FR2890477A1 - Vehicle e.g. motor truck, position determining method for detecting e.g. pedestrian, involves determining corresponding values of vehicle height with respect to ground, roll angle, and vehicle pitch angle for storing them in memory - Google Patents

Abstract

The method involves determining, by using a calculation unit (5), corresponding values of a height of a vehicle (2) with respect to the ground, roll angle and pitch angle of the vehicle, from an image of an observation zone (3), values of parameters stored in a memory and from data related to the shape of the ground at the level of the zone. The observation zone is illuminated by a light source (4) e.g. laser source, and the image is transmitted to the unit (5) by a charge coupled device (CCD) camera (1). The determined values are stored in the memory by the calculation unit.

Description

2890477 1

  METHOD AND DEVICE FOR EVALUATING THE POSITION OF A VEHICLE IN ITS ENVIRONMENT

DESCRIPTION TECHNICAL FIELD

  The technical field of the invention is that of devices and methods for identifying the position of a vehicle in an environment and the representation of this environment. The vehicle, such as for example an automobile, a truck, a construction machine, a vehicle mounted on rails or an unmanned vehicle, is able to move on a floor in an environment. The methods and devices concerned mainly use light projectors and video cameras to record images of the vehicle environment which are then transmitted to an image processing unit.

  STATE OF THE PRIOR ART The identification of the environment of a vehicle or that of the position of the vehicle in an environment is particularly useful in many situations, for example to detect obstacles and allow the vehicle to avoid them ( as in the case of parking assistance), to identify vehicle guidance areas or to locate the vehicle in a geographical environment (as in the case of vehicle route guidance).

  Patent DE 195 13 113 discloses a system for identifying the profile of a tunnel or a road where a vehicle is moving. The system is mounted on the vehicle, and includes a laser source, a video camera and an image processing unit. The image processing unit analyzes an image of an area scanned by the laser source, recorded and transmitted by the video camera, and determines the positions of the points of a profile of the area in a frame centered on the vehicle to be scanned. from a geometric transformation of the corresponding points of the recorded image. This profile tracking system does not, however, make it possible to evaluate the position of the vehicle relative to the ground, that is to say both the height of the vehicle and the inclination of the vehicle (the inclination of the plane of the frame for example) relative to the ground in the general case where these parameters are likely to vary (especially during the movement of the vehicle).

  Patent application WO 03/106924 discloses a device mounted on a vehicle for measuring the transverse profile of a road on which the vehicle is moving. The device comprises a laser source for vertically illuminating the road in a line transverse to the road and to the vehicle, a camera for recording an image of both the light line formed on the road and a reference line arranged on the road. the vehicle, and an image processing unit for detecting an angle between the line projected on the road and the reference line and deriving a transverse profile of the road. This device may also include a second laser source, disposed in a direction transverse to the vehicle and at a given distance from the first laser source, to form a second transverse light line on the road. The processing unit, from an image recorded by the camera comprising both the light lines and the reference line, determines an angle for each line and can thus calculate a longitudinal profile of the road between the two lines. transverse lights. In an attempt to improve this device, an accelerometer is also placed on the vehicle to measure an acceleration in the vertical laser plane. However, this device remains poorly adapted to the assessment of the position of the vehicle relative to the road because either it does not take into account the vertical movements of the chassis or it takes into account only via an acceleration measurement which then requires d perform a double time integration, a source of inaccuracy, to correct angles and profiles. In addition, since the vertical acceleration is only measured at one point, the device can not determine the inclination of the chassis relative to the ground. This device is also unsuited to determining the shape of the environment around and away from the vehicle as well as the detection of fixed or mobile obstacles (such as other vehicles or pedestrians for example).

  Much research is underway to improve vehicle safety by automatically processing data from sensors such as radars or cameras. This data processing aims at obtaining a three-dimensional representation of the environment around the vehicle in order to detect obstacles or dangers. The quality of the three-dimensional representation of the environment strongly depends on the accuracy and the calibration of the various sensors used. Unfortunately, as for movements of objects of the environment, acceleration, braking or cornering of the vehicle introduce changes in the position and orientation of the vehicle and, therefore, in the installation of sensors. In most data acquisition systems this problem is ignored by assuming that the motion of the vehicle does not introduce too much error into the three-dimensional representation of the environment. Moreover, in most of these systems, the world explored by the sensors is supposed to be flat to allow easier representation. However, these last two assumptions cause a degradation of the performance of these acquisition systems. Such assumptions may even prove to be dangerous in certain situations where good accuracy is required, for example to detect distant obstacles or very close obstacles (as in the case of parking the vehicle). Some of the devices used for the detection of nearby objects, such as ultrasonic sensors, besides not giving a precise three-dimensional representation, can not provide measurements at short distances (less than 20 or 30 cm).

  SUMMARY OF THE INVENTION The object of the present invention is therefore to overcome the drawbacks mentioned above, in particular as regards the determination of the position parameters of a vehicle relative to the ground and within an environment, or the detection of fixed or moving obstacles with respect to the vehicle.

  More precisely, one of the objectives of the present invention is to obtain a method, and a corresponding device, for determining the position of a vehicle in an environment.

  This objective is achieved thanks to a first general embodiment of a method according to the invention, which is a method for determining the position of a vehicle in an environment, for a vehicle able to move on a floor and equipped a first camera, mounted on the vehicle at a given and calibrated location (that is to say the values of the internal and external parameters have been determined), capable of recording an image of a first zone of observation located on the ground near the vehicle; a first light source capable of projecting light rays onto the first observation zone in order to illuminate it in a non-uniform manner, and mounted on the vehicle at a given location so that the directions of projection of the rays on the first observation zone do not coincide with the optical axis of the first camera recording an image the first observation zone; and - a calculation unit having a memory, and in which the first camera is able to transmit an image recorded to the computing unit, the computing unit is able to receive and process an image transmitted by the first camera, the memory is capable of storing values of the calibration and position parameters with respect to the vehicle of the first camera as well as position parameters with respect to the vehicle and projection of the first light source, said method being characterized in that, from an image of the first viewing area illuminated by the first light source recorded and transmitted to the computing unit by the first camera, values of the parameters stored in the memory, and from data relating to the shape of the ground at the level of the first observation zone stored in the memory, the calculation unit determines corresponding values from the top h of the vehicle relative to the ground, the roll angle 8 and the pitch angle 9 of the vehicle in a grounded R mark and stores these values in the memory.

  This first general embodiment is advantageous because all measurements are made optically, without contact with the ground.

  The objective is also achieved by means of a first corresponding device according to the invention, which is a general device for determining the position of a vehicle in an environment, the vehicle being able to move on a ground, the device capable of being embedded in the vehicle comprising: a first camera adapted to be mounted on the vehicle at a given and calibrated location, and capable of recording an image of a first observation area located on the ground close to the vehicle; a first light source capable of projecting light rays onto the first observation zone to illuminate it in a non-uniform manner, and able to be mounted on the vehicle at a given location so that the projection directions of the rays on the first observation zone do not coincide with the optical axis of the first camera recording an image of the reference points of the first observation zone; and a computing unit having a memory, wherein the first camera is adapted to transmit a recorded image to the computing unit; the computing unit is adapted to receive and process an image transmitted by the first camera; the memory is capable of storing values of the calibration and position parameters with respect to the vehicle of the first camera as well as position parameters with respect to the vehicle and projection of the first light source, said device being characterized in that , from an image of the first observation area illuminated by the first light source recorded and transmitted to the calculation unit by the first camera, values of the parameters stored in the memory, and from relative data. at the shape of the ground at the level of the first observation zone stored in the memory, the calculation unit is able to determine corresponding values of the height h of the vehicle relative to the ground, the roll angle A and the pitch angle cp of the vehicle in a ground-related locator R and storing these values in the memory.

  Thus, the first general embodiment makes it possible to determine the position parameters of the vehicle (h, e, cf) with great precision because the image processing of the first zone compensates for the distortion effects of the camera at through the use of distortion parameters. Indeed, for this image taken by the first camera can be exploited validly, the first camera is previously calibrated, that is to say that a correspondence is established between spatial positions (three-dimensional) of reference points of the first observation area, used as a calibration area, and two-dimensional positions of the corresponding control points on an image of this calibration area recorded in the camera image plane, and this correspondence takes into account the various distortions generated by the optical and / or recording media of the camera. The computing unit then uses the values obtained from these calibration parameters, stored in memory, to correct various distortion effects an image transmitted by the first camera.

  The first camera and the first light source are also arranged relative to each other so as to provide a stereoscopic view of the first viewing area, that is to say, so that the directions projection of the rays on the first observation zone do not coincide with the optical axis of the first camera when the latter records an image the first observation zone. This relative arrangement makes it possible always to be able to determine the position in space of a reference point from the corresponding control point (by of course using the values of the calibration parameters to correct the various distortion effects). The data of the values of the positions of the first camera and the first light source with respect to the vehicle as well as the projection parameters (the orientation of the projected rays) of the first light source, makes it possible to know the position and the direction the optical axis of the first camera as well as the relative position of the source and the camera. Thus, the rotation matrices between, respectively, a spatial reference linked to the first light source and a spatial reference linked to the vehicle, and between a spatial reference linked to the first camera (at its optical center, at its optical axis and its image plane for example) and the vehicle-related spatial repository are known, as are the spatial coordinates (in the vehicle-related reference frame) of a point of the camera and a point of the source . A two-dimensional repository linked to the image plane of the first camera is also known from these data. The computing unit can therefore make necessary changes of reference, taking into account the distortions, to go from a control point to a reference point.

  Finally, to remove any indeterminacy, the geometric shape of the soil at the level of the first observation zone is given. This form can also be approximated, for example by a plane. This plane can even be assumed horizontal (in an initial step of an iterative process of processing recorded images for example).

  Known methods, for example triangulation, then enable a control point in a two-dimensional reference frame of the image to be associated in a one-to-one way with a reference point in a ground-based three-dimensional coordinate system, and thus to calculate values of vehicle position parameters: height h relative to the ground, roll angle A (around any given direction of the vehicle), pitch angle p. More specifically, a second particular embodiment of the general mode according to the invention, in which the first light source is further capable of projecting a matrix of light spots on the first observation zone, comprises the following steps: (I) forming a matrix of at least three non-aligned light reference points on the first viewing area by means of the first light source; (II) recording, by means of the first camera, an image of the luminous reference points formed on the first observation zone in the image plane of the camera and transmitting the recorded image to the computing unit, the recorded image comprising control points and each control point corresponding to the image of a luminous reference point of the first observation zone; (III) determining by means of the calculation unit the values of the coordinates in the image plane of the first camera of at least three non-aligned control points of the image recorded by the first camera and storing the values determined in the first camera; memory of the computing unit; and (IV) from the determined values of the coordinates of the control points in the image plane of the first camera and the values of the calibration and position parameters with respect to the vehicle of the first camera as well as position parameters with respect to the vehicle and projection of the first light source, and from the data relating to the shape of the ground at the first observation zone, determine by means of the calculation unit the values of the spatial coordinates, in the reference R, reference points corresponding to the control points and deduce values of the height h of the vehicle relative to the ground, the roll angle 0 and the pitch angle cf) of the vehicle in this frame.

  The advantage provided by the luminous reference points is to allow a good localization of the corresponding control points on the recorded image of these luminous points, particularly with respect to the methods which require the determination of contours on the image which are particularly sensitive. to the illumination of the observation area. The method according to the invention is on the contrary very robust with respect to these conditions of illumination of the observation zone.

  In this second particular embodiment of the first general mode corresponds a second particular embodiment of the first general device, wherein the first light source is capable of projecting a matrix of at least three non-aligned light reference points on the first observation area; the first camera is able to record an image of luminous reference points formed on the first observation zone by the first light source in the image plane of the camera and transmit the recorded image to the computing unit, the image recorded having control points and each control point corresponding to the image of a light reference point of the first observation area; the calculation unit is capable of determining coordinate values in the image plane of the first camera of at least three non-aligned control points of an image recorded by the first camera and storing the determined values in the memory of the first camera; the unit of calculation; and from values stored in control point coordinate memory in the image plane of the first camera, calibration parameters and position relative to the vehicle of the first camera, position parameters with respect to the vehicle and projection of the first light source, and from data relating to the shape of the ground at the level of the first observation zone stored in memory, the computing unit is able to determine the values of the spatial coordinates, in a reference frame R ground points corresponding to the control points and to derive corresponding values of the height h of the vehicle relative to the ground, the roll angle θ and the pitch angle ρ of the vehicle in this mark.

  Another object of the invention is to allow the values of the position parameters h, 0, p corresponding to a given instant to be supplemented by two-dimensional position and orientation values of the vehicle on the portion of the ground on which it is located. at a corresponding moment. These last values make it possible to specify where the vehicle is located in the environment in a global manner. The two-dimensional position of the vehicle at ground level (in a plane locally parallel to the ground) makes it possible for example to locate the vehicle on a local (two-dimensional) map of the environment. From the measurement of the two-dimensional position of two distinct points of the vehicle, it is possible to obtain the orientation of a common axis at these two points, from which the two-dimensional orientation of the vehicle at the ground level is deduced. . The knowledge of this orientation then makes it possible to know that of the camera (that is to say of its optical axis), because the location of the camera on the vehicle is known. The images recorded by the camera, and the associated reference points, can then be placed in a more global context of the vehicle environment. There are many means, i.e., position sensors, for obtaining these two-dimensional position and orientation values of the vehicle at ground level, examples of which are position sensors using the GPS system (Global Positioning System), odometers, inertial units etc ...

  This object is achieved with a third embodiment of the invention, which is a variant of the previous second mode, wherein the vehicle is further equipped with a position sensor capable of determining a two-dimensional position and orientation of the vehicle. at ground level at a given instant and to transmit the data relating to this position and this orientation to the computing unit, the memory of the computing unit is able to store the position and orientation data of the vehicle 15 determined by the position sensor, the first camera, the first light source and the computing unit are able to operate synchronously with the position sensor by means of synchronization means, and step (IV) of determining the values of the height h of the vehicle relative to the ground, the roll angle 9 and the pitch angle W of the vehicle by the calculation unit and the determination of the values of the position and the orientation of the vehicle at ground level by the position sensor are synchronized by the synchronization means, the determined values of the position and the orientation of the vehicle being stored in the memory of the unit of calculation.

  This variant makes it possible to obtain a complete set of values making it possible to characterize the situation of the vehicle in its environment.

  Of course, it corresponds to this variant a third embodiment of the device according to the invention, which is a variant of the device according to the second embodiment, further comprising: a position sensor adapted to be mounted on the vehicle, to determining a two-dimensional position and orientation of the vehicle at ground level at a given moment and transmitting the data relating to this position and this orientation to the computing unit, the computing unit being able to store in the memory of the data position and orientation of the vehicle transmitted by the position sensor; and synchronization means adapted to synchronize the operation of the first camera, the first light source, the calculation unit and the position sensor, in which the synchronization means is able to synchronize the operation of determining the values of the height h of the vehicle relative to the ground, the roll angle e and the pitch angle cp of the vehicle by the calculation unit with the operation of determining the values of the position and the orientation of the vehicle at ground level by the position sensor.

  Another object of the invention is to make it possible to explore, using the first camera and a second light source, a second observation zone distinct from the first zone, for example further from the vehicle, for determining the spatial coordinates (in a reference linked to the ground) of the reference points of this second observation zone.

  This objective is achieved with a fourth embodiment of the method according to the invention, which is a variant of the third preceding mode, wherein the vehicle is further equipped with a second light source capable of projecting a matrix towards the ground. light reference points on a second observation zone distinct from the first observation zone, and the first camera is furthermore capable of recording an image of luminous reference points of the second observation zone, the second source of light being mounted on the vehicle at a given location so that the projection directions of the reference points on the second viewing area do not coincide with the optical axis of the first camera recording an image of the reference points of the second observation zone, the values of the position parameters with respect to the vehicle and the projection of the second source of light t stored in the memory of the computing unit, the first camera, the first and the second light source and the computing unit being able to operate synchronously with the position sensor by means of synchronization means, method further comprising the steps of synchronously with the position sensor: (a) forming a matrix of light reference points 30 on the second viewing area by means of the second light source; (b) recording, by means of the first camera, an image of the luminous reference points formed on the second observation zone in the image plane of the camera and transmitting the recorded image to the computing unit, the recorded image comprising control points and each control point corresponding to the image of a luminous reference point of the second observation zone; (c) determining by means of the computing unit the coordinate values in the image plane of the first image control point camera of the second viewing area recorded by the first camera and storing the determined values in the memory of the computing unit; and (d) from the determined values of the coordinates of the control points corresponding to the second observation area in the image plane of the first camera, values of the calibration and position parameters with respect to the vehicle of the first camera as well as that position parameters with respect to the vehicle and projection of the second light source, and from the values of the height h of the vehicle relative to the ground, the roll angle 8 and the pitch angle 9 of the vehicle calculated in step (IV) and data relating to the position and orientation of the vehicle, calculating by means of the calculation unit the values of the coordinates of the reference points of the second corresponding observation zone. to the control points whose coordinates were calculated in step (c) and store them in the memory of the computing unit.

  A fourth embodiment of the device according to the invention is associated with this method, which is a variant of the device according to the third embodiment, in which a second light source is able to project towards the ground a matrix of reference points. illuminated on a second observation zone distinct from the first observation zone; the first camera is furthermore capable of recording an image of luminous reference points formed on the second observation zone in the image plane of the camera and transmitting the recorded image to the computing unit, the recorded image comprising control points and each control point corresponding to the image of a luminous reference point of the second observation zone; the second light source is adapted to be mounted on the vehicle at a given location so that the directions of projection of the reference points on the second observation area do not coincide with the optical axis of the first recording camera an image of the reference points of this second observation zone; the memory of the computing unit is able to store position parameters with respect to the vehicle and projection of the second light source; the synchronizing means is operable synchronously with the position sensor for the first camera, the first and the second light source and the computing unit; and the calculation unit is able to: determine coordinate values in the image plane of the first camera of control points of an image of the second observation zone recorded by the first camera and store the values determined in the memory of the computing unit; and from values stored in control point coordinate memory corresponding to the second observation area in the image plane of the first camera, values of calibration parameters and position relative to the vehicle of the first camera as well as positional parameters with respect to the vehicle and projection of the second light source, and from values of the height h of the vehicle relative to the ground, the roll angle O and the pitch angle p of the vehicle and data relating to the position and orientation of the vehicle, to calculate values of the coordinates of the reference points of the second observation area corresponding to the control points and to store them in the memory of the control unit. calculation.

  Once the coordinates of the reference points of the second observation zone have been obtained, it is possible to use these data to determine the geometrical shape of the second observation zone. Many known methods make it possible to obtain a representation of the shape of this zone: for example using meshes or lines drawn by interpolation from reference points, by level lines, etc. The invention makes it possible to obtain a representation of the shape of this zone. then to represent, or to model, the environment corresponding to this second observation zone. Thus, in a fifth embodiment of the method according to the invention, variant of the fourth mode above, an additional step (e) consists in determining by means of the calculation unit () the geometric shape of the second observation area, from a set of reference point coordinate values calculated in step (d) at a given instant, and storing in the memory of the calculation unit () the data relating to this form. A fifth embodiment of the device according to the invention is associated with this fifth embodiment of the method according to the invention, which is a variant of the device according to the fourth embodiment, in which the calculation unit is suitable for perform step (e) above.

  It is possible to use the information represented by the geometrical shape calculated in step (e) during a displacement of the vehicle. Indeed, if the movement of the vehicle is such that over time the first observation zone covers a portion of the ground that was part of the second observation zone at a time when the geometric shape of the second zone of observation has been determined, then this geometrical form can be used as the shape of the ground at the first observation zone in step (IV) of the computation of h, 0 and p (in the second embodiment of the invention). For example, if the first and second viewing areas are at the rear of the vehicle, the second area being farther away from the vehicle than the first area, then if the vehicle backs off, the above-mentioned case of overlap can occur. produce. Similarly, if the first and second viewing areas are at the front of the vehicle, the second area being farther away from the vehicle than the first area, then if the vehicle is moving forward, the recovery case mentioned above may happen. Thus, accurate zone shape data stored in memory can be used to improve the accuracy of calculation of the values of h, 8 and (p.

  More precisely, such a use of a previously calculated geometric shape concerns a sixth embodiment of the method according to the invention, which is a variant of the fifth preceding mode in which, the second observation zone being further away from the vehicle than the first observation zone and the computing unit being further able to evaluate a displacement of the vehicle from position and orientation data of the vehicle transmitted by the position sensor, if the calculation unit determines that the vehicle moves so that, at a given instant, the first observation area covers a portion of the ground that has been part of the second observation zone at a previous time when said geometric shape of the second zone of observation has been determined, then this geometric shape is used as the shape of the ground at the level of the first observation zone at the step (IV) of calculating values of the height h of the vehicle relative to the ground, the roll angle 8 and the pitch angle c of the vehicle corresponding to the given instant.

  A sixth embodiment of the device according to the invention, variant of the device according to the fifth embodiment, corresponds to the method according to the sixth previous embodiment and has a calculation unit capable of implementing the steps of this method. Consequently, in this sixth embodiment, the second observation zone being further away from the vehicle than the first observation zone, the computing unit is able to evaluate a displacement of the vehicle from position data and from orientation of the vehicle transmitted by the position sensor and stored in memory, and, if the computing unit determines that the vehicle is moving so that, at a given moment, the first observation area covers a portion of the which has been part of the second observation zone at a previous time when the geometrical shape of said second observation zone has been determined, then the calculation unit uses this geometric shape stored in memory as the shape of the ground at the of the first observation zone for calculating the values of the height h of the vehicle relative to the ground, the roll angle e and the pitch angle p of the vehicle corresponding to the insta given.

  In the previous embodiments, the observation zone (s) are located in the field of view of the first camera. However, when the vehicle is required to move in different directions or to move in opposite directions of the same direction (for example if the vehicle moves forward and then backwards), it is advantageous to be able to observe a third zone (distinct from the first one). or the second observation area). For this purpose, a third light source may be mounted on the vehicle to project a matrix of light reference points on the third observation zone and a second camera to be able to record an image of this third observation zone in order to allow the computing unit to evaluate the coordinates of light reference points of this area. The third zone may for example be anywhere around the vehicle (front, rear or on one side), however it will be advantageous, in the case where the first camera is mounted at the rear of the vehicle. vehicle (respectively at the front) that this third observation area is located at the front of the vehicle (respectively at the rear). This arrangement will be particularly useful for vehicle parking operations or obstacle detection for example.

  Therefore, a seventh embodiment of the invention, variant of the method according to the third embodiment, is a method in which the vehicle is further equipped with: a third light source capable of projecting towards the ground a matrix of luminous reference points on a third observation zone distinct from the first observation zone; a second camera, mounted on the vehicle in a given and calibrated location, capable of recording an image of luminous reference points of the third observation zone, and in which the computing unit is furthermore able to receive and processing an image recorded by the second camera, the third light source is mounted on the vehicle at a given location so that the directions of projection of the reference points on the third observation area do not coincide with the optical axis of the second camera recording an image of the reference points of this third observation zone, the values of the position parameters relative to the vehicle and of the projection of the third light source being stored in the memory of the unit of calculation, the third light source, the second camera and the computing unit being able to operate in a synchronized manner by means of synchronous means wherein said method further comprises the steps of: synchronously with the position sensor: (a1) forming a matrix of light reference points on the third viewing area by means of the third light source; (bl) recording by means of the second camera an image of the luminous reference points formed on the third observation zone in the image plane of the camera and transmitting the recorded image to the computing unit, the recorded image comprising control points and each control point corresponding to the image of a luminous reference point of the third observation zone; (cl) determining by means of the calculation unit the values of the coordinates in the image plane of the second camera of the image control points of the third observation zone recorded by the second camera and storing the values determined in the memory of the computing unit; and (d1) from the determined values of the coordinates of the control points corresponding to the third observation zone in the image plane of the second camera, values of the calibration and position parameters with respect to the vehicle of the first camera as well as that position parameters with respect to the vehicle and projection of the third light source, and from the values of the height h of the vehicle relative to the ground, the roll angle e and the pitch angle p of the vehicle calculated in step (IV) and data relating to the position and the orientation of the vehicle, calculating by means of the calculation unit the values of the coordinates of the reference points of the third corresponding observation zone to the control points whose coordinates were calculated in step (c1) and store them in the memory of the calculation unit.

  Of course, the above method may further comprise the following step of: (el) determining by means of the computing unit the geometric shape of the third observation area, from a set of coordinate values reference points calculated in step (dl) at a given instant, and store in the memory of the computing unit the data relating to this form. In addition, the third observation zone being farther from the vehicle than the first observation zone and the computing unit being further able to evaluate a displacement of the vehicle from vehicle position and orientation data transmitted. by the position sensor, if the computing unit determines that the vehicle is moving so that, at a given moment, the first observation area covers a portion of the ground that has been part of the third zone of observation at a previous time when said geometrical shape of the third zone has been determined, then this geometrical shape is used as the shape of the ground at the level of the first observation zone in the step (IV) of calculating the values of the height h of the vehicle relative to the ground, the roll angle e and the pitch angle 9 of the vehicle corresponding to the given instant. Thus, the observation of the third zone can improve the accuracy of the calculation of the parameters h, 0 and p. It is clear that other complementary observation zones can be considered, in any variant of an embodiment using a second observation zone, as well as complementary cameras associated to observe these zones and complementary complementary light sources. to illuminate these areas (by projecting to the ground dots of light points). All this in order to further improve the possibilities of representing the surroundings of the vehicle or to improve the accuracy of the calculation of the position parameters of the vehicle relative to the ground.

  In the case of a complementary observation zone, this corresponds to a method according to the invention in which the vehicle is furthermore equipped with a complementary source of light able to project towards the ground a matrix of reference points of light on an associated complementary observation zone, of a complementary camera, mounted on the vehicle at a given and calibrated location, capable of recording an image of luminous reference points of the complementary observation zone, and in which the complementary source of light is mounted on the vehicle at a given location so that the projection directions of the reference points on the associated complementary observation area do not coincide with the optical axis of the complementary camera recording an image of the reference points of this complementary observation zone, the values of the calibration and position parameters with respect to the vehicle the complementary camera as well as position parameters with respect to the vehicle and projection of the complementary source of light being stored in the memory of the computing unit, the complementary source of light, the complementary camera and the unit with the synchronization means, said method further comprises the following steps, synchronously with the position sensor: (a2) forming a matrix of reference points illuminated on the complementary observation area by means of the complementary source of light; (b2) recording, by means of the complementary camera, an image of the luminous reference points formed on the complementary observation zone in the image plane of the camera and transmitting the recorded image to the computing unit, the recorded image comprising control points and each control point corresponding to the image of a luminous reference point of the complementary observation zone; (c2) determining, by means of the calculation unit, the values of the coordinates in the image plane of the first camera of image control points of the complementary observation zone recorded by the complementary camera and storing the values determined in the memory of the computing unit; and (d2) from the determined values of the coordinates of the control points corresponding to the complementary observation zone in the image plane of the complementary camera, the values of the calibration and position parameters relative to the vehicle of the complementary camera as well as only positional parameters relative to the vehicle and projection of the complementary source of light, and from the values of the height h of the vehicle relative to the ground, the roll angle A and the pitch angle g of the vehicle calculated in step (IV) and data relating to the position and orientation of the vehicle, calculating by means of the calculation unit the values of the coordinates of the reference points of the complementary observation zone corresponding to the control points whose coordinates were calculated in step (c2) and store them in the memory of the calculation unit.

  In order to exploit the acquired data, for example to model the vehicle environment, the above method may further comprise the following step of: (e2) determining by means of the computing unit the geometric shape of the area complementary observation, from a set of reference point coordinate values calculated in step (d2) at a given instant, and store in the memory of the calculation unit the data relating to this form.

  It is clear that the role of the complementary camera can be held by the first camera (when the complementary observation zone can be in the field of view of the latter). The first camera can also, for example, be movable on an axis so as to observe the third zone in a given position. In addition, the complementary source of light may also be the first or second light source. The role of the complementary camera can also be held by the second camera (when the complementary observation zone can be in the field of view of the latter), and in this case the complementary light source can also be the third source of light. Note that it is possible that in the previous modes the second light source is the first light source.

  In order to improve the quality of image recording by providing good contrast, it is advantageous in any of the preceding embodiments that at least one light source is a laser light source, or else to ensure the safety of the eyes of persons likely to be in the vehicle environment while enjoying a good contrast, that at least one light source is a source of pulsed light, and that a means of synchronization pulsation is able to synchronize each pulsed light source with the corresponding camera capable of recording an image of the observation area illuminated by the pulsed light source.

  The invention also relates to a method according to any one of the preceding modes implementing the recording of an image of the second observation zone, in which, if an object is in one of the observation zones illuminated by one of the light sources, the computing unit detects the presence of the object and the movement of the object relative to the vehicle from the processing of at least two images of this illuminated observation area by the light source and received consecutively in time, and stores in the memory the information relating to the detected motion and the location of the detected object. This embodiment makes it possible to detect the presence of obstacles in relative movement with respect to the vehicle and makes it possible, for example, to avoid a collision. This is especially important if the obstacle is a pedestrian. In the case where a pulsed light source is used to illuminate the observation zone in question, the value of the frequency of pulsation (possibly adjustable) and the time (possibly adjustable) separating two consecutive shots by the camera concerned determine the detectable maximum velocity of an object in motion relative to the vehicle.

  In another embodiment, variant of any one of the preceding modes using a position sensor, the computing unit calculates a graphical representation of the vehicle environment from the position data stored in the memory of the vehicle. calculation unit relating to the vehicle and reference points of at least one observation zone. This variant makes it possible to model the environment of the vehicle as observed from the different observation zones. In addition, in a complementary variant intended to provide a navigation aid, the computing unit associates a position of the vehicle stored in memory with a corresponding location on a map of the vehicle environment stored in memory, and stores in memory the result of this association.

  To ensure efficient control of the vehicle and its movement in the environment, an embodiment of the invention, variant of any of the preceding modes, is a method in which the computing unit formats data. stored in memory and transmits this formatted data to a display means for displaying a representation of these data. This allows an operator looking at the display means, on board the vehicle or outside (the display data is then transmitted by radio for example), to have access to the various data obtained by the calculation unit according to various representations. The operator can then view the position of the vehicle in the environment, possibly the situation of the vehicle on a map or the presence of obstacles (fixed or moving). Based on this information, the operator can guide the vehicle more safely.

  Of course, to each of the preceding embodiments of the method for determining the position of a vehicle in an environment, and its various variants, corresponds to a device according to the invention able to implement it, as indicated in the appended claims.

  Finally, the invention also relates to a vehicle equipped with any one of the devices for determining the position of a vehicle in an environment mentioned above.

  Other features and advantages of the invention will emerge from the following description of particular embodiments, given by way of example with reference to the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

  FIG. 1 is a simplified representation, corresponding to a view from above, of a first camera mounted at the rear of a vehicle, of a first light source for projecting a matrix of light spots on a first observation area.

  Figure 2 is a side view corresponding to Figure 1.

  Figure 3 is a simplified representation, corresponding to a top view, of an embodiment implementing a camera, mounted at the rear of the vehicle, with two light sources for illuminating two observation areas.

  Figure 4 is a side view corresponding to Figure 3.

  FIG. 5 is a simplified representation, corresponding to a view from above, of an embodiment implementing two cameras, one at the front of the vehicle and the other at the rear, with three light sources for illuminate three areas of observation.

  Figure 6 is a side view corresponding to Figure 5.

  Figure 7 is a simplified representation, corresponding to a top view, of an embodiment implementing a camera, at the rear, with four light sources for illuminating four observation zones.

  FIG. 8 is a simplified representation, corresponding to a view from above, of an embodiment implementing four cameras, one at the front of the vehicle, another at the rear and one on each side of the vehicle, with eleven light sources to illuminate eleven areas of observation.

  In the various figures, the identical elements have the same references.

  DETAILED DESCRIPTION OF A PARTICULAR EMBODIMENT A particular embodiment of the invention, described with reference to FIGS. 1 and 2, relates to a particular device of the second embodiment of the invention mentioned above comprising a first camera (1). ) CCD record mounted on the back of a vehicle (2) (here an automobile) for recording images of a first viewing area (3), located at the rear of the vehicle and in close proximity to that -this. Also mounted at the rear of the vehicle (2) the first light source (4), here a laser source, for projecting a matrix of light points to the first observation area (3). A calculation unit (5) is embedded in the vehicle (2) and makes it possible to process images received from the camera (1).

  The vehicle is located on a road, and it will be assumed that the shape of the ground at the first observation area is flat and even corresponds to a substantially horizontal plane. To simplify the presentation of the calculations and without restricting the generality, we suppose that this horizontal plane has the equation zp = 0 in a cartesian coordinate system linked to the camera (1). Thus, if Xp designates a point of the observation zone (3) in the coordinate system linked to the first light source (4), then the homogeneous coordinates of this point are Xp = [xp, yp, zp = 0, 1 ] T. The laser source (4) forms N (N> - 3) luminous reference points Pi (i = 1, ..., N) on the ground in the first observation zone (3) according to the step (I) of the method. The reference points Pi here correspond to the nodes of a light grid formed by the laser source (4) on the zone (3). The first camera (1) then records an image of the first observation zone (3) with the points Pi according to step (II). The coordinates of the points Pi are presented in the form of homogeneous coordinates for a matrix formulation of the problem: Pi = [Pix, Piy, Piz, 1] T, i = 1, ..., N, where the first three coordinates are the coordinates of the point Pi in a cartesian coordinate system R linked to the ground and 15 axes x, y and z. The camera (1) CCD records in its image plane the control points corresponding to the reference points according to step (II). Let Ci (i = 1, ..., N) denote the N control points on the recorded image that correspond to the N reference points Pi of the observation zone (3). These control points Ci, once corrected for the distortion effect by the calculation unit (5) (according to one of the many known methods, see for example the article by F.Devernay and O.Faugeras, Machine Vision and Applications (13), 14-24 2001), give points Di (i = 1, ..., N): we then have Di = F (Ci), i = 1, ..., N, where F denotes a distortion function formed with the distortion parameters of the camera (1) stored in the memory of the computing unit (5) In this embodiment, as in the others, the computing unit (5) ) can also implement a thresholding method when analyzing an image of the observation area received from the camera. This thresholding makes it possible to improve the accuracy of the determination by the calculation unit of the coordinates of the control points Ci detected on the image, and consequently also of the points Di associated with these points Ci during the distortion correction according to in step (III) of the method used.

  Let Rc denote the cartesian coordinate system of axes u, v of the image plane of the camera (1) (this image plane is perpendicular to the optical axis of the camera). The intrinsic calibration parameters of the first camera (1) are here four in number: a, the horizontal focal distance of the camera, b, the vertical focal distance of the camera, c, the horizontal coordinate of the optical center, and d , the vertical coordinate of the optical center. a represents the product of the length of the pixel by the focal length f of the camera; b represents the product of the pixel width by the focal distance f of the camera.

  The coordinates c and d in the reference Rc (of axes u, v) of the image plane of the camera (recording plane of the camera) represent the projection of the optical center along the optical axis in the image plane. These four parameters constitute the elements of a 0 c 0 matrix II

  (3x4): II = 0 bd 0 0 0 1 0 In the following, we designate the known matrix of rotation (3x3) of a reference linked to the laser source (4) relative to a reference linked to the vehicle by RL / v = [aij] (where the respective row and column indices i and j vary from 1 to 3), the known rotation matrix (3x3) of a reference linked to the camera (1) with respect to a reference mark linked to the vehicle by Rc / v = [bij] (where i, j = 1,2,3), and the rotation matrix (3x3) of a reference linked to the vehicle relative to the reference R linked to the ground by Rv / o = [vi, j] (where i, j = 1,2,3). The matrices RL / V and Rc / v are known because the locations of the camera (1) and the laser source (4), as well as the associated references, are known. The matrices RL / V and Rc / v are stored in the memory of the computing unit (5).

  In general, the coefficients vij of the rotation matrix Rv / o are expressed as a function of the angles of roll 0, pitch p of the vehicle (2) and a rotation angle 4 about a perpendicular axis by relative to the plane of the chassis of the vehicle (2), as follows: v1 1 = cosp cos * '1,2 = cosp sin * v1 3 = -sincp v2,1 = sine sincp cos * - cose sin * V2,2 = sine sincp sin * + cose cos * v2,3 = sine cosp v3,1 = cose sincp cos * + sine sin * V3,2 = cose sincp sin * - sine cos * v3,3 = cose cosp It is well known that if Ryix designates the rotation matrix of a Y mark relative to an X mark, then the Py / x passing matrix from the X mark to the Y mark can be written in the form (if any A vector expresses itself as a column vector A. in the X mark and as a column vector AY in the Y mark, then we have AY = PY / x AX): PY / x = [Râ'r (RY'1YX) 1, and the matrix of reverse passage.

  = (Py / x) -1 _ [(Ri1, XY where (RY / x) 'denotes 0 1 the transposed matrix of Ry / x.

  The respective coordinates of the origins of the marks of the laser source (4) and of the camera (1) in the vehicle mark (2), that is to say VL = [11, 12, 13] T and VC = [cl, c2, c3] T, are known and stored in the memory of the calculation unit (5).

  0V = [dx, dy, dZ] T denotes the coordinates of the origin of the vehicle reference mark in the reference frame R. Let D be a point (corrected for the distortion effects) corresponding to a reference point P of the zone d observation (3), whose homogeneous coordinates in the reference frame Rc are k. [u, v, 1] T, where k denotes the depth on the optical axis between the points P and D (ku and kv thus designating the coordinates corrected for the distortion of the camera (1), u, v designating the coordinates of the control point C corresponding to p determined by the calculation unit according to step (III)), then the equation (El) connecting the coordinates of D calculated by the calculation unit (5) to those of the corresponding reference point P (of the observation zone (3)) whose coordinates are expressed in the reference Px / Y bound to the light source ( 4) (see point XP above) is (with CL = VL VC): a 0 c 0- (El): D = II. Pc / v E (PL / v) -1. XP = 0 bd 0 0 0 1 0 Then, from equation (El) we extract: (E2): fxP-f (u, v) where f and g are known functions of y ,, = g ( u, v) their arguments, and from (E2) one arrives at: CRov (RL, v) 'RcXLI 0 1 J x ,, Op (E3): P = (Pv / o) 1. (PL / v ) 1. XP = C (Rvio) δ (Ruv) '(OV + (RPio) VL) 1 where P represents the reference point corresponding to the points D and Xi, in the reference frame R. It remains then to determine the values of the height h of the vehicle relative to the ground, the roll angle θ and the pitch angle ρ of the vehicle in the reference R. For this, it is observed that the equation (E3), which expresses the existing relation between corresponding points in the reference R and in the reference of the light source (4), can be written: Xi, where R = [rij], (i, j = 1, 2, 3), is a matrix P = (3x3) defined by: ri = aLk.vk, i (with the Einstein convention on indices), and where t denotes the vector OV + Rv / o. VL (of coordinates tX, ty, tZ). The matrix R is stored in the memory of the computing unit (5). So, (E3) is it equivalent to P x = rI, .xP + rl2.yp + tx (E4): P, = r2,! .XP + r2,2.yp + tY P = = r3.I.xP + r3,2.yp + tz To simplify the presentation of the computations (without restricting the generality of the method) one can suppose that the origin of the altitudes in the reference frame R corresponds to the coast z = 0. Moreover, because of the assumption of flatness of the observation zone made above (in a more general case it will suffice to give the equation representing the local form of this zone), we obtain that (E4) is equivalent to (E5): r3,1.xp + r3,2.yp + tZ = 0, which is also written V1,3 (a1,1Xp + a2, lyp + 11) + V2,3 (a1,2Xp + a2 , 2yp + 12) + V3, 3 (a1,3) Cp + a2,3yp + 13) + dz = 0, or else (E6): -m sincp + n sine cosp + p cose cosp = -dz with m = a1,1xp + a2,1yp + l1 n = al, 2xp + a2, 2yp + 12 p = a1, 3xp + a2, 3yp + 13 The values of xp and yp do not depend on the movement of the vehicle (2) (because Pv / o does not appear in (El)) and therefore m, n and p have constant values for a given point in the recorded image. In order to calculate the values of p, 0 and dz, at least three non-aligned points in the image recorded by the camera (1) are required: let D1, D2 and D3 be those three points (associated with the three control points C1, C2 and C3 detected by the calculation unit (5) on the image) whose corrected coordinates of the distortion effects are stored in the memory (these points therefore correspond to three points Xp1, Xp2 and Xp3 of the observation zone ), then from (E6), considered for each of these points, we obtain a system (E7) of three equations with three unknowns p, 0 and dZ: (E7): -m1 sincp + (ni sine + pi cose ) cosp = - dz 2890477 41 -m2 sincp + (n2 sine + p2 cose) coscp = - dz -m3 sincp + (n3 sine + p3 cose) cos = - dZ from which we obtain the result (E8): pie = [(m2-ml) (P3-pl) - (m3-m1) (P2-P1) [(m3-m1) (n2-n1) - (m2-ml) (n3-n1)] tan = (n2-n1) sine + (p2-pl) cose] / [m2-ml] dZ = misincp - (nlsine + p1cose) cosp where dZ is the height h of the vehicle the relative to the ground. The sought values of the angles 0 and cp are obtained by applying the inverse function Arc tan to the corresponding right-hand members in (E8). Thus, the calculation unit (5) is capable, according to step (IV) of the method, of calculating values of h, 0 and cp from at least three points D1, D2 and D3.

  In the general case, instead of assuming that the first viewing area (3) is flat and horizontal, it can be assumed that it has any given shape. For example, a point Xp of this area will belong to a given equation surface in parametric or explicit form. The equation of the surface (in the reference linked to the light source) then makes it possible to express one of the coordinates of the point as a function of the other two, for example zp = a (xp, yp). It is then enough to replace in (El) the vector xn by the vector nz ,, = a (xP7 yp) in (E3) the vector xp yP 0 1 by x ,, the vector (where xp = f (u, v) , yp = g (u, v) and zp = a (f (u, v), Y, zg (u, v))), then (E4) becomes P x = rt. I.xP + r] .2.yP + rt, 3.zP + t. (E4) ': PY = r2.1.xP + r2,2.yP + rz, 3.ZP + ty Pr. = R3.I.xP + r3.2.y, + r3,3.zp + tz But as the point P is on the observation zone whose surface equation is known (in the reference of the laser source it is zp = a (xp, yp)), there exists a known relation (corresponding to the equation of the area in the coordinate system R) between the coordinates of P, for example Pz = (3 (Px, Py) Therefore, the equation (E5) becomes a (E5) ': r3 1. xp + r3 2. yp + r3 3. a (xp, yp) + tz = 3 (Px, Py), where we have.

  xp = f (u, v), yp = g (u, v), Px = r1 1.xp + r1, 2E Yp + r1, 3Ea (xp, Yp) + tx, and Py = r2,1 .xp + r2,2EYp + r2,3Ea (xp, Yp) + ty In the same way as previously, the data of three points D1, D2 and D3 (and thus of the associated coordinates u, v) determined by the unit of calculation (5) makes it possible to form a system of three equations with three unknowns and to deduce from them values of h, A and T. Another particular embodiment of the invention, described with reference to FIGS. particular device of the fifth embodiment of the invention mentioned above comprising (the elements common with the previous particular mode have the same references) a vehicle (2) equipped with a first camera (1) mounted at the rear of the vehicle for recording images of a first viewing area (3) at the rear of the vehicle, a first laser light source (4) mounted at the rear of the vehicle for projecting reference light Pi (i = 1, ..., N) on the zone (3), a calculation unit (5) which makes it possible to process the images received from the camera (1), a second laser light source (6) mounted at the rear of the vehicle for projecting luminous reference points Qi (i = 1, ..., M) onto a second observation area (7) located at the rear of the vehicle; vehicle and more outside of it than the first zone (3). This second light source (6) is mounted on the vehicle (2) at a given location so that the directions of projection of the reference points on the second observation zone (7) do not coincide with the axis optical of the first camera (1) when it records an image of the reference points of the second observation area (7). The vehicle (2) is also equipped with a position sensor (8) able to determine a two-dimensional position and orientation of the vehicle at ground level at a given moment and to transmit the data relating to this position and this orientation to the vehicle. computing unit (5), and synchronizing means (9) for synchronously operating with the position sensor (8) the first camera (1), the first (4) and the second (6) light source as well as the computing unit (5). In this particular mode, the position sensor (8) is of the GPS type and makes it possible to record the positions of two points of the vehicle, located respectively at the front and rear ends of the vehicle (2), to determine a position and a position. two-dimensional orientation of the vehicle (2) at ground level at a given moment and transmit the data relating to this position and this orientation to the computing unit (5). This position sensor (8) makes it possible to know the direction of a longitudinal axis of the vehicle frame (2) (at the ground on which the vehicle is located) passing through the two points at the ends of the vehicle mentioned here. -above.

  Calculation by the calculation unit (5) of the values of h, 0 and cp from an image of the first observation zone (3) recorded by the camera (1) is carried out according to the same steps as those presented above for the first particular embodiment of the invention.

  To determine the values of the coordinates of each reference point Q of the second observation zone (7) in the ground-related reference frame R (see above with respect to the first particular embodiment of the invention with respect to the ground, camera, and first light source markers as well as rotation, passage and calibration matrices of the camera), the calculation unit (5) uses a method different from that seen above for determine the values of the coordinates of a reference point P of the first zone (3). Indeed, the values of h, 0 and cp have already been calculated (see above) from an image of the first observation area, and the position and orientation data of the vehicle (2) from the position sensor (8) directly calculates a corresponding value of the angle i.

  The known rotation matrix (3 × 3) of a reference linked to the second laser source (6) with respect to the reference linked to the vehicle by RL2 / v = [a'1] (where the respective row and column indices) is designated i and j vary from 1 to 3), this matrix is stored in memory. The coordinates of the origin of the reference of the second laser source (6) in the reference of the vehicle (2) are VL2_ [l ', 2r 1' 3] T, they are known and stored in the memory of the computing unit (5). If Xq designates a reference point of the observation zone (7) in the coordinate system linked to the second light source (6) (in the reference frame R this point is designated by Q, see step (a) of the fourth embodiment of the method according to the invention), then the homogeneous coordinates of this point are Xq = [xq, yq, zq, 1] T.

  At this reference point corresponds a control point Cq of the image of the zone (7) recorded by the camera (1) whose coordinates u, v in the reference Rc of the image plane are calculated by the calculation unit ( 5) (see steps (b) and (c) of the fourth embodiment of the method according to the invention), as well as a point Dq of homogeneous coordinates k. [U, v, l] T obtained by the calculation unit during distortion correction. The equation connecting Dq and Xq is = II. PC / V (PL2 / v) -1Eqq = 0c0-0-0 0 0 0 1 0 RC / V (RL2, V) 'RC / VCL, 0 1 Dq Xq Yq Zq PL2 / V is the matrix of passage of the mark of the vehicle where to the mark of the second laser source (6). This equation makes it possible to express Xq as a function of Dq. To obtain the corresponding point Q, it suffices then to apply the known relation Xq (R, o) '(Ri2') '(OV + (Rvo)' VL) 0 1 Yq Zq Q = (Pv / o) -1. (PL2 / v (where Q represents the reference point corresponding to the points Dq and Xq, in the frame R) The previous relation 10 can be written: R 't' Xq where R '= [r'iA], ( i, j = 1,2,3), is a 0 1 matrix (3x3) defined by: r'iA = a'i, k.vk, i (with the Einstein convention on indices), and where t 'denotes the vector OV + Rv / o VL, (of coordinates t'X, t'y, t'Z).

  vl l = coscp cos *, v1,2 = coscp sing, v1,3 = -since v2,1 = sine since cos * - cose sing V2,2 = sine sincp sing + cose cos *, v2,3 sine cozy v3, 1 = cos () sincp cos * + sine sing v3,2 = cose since sing - sine cos *, v3,3 cose coscp Here, the angles e and cp are known, the angle * (angle of rotation around a axis perpendicular to the plane of the chassis of the vehicle (2)) is also known, indeed

Q

  The position sensor (8) measured the position and the two-dimensional orientation (in the plane (x, y) of the reference R connected to the ground) of the vehicle (2), which makes it possible to deduce the value of the angle Therefore, all values vk, i are known.

  The matrix R 'is stored in the memory of the computing unit (5). This relation is equivalent to Qx = r '.. xq + r'.2.yq + r'.3.zq + t'x y = r2.I.xq + r2.2. yq + r2.3.zq + t Y Qz - r'3.,. xq + r'32. yq + r'3.3. zq + t 'z Thus, according to the fourth embodiment of the method according to the invention, step (d), the calculation unit (5) determines the value of the coordinates of the reference point Q of the second zone of observation (7). This operation is repeated for a set of reference points Qi (i = 1, ..., M). In addition, according to the fifth embodiment of the method according to the invention, step (e), the computing unit (5) determines the geometrical shape of the observation zone (7), for example by analytically interpolating ( linear or quadratic interpolation etc ...) between the reference points Qi (i = 1, ..., M) whose coordinates have been calculated by the calculation unit (5) as previously indicated. Thus there is a representation of the area of the observation area (7) which is stored and which can easily be represented on a screen by means of graphic software.

  Another particular embodiment of the device for determining the position of the vehicle (2) in the environment is illustrated in FIGS. 2890477 48 and 6. The device according to the fourth embodiment of the device according to the invention mentioned above. , shown mounted on the vehicle (2), comprises: - a first camera (1), here a CCD camera mounted at the rear of the vehicle and calibrated, to record images of a first (3) and a second (7) an observation area on each of which, respectively, a first laser light source (4) and a second laser light source (6) mounted at the rear of the vehicle can form a matrix of light reference points; a calculation unit (5), a position sensor (8) and an on-board synchronization means (9) (see the particular embodiment above), - a third light source (12), here a laser source mounted on the front of the vehicle (2), e t able to project towards the ground a matrix of luminous reference points on a third observation zone (11) (at the front of the vehicle and therefore distinct from the first observation zone (3) and the second zone d observation (7)), and - a second camera (10), here a CCD camera mounted at the front of the vehicle at a given and calibrated location, and capable of recording an image of light reference points of the third zone d observation (11) and transmitting a recorded image to the computing unit (5), wherein the computing unit (5) is further adapted to receive and process an image recorded by the second camera (10). ), the third light source (12) is mounted on the vehicle (2) at a given location so that the directions of projection of the reference points on the third observation area (11) do not coincide with the optical axis of the second camera (10) recording an image of the points of r reference of this third observation zone, the values of the position parameters with respect to the vehicle of the third light source (12) being stored in the memory of the computing unit (5), the third light source ( 12), the second camera (10) and the computing unit (5) being also able to operate synchronously by means of synchronization means (9), said device is adapted to synchronously with the position sensor ( 8): - forming a matrix of light reference points on the third observation zone (11) by means of the third light source (12); and - recording by means of the second camera (10) an image of the light reference points formed on the third observation zone (11) in the image plane of the camera and transmitting the recorded image to the computing unit ( 5), the recorded image having control points and each control point corresponding to the image of a light reference point of the third viewing area (11); and - determining by means of the calculation unit (5) the values of the coordinates in the image plane of the second camera (10) of control points of the image of the third observation zone recorded by the second camera ( 10) and storing the determined values in the memory of the computing unit (5); and - from the determined values of the coordinates of the control points corresponding to the third observation zone (11) in the image plane of the second camera (10), values of the calibration and position parameters relative to the vehicle of the first camera (1) as well as positional parameters with respect to the vehicle and projection of the third light source (12), and from the values of the height h of the vehicle relative to the ground, the angle of roll and the pitch angle cf) of the vehicle and data relating to the position and orientation of the vehicle, calculating by means of the calculation unit (5) the values of the coordinates of the reference points of the vehicle. third observation zone (11) corresponding to the control points whose coordinates have been calculated and store them in the memory of the calculation unit (5).

  In this particular embodiment, the calculation of the values of the coordinates of the reference points of the third observation zone (11), from the control points of the image of this zone recorded by the camera (10), is analogously to what has been seen above in the above particular embodiment (fourth mode) for computing the coordinates of the reference points of the second observation area (7). Indeed, in addition to the quantities already known (see the previous particular embodiment) and stored, the positions of the laser source (12) and the camera (10) are known; so we know too.

  the rotation matrix (3 × 3) of a reference linked to the third laser source (12) relative to the reference linked to the vehicle RL3 / v = [a '',] (where the respective row and column indices i and j vary from 1 to 3), this matrix is stored in memory; the coordinates of the origin of the reference of the third laser source (12) in the reference of the vehicle (2) are VL3 = [1 "1.1" 2.1 "3] T, they are known and stored in the memory of the computing unit (5), the coordinates of the origin of the camera mark (10) in the vehicle mark (2), that is to say VC2 = [c '1, c' 2r c: '3] T, which are stored in the memory of the computing unit (5) - the calibration matrix II' of the second camera (10), II '= a' 0 c '0- 0 b 'd' 0 0 0 1 0 - the components (t ', t "y, t" Z) of the vector t' '= 0V + Rv / o VL3, - the matrix R' '= [r "i, j], (i, j = 1, 2, 3), is a matrix (3x3) defined by: r "i, j = a" k.vk i (with the Einstein convention on indices). The matrix R "is stored in the memory of the computing unit (5), the vk, i being known because the angles 0, cp and 4 are known (see above the previous mode).

  Therefore, if A designates a reference point of the zone (11) in the ground-related reference frame R, and Xa that point in the reference of the laser source (12) (of homogeneous coordinates [xa, Ya, Za. ] T) L if Da designates the control point (corrected for the distortion by the computation unit) of the image recorded by the camera (10), it is similar to what has been seen above (mode PREFERRED EMBODIMENTS OF THE PREFERRED EMBODIMENTS OF THE PREFERRED EMBODIMENTS OF THE PREFERRED EMBODIMENTS OF THE PREFERRED EMBODIMENTS OF THE PREFERRED EMBODIMENT OF PC2 / V (PL3 / V). is the matrix of passage of the reference mark of the vehicle with the mark of the second camera (10), PL3 / v is the matrix of passage of the mark of the vehicle with the mark of the third laser source (121 (RL3 / v is the matrix rotation of the reference of the laser source (12) with respect to the reference of the vehicle, RC2 / v is the rotation matrix of the reference of the second camera (10) with respect to the reference of the vehicle, and C2L3 = VL3 -VC2). 'express Xa according to Da. To obtain the corresponding point A, it suffices then to apply the known relation xa 0 0 1 0 Y. za A = (Pv / o) 1. (PL3 / V) 1.Xa = [(Rv lo) '(RL3 (OV + (Ri, / 0) 'VL) (where A represents the reference point corresponding to the points Da and Xa, in the reference R). This relation is also written R "t" - A = from which one obtains the values Xa 0 1 sought: = r..x. + r, .z.ya + r.3.za + t x y = r z..xa + r z.2.y + r, 3. za + t Y A = r "3.1.xa + r'4z.ya + r" 33.ze + t "= This calculation is performed for each control point detected by the calculation unit (5) on the image of the area (11) (illuminated by the source (12)) recorded by the camera (10), and the coordinates obtained from the reference points of the area (11) are stored in the memory of the calculation unit. similarly, the computing unit may compute reference points formed on any observation area (using a suitable light source) from the control points of an image of that area recorded by a calibrated camera mounted on the vehicle at a specific location.These data can then be used to model the environment (three-dimensional representation, profile in a given direction, cut according to a given plane, etc.) of the vehicle at the zone level. concerned.

  The local modeling obtained can then be exploited in several ways: graphical representation on a screen (using a graphic software), shape detection, alert in case of distance from the vehicle to a detected object less than a minimum distance of security.

  The data from the various reference points detected on the images or from the position sensor are indexed by time in the case of devices having a means of synchronization, therefore the analysis of these data over time also makes it possible to detect and to estimate the displacement of an object in the environment of the vehicle (for example to allow the vehicle to avoid an obstacle). All these applications can be implemented on the computing unit.

  Other particular embodiments of the device according to the invention are illustrated in FIGS. 7 to 8 on which the elements already described with reference to the preceding figures are only mentioned.

  FIG. 7 illustrates a device according to the invention mounted on a vehicle (2) and comprising a first camera (1) at the rear of the vehicle, a first light source (4) for forming reference points on a first zone at the rear of the vehicle and in close proximity thereto, a second light source (6) for forming reference points on a second observation zone (7) at the rear of the vehicle. vehicle (farther from the vehicle) in the field of the first camera (1), an observation zone (13) illuminated by a light source (14) and an observation zone (15) illuminated by a light source (16) on which reference points can be formed, these two areas (13) and (15) being in the field of the first camera (1) at the rear of the vehicle and on the sides, a calculation unit ( 5), a position sensor (8) and a synchronizing means (9) (for synchronizing the operation of the camera, sensor and light sources). This device makes it possible to model the environment of the vehicle situated at the rear of the vehicle so that, for example, it is possible to detect the presence of the vehicles (27) and (28) on the sides of the vehicle (2) or an object (29). ) at the rear of the vehicle (2).

  FIG. 8 illustrates a device according to the invention mounted on a vehicle (2), more complete than that of FIG. 7, this device takes again that of FIG. 7 and completes it with: a camera (10) mounted at the front for recording images of the observation zones (11), (17) and (19) located at the front of the vehicle (these areas being respectively illuminated by the laser sources (12), (18) and (20)), a camera (25) and a camera (26) on each side of the vehicle for respectively recording images of the viewing areas (21) and (23) located on the sides of the vehicle (2). Light sources (22) and (24), on the sides of the vehicle, are able to form luminous reference points respectively on the observation zones (21) and (23). The computing unit (5) is able to process the images from the four cameras. The synchronization means (9) is able to synchronize the operation of the four cameras, the position sensor (8) and the eleven light sources.

  This device makes it possible to model the environment surrounding the vehicle so as to provide enhanced security.

  It is clear that many other related provisions of or cameras and / or light sources, for illuminating one or more viewing areas in the field of or cameras, are possible while remaining within the scope of the invention. 'invention.

2890477 57

Claims (47)

  1. Method for determining the position of a vehicle (2) in an environment in which the vehicle capable of traveling on a floor is equipped with: a first camera (1) mounted on the vehicle at a given location and calibrated, capable of recording an image of a first observation zone (3) located on the ground close to the vehicle; a first source (4) of light capable of projecting light rays onto the first observation zone (3) to illuminate it in a non-uniform manner, and mounted on the vehicle (2) at a given location so as to that the ray projection directions on the first observation zone (3) do not coincide with the optical axis of the first camera (1) recording an image the first observation zone; and a computing unit (5) having a memory, and in which the first camera (1) is able to transmit a recorded image to the computing unit (5), the computing unit (5) is suitable to receive and process an image transmitted by the first camera (1), the memory is able to store values of the calibration and position parameters relative to the vehicle of the first camera (1) as well as position parameters relative to to the vehicle and projection of the first light source (4), said method being characterized in that, from an image of the first observation area illuminated by the first light source recorded and transmitted to the unit calculation by the first camera, values of the parameters stored in the memory, and from data relating to the shape of the ground at the level of the first observation zone stored in the memory, the calculation unit (5) determines corresponding values of the height h of the vehicle relative to the ground, the roll angle e and the pitch angle of the vehicle in a ground-related reference R and stores these values in the memory.   The method according to claim 1, wherein the first light source (4) is further adapted to project a matrix of light spots onto the first viewing area (3), and comprising the following steps: ) forming a matrix of at least three non-aligned light reference points on the first viewing area (3) by means of the first light source (4); (II) recording by means of the first camera (1) an image of the luminous reference points formed on the first observation zone (3) in the image plane of the camera and transmitting the recorded image to the computing unit ( 5), the recorded image having control points and each control point corresponding to the image of a light reference point of the first viewing area (3); (III) determining by means of the calculation unit (5) the coordinate values in the image plane of the first camera (1) of at least three non-aligned control points of the image recorded by the first camera ( 1) and store the determined values in the memory of the computing unit (5); and (IV) from the determined values of the coordinates of the control points in the image plane of the first camera (1) and the values of the calibration and position parameters with respect to the vehicle of the first camera (1) as well as the positional parameters with respect to the vehicle and projection of the first light source (4), and from the data relating to the shape of the ground at the level of the first observation zone, determine by means of the calculation unit (5) the values of the spatial coordinates, in the reference frame R, of the reference points corresponding to the control points and to deduce from it values of the height h of the vehicle relative to the ground, the roll angle e and the pitch angle p of the vehicle in this frame.   3. The method of claim 2, wherein the vehicle is further equipped with a position sensor (8) adapted to determine a two-dimensional position and orientation of the vehicle at ground level at a given time and to transmit the data relating to at this position and this orientation to the calculation unit (5), the memory of the calculation unit (5) is able to store the position and orientation data of the vehicle determined by the position sensor (8) , the first camera (1), the first (4) light source and the computing unit (5) are able to operate synchronously with the position sensor (8) by means of synchronization means (9) , and the step (IV) of determining the values of the height h of the vehicle relative to the ground, the roll angle A and the pitch angle cp of the vehicle by the calculation unit and the determination values of the position and orientation of the vehicle at the end in the ground by the position sensor (8) are synchronized by the synchronization means (9), the determined values of the position and the orientation of the vehicle being stored in the memory of the calculation unit (5).   4. The method of claim 3, wherein the vehicle is further equipped with a second light source (6) adapted to project to the ground a matrix of light reference points on a second separate observation zone (7). of the first observation zone (3), and the first camera (1) is furthermore capable of recording an image of luminous reference points of the second observation zone (7), the second light source (6) being mounted on the vehicle (2) at a given location so that the directions of projection of the reference points on the second observation area (7) do not coincide with the optical axis of the first camera (1) recording an image of the reference points of the second observation area, the values of the position parameters with respect to the vehicle and the projection of the second light source (6) being stored in the memory of the calculation unit (5). ), the first camer a (1), 2890477 61 the first (4) and the second (6) light source and the computing unit (5) being able to operate synchronously with the position sensor (8) through the means of synchronization (9), said method further comprising the following steps, synchronously with the position sensor (8): (a) forming a matrix of light reference points on the second observation zone (7) at means of the second light source (6); (b) recording by means of the first camera (1) an image of the luminous reference points formed on the second observation zone (7) in the image plane of the camera and transmitting the recorded image to the computing unit ( 5), the recorded image having control points and each control point corresponding to the image of a luminous reference point of the second viewing area (7); (c) determining by means of the calculation unit (5) the values of the coordinates in the image plane of the first camera (1) of image control points of the second observation area recorded by the first camera (1) and storing the determined values in the memory of the computing unit (5); and (d) from the determined values of the coordinates of the control points corresponding to the second observation area (7) in the image plane of the first camera (1), values of the calibration and position parameters with respect to the vehicle of the first camera (1) as well as position parameters with respect to the vehicle and projection of the second light source (6), and from the values of the height h of the vehicle relative to the ground, of the roll angle e and the pitch angle p of the vehicle calculated in step (IV) and data relating to the position and orientation of the vehicle, calculate by means of the calculation unit (5) the values of the coordinates of the reference points of the second observation zone corresponding to the control points whose coordinates have been calculated in step (c) and store them in the memory of the calculation unit (5).   The method according to claim 4, further comprising the following step of: (e) determining by means of the computing unit (5) the geometric shape of the second viewing area, from a set reference point coordinate values calculated in step (d) at a given instant, and store in the memory of the computing unit (5) the data relating to that shape.   6. Method according to claim 5, wherein, the second observation zone (7) being further away from the vehicle (2) than the first observation zone (3) and the computing unit being further able to evaluate a movement of the vehicle (2) from position and orientation data of the vehicle transmitted by the position sensor (8), if the computing unit (5) determines that the vehicle is moving so that, at a given moment, the first observation zone (3) covers a portion of the ground that has been part of the second observation zone (7) at a previous moment when said geometrical shape of the second observation zone has been determined, then this geometric shape is used as the shape of the ground at the first observation zone in step (IV) for calculating the values of the height h of the vehicle relative to the ground, of the roll angle 0 and the pitch angle cp of the vehicle corresponding to the given instant ed.   7. The method according to claim 3, wherein the vehicle is further equipped with: a third light source (12) adapted to project to the ground a matrix of light reference points on a third observation zone (11); ) distinct from the first observation zone (3); a second camera (10), mounted on the vehicle (2) at a given and calibrated location, capable of recording an image of light reference points of the third viewing area (11), and in which the unit of calculation (5) is further adapted to receive and process an image recorded by the second camera (10), the third light source (12) is mounted on the vehicle (2) at a given location so that the directions of projection of the reference points on the third observation zone (11) do not coincide with the optical axis of the second camera (10) recording an image of the reference points of this third observation zone, the values positional parameters with respect to the vehicle and projection of the third light source (12) being stored in the memory of the computing unit (5), the third light source (12), the second camera ( 10) and the computing unit (5) both of which are capable of synchronously operating by means of synchronization means (9), said method further comprises the following steps, synchronously with the position sensor (8): (a1) forming a matrix of luminous reference points on the third viewing area (11) by means of the third light source (12); (bl) recording by means of the second camera (10) an image of the light reference points formed on the third viewing area (11) in the image plane of the camera and transmitting the recorded image to the computing unit (5), the recorded image having control points and each control point corresponding to the image of a light reference point of the third observation zone (11); (cl) determining by means of the calculation unit (5) the values of the coordinates in the image plane of the second camera (10) of image control points of the third observation zone recorded by the second camera (10) and storing the determined values in the memory of the computing unit (5); and (d1) from the determined values of the coordinates of the control points corresponding to the third observation zone (11) in the image plane of the second camera (10), values of the calibration and position parameters with respect to the vehicle of the first camera (1) as well as position parameters with respect to the vehicle and projection of the third light source (12), and from the values of the height h of the vehicle relative to the ground, of the roll angle e and the pitch angle W of the vehicle calculated in step (IV) and data relating to the position and orientation of the vehicle, calculating by means of the calculation unit (5) the values of the coordinates of the reference points of the third observation zone corresponding to the control points whose coordinates have been calculated in step (c1) and to store them in the memory of the calculation unit (5).   The method according to claim 7, further comprising the following step of: (el) determining by means of the calculation unit (5) the geometric shape of the third observation area, from a set reference point coordinate values calculated in step (d1) at a given instant, and store in the memory of the computing unit (5) the data relating to this form.   9. Method according to claim 8, wherein, the third observation zone (11) being further away from the vehicle (2) than the first observation zone (3) and the computing unit being further able to evaluate a movement of the vehicle (2) from position and orientation data of the vehicle transmitted by the position sensor (8), if the computing unit (5) determines that the vehicle is moving so that, at a given moment, the first observation zone (3) covers a portion of the ground which has formed part of the third observation zone (11) at a previous moment when said geometrical shape of the third zone has been determined, then this geometric shape is used as the shape of the ground at the first observation zone in step (IV) for calculating the values of the height h of the vehicle relative to the ground, the roll angle 8 and the pitch angle p of the vehicle corresponding to the given instant.   10. Method according to claims 6 and 9, in   which, the vehicle being able to move forwards or backwards, the first camera (1) is mounted at the rear of the vehicle (2) and the second camera (10) is mounted in front of the vehicle (2).   11. Method according to any one of claims 4 to 10, wherein the vehicle is further equipped with a complementary source of light (22) capable of projecting towards the ground a matrix of light reference points on an area of complementary complementary observation (21) of a complementary camera (25), mounted on the vehicle at a given and calibrated location, capable of recording an image of luminous reference points of the complementary observation zone (21), and in which the complementary light source (22) is mounted on the vehicle (2) at a given location so that the directions of projection of the reference points on the associated complementary observation area (21) do not coincide with the the optical axis of the complementary camera (25) recording an image of the reference points of this complementary observation zone, the values of the calibration and position parameters relative to to the vehicle of the complementary camera (25) as well as position parameters with respect to the vehicle and projection of the complementary source of light (22) being stored in the memory of the computing unit (5), the complementary source of light, the complementary camera and the computing unit being able to operate synchronously with the position sensor (8) by means of synchronization means (9), said method further comprises the following steps consisting of, synchronously with the position sensor (8): (a2) forming a matrix of luminous reference points on the complementary observation area by means of the complementary light source; (b2) recording, by means of the complementary camera, an image of the luminous reference points formed on the complementary observation zone in the image plane of the camera and transmitting the recorded image to the computing unit, the recorded image comprising control points and each control point corresponding to the image of a luminous reference point of the complementary observation zone; (c2) determining by means of the calculation unit the coordinate values in the image plane of the first camera (1) of image control points of the complementary observation area recorded by the complementary camera (25) and storing the determined values in the memory of the computing unit; and (d2) from the determined values of the coordinates of the control points corresponding to the complementary observation zone in the image plane of the complementary camera, the values of the calibration and position parameters relative to the vehicle of the complementary camera as well as than positional parameters with respect to the vehicle and projection of the complementary source of light, and from the values of the height h of the vehicle relative to the ground, the roll angle 6 and the pitch angle ( Vehicle ID calculated in step (IV) and vehicle position and orientation data, calculating by means of the calculation unit the coordinate values of the reference points of the complementary observation area corresponding to the control points whose coordinates were calculated in step (c2) and store them in the memory of the calculation unit.   The method of claim 11, further comprising the step of: (e2) determining by means of the computing unit the geometric shape of the complementary observation area, from a set of values of coordinates of reference points calculated in step (d2) at a given instant, and store in the memory of the computing unit the data relating to this form.   2890477 69 13. The method of claim 11 or 12, wherein the complementary camera is the first camera (1).   The method of claim 13, wherein the complementary light source is the first light source or the second light source.   The method of claim 11 or 12 and claim 7, wherein the complementary camera is the second camera (10).   The method of claim 15, wherein the complementary light source is the third light source (12).   The method of any one of claims 3 to 16, wherein the first light source is the second light source.   The method of any one of claims 1 to 17, wherein at least one light source is a laser source.   The method of any one of claims 1 to 18, wherein at least one light source is a pulsed light source, and a pulse synchronization means is adapted to synchronize each pulsed light source with the corresponding camera capable of recording an image of the viewing area illuminated by the pulsed light source.   20. A method according to any one of claims 4 to 19, wherein, if an object (29) is in one of the viewing areas illuminated by one of the light sources, the computing unit detects the presence of the object and the movement of the object relative to the vehicle (2) from the treatment of at least two images of this observation zone illuminated by the light source and received consecutively in time, and stores information about the detected motion and the location of the detected object in the memory.   Method according to any one of claims 3 to 20, wherein the computing unit (5) calculates a graphical representation of the environment of the vehicle (2) from the position data stored in the memory of the vehicle. calculation unit relating to the vehicle (2) and reference points of at least one observation zone.   22. Method according to any one of claims 3 to 21, wherein the computing unit associates a position of the vehicle stored in memory at a corresponding location on a map of the vehicle environment stored in memory, and stores in memory the result of this association.   2890477 71 The method of any one of claims 1 to 22, wherein the computing unit formats data stored in memory and transmits the formatted data to a display means for displaying a representation of that data.   24. A device for determining the position of a vehicle (2) in an environment, the vehicle being able to move on a ground, the device suitable for being on board the vehicle comprising: a first camera (1) adapted to be mounted on the vehicle at a given and calibrated location, and able to record an image of a first observation area (3) located on the ground close to the vehicle; a first source (4) of light capable of projecting light rays onto the first observation zone (3) so as to illuminate it in a non-uniform manner and able to be mounted on the vehicle at a given location so that the directions of projection of the rays on the first observation zone (3) do not coincide with the optical axis of the first camera (1) recording an image of the reference points of the first observation zone (3); and a computing unit (5) having a memory, wherein the first camera (1) is adapted to transmit a recorded image to the computing unit (5); the calculation unit (5) is able to receive and process an image transmitted by the first camera (1); the memory is capable of storing values of the calibration and position parameters with respect to the vehicle of the first camera (1) as well as position parameters with respect to the vehicle and projection of the first light source (4) , said device being characterized in that, from an image of the first observation area illuminated by the first light source recorded and transmitted to the computing unit by the first camera, values of the parameters stored in the memory, and from data relating to the shape of the ground at the first observation area stored in the memory, the calculation unit is able to determine corresponding values of the height h of the vehicle relative to the ground, of the roll angle e and the pitch angle p of the vehicle in a ground-related locator R and to store these values in the memory.   25. Device according to claim 24, wherein the first light source is capable of projecting a matrix of at least three non-aligned light reference points on the first observation zone; the first camera is able to record an image of luminous reference points formed on the first observation zone by the first light source in the image plane of the camera and transmit the recorded image to the computing unit, the image recorded having control points and each control point corresponding to the image of a light reference point of the first observation area; the calculation unit is capable of determining coordinate values in the image plane of the first camera of at least three non-aligned control points of an image recorded by the first camera and storing the determined values in the memory of the first camera; the unit of calculation; and from values stored in control point coordinate memory in the image plane of the first camera, calibration parameters and position relative to the vehicle of the first camera, position parameters with respect to the vehicle and projection of the first light source, and from data relating to the shape of the ground at the level of the first observation zone stored in memory, the computing unit is able to determine the values of the spatial coordinates, in a reference frame R linked to the ground, reference points corresponding to the control points and to deduce corresponding values of the height h of the vehicle relative to the ground, the roll angle 0 and the pitch angle cf) of the vehicle in this mark.   Apparatus according to claim 25, further comprising: a position sensor (8) adapted to be mounted on the vehicle, determining a two-dimensional position and orientation of the vehicle at ground level at a given time and transmitting the data. relating to this position and this orientation to the computing unit, the computing unit being able to store in the memory of the position and orientation data of the vehicle transmitted by the position sensor (8); and synchronization means (9) adapted to synchronize the operation of the first camera, the first light source, the computing unit and the position sensor, wherein the synchronizing means is adapted to synchronize the an operation for determining the values of the height h of the vehicle relative to the ground, the roll angle O and the pitch angle p of the vehicle by the calculation unit with the operation of determining the values of the position and orientation of the vehicle at ground level by the position sensor.   Device according to claim 26, wherein a second light source (6) is able to project towards the ground a matrix of luminous reference points on a second observation zone (7) distinct from the first observation zone. (3); the first camera (1) is furthermore capable of recording an image of luminous reference points formed on the second observation zone (3) in the image plane of the camera and transmitting the recorded image to the computing unit ( 5), the recorded image having control points and each control point corresponding to the image of a luminous reference point of the second viewing area (7); the second light source (6) is adapted to be mounted on the vehicle (2) at a given location so that the directions of projection of the reference points on the second observation zone (7) do not coincide with the optical axis of the first camera (1) recording an image of the reference points of this second observation area; The memory of the calculation unit is able to store position parameters with respect to the vehicle and projection of the second light source (6); the synchronization means (9) is able to operate synchronously with the position sensor (8) the first camera (1), the first (4) and the second (6) light source and the unit of calculation (5); and the calculation unit is able to: determine coordinate values in the image plane of the first camera of control points of an image of the second observation zone recorded by the first camera and store the values determined in the memory of the computing unit; and from values stored in control point coordinate memory corresponding to the second observation area in the image plane of the first camera, values of calibration parameters and positioning with respect to the vehicle of the first camera as well as positional parameters with respect to the vehicle and projection of the second light source, and from values of the height h of the vehicle relative to the ground, the roll angle θ and the pitch angle W of the vehicle and data relating to the position and orientation of the vehicle, calculating values of the coordinates of the reference points of the second observation zone corresponding to the control points and storing them in the memory of the calculation unit .   The apparatus of claim 27, wherein the computing unit is further adapted to determine the geometric shape of the second viewing area from a set of reference point coordinate values of the second observation zone stored in memory and corresponding to a given instant, and storing data relating to this form in the memory of the computing unit.   29. The device according to claim 28, wherein, the second observation zone being farther from the vehicle than the first observation zone, the computing unit is able to evaluate a displacement of the vehicle from position data. and vehicle orientation transmitted by the position sensor and stored in memory, and, if the computing unit determines that the vehicle is moving so that, at a given time, the first viewing area (3 ) covers a portion of the ground that has been part of the second observation zone (7) at a previous time when the geometric shape of said second observation zone has been determined, then the calculation unit uses this stored geometrical shape in memory as a shape of the ground at the first observation zone for calculating the values of the height h of the vehicle relative to the ground, the roll angle e and the pitch angle cp of the corresponding vehicle at the given moment.   30. Apparatus according to claim 26, further comprising a third light source (12) adapted to be mounted on the vehicle (2) and to project towards the ground a matrix of light reference points on a third observation zone. (11) distinct from the first observation zone (3); and a second camera (10) adapted to be mounted on the vehicle (2) at a given and calibrated location, and capable of recording an image of light reference points of the third viewing area (11) and transmitting an image stored in the computing unit (5), in which the computing unit is adapted to receive and process an image recorded by the second camera (10), the third light source (12) is adapted to be mounted on the vehicle at a given location so that the directions of projection of the reference points on the third observation zone (11) do not coincide with the optical axis of the second camera (10) recording an image of the points of reference of this third observation zone, the values of the position parameters with respect to the vehicle of the third light source being stored in the memory of the computing unit, the third light source, the second camera and the computing unit being able to operate in a synchronized manner by means of synchronization means (9), said device is adapted, in synchronism with the position sensor (8): to form a matrix of light reference points on the third observation zone by means of the third light source; and recording, by means of the second camera, an image of the luminous reference points formed on the third observation zone in the image plane of the camera and transmitting the recorded image to the computing unit, the recorded image comprising dots control and each control point corresponding to the image of a light reference point of the third observation zone; and determining by means of the calculation unit the values of the coordinates in the image plane of the second camera of the image control points of the third observation zone recorded by the second camera and storing the determined values in the memory the unit of calculation; and from the determined values of the coordinates of the control points corresponding to the third observation zone in the image plane of the second camera, values of the calibration and position parameters with respect to the vehicle of the first camera as well as parameters of position by: ratio to the vehicle and projection of the third light source, and from the values of the height h of the vehicle relative to the ground, the roll angle 0 and the pitch angle 9 of the vehicle and data relating to the position and orientation of the vehicle, calculating by means of the calculation unit the coordinate values of the reference points of the third observation zone corresponding to the control points whose coordinates have been calculated and store them in the memory of the computing unit.   Apparatus according to claim 30, wherein the computing unit is further adapted to determine the geometric shape of the third observation area from a set of reference point coordinate values stored in the memory. corresponding to a given instant, and to store in the memory the data relating to this form.   Apparatus according to claim 31, wherein the third viewing area is further away from the vehicle than the first viewing area and the computing unit is further adapted to evaluate a movement of the vehicle from position data. and orientation of the vehicle transmitted by the position sensor and stored in the memory, if the computing unit determines that the vehicle is moving so that, at a given moment, the first observation area covers a portion of the soil which has been part of the third observation zone at a previous time when said geometrical shape of the third zone has been determined, then this geometrical shape is used as the shape of the soil at the level of the first observation zone in the calculating the values of the height h of the vehicle relative to the ground, the roll angle θ and the pitch angle cp of the vehicle corresponding to the given instant.   33. Device according to claims 29 and 32, wherein, the vehicle being able to move forwards or backwards, the first camera is adapted to be mounted at the rear of the vehicle and the second camera is fit to be mounted at the front of the vehicle.   34. Device according to any one of claims 27 to 33, wherein a complementary light source (20) is adapted to be mounted on the vehicle and to project towards the ground a matrix of light reference points on a zone of complementary observation associated (19); a complementary camera is adapted to be mounted on the vehicle in a given and calibrated location, and to record an image of luminous reference points of the complementary observation area; the complementary source of light is adapted to be mounted on the vehicle at a given location so that the directions of projection of the reference points on the associated complementary observation area do not coincide with the optical axis of the complementary camera recording an image of the reference points of this complementary observation zone, the values of the calibration and position parameters relative to the vehicle of the complementary camera as well as the position parameters with respect to the vehicle and projection of the complementary source of light being stored in the memory of the computing unit, the complementary light source, the complementary camera and the computing unit being able to operate synchronously with the position sensor by means of synchronization, said device is furthermore capable of synchronously with the position sensor: forming a matrix of luminous reference points on the complementary observation zone by means of the complementary source of light; and recording by means of the complementary camera an image of the luminous reference points formed on the complementary observation area in the image plane of the camera and transmitting the recorded image to the computing unit, the recorded image comprising dots control and each control point corresponding to the image of a luminous reference point of the complementary observation zone; and determining by means of the calculation unit the coordinate values in the image plane of the first image control point camera of the complementary observation area recorded by the complementary camera and storing the determined values in the memory the unit of calculation; and from the determined values of the coordinates of the control points corresponding to the complementary observation zone in the image plane of the complementary camera, values of the calibration and position parameters relative to the vehicle of the complementary camera as well as the parameters of position relative to the vehicle and projection of the complementary source of light, and from the values of the height h of the vehicle relative to the ground, the roll angle e and the pitch angle cf) of the vehicle and data relating to the position and orientation of the vehicle, calculating by means of the calculation unit the values of the coordinates of the reference points of the complementary observation zone corresponding to the control points whose coordinates have have been calculated and stored in the memory of the computing unit.   Apparatus according to claim 34, wherein the computing unit is further adapted to determine the geometric shape of the complementary viewing area from a set of reference point coordinate values stored in the memory. corresponding to a given instant, and to store in the memory the data relating to this form.   36. Device according to claim 34 or 35, wherein the complementary camera is the first camera.   37. Device according to claim 36, wherein the complementary source of light is the first light source or the second light source.   Device according to claim 34 or 35, wherein the complementary camera is the second camera.   39. Device according to claim 38, wherein the complementary source of light is the third light source.   2890477 83 40. Device according to any one of claims 26 to 39, wherein the first light source is the second light source.   41. Device according to any one of claims 24 to 40, wherein at least one light source is a laser source.   42. The device of any one of claims 24 to 41, wherein at least one light source is a pulsed light source; a pulsation synchronization means is adapted to synchronize each pulsed light source with the corresponding camera capable of recording an image of the observation area illuminated by the pulsed light source.   43. Apparatus according to any one of claims 27 to 42, wherein if an object (29) is in one of the viewing areas illuminated by one of the light sources, the computing unit is able to detect the presence of the object and the movement of the object relative to the vehicle from the treatment of at least two images of this observation zone illuminated by the light source and received consecutively in time, and storing in the memory the information relating to the detected motion and the location of the detected object.   2890477 84 44. Apparatus according to any one of claims 26 to 43, wherein the computing unit graphical representation 5 from the data is able to calculate one of the position vehicle environment stored in the memory of the unit of vehicle calculation and reference points of at least one observation area.   45. Device according to any one of   claims 26 to 44, wherein   the calculation unit is capable of associating a position of the vehicle stored in memory with a corresponding location on a map of the environment of the vehicle stored in memory, and storing in memory the result of this association.   46. Device according to any one of   claims 24 to 45, wherein   the computing unit is able to format data stored in memory and to transmit this formatted data to a display means for displaying a representation of these data.   47. A vehicle (2) equipped with a position determining device in an environment according to any one of claims 24 to 46.