of instantaneous magnetic field measurements B b (t) having orthogonal components B x b (t), B y b (t), B z b (t) from the second magnetic field sensor 6 b . The first and second magnetic field sensors 6 a , 6 b are sensitive to respective first and second sub-volumes of space which intersect by a fraction which depends on the distance L. The volume of space S n to which the n th detector 2 is sensitive is a union of the first and second sub-volumes to which the respective first and second magnetic field sensors 6 a , 6 b of the n th detector are sensitive.
The controller 12 may additionally send and receive challenge/response data to the WPAN module 15 for exchanging information with a NFC transceiver on a vehicle 7 to identify the vehicle 7 . The device 2 communicates with the processing unit 3 via the network interface 13 and a bus 14 . The bus 14 is disposed within the bus cable 4 . The bus cable 4 and the interface with the detector 2 are encapsulated to the IP-68 certification standard.
Upon receiving a request from the processing unit 3 , the detector 2 sends the first and second sets of magnetic field measurements B a (t) and B b (t) to the processing unit 3 . The controller 12 is a microcontroller in the form of an ATMEL ATmega328P incorporating an Arduino® Bootloader microcontroller. However, the controller 12 may alternatively be provided by any other suitable microcontroller.
The magnetic field sensors 6 are anisotropic magnetoresistance sensors. Each magnetic field sensor 6 is a three-axis sensor providing instantaneous magnetic field measurements B (t) with three mutually orthogonal components B x (t), B y (t), B z (t). The first and second magnetic field sensors 6 a , 6 b take the form of Honeywell HMC5883L magnetic field 3D sensors, which have a resolution of 2×10−7 T and a magnetic field range of ±8×10−4 T. The Honeywell HMC5883L magnetic field 3D sensors can provide measurements at a maximum output rate of 160 Hz, using a 12-bit ADC.
The controller 12 may create and update a local cache (not shown) in local memory (not shown) to store a number of recent values of the six components of the magnetic field measurements B a (t), B b (t). The local cache may be updated at a frequency up to the maximum measurement frequency of the magnetic field sensors 6 , and may store recent values covering 128 sampling intervals. For example, with sampling every 0.3 milliseconds, a previous time interval of 38.4 milliseconds may be stored. The local cache may be updated at up to 160 Hz. The controller 12 may use the detector history table to filter out electromagnetic noise components having a noise frequency higher than 25 Hz and less than half of the maximum measurement frequency. The controller 12 may filter out electromagnetic noise components having frequencies between 25 and 80 Hz.
The WPAN module 15 includes an antenna capable of bi-directional communication with an WPAN transceiver (not shown) which may be attached to a vehicle 7 . The WPAN module 15 has a transmission power allowing communication with a transceiver up to 5 m away. The WPAN module is according to the Zigbee® standard.
When the processing unit 3 determines that a vehicle 7 has entered the bay P n corresponding to a detector 2 , the processing unit 3 may send challenge message data (not shown) for transmission by the WPAN module 15 . The controller 12 receives the challenge message data and directs the WPAN module 15 to transmit a challenge message (not shown). The detector 2 waits to receive a response message (not shown), and any response message data (not shown) received by the WPAN module 15 is relayed to the processing unit 3 . The communication protocol employed by the WPAN module 15 may be encrypted using conventional or application specific algorithms. The WPAN module 15 may communicate with WPAN transceivers which are external tokens employing Zigbee®, smartphones using Zigbee®, smartphones having Zigbee®-WiFi dongles or other similar devices.
The network interface 13 connects the controller 12 to the bus 14 . The network interface 13 need not be a separate module to the controller 12 , and may alternatively be integrated into the controller 12 . The network interface 13 may take the form of a Maxim MAX3440EESA+ Transmitter/Receiver RS-485 for interfacing with the RS-484 bus 14 . The Maxim MAX3440EESA+ Transmitter/Receiver RS-485 has a single supply voltage in the range 4.75 to 5.25 V and a data rate of 0.25 Mbits/s. The bus 14 is an RS-485 bus with a four-wire cable including two wires used for supplying power and two wires for data signalling.
One or more detectors 2 are arranged in a daisy chain configuration along bus cable 4 and connected in a bus configuration via the bus 14 to form a branch 5 . The detectors are connected in parallel to the two data wires and the two power wires. The branch 5 , including the one or more detectors 2 and the interconnecting bus cables 4 , is encapsulated to help protect the interconnections from ingress of particles or fluids.
Bus cable 4 and detectors 2 have connectors (not shown) which may be reversibly coupled to form a connection encapsulated to the IP-68 certification standard. In this way, the overall branch 5 is encapsulated to the same IP-68 certification standard as the devices. The branch 5 may be inserted into a pipe (not shown) which is subsequently sealed. Inserting the branch 5 into a pipe may help to protect the branch 5 , especially when the branch 5 is installed in a trench without back-filling the trench. For example, a pipe may prevent or reduce movement of a branch 5 due to, for example, flooding or subsidence.
Although the detectors include first and second sensors 6 a , 6 b which measure respective mutually orthogonal components of magnetic field B x (t), B y (t), B z (t), the measurement axes of the first and second magnetic field sensors 6 a , 6 b need not be aligned with each other and, in general, will not be aligned. The measurement axes of any detector 2 need not be aligned with those of any other detectors 2 in the same branch 5 , or in any other branches 5 .
The bus 14 need not be an RS-485 bus and other types of bus may be used instead. Alternatively, separate cables may provide power and Ethernet, Firewire or similar cables may be used for data transmission. Where separate power cables are used, the network interface 13 may connect the detectors 2 to the processing device using wireless communications such as, for example IEEE 802.11 or similar.
Up to fifty detectors 2 may be provided in a single branch 5 , with a bus cable 4 length of up to 800 metres. However, larger numbers of detectors may be supported by a single branch 5 when data cables with a higher bandwidth are used. Total length of a bus cable 4 may be extended by using, for example, larger voltages, lower impedance cables or repeaters (not shown) spaced along the length of the bus cable 4 .
Referring to FIG. 5 , the processing unit 3 includes a processor 16 , memory 17 , an external network interface 18 , one or more bus network interfaces 19 and a power supply 20 .
The external network interface 18 connects the processing unit 3 to local networks or the internet, using wired or wireless connections. A server (not shown) may execute a client (not shown) which communicates with one or more processing units 3 via a local network or the internet. In this way several processing units 3 , each corresponding to a different section or area of a parking environment, may be monitored or controlled by an operative. In this way geographically separated or distributed parking environments may be monitored by an operative. Additionally or alternatively, users or prospective users of a parking environment may execute a client, for example on a mobile device, which may inform them of the availability of vehicle parking bays.
Each bus network interface 19 communicates with a bus 14 which connects to a branch 5 including one or more detectors 2 . The processor 16 sends requests addressed to specific detectors 2 belonging to the branch 5 , and receives the requested magnetic field measurements B a (t), B b (t) via the bus network interface 19 . The bus network interface 19 may include a conversion stage between the RS-485 bus on the branch 5 side and a different type of bus on the processor 16 side, such as, for example, USB. The bus network interface may take the form of a Maxim MAX3440EESA+ Transmitter/Receiver RS-485.
The processing unit 3 includes M bus network interfaces 19 (where M is a positive non-zero integer) communicating with a corresponding number M of branches 5 . The processing unit 3 supports up to sixteen branches such that M≦16. However, the processing unit 3 may support a larger or smaller number of branches, depending on the capabilities of the processor 16 and memory 17 used.
The power supply 20 receives power from an external power source 21 and supplies it to the processing unit 3 . The power supply 20 powers the processor 16 , the external network interface 18 and the bus network interfaces 19 . The bus network interfaces 19 supply power to the individual detectors 2 via the buses 14 . The power source 21 may be a mains electrical supply, or it may be a local source such as, for example, a photovoltaic panel (not shown) or similar energy harvesting device.
The processing unit 3 may optionally include energy storage 22 , for example, a secondary battery such as a storage or rechargeable battery. The energy storage 22 may provide backup power to the power supply 20 when the power source 21 is mains or grid electricity. Alternatively, when the power source 21 is an energy harvesting device, the power supply 20 may store energy in the energy storage 22 when the power source 21 is providing surplus power, for example during the day, so that the energy storage 22 may power the processing unit 3 when the power source does not provide power, for example at night.
The memory 17 stores the history table 8 , application software 23 for the processing unit 3 , a baseline table 24 , an occupancy table 25 , a dependency table 26 , and a correction table 27 . When the detectors 2 include a WPAN module 15 , the memory 17 additionally includes a transceiver identity table 28 . Any additional variables, arrays, data structures, data pages or data objects described herein in relation to methods of detecting vehicles 7 may, unless otherwise described, be assumed to be created by the processor 16 and stored in or retrieved from the memory 17 as required.
The transceiver identity table 28 includes a list of challenge/response message data items (not shown). When the processing unit 3 determines that a vehicle 7 has entered a bay P n containing a detector 2 , the processing unit 3 sends challenge message data (not shown) to the detector 2 for transmission by the WPAN module 15 . The detector 2 waits to receive a response message (not shown), and any response message data (not shown) received by the WPAN module 15 is sent to the processing unit 3 . If no response message data is received within a configurable elapsed time period, or if the response message data does not match a correct response stored in the transceiver identity table 28 , then an alarm flag (not shown) is set on the processing unit 3 . The alarm flag may be communicated or reported to an operative monitoring the vehicle detection system 1 via the external network interface 18 and a local network or the internet.
The challenge/response message data items stored in the transceiver identity table 28 can be user specific and may be created, delivered and/or monitored by an operative via the external network interface 18 and local networks or the internet. For example, the challenge message data may simply be a request for an identity string (not shown) and specific identity strings may be assigned to specific vehicles 7 operated by authorised users. In such a case, the transceiver identity table may store a list of the identity strings which are authorised for each bay P n , or for a group of bays P j , . . . , P k , in which j, k are integers less than the total number N of bays P n /detectors 2 . For example, the vehicles 7 registered to employees of company A may be authorised to park in any bay from P j , . . . , P k , whereas other vehicles 7 may not be authorised. A set of allowed bays P n for a specific identity string need not be consecutively numbered.
Referring also to FIGS. 6 and 7 , the history table 8 stores a time history for the six magnetic field measurements B x a (t), B y a (t), B z a (t), B x b (t), B y b (t), B z b (t) corresponding to the pair of magnetic fields sensors 6 a , 6 b for each individual detector 2 . The processing unit 3 receives magnetic field measurements B a (t), B b (t) from all M branches 5 independently and simultaneously, and updates the history table 8 with the most recently received values. Each branch 5 reports magnetic field measurements B a (t), B b (t) at a maximum rate of 75 Hz/N m , in which N m is the total number of detectors 2 inside the m th branch 5 out of M branches 5 in total. The overall update rate of the history table 8 is determined according to the largest number of detectors N m in a single branch 5 . It is preferable for the numbers of detectors 2 connected in each branch 5 to be balanced in order to maximise the refresh rate for a given total number N of detectors 2 .
The history table 8 includes a data structure 29 also referred to herein as recent(τ,n,i), in which r is an integer index which runs 1≦τ≦T such that τ=T corresponds to the most recent measurements B a (t), B b (t) and τ=1 corresponds to the furthest point in the past for which measurements B a (t), B b (t) are stored, in which n is an integer index which runs 1≦n≦N such that n refers to the n th detector 2 out of N detectors in total, in which i is an integer index such that i=1 corresponds to B x a (t), i=2 corresponds to B y a (t), i=3 corresponds to B z a (t) and 4≦i≦6 corresponds to B x b (t), B y b (t) and B z b (t) respectively. In this case, T=128 and one detector 2 corresponds to each of N vehicle bays P n . Alternatively, the history table 8 may store magnetic field measurements B a (t), B b (t) corresponding to more or fewer than T=128 different time intervals, limited only by the available capacity of the memory 17 .
The data structure 29 recent(τ,n,i) includes a number, N, of page objects 30 . Each page object 30 stores the magnetic field measurements B a (t), B b (t) received from the n th detector 2 , for all time intervals and all field components, i.e. for 1≦τ≦T and 1≦i≦6. Each page object 30 includes a number, T, of data objects 31 . Each data object 31 stores the magnetic field measurements B a (t), B b (t) received from the n th detector 2 at the (T−τ) th time interval before the most recent measurement. Each data object 31 includes the six individual measured magnetic field values B x a (t), B y a (t), B z a (t), B x b (t), B y b (t), B z b (t) measured by the n th detector 2 at the (T−τ) th time interval.
For example, if measurements B a (t), B b (t) are reported at intervals of 6 t and the most recent measurements occurred at a time t=t T , then recent(τ,n,i) stores the values corresponding to t=t T −δt.(T−τ). The sequential reporting of magnetic field measurements B a (t), B b (t) from the detectors 2 within the m th branch 5 means that the same r value may correspond to slightly different (shifted) actual measurement times in each of the respective data objects 31 for the m th branch.
Referring also to FIGS. 8A to 8C , the procedure for updating a page object 30 of the data structure 29 recent(τ,n,i) to include a data object 31 storing the most recently received measured magnetic field values B a (t), B b (t) is illustrated. Until the updating procedure is completed, reference will be made to the r index values of data objects 31 prior to the updating procedure.
Referring to FIG. 8A , the processing unit 3 receives magnetic field measurements B a (t), B b (t) from each of N detectors 2 . The processing unit 3 creates a new data object 31 , recent(T+1,n,i) for the n th detector 2 , to store most recently measured magnetic field values B x a (t), B y a (t), B x a (t), B x b (t), B y b (t), B z b (t) measured by the n th detector 2 . The processing unit 3 increments each existing data object 31 stored on the page object 30 corresponding to the n th detector to a lower value of the index τ. For example, the processing unit 3 overwrites the oldest stored data object 31 , recent(1,n,i) with the second oldest data object 31 , recent(2,n,i), i.e. the τ=2 entry is shifted to the τ=1 position. The processing unit 3 also overwrites recent(2,n,i) with recent(3,n,i), the processing unit overwrites recent(3,n,i) with recent(4,n, i), and continues in this way until recent(T−1,n,i) has been overwritten by recent(T,n,i).
Referring to FIG. 8B , the page object 30 corresponding to the n th detector has the data object recent(2,n,i) corresponding to the index τ=1, the data object recent(3,n,i) corresponding to τ=2 and so on until the indexed τ=T and τ=T−1 both correspond to the same data object recent(T,n,i). The data object 31 storing the most recent magnetic field measurements B a (t), B b (t) from the n th detector 2 , recent(T+1,n,i), is written to the τ=T index position.
Referring to FIG. 8C , the page object 30 corresponding to the n th detector has been updated so that each data object 31 has been incremented by an amount corresponding to one time interval between obtaining magnetic field measurements B a (t), B b (t). Each time the processing unit 3 receives new magnetic field measurements B a (t), B b (t), the page object 30 corresponding to each of the N detectors 2 is updated in this way, such that the history table 8 always holds the most recent magnetic field measurements B a (t), B b (t) and T−1 prior measurements for each detector 2 .
Referring again to FIG. 5 , the baseline table 24 stores a data structure referred to herein as ref(n,i), which is an array with N rows and 6 columns, in which n and i are integer indexes defined in the same way as for the data structure recent(τ,n,i). The n th row of ref(n,i) stores the six individual measured magnetic field components B x a0 , B y a0 , B z a0 , B x b0 , B y b0 , B z b0 of the baseline field B 0 measured by the n th detector 2 when there are no vehicles 7 present in the n th bay P n or in any other bay P p which is at least partially encompassed by the volume of space S n to which the n th detector 2 is sensitive. For example, when no vehicles 7 are present in the n th bay P n or any other bays P p which are neighbouring/adjacent or otherwise coupled to the n th bay P n , as described hereinafter in relation to the dependency table 26 . The six individual measured magnetic field components B x a0 , B y a0 , B z a0 , B x b0 , B y b0 , B z b0 of the baseline field B 0 measured by the n th detector 2 and stored in the array ref(n,i) will generally be different for each sensor 6 of each detector 2 , even when the baseline field B 0 itself is substantially invariant with position. This is because the detection axes of the sensors 6 and detectors 2 need not be aligned with one another and, in general, are not aligned with one another.
The values stored by the baseline table 24 in the data structure ref(n,i) may be measured at times when the parking environment is empty, for example, when the parking environment is closed. The values stored by the baseline table 24 may be periodically updated during use of the vehicle detection system 1 , for example, by measuring magnetic field values from the n th detector 2 when there are no vehicles 7 present in the n th bay P n or in any other bay P p which is neighbouring/adjacent or otherwise coupled to the n th bay P n , as described hereinafter in relation to the dependency table 26 .
The occupancy table 25 stores a data structure referred to herein as occupy(n), which is an array with N entries, in which n is an integer index defined in the same way as for the data structure recent(τ,n,i). The n th entry of occupy stores the occupancy state of the n th bay P n such that occupy(n)=0 if the n th bay P n is unoccupied and occupy(n)=1 if the n th bay P n is occupied.
The dependency table 26 stores a data structure referred to herein as depend(n,p), which is an array with N rows and N columns, in which n is an integer index defined in the same way as for the data structure recent(τ,n,i) and in which p is an integer index which runs 1≦p≦N, such that n and p refer to the n th and p th detectors 2 and bays P n , P, respectively. If the n th bay P n is sufficiently close to the p th bay P p that when a vehicle is parked in the n th bay P n the magnetic fields in the p th bay P p will measurably deviate from the baseline field B 0 , and vice versa, then depend(n,p)=1 and also depend(p,n)=1. The entries depend(n,n)=0 along the array diagonal. When the n th and p th bays P n , P, interact in this way, they may also be referred to herein as “coupled”. Another way to describe such coupling is that if the n th bay P n is sufficiently close to a p th bay P p that the volume of space S n to which the n th detector is sensitive at least partially encompasses the p th bay P p , then the n th and p th bays P n , P p are coupled.
Referring to FIG. 9A , the data structure depend(n,p) for the bays P 1 , P 2 , . . . , P 16 is illustrated for the example shown in FIG. 1 in a case where only bays P n which are directly adjacent (nearest neighbours) are coupled and those which are diagonally adjacent (next nearest neighbours) are not coupled.
Referring to FIG. 9B , the data structure depend(n,p) for the bays P 1 , P 2 , . . . , P 16 is illustrated for the example shown in FIG. 1 in a case where diagonally adjacent bays P n (next nearest neighbours) are coupled in addition to directly adjacent bays P n (nearest neighbours).
Coupling between n th and p th bays P n , P p need not be limited to nearest neighbour or next nearest neighbour bays P p . For example, any pair of bays P n , P p may be coupled to one another in the event that a vehicle parked in one would produce a measurable deviation of the measured magnetic field values B a (t), B b (t) away from B 0 at the detector 2 corresponding to the other. In other words, if the volume of space S n to which the n th detector 2 is sensitive at least partially encompasses the p th bay P p , then the n th and p th bays P n , P p may be coupled.
Referring to FIG. 5 , the correction table 27 stores a data structure referred to herein as correct(p,n,i), in which n and p are both integer indexes defined in the same way as for the data structure depend(n,p) and i is an integer index defined in the same way as for the data structure recent(τ,n,i). The entry of the data structure correct (p,n,i) corresponding to the n th and p th bays P n , P p only holds non-zero values if the corresponding entry of the data structure depend(n,p) is equal to one. The entry of the data structure correct(p,n,i) stores six individual magnetic field correction values δB x a , δB y a , δB z a , δB x b , δB y b , δB z b for adjusting the baseline values B x a0 , B y a0 , B z a0 , B x b0 , B y b0 , B z b0 stored by ref(n,i) for the n th detector 2 in the case that the p th bay P p is occupied by a vehicle 7 .
Referring also to FIG. 10A , a bay P n and an adjacent bay P p are both unoccupied and are delimited by parking bay markings 32 . An n th detector 2 is disposed under the n th bay and is sensitive to an n th volume of space S n . A p th detector 2 is disposed under the p th bay and is sensitive to a p th volume of space S p . The n th and p th detectors 2 are consecutively connected as part of the same branch 5 . However, a pair of coupled bays P n , P p need not correspond to consecutive detectors 2 on the same branch 5 . It should be understood that the p th bay P p may be any bay P p which is coupled to the n th bay P n such that depend(p,n)=1 and depend(n,p)=1. The detectors 2 are disposed beneath the respective bays P n , P p by emplacing the detectors 2 underneath the floor of the parking bays P n , P p .
In the absence of any nearby vehicles 7 , the first and second magnetic field sensors 6 a , 6 b of the n th and p th detectors 2 measure the baseline magnetic field B 0 . The baseline magnetic field B 0 arises from the Earth's magnetic field, potentially modified by any magnetic materials present in the ambient environment. For example, steel girders in a building structure. The baseline magnetic field B 0 is shown as being substantially invariant with position. However, in some parking environments the baseline magnetic field B 0 in the absence of any vehicles may vary between or even within adjacent bays P n , P p due to ambient magnetic materials, i.e. the baseline magnetic field may be position dependent B 0 ( x ).
Referring also to FIG. 10B , when a vehicle 7 occupies the n th bay P n , elements of the vehicle 7 having high relative magnetic permeability μ/μ 0 will distort the baseline magnetic field B 0 , such that the instantaneous magnetic fields B a (t), B b (t) measured by the n th detector 2 , and by coupled detectors 2 such as the p th detector 2 , will deviate from the baseline field B 0 . In particular, steel components of the vehicle 7 may act as magnetic conductors, causing a focussing of the Earth's magnetic field lines below the steel components 7 and a relative weakening to either side. The precise pattern of distortions to the magnetic field observed will depend on the quantity and distribution of magnetic materials in a vehicle 7 .
As a result of the vehicle 7 distorting the magnetic field B , the first and second magnetic field sensors 6 a , 6 b of the n th detector 2 measure first and second magnetic field differences 35 , 36 respectively. The first magnetic field sensor 6 a of the p th detector 2 also measures a third magnetic field difference 37 , and the second magnetic field sensor 6 b of the p th detector 2 does not measure a substantial deviation from the baseline magnetic field B 0 . However, depending on the quantity and distribution of magnetic materials in the vehicle 7 , the second magnetic field sensor 6 b of the p th detector 2 may measure a fourth deviation (not shown).
The first and second magnetic field differences 35 , 36 may be used by the processing unit 3 to determine that the n th bay P n is occupied. The third 37 and fourth magnetic field differences are stored to the correction table 27 in the data object correct(n,p,i) as δB x a and δB x b (as entries correct(n,p,1) and correct(n,p,4)) respectively. Similar magnetic field differences are stored to the data object correct(n,p,i) as δB y a , δB z a , δB y b , δB z b . If a further vehicle 7 subsequently enters the p th bay P p , the presence of that further vehicle 7 may be detected in dependence upon the deviation of the measured magnetic field values B x a (t), B y a (t), B z a (t), B x b (t), B y b (t), B z b (t) from a corrected baseline provided by summing the magnetic field difference values δB x a , δB y a , δB z a , δB x b , δB y b , δB z b stored in correct(n,p,i) and the baseline values B x a0 , B y a0 , B z a0 , B x b0 , B y b0 , B z b0 stored in ref(n+1,i).
Referring to FIGS. 5 and 11 , a first method of detecting vehicles 7 using a vehicle detection system 1 will now be described.
At start up, the processing unit 3 receives magnetic field measurements B a (t), B b (t) from all N detectors 2 until the history table 8 is fully populated with the T most recently received values (step S 1 ). The processing unit 3 has a defined time period stored by a variable referred to herein as stdSize. In this case, stdSize=1 s. However, other time periods may be set, such as, for example, 2 s or up to 10 s. The number T of previous time periods stored should be large enough to span at least 10×stdSize seconds.
The processing unit 3 initialises an array referred to herein as stdCalm(n), which is an array with N entries, in which n is an integer index defined in the same way as for the data structure recent(τ,n,i). The n th entry of stdCalm(n) stores the standard deviation of the measured magnetic field values B x a (t), B y a (t), B z a (t), B x b (t), B y b (t), B z b (t) from the n th detector across the 10×stdSize seconds after start up. The n th entry of stdCalm(n) is calculated across all six individual measured magnetic field values B x a (t), B y a (t), B z a (t), B x b (t), B y b (t), B z b (t), such that stdCalm(n) is not specific to any axis (1≦i≦6) or the first or second magnetic field sensors 6 a , 6 b of the n th detector 2 .
After start up, the processing unit 3 begins monitoring for vehicles 7 . The processing unit 3 addresses each detector 2 in turn to request measurements, and each detector 2 sends magnetic field measurements B x a (t), B y a (t), B z a (t), B x b (t), B y b (t), B z b (t) to the processing unit 3 (step S 2 ). The processing unit 3 receives magnetic field measurements B x a (t), B y a (t), B z a (t), B x b (t), B y b (t), B z b (t) from M branches 5 independently and simultaneously. Each branch 5 reports magnetic field measurements B x a (t), B y a (t), B z a (t), B x b (t), B y b (t), B z b (t) at a maximum rate of 75 Hz/N m , in which N m is the number of detectors 2 included in the m th branch 5 out of M branches 5 in total. The overall update rate of the history table is determined according to the largest number of detectors N m in a single branch.
The processing unit 3 stores the most recently received magnetic field measurements B x a (t), B y a (t), B z a (t), B x b (t), B y b (t), B z b (t) to the history table 8 . For each page 30 , corresponding to each of N detectors 2 , the processing unit 3 increments the previously stored measurements in the data structure 29 recent(τ,n,i) by one, overwriting the τ=1 data object 31 corresponding to the oldest measurements with the τ=2 data object 31 , and writing a data object 31 storing the most recent magnetic field measurements B x a (t), B y a (t), B z a (t), B x b (t), B y b (t), B z b (t) to the τ=T position (step S 3 ).
The processing unit 3 determines the presence or absence of a vehicle 7 in the volume of space S n to which each detector 2 is sensitive in dependence upon the magnetic field measurements in the history table 8 (step S 4 ). The processing unit 3 may be configured to determine the presence of a vehicle in each volume in dependence upon the magnetic field measurements B x a (t), B y a (t), B z a (t), B x b (t), B y b (t), B z b (t) from detectors 2 corresponding to two or more bays P n .
The processing unit 3 determines the presence of absence of a vehicle 7 in the volume of space S n to which each detector 2 is sensitive in dependence upon an N by 6 array referred to herein as signalExt(n,i), which is defined as:








signalExt


(

n
,
i

)


=

abs


(


recent


(

T
,
n
,
i

)


-

ref


(

n
,
i

)


-


∑

p
=
1

N







(


depend


(

n
,
p

)


×

correct


(

p
,
n
,
i

)



)



)






(
1
)







in which n and i are integer indexes defined in the same way as for the data structure recent(τ,n,i) and in which T refers to τ=T such that signal(n,i) is based upon the most recently received magnetic field measurements B x a (t), B y a (t), B z a (t), B x b (t), B y b (t), B z b (t), in which p is an integer index which runs 1≦p≦N such that p refers to the p th detector 2 out of N detectors in total.
Based upon signalExt(n,i), an array with N entries referred to herein as signal(n) is defined according to:








signal


(
n
)


=


∑

i
=
1

6







{




signalExt


(

n
,
i

)






if






signalExt


(

n
,
i

)



>

stdCalm


(
n
)







0




if






signalExt


(

n
,
i

)



≤

stdCalm


(
n
)












(
2
)







in which n and i are integer indexes defined in the same way as for the data structure recent(T,n,i). During an iteration, if the value of signal(n) exceeds a threshold stored by a variable referred to herein as thresh, the corresponding volume of space S n is flagged as occupied, i.e. occupy(n)=1. The same threshold thresh may be applied to each of N detectors 2 . Alternatively, thresh may be an array with N entries such that thresh(n) may be applied to the n th of N detectors 2 . In such a case, the threshold thresh(n) is specific to each detector, and may be defined as:
thresh( n )=stdcalm( n ) (3)

in which n is an integer index defined in the same way as for the data structure recent(T,n,i).
By using the data structures depend(n,p) and correct(p,n,i) to include the effects of coupling and distortions of magnetic field resulting from vehicles 7 occupying adjacent bays P n , the vehicle detection system 1 can perform more reliably when magnetic field sensors 6 are used to monitor parking bays P n which are densely packed such that the corresponding volumes of space S n to which a detector 2 is sensitive at least partially encompass one or more adjacent bays P p .
The occupancy of a given bay P p is not altered if a neighbouring/adjacent bay P n coupled to that bay P p by the data structure depend(n,p) has a larger value of signal(n) than the given bay P p during the current monitoring iteration, i.e. if signal(n)>signal(p). Instead, only the neighbouring/adjacent bay P n having the larger value of signal(n) will be flagged as occupied. In this way, other bays P p which are coupled to an occupied bay P n by the data structure depend(n,p) are prevented from being erroneously flagged as occupied during a monitoring iteration in which the bay P n become occupied.
When the occupancy of a bay P n is altered, the correction values stored in the data structure correct(p,n,i) are updated for every other bay P p which is flagged as coupled to the bay P n in the data structure depend(n,p). When a bay becomes occupied, the correction values correct(p,n,i) are updated for each other bay P p which is flagged as coupled to the bay P n in the data structure depend(n,p), using the field differences measured at that coupled bay P p , i.e. the signalExt(p,i) values. When a bay P n becomes unoccupied, the corresponding correction values correct(p,n,i) are set to zero for every other bay P p which is flagged as coupled to the bay P n in the data structure depend(n,p).
As a result, when a vehicle 7 enters a bay P n , the updating of the correction values correct(p,n,i) prevents any coupled bays P p from being incorrectly flagged as occupied during subsequent monitoring iterations. In this way, the precise magnitudes of the corrections applied to the other bays P p which are coupled to an occupied bay P n are updated to reflect the magnitude of the distortion caused by the vehicle 7 occupying the bay P n .
In this way, the vehicle detection system 1 can accurately correct for the effects on a coupled bay P p when a bay P n is occupied by a large vehicle such as, for example, a coach, bus, mini-bus, sports utility vehicle (SUV), pickup truck, truck, lorry or van, or when a bay P n is occupied by a small vehicle such as, for example, a small hatchback, ultra-compact car, city car, super-mini car or a micro car.
Equations 1 and 2 above define the arrays signalExt(n,i) and signal(n) on the basis of the most recently received magnetic field values. However, the processing unit 3 may alternatively determine the presence or absence of a vehicle 7 in the bay P n corresponding to the n th detector 2 in dependence upon the mean of the array signal(n) taken across two or more consecutive points in the time series stored in the history table. For example, the processing unit 3 may alternatively determine the presence or absence of a vehicle in the bay P n corresponding to each detector 2 in dependence upon a data structure referred to herein as avg(T 0 ,n,i) and defined as:








avg


(


T
0

,
n
,
i

)


=



∑

τ
=

T
-

T
0



T







abs


(





recent


(

τ
,
n
,
i

)


-

ref


(

n
,
i

)


-







∑

p
=
1

N







(




depend


(

n
,
p

)

×






correct


(

p
,
n
,
i

)





)





)




T
-

T
0







(
4
)







in which r, n and i are integer indexes defined in the same way as for the data structure recent(τ,n,i), T 0 is an integer such that 1≦T 0 ≦T−1. The data structure avg(T 0 ,n,i) is an N by 6 array parameterised by T 0 . In such a case, the array signal(n) may be alternatively defined by substituting the data structure avg(T 0 ,n,i) for the array signalExt(n,i).
Determining the presence or absence of a vehicle 7 based upon the data structure avg(T 0 ,n,i) defined with respect to two or more consecutive time points may help to reduce the incidence of false detections caused by electromagnetic noise.
Optionally, when the processing unit 3 determines that a vehicle has entered a bay P n , the processing unit 3 may send challenge message data (not shown) to the corresponding detector 2 for transmission by an WPAN module 15 (step S 5 ). The detector 2 waits to receive a response message (not shown), and any response message data (not shown) captured by the NFC module 15 is relayed to the processing unit 3 . If no response message data is received within a configurable elapsed time period, or if the response data does not match the correct response stored in the transceiver identity table 28 , then an alarm flag (not shown) is set on the processing unit 3 .
The processing unit 3 checks whether to continue monitoring for vehicles (step S 6 ). By default, the processing unit 3 continues monitoring by addressing each detector 2 in turn to request measurements, and each detector 2 sends the first and second sets of magnetic field measurements B x a (t), B y a (t), B z a (t), B x b (t), B y b (t), B z b (t) to the processing unit 3 (Step S 1 ). However, the processing unit 3 may stop monitoring in response to, for example, a command from a monitoring operative, or in order to save power during hours when a parking environment is closed or is not controlled.
Referring to FIGS. 5, 11 and 12 , a second method of detecting vehicles 7 using a vehicle detection system 1 will now be described.
The second method is the same as the first method, except that the determining the presence of a vehicle 7 in a volume of space is additionally based on the detection of events which correspond to occurrences such as vehicles arriving into or departing from a volume of space.
The processing unit 3 populates the history table (step S 1 ), receives (step S 2 ) and stores (step S 3 ) magnetic field measurements B x a (t), B y a (t), B z a (t), B x b (t), B y b (t), B z b (t) in substantially the same way as the first method.
The processing unit 3 determines whether a local event has commenced at each of the N detectors (step S 4 - 1 ). The processing unit calculates an array referred to herein as signalStd(n), which is an array with N entries, in which n is an integer index defined in the same way as for the data structure recent(τ,n,i). The n th entry of signalStd(n) stores the standard deviation of the measured magnetic field values B x a (t), B y a (t), B z a (t), B x b (t), B y b (t), B z b (t) from the n th detector across the previous 10×stdSize seconds. In the same way as stdCalm(n), signalStd(n) is not specific to any axis (1≦i≦6) or the first or second magnetic field sensors 6 a , 6 b of the n th detector 2 , that is to say that the signalStd(n) values are averaged across all six axes of the pair of sensors 6 a , 6 b.
The values of signalStd(n), which are calculated using values spanning 10×stdSize seconds, may be significantly larger than the difference in signal before/after the vehicle 7 has finished parking in a bay P n . This is because signalStd(n) values incorporate the effect of measured magnetic field values collected when the vehicle 7 is moving on top of the sensor, which typically includes the maximum deviations. Using values of signalStd(n) to determine local events reduces the possibility failing to detect a vehicle 7 .
When a local event is initialised, a flag is set in an array referred to herein as inEvent(n), which stores inEvent(n)=0 when the n th detector is not in a local event and inevent(n)=1 when the n th detector is in a local event. A local event is not initialised for the n th detector during an iteration if it is already in a local event, i.e. when inevent(n)=1.
A local event is initialised for the n th detector 2 if signalStd(n) exceeds 3×stdcalm. If the n th detector 2 is coupled to the p th detector 2 according to the data structure depend(n,p), then initialising a local event for the n th detector is cancelled if signalStd(p) exceeds signalStd(n). When a local event is initialised, the processing unit 3 stores the value of signalExt(n,i) calculated according to Equation 1 to an array referred to herein as eventStart(n,i), in which n and i are integer indexes defined in the same way as for the data structure recent(τ,n,i). The values of signalExt(p,i) are also calculated and stored to eventStart(p,i) for every p th detector which is coupled to the n th detector according to the data structure depend(n,p). A separate instance of eventStart(n,i) is created for each local event. Alternatively, eventStart(n,i) may store values of avg(T 0 ,n,i) calculated according to Equation 4 above.
The processing unit 3 determines whether a local event has terminated at each of the N detectors, based upon the value of signalStd(n) (step S 4 - 2 ). If inEvent(n)=1 and signalStd(n) is less than 3×stdcalm then a local event is terminated. When a local event is terminated, the processing unit 3 stores the value of signalExt(n,i) calculated according to Equation 1 to an array referred to herein as eventEnd(n,i), in which n and i are integer indexes defined in the same way as for the data structure recent(τ,n,i). The values of signalExt(p,i) are also calculated and stored to eventEnd(p,i) for every p th detector which is coupled to the n th detector according to the data structure depend(n,p). Alternatively, eventEnd(n,i) may store values of avg(T 0 ,n,i) calculated according to Equation 4 above.
When a local event corresponding to the n th detector is closed, the processing unit determines whether the occupancy of the respective bay P n has changed based on the values stored in an array referred to herein as eventDiff(n), defined as:








eventDiff


(
n
)


=



∑

i
=
1

6







abs


(

eventEnd


(

n
,
i

)


)



-


∑

i
=
1

6







abs


(

eventStart


(

n
,
i

)


)








(
5
)







in which n and i are integer indexes defined in the same way as for the data structure recent(τ,n,i). The array eventDiff(n) has N entries. The n th entry of eventDiff(n) stores a measure of the change in measured magnetic field values B x a (t), B y a (t), B z a (t), B x b (t), B y b (t), B z b (t) between the start and end of an event. A second array with N entries, referred to herein as eventCompare(n) is defined as:








eventCompare


(
n
)


=


∑

i
=
1

6







abs


(


recent


(


T
end

,
n
,
i

)


-

recent


(


T
start

,
n
,
i

)



)







(
6
)







in which n and i are integer indexes defined in the same way as for the data structure recent(τ,n,i), T end is the value of the index τ corresponding to the iteration in which the event was closed, which will by T end =T and T start is the value of the index τ corresponding to the iteration in which the event was opened, which will be T start <T. The n th entry of eventDiff(n) stores the overall difference in the measured magnetic field values B x a (t), B y a (t), B z a (t), B x b (t), B y b (t), B z b (t) between the start and end of an event. Alternatively, where eventStart(n,i) and eventEnd(n,i) are based upon avg(T 0 ,n,i), eventCompare(n) may be calculated based on averaging the values recent(T end ,n,i) to recent(T end −T 0 ,n,i) and averaging the values recent(T start ,n,i) to recent(T start −T 0 ,n,i).
Whether there is a change of occupancy for the n th bay P n is determined based upon the ratio of eventDiff(n)/eventCompare(n). The ratio of eventDiff(n)/eventCompare(n) is a number between −1 and 1. If the n th bay P n is flagged as occupied, occupy(n)=1, and the ratio of eventDiff(n)/eventCompare(n) is less than −0.8, then the occupancy for the n th bay P n is changed to unoccupied, occupy(n)=0. If the n th bay P n is flagged as unoccupied, occupy(n)=10, and the ratio of eventDiff(n)/eventCompare(n) is greater than 0.8, then the occupancy for the n th bay P n is changed to occupied, occupy(n)=1.
When the occupancy of the n th bay P n is changed to occupied, the corresponding correction values stored in the data structure correct(p,n,i) for each p th detector which is coupled to the n th detector by the data structure depend(n,p) are updated based on the differences event.end(p,i)−event.start(p,i). When the occupancy of the n th bay P n is changed to unoccupied, the corresponding correction values stored in the data structure correct(p,n,i) for each p th detector which is coupled to the n th detector by the data structure depend(n,p) are set to zero. The processing unit 3 may wait until there are no active local events corresponding to any of the n th bay and every p th bay coupled to the n th bay before updating the respective values stored in the data structure correct(p,n,i).
The termination of a local event need not be determined based upon the value of signalStd(n). Instead, the history table 8 may additionally include a data structure recentStd(τ,n), in which T and n are integer indexes defined as for recent(τ,n,i). In such a case, the termination of a local event may be based upon the values stored in recentStd(τ,n) having been less than 3×stdCalm for a period of time such as, for example, stdSize seconds, or 3×stdSize seconds, or 5×stdSize seconds or longer.
Optionally, the processing unit 3 may perform a policing check when there are no local events active (i.e. inEvent(n)=0 for 1≦n≦N) (step S 4 - 3 ). For example, the processing unit 3 may store an array referred to herein as policeUp(n) which is an array having N entries. If the n th bay P n is unoccupied, i.e. occupy(n)=0, then if signal(n) exceeds thresh(n) during an iteration in which signalStd(n) is less than stdCalm(n), the corresponding value of policeUp(n) is increased by one. If any of these conditions are not satisfied on a subsequent iteration, then the value of policeUp(n) is reset to zero. If the value of policeUp(n) exceeds a predetermined number, for example 20, then the occupancy of the n th bay P n is changed to occupied.
Similarly, the processing unit 3 may store an array referred to herein as policeDown(n) which is an array having N entries. If the n th bay P n is occupied, i.e. occupy(n)=1, then if signal(n) does not exceed thresh(n) during an iteration in which signalStd(n) is less than stdcalm(n), the corresponding value of policeDown(n) is increased by one. If any of these conditions are not satisfied on a subsequent iteration, then the value of policeDown(n) is reset to zero. If the value of policeDown(n) exceeds a predetermined number, for example 20, then the occupancy of the n th bay P n is changed to unoccupied.
In this way, even where an event of a vehicle arriving or leaving a bay P n is missed or not captured at the time, the vehicle detection system 1 can subsequently correct the problem to maintain the accuracy of the monitoring.
Traffic Monitoring System
Referring to FIG. 13 , the vehicle detections system 1 may be used to monitor the volume and speed of traffic on a road. The road has four lanes, L 1 , L 2 , L 3 and L 4 and detectors 2 are sensitive to volumes of space which encompass at least part of, for example all of, the lanes L 1 , L 2 , L 3 , L 4 . Detectors 2 are connected in series to form four branches 5 , m=1, m=2, m=3 and m=4. Each branch 5 includes one detector 2 corresponding to each of the four lanes L 1 , L 2 , L 3 , L 4 , and the detectors 2 belonging to the different branches 5 are spread out along each lane L 1 , L 2 , L 3 , L 4 .
Alternatively, more or fewer lanes L may contain detectors 2 . Branches 5 need not be connected so as to span the lanes L, and may alternatively be connected in series such that several detectors 2 corresponding to a single lane L are connected in the same branch 5 .
When used for traffic monitoring, the vehicle detection system 1 has the same processing unit 3 and detectors 2 as are used to monitor a parking environment. The operation of the processing unit 3 is substantially the same as when the vehicle detection system monitors a parking environment, except that the processing unit does not determine whether parking bays P are occupied based upon the magnetic field measurements stored in the history table 8 by the data structure recent(T,n,i). Instead, the processing unit 3 determines the speeds of vehicles based on the magnetic field measurements stored in the history table 8 by the data structure recent(T,n,i). The data structure recent(τ,n,i) is used to calculate the array signalStd(n) in the same way as for the vehicle parking system. However, in the case of a traffic monitoring system, the total number of detectors may be lower and the refresh rate of the history table 8 may be higher than for a vehicle parking system.
The processing unit 3 determines that a vehicle 7 has arrived at the n th detector 2 when the corresponding value of signalStd(n) exceeds 3×stdCalm(n). The processing unit 3 determines that a vehicle 7 has departed from the n th detector 2 when the corresponding value of signalStd(n) drops back below 3×stdCalm(n). The processing unit 3 may determine the delay between the vehicle 7 arriving/departing from successive detectors 2 in the direction of travel in order to determine a speed of the vehicle 7 .
The processing unit 3 may also count the numbers of vehicles 7 which pass along each lane L 1 , L 2 , L 3 , L 4 . The processing unit 3 may also estimate the number of heavy goods vehicles, trucks or lorries using the road based upon the magnitudes of the detected disturbances to the measured magnetic field values.
When the vehicle detection system 1 is used for traffic monitoring, the data structure depend(n,p) may record which detectors 2 are coupled, for example, detectors 2 disposed under adjacent lanes L 1 , L 2 , L 3 , L 4 When monitoring traffic, the vehicle detection system 1 may reject a signal from a detector 2 if an adjacent coupled detector 2 records a larger value of signalStd(n). The branches 5 and detectors 2 may be configured so that coupled detectors may report to the processing unit 3 substantially simultaneously.
When the vehicle detection system 1 is used for traffic monitoring, an event may be initialised for a lane L 1 , L 2 , L 3 , L 4 at the time when a first detector 2 within that lane in the direction of vehicle 7 travel has the corresponding value of signalStd(n) exceed 3×stdCalm(n) and terminated at the time when a last detector 2 within the same lane has the corresponding value of signalStd(n) drop back below 3×stdCalm(n).
Traffic Signal Management
Referring to FIG. 14 , the vehicle detection system 1 may be used for traffic signal management at a junction controlled by automated traffic signals 38 . When an entrance lane E 1 , E 2 , E 3 , E 4 is signalled to stop by the signals 38 , the vehicle detection system 1 may monitor whether, and how many, vehicles are waiting in the entrance lane E 1 , E 2 , E 3 , E 4 in substantially the same way as the vehicle detection system 1 monitors a parking environment. When an entrance lane E 1 , E 2 , E 3 , E 4 is signalled to proceed by the signals 38 , the vehicle detection system 1 may detect traffic volumes and speeds in substantially the same way as the vehicle detection system 1 monitors traffic on a road.
By providing information about both the existence of queuing vehicles and additionally the volume of traffic passing down each approach lane when that lane is signalled to proceed, the traffic signals 38 are enabled to more accurately and reliably perform dynamic adjustments of traffic signal scheduling to maintain traffic flows.
The vehicle detection system 1 is shown in FIG. 14 with detectors 2 emplaced only in the approach lanes to the traffic signal 38 controlled junction. Alternatively, detectors may additionally be emplaced in the exit lanes, allowing the processing unit 3 to infer the distribution of traffic passing from each entrance lane to each of the exit lanes.
Numerous vehicle detection systems 1 , each controlling different junctions, may communicate with each other or with a central client via local networks or the internet to optimise traffic flows in an urban area.
Second Vehicle Detection System for a Parking Environment
Referring to FIGS. 15 and 16 , a second vehicle detection system 39 is similar to the vehicle detection system 1 , except that the second vehicle detection system 39 includes transceiver devices 40 for providing authentication and verification of vehicle 7 identities. Optionally, transceiver devices may also include one or more signalling devices 41 for providing a visual indication of the occupancy status of a space for receiving a vehicle, for example in the form of parking bay P n .
The second vehicle detection system 39 is used for detecting the occupancy of spaces for receiving a vehicle, for example in the form of bays P 1 , P 2 , . . . , P 16 in a vehicle parking environment. The second vehicle detection system 39 comprises at least one detector 2 , one transceiver device 40 and a processing unit 3 . Each detector 2 and each transceiver device 40 form a pair corresponding to a particular parking bay P n . Each bay P 1 , P 2 , . . . , P 16 contains a corresponding detector 2 . A transceiver device 40 corresponding to a bay P n is located proximate to and just outside the bay P n so as to remain visible when the bay P n is occupied by a vehicle 7 . For example, a transceiver device 40 corresponding to a parking bay P n may be arranged centrally with respect to the bay P n width and just outside the bay P n length. The detector(s) 2 are connected to the processing unit 3 by a wired network in the form of one or more detector bus cables 4 a . The transceiver device(s) 40 are connected to the processing unit 3 by a wired network in the form of one or more transceiver bus cables 4 b . The detector(s) 2 connected to a single detector bus cable 4 a and the corresponding transceiver device(s) 40 connected to a single transceiver bus cable 4 b collectively constitute a second branch 42 . Detectors 2 which are connected to a single detector bus cable 4 a and transceiver devices 40 which are connected to a single transceiver bus cable 4 b are connected in a daisy chain configuration by the corresponding bus cables 4 a , 4 b . The detector/transceiver bus cables 4 a , 4 b are substantially the same as the bus cables 4 . However, the detector bus cables 4 a and transceiver bus cables 4 b need not be the same type of bus, and different buses may be used if the transceiver devices 40 and detectors 2 have differing power and/or data requirements.
The detectors 2 are disposed underneath the corresponding bays P n in the same way as for the vehicle detection system 1 . The transceiver devices 40 are emplaced proximate to the corresponding bays P n such that at least a part of each transceiver device 40 is preferably visible above the ground or floor. For example, the transceiver bus cable 4 b may be placed in a slot or trench and the transceiver devices 40 may extend out of the slot or trench above and/or onto the surface of the road or floor. The slot or trench may be filled in or covered over to protect the transceiver bus cable 4 b and connections to the transceiver devices 40 . The individual transceiver devices 40 may also be fastened to the road or floor, for example using bolts.
Each transceiver 40 includes a controller 43 , a network interface 44 connecting the detector to a bus 14 and a WPAN module 15 supporting wireless personal area network communications such as Bluetooth®, ZigBee® or Z-Wave®. The bus 14 connects to a bus network interface 19 in the processing unit 3 in order to allow communication between the controller 43 and the processor 16 ( FIG. 5 ). Optionally, each transceiver device 40 also includes at least one signalling device 41 , for example first and second light emitting diodes (LEDs) 45 a , 45 b.
The transceiver(s) 40 are sealed to prevent the ingress of particles such as particles of dirt and/or dust. The transceiver(s) 40 are sealed so as to be waterproof. The transceiver device(s) 40 are encapsulated to the IP-68 certification standard, in which IP stands for Ingress Protection. The IP-68 standard refers to dust tight protection against particle ingress and waterproof to immersion beyond 1 m depth of water.
The controller 43 sends and receives challenge/response data to/from the WPAN module 15 for exchanging information with a WPAN transceiver on or in a vehicle 7 or carried by a vehicle operator, in order to authorise and identify a vehicle 7 entering a parking bay P n . Each transceiver device 40 communicates with the processing unit 3 via the network interface 44 and a bus 14 . The bus 14 is disposed within the transceiver bus cable 4 b . The transceiver bus cable 4 b and the interface with the transceiver device are encapsulated to the IP-68 certification standard.
The WPAN module 15 includes an antenna capable of bi-directional communication with a WPAN transceiver (not shown) on or in a vehicle 7 or carried by the vehicle operator. The WPAN module 15 has a transmission power allowing communication with a transceiver in the range of 5 to 10 m away. In the second vehicle detection system 39 the WPAN module 15 preferably communicates according to the Bluetooth® standard.
The controller 43 may also be configured to receive status indications from the processor 16 and to change the output state of the signalling device(s) 41 , for example to illuminate the LEDs 45 a , 45 b in response to the status indications. Each LED 45 a , 45 b may be a single colour LED, a multi-colour LED or an LED array. For example, the processor 16 may send a signal to a transceiver device 40 to communicate that the corresponding detector 2 has detected that the respective parking bay P n has become occupied, and in the response the controller 43 may cause one or more LEDs 45 a , 45 b to switch from green illumination to red illumination. Alternatively, when single colour LEDs 45 a , 45 b are used the LEDs 45 a , 45 b may be illuminated when the corresponding parking bay P n is unoccupied and switched off when the bay P n is occupied. In this way, users of the parking environment can clearly see whether a row of parking bays P n has available space from a distance, without needing to drive along a row.
Further information may be communicated by the signalling device(s), for example first and second LEDs 45 a , 45 b . For example multi-colour or array LEDs 45 a , 45 b may be illuminated using a third colour to indicate that a vehicle 7 has overstayed a maximum duration, for example LEDs 45 a , 45 b may be illuminated to provide orange light. Alternatively, multi-coloured, array or single colour LEDs 45 a , 45 b could be controlled to blink or flash to indicate overstaying. This may provide an easily understood and visible prompt for parking enforcement authorities. Alternatively, the LEDs 45 a , 45 b may be used to provide similar visual indications that a vehicle 7 is not identified or unauthorised for the corresponding parking bay P n .
Referring also to FIGS. 11 and 12 , the processing unit 3 determines that a vehicle 7 is entering or has entered the bay P n corresponding to a transceiver device 40 using signals from the respective detector 2 and also any coupled detectors according to the first or second methods of detecting the presence of a vehicle in a volume of space. In response, the processing unit 3 sends challenge message data (not shown) for transmission by the WPAN module 15 . The controller 43 receives the challenge message data and directs the WPAN module 15 to transmit a challenge message 46 . The transceiver device 40 waits to receive a response message (not shown), and any response message data (not shown) received by the WPAN module 15 is relayed to the processing unit 3 . The communication protocol employed by the WPAN module 15 may be encrypted using conventional or application specific algorithms. The WPAN module 15 may communicate with WPAN transceivers which are external tokens employing, for example Bluetooth®, smartphones using Bluetooth®, smartphones having Bluetooth®-WiFi dongles or other similar devices.
In this way, the second vehicle detection system 39 may be operated in a substantially similar way to the vehicle detection system 39 , except that the WPAN module 15 is included within the transceiver device(s) 40 instead of within the detector(s) 2 . Because the transceiver device(s) 40 are arranged proximate to the bays P n instead of within or under the bays P n , the vehicle 7 is less likely to block or interfere with transmission from the WPAN module 15 . In this way, a WPAN module 15 provided in a transceiver device 40 may communicate more reliably with a transceiver or token in or on the vehicle 7 or carried by the vehicle operator.
Third Method Including Vehicle Detection
Referring to FIGS. 11, 12, and 15 to 17 , a third method of detecting the presence of a vehicle in a volume of space is the same as the second or third methods except for the step (step S 5 ) of identifying a vehicle 7 . Steps relating to detection of vehicles 7 using the detectors 2 according to the first or second methods (steps S 1 , S 2 , S 3 , S 4 , S 4 - 1 , S 4 - 2 , S 4 - 3 , S 6 ) are the same in the third method and detailed description is not repeated. The third method is carried out using an example of the second vehicle detection system 39 which includes transceiver devices 40 corresponding to each detector 2 .
The processing unit 3 is in communication with an external device 47 via the external network interface 18 . The external device 47 may be, for example, a data processing apparatus having a display (not shown) and located in a control room (not shown) for monitoring by an operator. The external device 47 may be, for example, a server which is in further communication with a data processing apparatus having a display (not shown) and located in a control room (not shown). Alternatively, the external device 47 may be a portable device, for example a mobile phone, tablet computer or similar device, which is carried by a parking enforcement operative. The external device 47 may be in communication with several second vehicle detection systems 39 at the same time.
When the processing unit 3 has determined that a vehicle 7 has arrived or is arriving into a space for receiving a vehicle 7 , for example a parking bay P n corresponding to a detector 2 (e.g. step S 4 of the first method, steps S 4 - 1 or S 4 - 2 of the second method), one or more transceiver devices 40 belonging to the same second branch 42 as the respective detector 2 broadcast a randomly selected challenge message 46 (step S 5 - 1 ). For example, at least the transceiver device 40 corresponding to the detector 2 which detects an arriving vehicle 7 broadcasts a challenge message 46 . In another example, the transceiver devices 40 corresponding to adjacent or coupled parking bays P n may broadcast the challenge message 46 . Preferably, all transceiver devices 40 connected to the same second branch 42 broadcast the challenge message 46 . The one or more transceiver devices 40 all broadcast the same challenge message 46 concurrently, during overlapping time-periods or during separate time-periods. Because multiple transceiver devices 40 connected to a single transceiver bus cable 4 b may broadcast the same challenge message 46 , there is an increased probability that one of the transceiver devices 40 will have a favourable transmission path to a WPAN transceiver or token in or on the vehicle 7 or carried by the operator (if the vehicle or operator are authorised). The processing unit 3 randomly selects the challenge message 46 from information stored in the transceiver identity table 28 . The transceiver identity table 28 stores a large number, for example hundreds or thousands of challenge messages 46 and the corresponding correct response messages.
The WPAN module 15 of every transceiver device 40 connected to the respective second branch 42 listens for a reply (step S 5 - 2 ). All of the transceiver devices 40 within transmission range of the vehicle 7 may listen for and receive any response message. If no reply is received within a predetermined duration (step S 5 - 2 ; No), for example ten seconds, the processing unit 3 checks whether the total elapsed time since activity was detected has exceeded a maximum allowed time (step S 5 - 11 ). If the maximum allowed time has not elapsed (step S 5 - 11 ; No), then a new challenge message is selected by the processing unit 3 and communicated to the one or more transceiver devices 40 for broadcasting (step S 5 - 1 ). If the maximum allowed time has elapsed (step S 5 - 11 ; Yes), then the processing unit 3 flags the vehicle 7 occupying or entering the respective bay P n as unauthorised and transmits a message to the external device 47 (step S 5 - 10 ).
However, if a response message (not shown) corresponding to the challenge message 46 is received (step S 5 - 2 ; Yes), then the controller 44 of the transceiver device 40 communicates the response to the processing unit 3 which checks whether the received response matches the answer stored in the transceiver identity table 28 (S 5 - 3 ). If the response is correct (step S 5 - 3 ; Yes), then the processing unit 3 increments an internal counter of correct responses by one (step S 5 - 5 ). However, if the response is not correct (step S 5 - 3 ; No), then the processing unit 3 increments an internal counter of incorrect responses by one (step S 5 - 4 ). Alternatively, the processing unit 3 may store a first counter which is the total number of challenge messages 46 in reply to which any response message had been received and a second counter which is the total number of correct or incorrect responses received.
If the same response message to a particular challenge message is received by multiple transceiver devices 40 , the response message is only counted the first time it is received.
The processing unit 3 checks whether at least a sufficient or threshold total number of responses have been received in order to confirm/refuse authorisation of the vehicle 7 (step S 5 - 6 ). For example, the threshold number may be at least five, at least ten or at least fifteen responses (correct or incorrect). If the minimum number of responses has not yet been received (step S 5 - 6 ; No), then the processing unit 3 checks whether the maximum allowed time has elapsed (step S 5 - 11 ).
If the minimum number of responses has been received, the processing unit 3 checks whether a sufficiently high fraction of the responses received are correct (step S 5 - 7 ). For example, the processing unit may require that 75%, 80%, 90% or more of received responses are correct. If a sufficiently high fraction of responses received are correct (step S 5 - 7 ; Yes), then the processing unit 3 transmits a message to the external device 47 to indicate that the detected vehicle 7 has been authenticated. The processing unit 3 also directs the one or more transceiver devices 40 in the corresponding second branch 42 to broadcast a message requesting the identity or name of the authorised vehicle (step S 5 - 9 ). When one or more transceiver devices 40 receive a response including the identity or name of the vehicle, for example a vehicle registration number of an operator or owner name, the processing unit 3 logs the identity. Any transceiver device 40 may receive a response, not limited to only the one or more transceiver devices 40 which broadcast the identity request. The processing unit 3 may also transmit the identity information to be transmitted to the external device 47 , where the identity of the vehicle may be shown on a display (not shown) to a person responsible for monitoring the second vehicle detection system 39 .
If a sufficiently high fraction of correct responses has not been received (step S 5 - 7 ; No), for example if only 65% of response are correct and the threshold for authentication is 80%, the processing unit 3 checks whether a sufficiently high fraction of responses received are incorrect (step S 5 - 8 ). For example, the processing unit 3 may require that 75%, 80%, 90% or more of received responses are incorrect. If a sufficiently high fraction of response received are incorrect (step S 5 - 8 ; Yes), then the processing unit 3 flags the vehicle 7 which is entering or has entered the parking bay P n as unauthorised (step S 5 - 10 ). The processing unit 3 may also transmit a message indicating that detected vehicle is not authenticated to the external device 47 , where the unauthorised status of the vehicle may be shown on a display (not shown) to a person responsible for monitoring the second vehicle detection system 39 . Additionally or alternatively, the processing unit 3 may cause an e-mail or SMS message to be sent to a parking enforcement operator responsible for the parking environment, either directly through the external network interface 18 ( FIG. 5 ) or indirectly via the external device 47 .
If a sufficiently high fraction of incorrect responses has not been received (step S 5 - 8 ; No), the processing unit 3 checks whether the maximum allowed time has elapsed (step S 5 - 11 ). In this way, the one or more transceiver devices 40 continue to transmit challenge messages to the newly arriving or arrived vehicle 7 until either a sufficiently high fraction of responses are confirmed correct, a sufficiently high fraction of responses are confirmed incorrect or the maximum allowed time expires. This can allow the authentication and identification of a vehicle 7 to be performed robustly even in circumstances where many broadcast messages may be interrupted or blocked due to stationary and/or moving vehicles 7 .
In the second vehicle detection system 39 operating according to the third method, the challenge messages 46 and corresponding correct responses stored in the transceiver identity table 28 may be updated according to a predetermined schedule, for example daily, weekly or monthly. The challenge messages and corresponding responses may be updated remotely via the external network interface 18 ( FIG. 5 ) of the processing unit 3 .
Alternatively, the transceiver identity table 28 need not contain entirely predetermined challenges and corresponding responses. Instead, the challenge messages may be partly or fully made up of information which is generated by the processing unit 3 and/or a WPAN transceiver in or on the vehicle 7 . For example, the processing unit 3 may read data such as a number or a string (not shown) from the transceiver identity table and/or from a system clock. The processing unit 3 may apply a first mathematical operation to the number or string and broadcast the output number or string of the first operation. The processing unit 3 may then apply a second mathematical operation to the output number or string to generate the correct response. The WPAN transceiver in or on the vehicle 7 received the output number or string, applies one or more second mathematical operations to the received output number or string, and broadcast a message including one or more response numbers or strings. When one or more transceiver devices 40 receives a message including response numbers or strings, these are checked against the correct response generated in the processing unit 3 to determine whether any of the responses are correct. In some examples, the transceiver identity table 28 may include a list of first mathematical operations and corresponding second mathematical operations.
The transceiver devices 40 may additionally record information about the relative signal strength of messages received from a WPAN transceiver associated with a vehicle 7 . The processing unit 3 may use the relative signal strengths of received messages in order to estimate ranges from each transceiver device 40 to the replying vehicle 7 . Range information may be used by the processing unit 3 for the purpose of distinguishing between different vehicles 7 when two or more vehicles arrive close together or simultaneously to nearby parking bays P n , for example to coupled bays P n .
As explained hereinbefore, the controller 43 of each transceiver device 40 may be configured to receive status indications from the processing unit 3 and to illuminate the LEDs 45 a , 45 b in response to the status indications. For example, LEDs 45 a , 45 b may be illuminated in one or more colours and/or flashed, blinked or strobed in order to indicate that a corresponding parking bay P n is occupied, unoccupied, occupied by an unauthorised vehicle or occupied by an overstaying vehicle etc.
First Transceiver Device
Referring to FIGS. 16, 18 and 19A to 19C , a first example of a transceiver device 40 , 48 includes a low-profile dome-like casing 49 . An electronics module 50 including the controller 42 , WPAN module 15 , first and second LEDs 45 a , 45 b and a network interface 44 is received into the casing 49 and secured in place by fastening a circular base plate 51 to the casing 49 . The electronics module 50 may be substantially cuboidal. However, the electronics module 50 may take other shapes. The interior surface of the casing 49 , which in general need not correspond to the exterior low-profile dome shape of the casing 49 , conforms to the shape of the electronics module 50 .
The casing 47 is formed with a dome-portion 52 extending upwards from a cylindrical portion 53 . The dome-portion 52 extends upwards for a greater distance than the cylindrical portion 53 . A pair of indentations 54 are formed on opposite sides of the dome-portion 52 . Each indentation 54 is formed with a flat base 55 extending perpendicular to the external surface of the cylindrical portion 53 and a curved wall 56 extending perpendicularly upwards from the flat base 55 and ending at the exterior surface of the dome portion 52 . The curved wall 56 of the indentations 54 is bounded by the flat base 55 and the exterior surface of the dome-portion 52 but is not bounded by an interior surface of the casing 49 . A through hole 57 is formed through each flat base 55 which connects to the underside of the casing 49 . When assembled, the through holes 57 are aligned with corresponding through holes 58 formed through the base plate 51 . The casing 49 and base plate 51 are secured using bolts 59 received by the through holes 57 in the indentations 54 and the through holes 58 in the base plate 51 . Optionally, washer(s) 60 may be used. The bolts 59 may be secured using nuts (not shown) tightened against an underside of the base plate 51 , or the through holes 58 in the base plate 51 may have an internal thread formed therein. The base plate 51 may be received into a stepped recess 61 formed into the underside of the cylindrical portion 51 of the casing 47 , such that the undersides of the casing 49 and base plate 51 are substantially flush.
The casing 49 also includes a pair of windows 62 . The windows 62 are formed opposite to one another and at positions rotated ninety degrees about the axis of the dome portion 52 from the indentations 54 . Similarly to the indentations 54 , each window 62 is formed with a flat base 63 . However, unlike the indentations 54 , each window 62 is formed with a curved surface 64 which is bounded by the flat base 63 , the exterior surface of the dome portion 52 and an interior surface 65 of the casing 49 . In this way, the window 62 connects the interior and exterior of the casing 49 . When assembled with the electronics module 50 , the windows 62 align with the LEDs 45 a 45 b to allow light emitted from the LEDs 45 a , 45 b to exit the first transceiver device 48 . A first LED 45 a is visible through one window 62 and a second LED 45 b is visible through the opposite window 62 .
The circular base plate 51 includes a circular aperture 66 formed concentrically with the outer perimeter of the base plate 51 . A connector 67 descends perpendicularly from the underside of the electronics module 50 and is received through the circular aperture 66 when the first transceiver device 48 is assembled. The connector 67 is for connecting to a corresponding connector 68 extending from the transceiver bus cable 4 b . The casing 49 and base plate 51 lie on the surface of the road or floor and the connector 67 projects down into a trench or slot cut into the road/floor surface and into which the transceiver bus cable 4 b is received. The first transceiver device 48 is installed so that the first and second LEDs 45 a , 45 b are directed along a road or a row of parking bays, so that vehicle operators can easily spot an unoccupied space at a glance.
The casing 49 must be mechanically strong enough to tolerate the tyres of a vehicle 7 passing directly over the first transceiver device 48 . The casing 49 may be reinforced internally by struts, trusses or similar reinforcing structural elements. The casing 49 and base plate 51 may be formed from suitable metallic materials. For example, aluminium 6061-T6 or other aluminium alloys having comparable mechanical and corrosion properties. Alternatively, the casing 47 and base plate 49 may be made from high strength polymeric materials or polymer composite materials.
Second Transceiver Device
Referring also to FIGS. 20A to 20D , a second transceiver device 40 , 69 is the same as the first transceiver device 48 except that the second transceiver device 69 includes a WPAN module 15 attached on the outside of the electronics module 50 and the casing 49 of the second transceiver device 69 is modified to include a transmission window 70 which may help to reduce or prevent the casing 49 material from blocking or interfering with transmissions to and from an antenna 71 of the WPAN module 15 .
The transmission window 70 is received into an aperture 72 formed through the metal casing 49 . The transmission window 70 includes a lip 73 which is received into a corresponding slot 74 formed into the side of the aperture 72 . The boundary between the transmission window 70 and the sides of the aperture 72 may be sealed against dust and water using, for example, waterproof adhesive (not shown), silicone rubber (not shown) or similar compounds.
The WPAN module 15 takes the form of a Bluetooth® module including a substrate 75 , a Bluetooth® circuit board 76 mounted on the substrate 75 and connectors 77 extending away from the substrate 75 . Each connector 77 is electrically connected to the Bluetooth® circuit board 76 by conductive traces (not shown) disposed on the substrate 75 . The Bluetooth® circuit board 76 includes an antenna 71 in the form of a conductive region 78 disposed on the Bluetooth® circuit board 76 . When the second transceiver device 69 is assembled, the Bluetooth® WPAN module 15 is connected to the electronics module 50 and projects from the upper surface of the electronics module 50 . The electronics module 50 is received into the casing 49 such that the Bluetooth® WPAN module 15 is received in a recess 79 formed in the upper portion of the casing 49 and connecting with the aperture 72 . When the second transceiver device 69 is assembled, the antenna 71 , 78 is arranged substantially beneath the transmission window 70 . In this way, blocking or interference of signals to and from the antenna 45 by the casing 49 material may be reduced or prevented.
Modifications
It will be appreciated that many modifications may be made to the embodiments hereinbefore described. Such modifications may involve equivalent and other features which are already known in the design, manufacture and use of magnetic field sensors and which may be used instead of or in addition to features already described herein. Features of one embodiment may be replaced or supplemented by features of another embodiment.
Detectors 2 have been described as comprising a pair of magnetic field sensors 6 a , 6 b , housed in respective first and second units 9 , 10 . However, detectors may employ more than two sensors 6 . Vehicles may not be positioned precisely in a volume of space, and magnetic materials may be unevenly distributed in a vehicle 7 , such that the size of magnetic field variations within a volume of space may be uneven. A single sensor may have blind spots and fail to detect the largest/most significant magnetic field variations. Using a pair of sensors 6 a , 6 b may help to improve the accuracy and reliability of detecting a vehicle by reducing or eliminating sensor blind spots. Adding further sensors, for example a total of three sensors 6 , four sensors 6 or more, may further improve accuracy and reliability. Where more than two sensors 6 are used, each may be housed in a separate unit, and each unit being spaced from the other units by one or more link members 11 .
The detectors 2 have been described as being encapsulated to the IP-68 certification standard. However, other standards or levels of ingress protection may used for the detector(s) 2 , depending upon the physical and regulatory environment in which the detector(s) 2 are to be deployed.
The controller 12 has been described as being in the form of an Atmel ATmega328P-AU incorporating an Arduino® Bootloader microcontroller. However, the controller 12 may alternatively be provided by any other suitable microcontroller. The controller 12 need not be a microcontroller and may alternatively be provided by a field programmable gate array, a microprocessor or a dedicated integrated circuit.
The magnetic field sensors 6 have been described as being in the form of Honeywell HMC5883L magnetic field 3D sensors. However, the magnetic field sensors 6 need not be Honeywell HMC5883L magnetic field 3D sensors. Other types of anisotropic magnetic field sensors may be used instead.
Alternatively, the magnetic field sensors 6 need not be anisotropic magnetoresistance sensors, and any magnetic field sensor having sufficient sensitivity to detect changes in local magnetic field due to a vehicle 7 may be used such as, for example, Hall probes or giant magnetoresistance sensors. The magnetic field sensors 6 may be provided by any type of sensor having an output rate of more than 1 Hz, between 1 and 10 Hz, between 10 and 100 Hz or more than 100 Hz. The magnetic field sensors 6 may be provided by any type of sensor having a resolution equal to or better than 1×10 −7 T. The magnetic field sensors 6 need not use a 12-bit ADC, and may use, for example, an ADC operating at 8-bit, 16-bit, 32-bit or greater precision.
The optional WPAN module 15 has been described as having a transmission power allowing communication with a transceiver up to 5 m away. However, the WPAN module 15 may have a transmission power to allow communication up to 10 m away or further.
The optional WPAN module has been described as being a Zigbee® module. However, the WPAN module need not be a Zigbee® module, and other protocols may be used for WPAN communications such as, for example, Bluetooth®, IEEE 802.11, IEEE 802.15.4, Z-wave® or similar.
Vehicles 7 may include internal combustion engine(s). Vehicles 7 may include electric motor(s). Vehicles 7 may be hybrid vehicles including internal combustion engine(s) and electric motor(s). Vehicles 7 may be unpowered vehicles designed to be towed by another vehicle such as, for example, caravans, trailers, construction equipment or agricultural equipment. Trailers may be small trailers for cars or large trailers for trucks.
The second vehicles detection system 39 has been described as including a second branch 42 made up of a number of detectors 2 connected in a daisy chain configuration by a detector bus cable 4 a and a number of corresponding transceiver devices 40 connected in a daisy chain configuration by a transceiver bus cable 4 b . With this arrangement, the detectors 2 and transceiver devices 40 may by connected using different types of bus cable. However, if the same type of bus cable is used for detectors 2 and transceiver devices 40 , then separate bus cables 4 a , 4 b need not be used. Instead, the detectors 2 and transceiver devices 40 forming a second branch 42 could all be connected to the same bus cable 4 , for example by connecting together all of the detectors 2 in a first trench, then doubling the bus cable 4 back through a second trench to connect the corresponding transceiver devices 40 .
Transceiver devices 40 , 48 , 69 have been described which include WPAN module 15 and which optionally include one or more signalling devices 41 , for example first and second LEDs 45 a , 45 b . Alternatively, instead of transceiver devices 40 , 48 , 69 , indicator devices (not shown) may be used having similar construction but omitting the WPAN module 15 and including at least one or more signalling devices 41 . Such indicator devices may be used in the same way as transceiver devices 40 , 48 , 69 to provide visible indications of the status of a parking bay P n such as unoccupied, occupied or that an occupying vehicle has overstayed. When an indicator device having no WPAN module 15 is used, the first or second vehicle detection methods may be used, omitting steps of authenticating and/or identifying a vehicle 7 .
Although claims have been formulated in this application to particular combinations of features, it should be understood that the scope of the disclosure of the present invention also includes any novel features or any novel combination of features disclosed herein either explicitly or implicitly or any generalization thereof, whether or not it relates to the same invention as presently claimed in any claim and whether or not it mitigates any or all of the same technical problems as does the present invention. The applicant hereby gives notice that new claims may be formulated to such features and/or combinations of such features during the prosecution of the present application or of any further application derived therefrom.