GB2408372A - Apparatus and method for determining the presence of a vehicle on a highway - Google Patents

Abstract

A vehicle detector for detecting vehicles on a highway comprises an inductive loop and an inductive detection means. The loop will preferably be at least double the length of a single vehicle and may be over 20m long. In one embodiment a counter is attached to the induction detection means and functions by comparing a current inductance value to one when no vehicles are present and divides the inductance value by a predetermined amount corresponding to the change caused by a single vehicle. In a another embodiment the sensitivity of the loop may vary as a vehicle moves along the roadway. The width and depth of the loop may vary along the length. The loop may also have additional conductors for changing the sensitivity of the loop in specific areas. In both embodiments the loop may be able to determine the direction of a vehicle. In use the loop is disposed along a roadway or hard shoulder to detect the presence of vehicles and may be connected to a traffic management system and a plurality of electronic signs for indicating if a lane may be used by running traffic.

Description

DETERMINING THE STATUS OF AN AREA OF Hl(lHWAY The present invention

relates to highway traffic detection systems, and in particular to the reliable detection of the status of a single lane, including factors such as the occupancy of the lane and the direction of movement and approximate velocity of vehicles in the lane.

Highway operators are interested in the safe operation of the highways under their control. In England, the largest inter-urban highway operator is the Highways Agency, part of the Department for Transport. The Highways Agency also seeks to obtain the maximum utility and throughput from its network of approximately 6,000 kilometres of dual carriageway roads. Such roads have a central reservation so that the carriageways carrying traffic in opposite directions are separated. Generally speaking, such roads also include a shoulder lane, sometimes known as a "hard shoulder", at the outside edge of each carriageway. These shoulder lanes are traditionally used as an emergency lane in which vehicles which have broken down can stop, leaving the main carriageway clear for moving traffic.

Recently the Highways Agency has decided to assess various forms of Active Traffic Management (ATM) including the possibility of using the hard shoulder of a motorway as a running lane during peak hours, when the normal running lanes are already full of traffic. The hard shoulder may be opened to running traffic at times when demand exceeds supply for the normal traffic lanes, and closed when sufficient capacity is available in the normal lanes. Signs may be fitted on overhead gantries over each lane, and when the hard shoulder is available for use by traffic as a running lane, a green arrow may be displayed above it. This will change to a red cross when the lane is closed. Emergency rescue areas (ERA) would be installed at approximately 500 metro intervals so that vehicles breaking down can use these EKAs to pull off the main carriageway, and make use of an emergency telephone to summon help. More details of 3() the proposed operation can be found on the Highways Agency website, whose URI, is http://www.highwaysagency gov uk In order to operate such a system, it is necessary to detect the presence of stationary or slow moving vehicles on the hard shoulder before it is opened to traffic. If the hard shoulder was to be opened when a stationary or very slow moving vehicle was within S the lane, an accident could occur as vehicles moved into the lane, expecting it to be open and clear, and came across the stationary vehicle.

Furthermore, once the hard shoulder has been opened for use as a running lane, it is useful to be able to determine when a vehicle has stopped and not removed itself to the I0 ERA. The hard shoulder must then be temporarily closed, for example by means of a sign showing a red cross above the hard shoulder.

Due to the intensity of traffic flow and density in areas where this technique is useful, there is a need for particularly reliable sensing of vehicles in the hard shoulder. A sensor which can give an indication of its own status is desirable. In addition, the sensor must never indicate that the lane is empty, when a stationary vehicle is present, and at best very rarely give an output indicating the presence of a vehicle when there is none there.

It will be appreciated that these requirements apply not only to shoulder lanes, but also to any arrangement in which lanes can be opened and closed. Other examples include normal travelling lanes in a tunnel, or elsewhere where it is necessary to make more use of lanes than is presently the case, for example in the case of "tidal flow" lanes where a lane is used in a different direction at different times of the day, or elsewhere where a higher level of requirement t'or safety exists.

There are a number of devices already in use t'or the determination of the status of an area of highway. These include Video Image Processing (VIP), microwave detectors, infrared detectors, scanning lasers and spot loop detectors.

Video image Processing (VIP) uses a closed-circuit television (CCTV) and image analysis to determine whether vehicles are present in, or entering, a zone of interest.

VIP works reasonably well but has a number of shortcomings. These include: Obscuration: If the camera is not at a suitable height, vehicles in the foreground may obscure vehicles in the zone of interest.

Distance: The effectiveness of a VIP system is greatly reduced with increasing distance from the area of interest. VIP systems generally do not function well unless they are spaced at 100-200 metro intervals when mounted at typical heights.

Environmental obscuration: Rain, snow, fog, smoke, haze and spray can affect the data recorded by the camera. This can cause partial failure or even total disability. Often conditions of bad visibility occur just when the system is most needed to function correctly.

Microwave detectors use a beam of microwave energy which is radiated over the zone of interest. A portion of the beam is reflected by objects in the zone, especially from metallic vehicles which are good reflectors. If the vehicle is moving, there is a Doppler shift in the frequency of the return signal which is detected by a mixer chamber in the detector. An output is then given indicating the presence of the Doppler shift (indicating that a moving vehicle is present in the zone of interest), and the output usually also indicates the speed of the vehicle. There are again shortcomings in the application of microwave technology to lane or area status monitoring, and these include: Detection of stationary vehicles: If a vehicle stops, there is no Doppler shift in the signal received at the receiver. The receiving system may therefore conclude that no vehicle is present when there is a stationary vehicle in the zone of interest.

Distance: Microwave systems typically have a limited range of up t-'200 metres.

Obscuration: As with VIP technology, if the microwave detector is not at a suitable height, vehicles in the foreground may obscure vehicles in the zone of interest.

Environmental obscuration: Very heavy rain or snow may reduce the signal at the receiver, causing partial failure or even total disability. As with VIP technology, conditions of bad visibility often occur just when the system is most needed to function correctly.

Infrared detectors rely either on naturally emitted infrared emissions from a vehicle, caused by the difference between the background temperature and the temperature of the vehicle, (so-called "passive detectors"), or reflected infrared radiation from an actively scanned or illuminated area (so-called "active detectors"). The shortcomings of infrared detection for stationary vehicle detection are: Distance: Such sensors only operate over a limited distance, typically up to a maximum of 100 metros.

A vehicle with the same overall temperature as the background may not be detected at times with a passive device. This is often a problem with motorcycles, energy efficient vehicles and certain battery driven vehicles.

Due to the dynamic nature of typical passive infrared detectors, vehicles are "lost" after remaining stationary for a period of time. This is the same effect as seen in a burglar alarm sensor, where standing still for a period will cause the detector to "lose" the intruder.

Obscuration: As with VIP and microwave technology, if the infrared detector is not at a suitable height, vehicles in the foreground will obscure vehicles in the zone of interest.

Environmental obscuration: To a similar extent as with microwave detectors, very heavy rain and snow may have varying effects from total disability to partial failure. Again, this occurs jLlSt when the system is most needed to function correctly.

Scanning laser detectors use a narrow beam of infrared energy which is scanned at high speed over the zone of interest. A portion of the beam is reflected by objects in the zone, especially shiny objects such as cars, which are generally good reflectors. Such systems have a very limited range of up to 50-100 metros. Other shortcomings of the application of laser scanning technology to lane or area status monitoring include: Obscuration: As with VIP technology, if the laser scanning detector is not at a suitable height, vehicles in the foreground will obscure vehicles in the zone of interest. One way to address this is to mount increase the spatial frequency of 10the units along the side of the highway, but other roadside furniture may obstruct the beam, and the lasers themselves become a form of roadside clutter and/or attraction for vandals.

Environmental obscuration: In rain, spray and snow conditions there may be many false alarms. Performance in fog is degraded.

15 Cost: The cost is very high and the scanning device is more likely to fail for mechanical reasons than the other sensors described, which are all solid state devices.

Finally, spot loop sensors involve the provision of conducting loops embedded in the roadway at discrete points along the lane. An alternating current is supplied to each loop, setting up a resonant circuit and enabling the inductance of the loop to be measured. The inductance ol the loop changes when a metal chassis passes over the loop, and this change is detected by a loop detector. The presence of a vehicle passing over the loop can thus by identified. Interpolation and/or estimation can be used as to what is happening between consecutive sensors.

Typically pairs of normal loops, each 2 metros square, are placed at 100 metro intervals.

A vehicle is detected and its sp,eetl calculated by timing the leading edge actuation ol each loop. Since the systc,n knows the distance between the spot sites, it can calculate the expected time of arrival of the vehicle at the next pair of loops, and if the vehicle does not arrive, deduce that the vehicle must have stopped between the two spot sites.

A variant of this approach is to count vehicles at both spot sites between periods when there are no vehicles detected. If a vehicle is "lost" at the second station, then it is presumed that the vehicle has stopped.

Spot loop sensors overcome many of the distance and environmental factors suffered by the devices described above, but suffer from further shortcomings, including: The system does not actually detect in the zone of interest, but deduces from the readings of the spot sensors that a vehicle has stopped between them. There is a lack of affirmative evidence of the presence of the vehicle.

Vehicles may change lanes between the spot loop sites and give false actuations of a stopped vehicle.

A vehicle such as a motorcycle may enter the zone of interest between two spot loop lanes, and stop without passing over any loops. It is therefore possible for a stationary vehicle to remain undetected.

Thus it can be seen that all the known devices have significant disadvantages for use in determining the status of a shoulder lane which may contain stationary vehicles. If a shoulder lane is to be used as a running lane for traffic, it is vital that the presence of a stationary or slow-moving vehicle can be reliably identified. With such basic disadvantages, the known devices cannot be relied on to identify stationary vehicles with a sufficient level of confidence.

In accordance with a first aspect of the present invention, there is provided a vehicle detector for detecting vehicles on a roadway, comprising: a conductive loop for location in or on the roadway, having a dimension in the direction of travel of vehicles along the roadway of at least double the length of a single vehicle so that a vehicle above at least a porlio,1 of the loop will change the inductance of the loop; inductance detection means S() for determining the inductance al the conductive loop; anti counting means arranged to count how many vehicles are in the detection area of roadway above the loop by comparing the inductance of the loop with the inductance when no vehicles are in the detection area, and dividing by a predetermined value corresponding to the typical change in inductance caused by a single vehicle in the detection area. A typical such loop will be 20 to 100 m in length in the direction of travel.

Thus a single loop sensor can be used to cover the whole length of a section of roadway, enabling any vehicles on that section to be detected.

The loop sensor may be excited in a resonant arrangement in the same manner as known loop sensors, and changes in inductance monitored. The presence of sharp changes in inductance may be associated with vehicles entering and leaving the detection area of roadway covered by the loop. In this way a count of the number of vehicles in the zone of detection may be determined, and communicated to a further downstream system. In addition or alternatively, the change of inductance generated by a single vehicle may be known, enabling an independent calculation of the number of vehicles over the loop from the total change in inductance of the loop from its known value when no vehicles are located over it.

In a preferred embodiment, the loop sensitivity may be non-homogeneous in the direction of travel of vehicles on the roadway, so as give a timevarying signal while a vehicle is moving along the loop. This signal will return to a steady state signal when all vehicles in the zone of detection stop moving. The speed of vehicles may be determinable from the rate of change of the time-varying signal.

The longitudinal variation in the loop may be achieved in a number of ways, including varying the width or the depth of the loop, or introducing additional conductors, or secondary loops, at points along the loop. These further conductors may be attached to the loop itself, or arranged adjacent to the loop (for example above or below the loop).

impedance variation means, for example in the loran of a switch, may be associated with the additional conductors. This enables the sensitivity of different portions of the loop to be changed to meet different requirements.

The loop may comprise a plurality of alternating high and low sensitivity sections, so that a vehicle located above a high sensitivity section will cause a greater change in the inductance of the loop than a vehicle located above a low sensitivity section. A vehicle moving along such a loop will give rise to an alternating signal superimposed on the step signal generated by the presence of the vehicle above the loop. The speed of the vehicle can be estimated from the frequency of the alternating component of the signal.

In one embodiment the different sensitivity of the sections is achieved by their being different widths.

In a further preferred embodiment, the sensitivity of at least a portion of the loop sensor is not reflectively symmetrical in the direction of travel of vehicles, so that the changes of inductance caused by a vehicle travelling forwards over the loop can be distinguished from those caused by a vehicle travelling backwards. The loop may have a portion where the sensitivity varies linearly with length along that portion, for example by linearly changing the width along that portion. The direction of travel of a vehicle can then be determined from whether the signal increases or decreases as a vehicle moves over that portion.

In one embodiment, the sensitivity increases gradually along the whole length of the loop. Alternatively, the loop may be divided into two or more sections, within each of which the sensitivity increases along the length of that section.

The invention also provides a device for determining the status of a section of roadway, comprising a vehicle detector as described above having a conductive loop installed along substantially the whole length of that section of roadway. The section of roadway may comprise a single lane, for examE,lc a shoulder lane, of a multi-lane road. 3()

The invention further provides a traffic management system comprising a roadway having a plurality of lanes, a device as just descriLcd installed in a lane of the roadway for determining the status of that lane, and a plurality of electronic signposts located over that lane for indicating whether or not it may be used by running traffic.

In accordance with a second aspect of the present invention there is provided a method of determining whether a section of roadway longer than two vehicle lengths is occupied by vehicles, the method comprising measuring the inductance of a conductive loop installed in the surface of the roadway along substantially the whole length of the section, and determining from the measured inductance whether a vehicle is located above the loop.

Some preferred embodiments of the invention will now be described by way of example only and with reference to the accompanying drawings, in which: Figure l is a perspective view of two lanes of a road, with a loop sensor in one of the lanes; Figure 2 is a plan view of the lanes and loop sensor of Figure 1; Figure 3 is a section view of the road surface of Figure 1; Figure 4 is a graph showing how the inductance of the loop of Figure l changes with time as a vehicle passes over the loop; Figure 5 is a graph showing how the inductance of the loop of Figure l changes with time as two vehicles pass over the loop; Figure 6 is a plan view of a modulated loop sensor in a roadway; Figure 7 is a graph showing how the inductance of the modulated loop of Figure 6 changes with time as a vehicle passes over the loop; Figure 8 is a graph showing how the inductance of the modulated loop of Figure 6 S changes with time as a vehicle passes onto the loop and stops; Figure 9 is a plan view showing three alternative configurations to the modulated loop of Figure 6; Figure 10 is a plan view of a further alternative modulated loop sensor in a roadway; and Figure 11 is a plan view showing three further alternative configurations to the modulated loop of Figure 10.

Figure 1 is a perspective view of a two lanes 111, 112 of a roadway 113. In this example, the roadway is a dual carriageway, and the outside lane 111 is a shoulder lane generally kept for emergency use only. The other lane shown 112 carries slower moving traffic in normal use. The roadway also comprises an overtaking lane (not shown), central reservation (not shown), and two lanes and a hard shoulder on the opposite carriageway (not shown) for traffic travelling in the opposite direction. The direction of travel of traffic along the carriageway 113 is shown by the arrow 114.

A loop sensor 115 is buried in the surface of the shoulder lane 111. The loop sensor 115 comprises a coil of wire or conducting material which extends 2-3 metros across the shoulder lane 111, and extends along the lane in the direction of travel of traffic. In one embodiment the loop stretches 40m in the direction of travel, but it will be appreciated that any large distance is possible subject to the constraints of obtaining a usable signal.

loop detector 116 is connected to the loop sensor 115 and comprises circuitry for providing an alternating current to the loop sensor l 15. Since the loop sensor includes a conducting coil, it forms a resonant circuit with a measurable inductance. The loop detector also comprises circuitry arranged to monitor the inductance of the loop sensor.

The theory of loop detection systems is described in detail in GB patent publication no. 2149952. While the detailed design of loop detectors may vary, this patent describes the method of operation and aspects of detection which apply to most detectors available today.

The constraints which apply to the length of the loop are dominated by the need to generate a relatively large signal from any vehicle which is to be detected. Whilst a common rectangular loop may generate a 4% signal swing with a passenger car, this falls to less than 0.5% for a motorcycle. When the length of the loop is extended, as under the present invention, this signal falls roughly in proportion to the length of the loop occupied by the vehicle in question. For example, a car, whose length is 5 metros, will typically give a 0.5% signal swing, instead of 4% previously, with a 40 metro long loop. Clearly this effect calls for a more sensitive detector than would normally be the case, by a factor of about 8 to 10.

Figure 2 shows a plan view of the two lanes ill, 112 of the roadway 113 shown in Figure 1. The loop sensor 115 is buried in the surface of the shoulder lane 111. Two cars 211, 212 are shown travelling along the usual slow lane for traffic. A third car 213 is shown stopped in the shoulder lane 111, over the loop sensor 115.

Figure 3 is a section taken along the line 3-3 of Figure 2, and shows the detection zone of the loop sensor 115. The alternating current passing through the loop wires 115 generates a magnetic field 311 which coincides with the car 213 above the loop sensor 115. The presence of the conducting metal in the chassis of the car 213 affects the inductance of the coil in the loop sensor 115, and this change in inductance compared to the situation when no car is above the loop 115 is detected by the loop detector 116.

Figure 4 shows the behaviour of the inductance of the loop sensor 115 with time as a single vehicle passes over the loop. For clarity, the behaviour of the reciprocal of the inductance (I/I) is shown on the graph.

With reference to Figure 1, as the vehicle crosses the leading edge 117 of the loop sensor, the inductance decreases sharply 410 (shown as an increase in 1/1 in Figure 4).

The inductance then remains relatively steady at a constant value 411, lower than the value when no car is present, as the vehicle proceeds along the length of the loop. As the vehicle crosses the trailing edge 118 of the loop sensor 115 the inductance of the loop rises sharply 412 to its original value (shown as a decrease in 1/I in Figure 4). It will be appreciated that if the vehicle stops while located over the loop sensor, as with the vehicle 213 shown in Figure 2, the inductance will remain at the lower steady state level 411. The detector can therefore determine that there is a vehicle somewhere in the zone of interest.

Figure 5 shows the inductance plot which is produced if a second vehicle moves above the loop sensor 115 while the first vehicle is still moving along above the loop 115.

When the first vehicle crosses the leading edge 117, 1/1 increases sharply 510 to a new steady state level. As the second vehicle crosses the leading edge 1/I sharply increases further 512. It will be noted from Figure 4 that the change in value 511 of 1/I which occurs 510 as the first vehicle crosses the leading edge 117 is smaller than the change in value 513 which occurs 512 as the second vehicle crosses the leading edge 117. This is caused by the fact that the two vehicles are not identical. The change in inductance is a function of the length of each vehicle divided by the height of its chassis above the roadway. In the example illustrated in Figure 5, the larger change in 1/1 513 caused by the second vehicle indicates that it has a lower chassis height or a longer length than the first vehicle.

When the first vehicle crosses the trailing edge l 18 to move off the loop sensor 115,1/1 decreases 514 by the same amount as the original change 511 when the first vehicle crossed the leading edge, before settling at a new steady state level 515 caused by the presence of the second vehicle only over the loop sensor 115. When the second vehicle crosses the trailing edge 118 to move off the loop 115, the inductance returns 516 to its starting point.

Figure 6 is a plan view of two lanes 611, 612 of a roadway 613, similar to the roadway 113 shown in Figures 1 to 3. An alternative loop sensor 615 is inserted into the surface of the shoulder lane 611 and connected to a loop detector 616. The loop sensor is "modulated" so that its width across the lane varies periodically along the length of the lane. In the example shown, the loop sensor has alternating sections 621, 622 having different widths 623, 624. The magnetic field over each section 621, 622 will therefore be different. As a vehicle passes over the loop 615, the interaction of the vehicle chassis with the different magnetic fields over the alternating sections 621, 622 will mean that the induction also changes as the vehicle passes over the sections 621, 622.

The behaviour of the inductance of the modulated loop sensor 615 as a vehicle passes thereover is shown in Figure 7. As the vehicle crosses the leading edge 618 of the loop sensor 615, the inductance decreases sharply 610 in a similar manner to that exhibited by the loop sensor 115 shown in Figures 1 to 3. Again, this decrease in inductance appears as an increase in 1/I in Figure 7.

As the vehicle moves along the loop sensor 615, it crosses the sections 621,622 having alternating widths 623, 624, resulting in a periodically varying inductance in the loop 615. The signal 1/I shown in Figure 7 thus has a high frequency component 713 superimposed on the plateau 711 in 1/I caused by the presence of the vehicle over the loop 615. The peaks 721 in the high frequency component occur as the vehicle passes over the wider sections 621 and the troughs occur as the vehicle passes over the narrower sections 622. As the vehicle crosses the trailing edge 618 to move off the loop the valtie of 1/l returns 712 to its original value.

The loop cletector 615 includes a high bandpass filter, which separates the high frequency signal 713 from the steady state "platcat'" 711. T he frequency ot the signal modulation is measured, and the velocity of the vehicle can be determined by dividing this frequency by the known separation of the alternating sections 621, 622 of the loop sensor 615. This method of determining vehicle speed is not as accurate as using two separate loops and measuring time of flight, but is nevertheless sufficiently accurate for use in determining the status of the lane.

Figure 8 shows the behaviour of the inductance if a vehicle moving along the loop sensor 615 shown in Figure 6 comes to a stop while still over the sensor. The signal begins in the same manner, with a sharp increase 710 in 1/I as the vehicle crosses the leading edge 617, followed by a high frequency component 713 superimposed on a plateau 711. However, when the vehicle stops the high frequency component 713 of the signal also stops. The signal 811 remains steady thereafter, at the level of the peaks 721 if the vehicle has stopped over one of the wider sections 621 of the loop sensor 615, or at the level of the troughs 722 of the vehicle has stopped over one of the narrow sections Is 622. The number of peaks 721 and troughs 722 in the preceding high frequency portion 713 indicate how many sections 621, 622 of the loop 615 have been traversed, and thus the position of the stationary vehicle.

Figure 9 shows a plan view of a lane 91 l having three alternative configurations (a), (b) and (c) of loop sensors 912, 913, 914 which operate in a similar manner to the loop sensor 615 of Figure 6. A vehicle traversing any of these loop sensors will give rise to an inductance signal comprising periodic peaks superimposed on a steady state change in inductance level. It will be appreciated that many more geometries are possible, and any geometry which gives rise to a change in induction of a loop as a vehicle moves along above that loop may be suitable for use in identifying the presence and speed of that vehicle, and discriminating between stationary and moving vehicles.

The loop sensors 615, 9l 1, 912, 913 are cllective at discriminating between moving and stationary vehicles, but provide no information about the direction of travel. It may be necessary to know not only that a vehicle is in a particular area, but also in which direction it is travelling. For example, a vehicle may partially break down, pull into the shoulder lane and come to a halt. The driver may then decide to reverse his vehicle to the nearest emergency telephone and/or ERA. A reversing vehicle would be an extreme hazard for other vehicles entering theshoulder lane, and it is therefore very important that this situation should be identified.

Figure 10 shows a lane 1011, like the lanes 111, 611 and 911, in which is installed a loop sensor 1015 having a geometry which enables the direction of travel of vehicles over the loop sensor 1015 to be established. The loop is connected to a loop detector 1016, like the loop detectors 116 and 616, for providing an alternating current to the loop sensor IO]S and detecting changes in inductance.

The loop sensor 1015 is formed in two sections 1021, 1022. The widest point of the loop 1015 is at its leading edge 1017. From there, the loop gradually narrows along the whole length of the first section 1021 as far as the middle 1023 of the loop. Here the loop widens in a step change back to the width of the leading edge 1017. Along the length of the second section 1022 the loop again narrows gradually, until at the trailing edge 1018 it is again the same width as at the middle 1023 if the loop 1015. In one example, the loop 1015 is 40m long, 3m wide at the leading edge 1017, narrowing to Im at the middle 1023, at which stage a step change of 2m restores the total width to 3m, and the loop again narrows through the second section 1022 to Im at the trailing edge 1018.

Consider a vehicle which crosses the leading edge 1017 of the loop 1015 of Figure 10 from the left and proceeds to the right. The detected signal (1/1) will initially increase sharply as the vehicle crosses the leading edge 1017, and then gradually decrease, in proportion to the width of the loop, as the vehicle moves along the first section 1021.

When the middle 1023 of the loop 1015 is reached, the signal will again increase sharply, before starting to decrease again gradually. As the vehicle leaves the loop the signal, already small, will fall back to the level indicating no vehicle above the loop.

The loop detector 1016 identifies this low frequency modulation, and determines from the shape of the detected signal that the direction of the vehicle is from left to right. Its speed may also be approximated from the rate of decrease of the inductance level.

If the vehicle stops before it reaches the half way point 1023, the signal will stop decreasing and remain constant at whatever level has been reached. The detection circuitry can then determine that the vehicle has stopped moving while located over the first section 1021 of the loop sensor 1015.

If the vehicle then starts to reverse in the lane, the signal will gradually start to increase, as it moves back towards the leading edge 1017 and the wider portions of the first section 1021. The loop detector 1016 will identify from the rising signal that a vehicle is reversing along the shoulder lane, and the "reversing vehicle" alarm will be raised. It is apparent that, wherever the vehicle is above the loop 1015, the loop detector will be able to identify from the form of the signal received in which direction a vehicle is travelling, or whether it is stationary.

It will be obvious to the skilled person that similar geometries will provide similar effects, with other attributes which may be of interest for certain applications and situations.

For example, Figure ll(a) shows a loop sensor 1112 in a lane 1111 in which the modulation can be effected in one single reach along the length of the loop 1112.

Figure I l(b) shows a loop sensor 1113 divided into three sections of modulation 1121, 1122, 1123. Figure Il(c) shows a loop 1114 in two sections 1124, 1125. The loop 1114 narrows along the first section 1124 from the leading edge 1117 to the mid-point 1119, and then widens along the second section 1125 to the trailing edge 1118. Such a loop sensor would require more precise processing to obtain directional information, since it would not he clear with a memory circuit element if the vehicle was in the first section 1124 going forward, or in the second section 1125 going in reverse.

It will be appreciated that variations from the above described embodiments may still fall within the scope of the invention. For example, all of the loop sensors shown in Figures 6, 9, 10 and 11 have geometries whose effect is to modulate the sensitivity of the loop sensor along the loop in the longitudinal direction. All of the loops shown achieve this by changing the width of the loop. It will be appreciated that the same effect can be achieved in a variety of other ways. For example, the depth of the loop could be varied. For example, the loop sensor 615 shown in Figure 6 could be replaced by a rectangular loop having alternating sections at different depths. The magnetic field at the height of a vehicle chassis will be less from a section of loop buried deeper in the road surface than from a section close to the surface. The deeply buried sections will thus have a lower sensitivity than the shallower portions.

This same principle can be applied by installing the loop vertically along the centre of the lane. In other words, the loops shown in plan view in Figures 6, 9, 10 and l l could lS be rotated through 90 so that the loop lies in a slot perpendicular to the road surface.

In a further alternative, the sensitivity of the loop in some sections may be altered by adding additional conductors to those sections. The different current paths will create different magnetic fields. For example, a mat of light weight concrete reinforcing mesh, having conductors with spacing varied in a controlled manner, could be placed above or below a homogeneous loop such as that shown in Figure 1. The sensitivity of the sensor and the shape of the field will be altered along the length of the loop, and this will cause a modulation in the signal output as the vehicle proceeds along the sensor. A variable resistance membrane or plastic carbon mesh may be used to achieve the same effect.

A variable sensitivity may be also be obtained using the concepts described in GB-A- 2385138 entitled "Validation and calibration of loop detection systems", which is incorporated herein by reference. This document describes the installation of secondary loops, or conducting paths, above a loop sensor. Switches and a control system are included so that the location of closed conducting paths above the loop sensor is controllable. In the document just rcicrred to, this enables the simulation of the movement of a vehicle above the loop sensor, enabling the response of the detection system to be assessed. A similar principle may be applied to the present invention, by providing adjustable conducting paths near the loop sensor. When a complete conducting loop is formed near a portion of the loop sensor, the sensitivity of that portion will be reduced. This enables the installation of a loop sensor whose spatial inhomogeneity in sensitivity may be controlled after installation of the loop. Such control need not be static, and may be varied either from site to site to accommodate local requirements or effects, or may be driven at a rate faster than the vehicle velocity, in order to provide a continuously adjustable sensitivity level. When coupled with a suitable detector circuitry, this can be adapted to provide a very precise sensitivity level with reference to particular vehicles and their longitudinal position along the loop.

Such conductive elements may be above or below the loop sensor. In such a case it may not be necessary to modify the width of the detector as shown in Figures 6 and 9 to ll. However, the conductors would need to be installed in the pavement during construction, and this would lead to an increase in complexity of installation.

The concepts of GB-A-2385138 have a further application to the present invention. In view of the application of the invention into safetycritical situations, the sensor may be enhanced by combination with the verification system described in GB-A-2385138.

Using this system, conductors placed above the loop sensor, possibly in the roadway, are connected in such a way as to simulate a vehicle passing over the loop sensor, which should cause a detection output. In this manner the fail-safe operation of any of the loop sensors can be verified easily, and in the event of a failure a repair to the sensor or a withdrawal of the automated system effected. These two inventions thus form a tightly coupled safe system element whose output can be confirmed.

If will be clear to practitioners that the invention as described is not limited to the field of road traf'l'ic detection, hut may also have application for the detection ol'taxiing or 3() parked aircraft as well as trains and other rail based vehicles.

Claims (47)

1. CLAIMS: 1. A vehicle detector for detecting vehicles on a roadway,

   comprising: a conductive loop for location in or on the roadway, having a dimension in the direction of travel of vehicles along the roadway of at least double the length of a single vehicle so that a vehicle above at least a portion of the loop will change the inductance of the loop; inductance detection means for determining the inductance of the conductive loop; and counting means arranged to count how many vehicles are in the detection area of roadway above the loop by comparing the inductance of the loop with the inductance when no vehicles are in the detection area, and dividing by a predetermined value corresponding to the typical change in inductance caused by a single vehicle in the detection area.
2. 2. A vehicle detector as claimed in claim l, arranged to identify when a vehicle enters or leaves the detection area of roadway above the loop by a sharp change in inductance.
3. 3. A vehicle detector as claimed in claim 2, arranged to count how many vehicles are in the detection area by counting the number of vehicles entering and leaving the detection area.
4. 4. A vehicle detector for detecting vehicles on a roadway, comprising: a conductive loop for location in or on the roadway, having a dimension in the direction of travel of vehicles along the roadway of at least double the length of a single vehicle, so that a vehicle above at least a portion of the loop will affect the inductance of the loop; and inductance detection means for determining the inductance of the conductive 3(:) loop; wherein the sensitivity of the inductance of the loop to a vehicle above a portion of the loop is non-homogeneous in the direction of travel of vehicles on the roadway.
5. 5. A vehicle detector as claimed in claim 4, wherein the inhomogeneity of the loop S sensitivity is arranged so that the inductance of the loop will vary with time as a vehicle moves along the roadway above the loop.
6. 6. A vehicle detector as claimed in claim 5, comprising speed determination means for determining the speed of a vehicle moving along the roadway above the loop from the rate of change of the inductance of the loop.
7. 7. A vehicle detector as claimed in claim 4, 5 or 6, wherein the width of the loop perpendicular to the direction of travel of vehicles on the roadway varies along the length of the loop in the direction of travel.
8. 8. A vehicle detector as claimed in claim 4, 5 or 6, wherein the depth of the loop beneath the surface of the roadway varies along its length in the direction of travel of vehicles along the roadway.
9. 9. A vehicle detector as claimed in claim 4, 5 or 6, further comprising additional conductors attached to sections of the loop for changing the magnetic field near those sections.
10. to. A vehicle detector as claimed in claim 4, 5 or 6, further comprising additional conductors located adjacent to sections of the loop for changing the sensitivity of the loop near those sections.
11. 11. A vehicle detector as claimed in claim 10, wherein the additional conductors include impedance variation means for varying the impedance of the additional () conductors.
12. 12. A vehicle detector as claimed in claim 11, wherein the impedance variation means includes a switch for completing a closed conducting path around some or all of the additional conductors.
13. 13. A vehicle detector as claimed in any of claims 4 to 13, wherein the loop comprises a plurality of alternating high sensitivity sections and low sensitivity sections, so that a vehicle located above a high sensitivity section will cause a greater change in the inductance of the loop than a vehicle located above a low sensitivity section.
14. 14. A vehicle detector as claimed in claim 13, wherein the width of the high sensitivity sections is greater than the width of the low sensitivity sections.
15. 15. A vehicle detector as claimed in any preceding claim, wherein the inductance response of at least a portion of the loop is not reflectively symmetrical in the direction of travel of vehicles, so that changes of inductance caused by a vehicle travelling forwards over the portion of the loop are distinguishable from those caused by a vehicle travelling backwards.
16. 16. A vehicle detector as claimed in claim 15, wherein the effect of a vehicle above at least a portion of the loop on the inductance thereof varies substantially linearly along that portion.
17. 17. A vehicle detector as claimed in claim 16, wherein the direction of travel of a vehicle over the portion of loop is determinable from whether the inductance of the loop increases or decreases as the vehicle moves over the portion.
18. 18. A vehicle detector as claimed in claim 16 or 17, wherein the ettect of a vehicle above the loop on the inductance thereof varies substantially linearly along the whole length of the loop.
19. 19. A vehicle detector as claimed in claim 16 or 17, wherein the loop is divided into two or more sections, within each of which the effect of a vehicle above that section on the inductance of the loop varies substantially linearly along the length of that section.
20. 20. A vehicle detector as claimed in any of claims 15 to 19, wherein the loop varies substantially linearly in width along at least a portion thereof.
21. 21. A vehicle detector as claimed in any preceding claim, wherein the loop is at least metros long in the direction of travel.
22. 22. A device for determining the status of a section of roadway, comprising a vehicle detector as claimed in any preceding claim, wherein the conductive loop has been installed along substantially the whole length of that section of roadway.
23. 23. A device as claimed in claim 22, arranged to determine the status of a lane of the roadway, wherein the loop has been installed along the length of that lane of roadway.
24. 24. A device as claimed in claim 23, wherein the lane is a shoulder lane.
25. 25. A traffic management system, comprising a roadway having a plurality of lanes, a device as claimed in claim 23 or 24 installed in a lane of the roadway for determining the status of that lane, and a plurality of electronic signposts located over that lane for indicating whether or not it may be used by running traffic.
26. 26. A method of determining whether a section of roadway longer than two vehicle lengths is occupied by vehicles, the method comprising measuring the inductance of a conductive loop installed in the surface of the roadway along substantially the whole length of the section, and determining from the measured inductance whether a vehicle is located above the loop. 3()
27. 27. A method as claimed in claim 26, wherein the section of roadway comprises a lane of a multi-lane road.
28. 28. A method as claimed in claim 26 or 27, including identifying from the measured inductance how many vehicles are located above the loop.
29. 29. A method as claimed in claim 26, 27 or 28, including identifying when vehicles enter or leave the section of roadway by sharp changes in the inductance of the loop.
30. 30. A method as claimed in claim 29, including identifying the number of vehicles in the section by counting the vehicles entering and leaving the section.
31. 31. A method of identifying that a stationary vehicle is located on a section of roadway, comprising determining the occupancy of the section using a method as claimed in any of claims 26 to 3(), and identifying that the inductance of the loop remains at a level indicating the presence of a vehicle for more than a predetermined period of time.
32. 32. A method as claimed in any of claims 26 to 30, wherein the response of the inductance of the loop to a vehicle located above a portion thereof is non-homogeneous in the direction of travel of vehicles along the roadway, so that the inductance varies with time as a vehicle moves along the section of roadway.
33. 33. A method as claimed in claim 32, wherein the loop includes alternating high sensitivity sections and low sensitivity sections, wherein a vehicle located above a high sensitivity section will cause a greater change in the inductance of the loop than a vehicle located above a low sensitivity section, so that as the vehicle moves along the loop the measured inductance has an alternating component.
34. 34. A methoti as claimed in claim 33, wherein the high sensitivity sections arc wider than the low sensitivity sections.
35. 35. A method as claimed in claim 33 or 34, including determining the speed of the vehicle from the frequency of the alternating component.
36. 36. A method as claimed in claim 32, wherein the inductance response of at least a section of the loop is not reflectively symmetrical in the direction of travel of vehicles, so that changes of inductance caused by a vehicle travelling forwards over the section of the loop are distinguishable from those caused by a vehicle travelling backwards.
37. 37. A method as claimed in claim 36, including identifying the direction of travel of a vehicle from the change in measured inductance.
38. 38. A method as claimed in claim 36 or 37, wherein the effect of a vehicle on the inductance of the loop is proportional for at least a section of the loop to the distance of the vehicle from the end of that section.
39. 39. A method as claimed in any of claims 29 to 35, wherein the width of the loop varies along its length for providing the inhomogeneity in the direction of travel.
40. 40. A method of identifying that a stationary vehicle is located on a section of roadway, comprising determining whether the section is occupied using a method as claimed in any of claims 32 to 39, identifying that the inductance of the loop is different to the inductance when no vehicle is located above the loop, and identifying that the inductance of the loop does not vary with time.
41. 41. A method of regulating traffic flow on a multi-lane roadway, comprising determining whether any stationary vehicles are present in an unused lane of the roadway using a method as claimed in claim 31 or 4(), and opening the unused lane to running traffic il no stationary vehicles are identified.
42. 42. A vehicle clctcctor as described herein with reference to Figures I to 5.
43. 43. A vehicle detector as described herein with reference to Figures 6 to 9.
44. 44. A vehicle detector as described herein with reference to Figures 10 and 11. s
45. 45. A method of determining the status of an area of highway as described herein with reference to Figures 1 to 5.
46. 46. A method of determining the status of an area of highway as described herein with reference to Figures 6 to 9.
47. 47. A method of determining the status of an area of highway as described herein with reference to Figures 10 and 11.