EP0277050B1 - Determination method for the trajectory of a body suitable to move about a road, and device using this method - Google Patents

Abstract

A method of determining the trajectory of a body (1) such as a vehicle body on a relatively plane path portion (2), the method being characterized in that it consists: in forming a main real image (4) of the path portion in a plane (19) at a non-zero angle (5) with the path portion; in decomposing said main image as formed into a plurality of points (21); in determining the relationship between the size of a unit length (30) taken substantially at the level of the path portion and the size of its image formed in the main image the size being a function of the number of points covered by the image and of the location of the unit length on said path portion; in determining a secondary image (32) in the main image, the secondary image corresponding to a longitudinal reference mark (31) related to the vehicle on the path portion; and in determining the various successive positions (32, 33, 34) of the secondary image by correlation of the number of points covered by the secondary image, given that the secondary image corresponds, according to the relationship, to a length which is constant on the path portion.

Description

The present invention relates to the methods for determining the trajectory of a body capable of moving over a portion of track, and more particularly to the methods making it possible to be able to determine the trajectory of vehicles of the automobile type on tracks such as roads, motorways, etc ..., over a relatively large distance and displacement surfaces of different shapes such as a rectilinear portion, two portions forming a crossroads, etc ... The present invention also relates to the devices making it possible to implement these methods.

Motor vehicle traffic has been increasing for a few years and this increase is not followed, in certain regions, by an adapted evolution of the road network. This means that, in certain circumstances, bottlenecks are produced which undoubtedly impair circulation. It was therefore thought that these drawbacks could be remedied by controlling the circulation of vehicles.

In order to be able to carry out these checks, it was necessary to produce sensors which can give an image of the movement of vehicles. Many sensors have been developed. For example, a sensor based on light rays directed towards the tracks traveled by vehicles has been developed. With these light beams generally returned by reflective surfaces arranged for this purpose on the roadways, are associated photosensitive receivers which deliver representative signals to their outputs, each time a vehicle cuts these light beams.

This technique gives good results. However, the signals delivered are only representative of the traffic at a given point and the sensors used are not flexible to use because they require elements to be placed on the roadway. It is therefore imposed on them a defined location and they cannot be moved without problems. In addition, the elements placed on the road require frequent interventions, if only for the cleaning of reflective surfaces.

Other sensors have been made to increase the surface of surveillance. Such is the case of a sensor constituted by a magnetic loop embedded in the roadway. This sensor overcomes some of the drawbacks mentioned above, but its usefulness is still too punctual and it is always linked to a specific location on the road.

Therefore, the object of the present invention is to implement a method for determining the trajectory of a body such as, for example, a motor vehicle, on a portion of track, which allows the control of a large track surface, which does not require any particular addition to the portion of track to be monitored, and which can give a plurality of results defining all the parameters of traffic, in particular of motor vehicles.

The present invention also aims to provide a device for implementing the method.

• à former une image réelle principale de ladite portion de voie, dans un plan formant un angle non nul avec celui de ladite portion de voie,
• à décomposer ladite image principale formée en une pluralité de points,
• à déterminer la relation entre la dimension d'une longueur unitaire prise sensiblement au niveau de ladite portion de voie et la dimension de son image formée dans ladite image principale, en fonction du nombre de points recouverts par ladite image et de l'emplacement de ladite longueur unitaire sur ladite portion de voie,
• à déterminer une image secondaire dans ladite image principale, cette image secondaire correspondant à un repère de longueur constante lié audit véhicule se trouvant sur ladite portion de voie,
• à déterminer les différentes positions successives de ladite image secondaire par corrélation du nombre de points recouverts par cette dite image secondaire, sachant que cette dite image secondaire correspond, suivant ladite relation, audit repère de longueur constante au niveau de ladite portion de voie.

More specifically, the subject of the present invention is a method for determining the trajectory of a body like a vehicle on a relatively flat portion of track, characterized in that it consists:

• forming a main real image of said portion of track, in a plane forming a non-zero angle with that of said portion of track,
• decomposing said main image formed into a plurality of points,
• determining the relationship between the dimension of a unit length taken substantially at the level of said track portion and the dimension of its image formed in said main image, as a function of the number of points covered by said image and of the location of said unit length on said portion of track,
• determining a secondary image in said main image, this secondary image corresponding to a reference of constant length linked to said vehicle located on said portion of track,
• to determine the different successive positions of said secondary image by correlation of the number of points covered by this said secondary image, knowing that said secondary image corresponds, according to said relationship, to said reference frame of constant length at said portion of track.

The present invention also relates to a device making it possible to implement said method.

• les Figures 1 à 6 représentent des schémas permettant d'expliciter la mise en oeuvre du procédé selon l'invention, et
• la Figure 7 représente un exemple d'un résultat obtenu avec la mise en oeuvre du procédé selon l'invention.

Other features and advantages of the present invention will appear in the course of the following description given with reference to the drawings annexed by way of illustration, but in no way limiting, in which:

• FIGS. 1 to 6 represent diagrams making it possible to explain the implementation of the method according to the invention, and
• Figure 7 shows an example of a result obtained with the implementation of the method according to the invention.

It is firstly specified that the set of Figures represents the same set of elements making it possible to explain an implementation of the method according to the invention. Consequently, the same references designate the same elements there, whatever the Figures on which they appear.

The method makes it possible to determine the trajectory of a body such as a motor vehicle 1 over a portion 2 of track 3 (Figure 1). Preferably, this portion of track is chosen so that its surface is substantially flat, regardless of its slope.

From this portion of track, an image 4 is produced in a second plane 19 which makes a non-zero angle 5 with that of the portion of track and such that the line of intersection 6 of these two planes is outside the portion of way. This image is advantageously produced by focusing means 7 such as, for example, a converging lens 8 arranged so that its optical axis 9 passes substantially through the center 10 of the track portion 2.

As, in general, the edges 11, 12 of the track portion are substantially parallel and as the surface of this track portion is chosen to be flat, the image 4 formed is a trapezoid 13 whose large base 14 corresponds to the line of delimitation 15 of the portion of track which is closest to the line of intersection 6 of the two planes, while the small base 16 corresponds to the other line 17 of transverse delimitation of the portion of track 2. These two lines of delimitation are, in fact, arbitrarily defined by the field of the lens 8 and also by the photosensitive surface 20 for receiving the image as defined below, in particular with regard to FIG. 2.

Indeed, the image 4 is received on a photosensitive receiving surface 20 decomposed into a plurality of photosensitive points 21, each being perfectly addressed in a repository 22.

Advantageously, for the sake of ease, in particular for carrying out the means for implementing the method, the reference system 22 is an orthonormal reference and the photosensitive points are distributed uniformly in lines 23 and in columns 24 such as, for example, in a frame hexagonal, to form a perfectly defined matrix. The number of points 21 will be as large as possible per unit of area of the photosensitive receiving surface 20.

The method then consists in determining at the level of the track portion 2 the dimension of a unit length 30 and of measuring the dimension of its image on the receiving surface in number of points on this surface. In general, a relationship is determined between the dimension of the unit length taken substantially at the level of the track portion and the dimension of its image formed in the main image, as a function of the number of points covered by this image. and the location of the unit length on the portion of track between the two delimitation lines 15, 17.

This relationship thus makes it possible to establish a one-to-one correspondence law between a secondary longitudinal image in the main image 4 and an actual length situated substantially at the level of the track portion.

As a practical example, it is assumed that the length of a bumper 31 of the vehicle 1 is defined. If the vehicle travels the portion of track 2 between the two limits 15 and 17, in the direction going from the limit 15 towards limit 17, it is obvious that the length of its bumper remains constant. On the other hand, its image 32 formed on the receiving surface will vary by a length corresponding to a large number of points 33 located on the large base 14 of the trapezoid, up to a lower number of points 34 located on the small base 16 This correspondence is established according to a law translated by a formula relating the value of the angle 5 equal to "a", the distance "h" from the optical center 35 of the lens 8 to the plane of the track portion 2, a length "L" taken at the level of the portion of track 2 (for example the length 30 represented in FIG. 1), "x" the length of its corresponding image in number of points on the plane of image 4, the distance " f "from the plane of the image 4 to the optical center 35 (that is to say substantially the focal distance of the lens 8), and the distance" d "representing the position of the length" L "on the channel by compared to a point of origin which is the projection of the optical center 35 on the plane of channel 2.

This law is of the form:

Under these conditions, let a marker linked to a vehicle and its image in the main image 4. It is then possible to determine the trajectory of the vehicle on the track portion by following the evolution of the image of the marker. The evolution of this image of the benchmark essentially results in a variation in length corresponding to a constant length at the level of the track, this variation being defined by the law given above, which makes the parameters of the vehicle trajectory known. to which the coordinate system whose image has been analyzed is linked.

As mentioned above, it is therefore necessary to know the length values of a secondary image in the main image. To facilitate the implementation of the method, the photosensitive definition points 21 of the image are given by the target 50 of a video camera 51 whose objective 52 is equivalent to the focusing lens 8 defined above. The target 50 is made up of photosensitive dots which technicians know by the name of "pixels" and which can be read very easily and very quickly by the video technique of line-by-line scanning, each pixel having a perfectly defined address in the orthonormal coordinate system. 22.

Thus, for each point or pixel, it is very easy to know electronically the information concerning its state of illumination and its address. For this, an organ 54 for acquiring and processing data, and for producing result signals is connected to the output 53 of the video camera. The organ 54 has its output 55 connected to the input 56 of a system for highlighting the results signals 57 such as, for example, a paper recorder or an afterglow screen, etc. The organ 54 is, for example, a processor specialized in processing.

As mentioned above, it is necessary to be able to assign to each vehicle passing on the track portion 2 a characteristic reference being linked to it, of a constant length and dissociating perfectly from the whole of the general view of the track. Figure 4 shows, by way of example, vehicles moving on the portion of track lit naturally by the sun, or else artificially. It can be seen that, in general, the portion of track 2 is of a relatively gray color, and that two types of contrast values appear when a vehicle moves on the track, and this regardless of the type lighting, this contrast however being more marked by natural lighting.

These two contrasting images are in fact the shadow 60 of a car projected onto the track and / or onto itself. While the car is traveling the entire portion of track 2, we can consider that the position of the sun and that the orientation of the sun's rays do not change, the distance traveled by the vehicle being of the order of a few tens of meters, or even few hundreds. In addition, this shadow, apart from very rare exceptions, has a length which can be considered as perfectly defined. Whether they are the shadows of cars or trucks, their length is around 1.5 to 2.5 meters.

On the other hand, the metal roofs 62 of the vehicles, although painted, have a power of reflection very much greater than that of the track portion, with the possible exception of continuous or discontinuous lines 61 painted on the roadways to delimit the tracks of circulation. However, these lines being very narrow, they can be discriminated by their width, as can small objects located on vehicles and producing parasitic reflections, such as for example mirrors.

It is thus very easy to spot the vehicles by taking into consideration the shade 60 that they project, or the lighter light emission that they emit, or a combination of the two phenomena.

FIG. 5 represents a curve 70 giving the quantity of light on a line 23 of pixels of the target 50 of a video camera 51. By way of example, two portions 71, 72 of lower light intensity are represented which correspond to shaded areas on the portion of track 2 whose image is formed on the target. However, the law defined above makes it possible to know the real length of an object at the level of the portion of track to which an image in the main image corresponds. From these portions 71, 72, it is therefore possible to eliminate those which cannot correspond to the transverse dimension of a motor vehicle mentioned above. In the example illustrated, the portion 71 must be rejected because it corresponds to an object whose dimension is not included in the predetermined range.

On the other hand, if the portion 72 has a length corresponding to an object dimension included in the predetermined range, it is very likely that it is representative of the shadow of a vehicle.

Of course, since the shadows of the vehicles have a certain height, a number of successive curves 70 will be taken into consideration. If the portion 72 is found in almost all of these curves, then it is almost certain that a vehicle has been spotted on the portion of track 2.

This portion 72 therefore corresponds to a benchmark linked to a vehicle and the analysis of the evolution of this portion 72 along the main image 4 makes it possible to determine the trajectory of the vehicle on the track portion 2.

The processing unit 54 generally comprising a clock, it makes it possible to date the position of the vehicle on the track portion.

The analysis of the recognition of a vehicle as a function of a dark marker has been described above. But one can also use a marking starting from clear zones. Figure 6 shows a curve 90 representing the amount of light on a line 23 of pixels of the photosensitive target of a video camera. By way of illustration, this curve comprises two portions 91, 92 corresponding to light-colored objects at the level of the track portion 2. These two portions could be used to discriminate objects whose dimensions are included within a certain range value, in the same way as described with reference to Figure 5. A clear portion, like the portion 92, if it corresponds to a length equivalent to a transverse dimension of a vehicle, may be used to determine the trajectory of a vehicle, in the same manner as described above.

Of course, the method can be implemented by using vehicle marking cumulatively in dark and light areas with respect to the roadway, in order to determine with more certainty the presence of a vehicle on the portion of track.

The analysis of the various images of the benchmarks linked to the vehicles can be carried out continuously but also sequentially, this latter way allowing the electronic circuits to be able to work out the signals representative of the results between each sequence, which makes it possible to produce a treatment 54 of a structure less complex than that necessary for continuous analysis.

• la trajectoire 82 est celle d'un véhicule qui a une vitesse constante entre les instants t1 et t7, du fait que la pente de cette trajectoire est constante.
• la trajectoire 85 est celle d'un véhicule ayant pénétré sur la portion de voie à l'instant t5.
• la trajectoire 84 est celle d'un véhicule qui a eu une vitesse constante entre les instants t1 et t5 et qui a accéléré après cet instant t5, du fait de l'augmentation de la pente de cette trajectoire.
• la trajectoire 88 est celle d'un véhicule qui était sur la portion de voie jusqu'à l'instant t3 et qui l'a quittée à cet instant pour doubler le véhicule devant lui qui a, par exemple, la trajectoire 83 montrant qu'il a ralenti à l'instant t3. Si deux portions de voie correspondant à deux bandes de circulation comme illustré sur la Figure 4 sont surveillées simultanément de la même façon, la trajectoire 88 apparaîtrait à l'instant t3 sur le graphique de l'autre portion de voie, en continuité de celle représentée sur la Figure 7.
• la trajectoire 86 est celle d'un véhicule qui s'est arrêté sur la voie entre les instants t3 et t4 du fait que son ordonnée est constante.
• la trajectoire 87 est celle d'un véhicule que est sorti de la portion de voie entre les instants t2 et t3.

By way of illustration, FIG. 7 represents the graphic results of a sequential analysis on a portion of track between an origin "0" and an end "Xm", for seven successive sequences from t1 to t7, as they would be viewed on the paper unrolling continuously from a graphic recorder 57. This graph shows, on the ordinate, the positions of the vehicles on the track portion and, on the abscissa, the dates of the sequences. Thus, at time t1, the track portion included six vehicles 80. This graph makes it possible to determine the different trajectories 81 of the vehicles on this track portion:

• the trajectory 82 is that of a vehicle which has a constant speed between the instants t1 and t7, owing to the fact that the slope of this trajectory is constant.
• the trajectory 85 is that of a vehicle having entered the portion of track at time t5.
• the trajectory 84 is that of a vehicle which has had a constant speed between the instants t1 and t5 and which has accelerated after this instant t5, due to the increase in the slope of this trajectory.
• the trajectory 88 is that of a vehicle which was on the portion of the track until the instant t3 and which left it at this instant to overtake the vehicle in front of it which has, for example, the trajectory 83 showing that it slowed down at time t3. If two portions of track corresponding to two lanes of traffic as illustrated in Figure 4 are simultaneously monitored in the same way, the trajectory 88 would appear at time t3 on the graph of the other portion of track, in continuity of that shown in Figure 7.
• the trajectory 86 is that of a vehicle which stopped on the track between the instants t3 and t4 because its ordinate is constant.
• the trajectory 87 is that of a vehicle which has left the portion of track between the instants t2 and t3.

In the preceding description, it can be seen that it is therefore possible to continuously monitor the traffic on a large portion of track, by determining a large number of parameters which are, in particular, the density, the instantaneous and average speed of the vehicles, the position of vehicles, their change of direction, without the need to install special elements on the road. The device for implementing the method essentially comprises a video type camera, for example black and white, positioned on a bridge or a pylon, the processing electronics having no particular bulk and being of an implementation. relatively easy for those skilled in the art of computer science.

In addition, with the method as described above, it is possible to analyze the traffic simultaneously on several portions of lanes, for example portions making non-zero angles between them such as highway ramps with highways, and even intersecting portions like, for example, crossroads.

Claims (14)

1. A method of determining the trajectory of a body (1) such as a vehicle on a relatively plane path portion (2), characterized by the fact that it consists :

   - in forming a main real image (4) of said path portion, in a plane (19) at a non-zero angle (5) with said path portion,

   - in decomposing said main image as formed into a plurality of points (21),

   - in determining the relationship between the size of a unit length (30) taken substantially at the level of said path portion and the size of its image formed in said main image, as a function of the number of points covered by said image and of the location of said unit length on said path portion,

   - in determining a secondary image (32) in said main image, said secondary image corresponding to a constant length reference mark (31) related to said vehicle on said path portion, and

   - in determining the various successive positions (32, 33, 34) of said secondary image by correlation of the number of points covered by said secondary image, given that said secondary image corresponds, according to said relationship, to said constant length reference mark on said path portion.
2. A method according to claim 1, characterized by the fact that the images are formed by focusing using a converging lens optical system (8).
3. A method according to claim 1, characterized by the fact that said plurality of points are photosensitive points (21).
4. A method according to claim 3, characterized by the fact that said plurality of points are distributed on a raster (23, 24) of a matrix defined relative to a frame of reference (22).
5. A method according to claim 4, characterized by the fact that said raster is hexagonal.
6. A method according to any one of claims 1 to 5, characterized by the fact that said secondary image (32) is determined from at least one reference mark (31, 60, 62) related to the vehicle, said reference mark being optically contrasted relative to said path.
7. A method according to claim 6, characterized by the fact that said reference mark is a dark reference mark (60).
8. A method according to claim 7, characterized by the fact that said dark reference mark is a portion of the shadow cast by said vehicle.
9. A method according to claim 6, characterized by the fact that said reference mark is a pale reference mark (62).
10. A method according to claim 9, characterized by the fact that said pale reference mark is given by light reflected from a portion of the bodywork of said vehicle.
11. A method according to claim 6, characterized by the fact that said reference mark is constituted both by a dark reference mark and by a pale reference mark.
12. A method according to any preceding claim, characterized by the fact that the various successive positions of said secondary image by correlation of the number of points covered by said secondary image are performed by comparing the number of said points with a range of point numbers determined as a function of the position of said secondary image (32) in said main image (4).
13. A method according to any preceding claim, characterized by the fact that the various successive positions of said secondary image by correlation of the number of points covered by said secondary image are performed by one of the following methods : continuous analysis, and sequential analysis.
14. Apparatus for implementing the method according to any preceding claim, characterized by the fact that it comprises a video camera (51) whose target (50) is defined by a plurality of pixels, a processor (54) for processing the signals delivered at the output (53) from said camera, and a recorder (57) whose input (56) is connected to the output (55) of said processor.

Images