Vehicle maneuvering and parking assistance system within an infrastructure

Abstract

Vehicle maneuvering and parking assistance system (1) in an infrastructure Vehicle maneuvering assistance system (10) characterized in that it comprises: - a mobile image acquisition means (12) configured to acquire images at the front of the vehicle (1) itself equipped with an image acquisition means disposed at the front of said vehicle (1); - a steering system (16) configured to move the mobile image acquisition means (12) to follow the vehicle (1);- a central unit (20) comprising a communication module (23) configured to receive images and apply processing to them to identify an obstacle (40) on the trajectory of the vehicle (1) and to communicate with a communication module (3) of the vehicle (1), and configured to communicate with a communication module (13) of the steering system (16), the communication module (23) of the central unit (20) being configured to transmit positioning coordinates to the communication module (3) of the vehicle (1) and to the communication module (13) of the steering system (16). Figure for the abbreviation: Fig.1;

Description

Title of the invention: System to assist in maneuvering and parking a vehicle in an infrastructure

[0001] The invention relates generally to the field of methods for assisting the piloting of motor vehicles. More particularly, the invention relates to a system for assisting the maneuvering and parking of a vehicle within an infrastructure. The invention further relates to a method for assisting the maneuvering and parking of a vehicle using such a system.

[0002] Various car manufacturers offer systems that automate the task of parking a vehicle and certain types of vehicle movement, such as driving forward in a straight line, reversing in a straight line, turning with a predetermined radius of curvature, etc. These driver assistance systems are generally embedded within a vehicle itself.

[0003] These driver assistance systems are generally used in a defined environment such as a parking area, for example. Such a parking area, commonly called a "parking lot", generally includes delimited traffic lanes and a plurality of individual spaces reserved for parking vehicles.

[0004] When a vehicle equipped with a driver assistance system is operating autonomously, that is, without driver intervention in such a defined environment, the vehicle's trajectory is generally free of obstacles. However, if an obstacle is present on the vehicle's trajectory, detection of this obstacle is desirable, or even necessary, to prevent the vehicle from colliding with it.

[0005] Generally, obstacle detection is performed using one or more sensor systems located on the vehicle, and a computer method for processing the sensor data received from the sensor system(s). A sensor system may include a transmitter and one or more sensors. A non-exhaustive list of sensor systems may include: a camera, a LiDAR, radar, and ultrasonic sensors, among others. However, despite the available sensor data, the accurate and efficient detection of an obstacle on a trajectory presents certain challenges. Furthermore, a large number of sensors within the sensor system significantly increases the final cost of the motor vehicle equipped with such a sensor system.

[0006] The object of the invention is to overcome the drawbacks described above. The objective of the invention is to use a limited number of sensors on an infrastructure in parallel with an on-board sensor on the motor vehicle traveling within the infrastructure. The various sensors provide data which, once processed, makes it possible to detect a potential obstacle on the path taken by the motor vehicle.

[0007] To this end, the invention relates to a system for assisting in the maneuvering and parking of a vehicle in an infrastructure, the vehicle being equipped with an image acquisition means disposed at the front of said vehicle and a first communication module, the system being characterized in that it comprises: - at least one means for acquiring mobile images in translation along a first direction on a support of the infrastructure, the mobile means being configured to acquire images of an environment in front of the vehicle traveling autonomously in the infrastructure along a predetermined trajectory; - a control system configured to move the mobile image acquisition means on the infrastructure support so as to follow the vehicle moving in the infrastructure, the control system and at least one mobile image acquisition means being equipped with a second communication module; - a central unit comprising a third communication module configured to receive images acquired by the image acquisition means located at the front of said vehicle and images acquired by the mobile image acquisition means, the central unit being configured to apply image processing to said images enabling the identification of an obstacle on the trajectory of the vehicle traveling in the infrastructure, the third communication module of the central unit also being configured to communicate with the first communication module of the vehicle to exchange data concerning said vehicle, and configured to communicate with the second communication module of the steering system and the mobile image acquisition means, the third communication module of the central unit being configured to transmit positioning coordinates, in particular GPS coordinates of a trajectory,to the first communication module of the vehicle and / or positioning coordinates to the second communication module of the steering system and the mobile image acquisition device.

[0008] According to one embodiment, the mobile image acquisition means is configured to reproduce the image focal length and the horizontal and vertical image resolutions of the vehicle image acquisition means.

[0009] According to one embodiment, the image acquisition means, which is mobile in translation, is also mobile in rotation about a first axis of rotation such that the the control system can rotate the mobile image acquisition means around said first axis of rotation.

[0010] According to one embodiment, the mobile image acquisition means is also mobile in rotation about a second axis of rotation so that the control system can pivot the mobile image acquisition means about said second axis of rotation, this second axis of rotation is substantially perpendicular to the first axis of rotation.

[0011] According to one embodiment, the mobile image acquisition means is also mobile in translation along a second direction on a support of the infrastructure, the second direction being substantially perpendicular to the first direction.

[0012] According to one embodiment, the system comprises two mobile image acquisition means, the two image acquisition means being arranged on a support located above the vehicle, the two mobile image acquisition means being oriented towards the ground of the infrastructure so as to acquire images of an environment in front of the vehicle traveling in the infrastructure, a first mobile image acquisition means being aligned vertically with the image acquisition means disposed at the front of said vehicle, a second mobile image acquisition means being juxtaposed longitudinally in front of the first mobile image acquisition means.

[0013] According to one embodiment, the control system is configured to move the second mobile image acquisition means independently of the first mobile image acquisition means.

[0014] According to one embodiment, the support comprises a rail or an arrangement of rails and the control system is configured to move at least one mobile image acquisition means along said rail or along said rails of the arrangement.

[0015] The invention also relates to a method for assisting in the maneuvering and parking of a vehicle using a system as described above, the method being characterized in that it comprises the following steps:

[0016] - the first communication module of the vehicle sends data from said vehicle to the third communication module of the central unit; - the mobile image acquisition means is positioned at the same longitudinal coordinate as the image acquisition means arranged on the vehicle; - the image acquisition system mounted on the vehicle acquires a first image of the environment in front of the vehicle; - the vehicle's first communication module sends the first acquired image to the third communication module of the infrastructure's central unit; - the central unit generates a distortion-free image from the image sent by the vehicle's first communication module; - the position of the mobile image acquisition device is updated; - the mobile image acquisition means acquires a second image of the environment in front of the vehicle; - the second communication module of the mobile image acquisition means sends the second acquired image to the third communication module of the central unit of the infrastructure; - the central unit generates a distortion-free image from the received image; - The central unit applies image processing to the distortion-free images to detect one or more potential obstacles in the vehicle's path.

[0017] According to one embodiment, the method also includes the following steps when an obstacle is detected in the vehicle's path: - the third communication module of the central unit sends a command to the second communication module of the mobile image acquisition means's control system to rotate said mobile image acquisition means around the second axis of rotation so as to center the images acquired by the mobile image acquisition means on the detected obstacle; - the mobile image acquisition means acquires a third image of the environment in front of the vehicle; - the second communication module of the mobile image acquisition means sends the third acquired image to the third communication module of the central unit of the infrastructure; - the central unit generates a distortion-free image from the images received by the second communication module of the mobile image acquisition means and by the first communication module of the vehicle image acquisition means; - The central unit applies a stereovision method to evaluate the distance between the vehicle and the detected obstacle.

[0018] These objects, features and advantages of the present invention will be described in detail in the following description of a particular embodiment, given by way of non-limiting example, with reference to the accompanying figures, among which:

[0019] Fig. 1 illustrates a system for assisting the maneuvering and parking of a vehicle in an infrastructure according to a first embodiment.

[0020] Figure [Fig. 2] illustrates an example of an image acquired by the vehicle's image acquisition means.

[0021] Figure [Fig.3] illustrates an example of an image acquired by the moving image acquisition means of the system in [Fig.1].

[0022] Fig. 4 illustrates an example of a distortion-free image generated by the central unit from the image in Fig. 2.

[0023] Fig. 5 illustrates an example of a distortion-free image generated by the central unit from the image in Fig. 3.

[0024] [Fig.6] illustrates an example of an image obtained following a change of focal length from the image in [Fig.5].

[0025] Figure 7 illustrates an example of a distortion-free image generated by the central unit from an image acquired by an image acquisition means of a reference vehicle during a calibration phase.

[0026] Fig. 8 illustrates an example of a result obtained following image processing applied to the image in Fig. 7.

[0027] Fig. 9 illustrates an example of a distortion-free image generated by the central unit from an image acquired by the vehicle's image acquisition means in the case where an obstacle is detected on the vehicle's trajectory.

[0028] Fig. 10 illustrates an example of a distortion-free image generated by the central unit from an image acquired by the moving image acquisition means of the system in the case where an obstacle is detected on the vehicle's path.

[0029] Fig. 11 illustrates an example of a result obtained following image processing applied to the images in Figures 9 and 10.

[0030] Fig. 12 illustrates an example of an image acquired by the system's mobile image acquisition means following a repositioning of said mobile image acquisition means.

[0031] Fig. 13 illustrates an example of images obtained with a stereovision method applied by the central unit.

[0032] Figure 14 illustrates a system according to a second embodiment.

[0033] Figure 15 illustrates examples of images acquired by the acquisition means moving image of the system of the [Fig. 14].

[0034] The maneuvering and parking assistance system 10 that is the subject of this patent application incorporates various elements of an infrastructure 100 with elements of a motor vehicle 1, and in particular of a motor vehicle 1 that is at least partially automated. In the embodiments of the systems 10 described herein, some elements are specific to the infrastructure 100 while other elements are specific to the motor vehicle 1.

[0035] Figure 1 schematically illustrates a motor vehicle 1 traveling within an infrastructure 100. The infrastructure 100 may be in the form of a parking area, a garage, or any other environment with a set of parking spaces for motor vehicles 1. The infrastructure 100 may be above ground or underground. Such an infrastructure 100 is made available of vehicle users 1 motor vehicles to be able to park their vehicle 1. In the case where vehicle 1 is at least partially automated, a user can in particular leave their vehicle 1 at the entrance of the infrastructure 100 for example and the maneuvering and parking assistance system 10 then takes care of moving the vehicle 1 motor vehicle to a suitable parking space.

[0036] The vehicle 1 may, for example, include a control device configured to control a steering device and a drive and braking device of the vehicle 1. In a preferred embodiment, the maneuvering and parking assistance system 10 of the vehicle 1 may allow the vehicle 1 to autonomously follow a predetermined trajectory and to autonomously perform a maneuver to park said vehicle 1 in an unoccupied parking space.

[0037] The trajectory that vehicle 1 is required to follow when traveling within infrastructure 100 is predetermined, in the sense that GPS coordinates representative of said trajectory are known in advance. These coordinates were, for example, established and recorded during a system calibration phase in which a reference vehicle traveled said trajectory before infrastructure 100 was put into service.

[0038] Vehicle 1 is equipped with an image acquisition means 2 located at the front of said vehicle 1. Such an image acquisition means 2 is, for example, arranged at the level of the grille of vehicle 1, for example behind said grille, above a license plate of vehicle 1, a few tens of centimeters from the ground (for example 60 cm). The image acquisition means 2 is preferably laterally centered, that is to say, arranged in the middle of the front face of vehicle 1.

[0039] Generally, such an image acquisition means 2 is oriented forwards with an angle of inclination, for example 40°, so as to acquire images of the ground just in front of the front face of the vehicle 1.

[0040] The image acquisition means 2 is, for example, a camera with a wide-angle lens (a "fisheye" lens), which allows images to be acquired with a horizontal viewing angle close to 180°, or even greater than 180°. The images obtained with such an image acquisition means 2 make it possible to simultaneously see the ground directly in front of the front of the vehicle 1 and the surrounding area at a greater distance. Figure 2 shows an example of an image acquired with such an image acquisition means. A wide-angle lens generally introduces distortion into the images obtained; this distortion is also visible in Figure 2.

[0041] Furthermore, vehicle 1 is equipped with a first communication module 3 which allows these images to be transferred, that is, sent to a receiver external to the vehicle 1. More generally, the first communication module The first communication module 3 of vehicle 1 allows for data exchange; that is, it can send not only images acquired by the image acquisition device 2 located at the front of said vehicle 1, but also other types of data (examples of data will be given later in the description). Furthermore, the first communication module 3 is also configured to receive data sent to it by another communication module located outside of vehicle 1.

[0042] The maneuvering and parking assistance system 10 comprises at least one mobile image acquisition means 12, which may be a camera with a wide-angle lens. The mobile image acquisition means 12 is configured to acquire images of the environment in front of the vehicle 1 traveling autonomously within the infrastructure 100 along a predetermined trajectory. Figure 3 shows an example of an image obtained with such a mobile image acquisition means 12.

[0043] The mobile image acquisition means 12 is mobile in translation along a first direction on a support 14 of the infrastructure 100. More specifically, the system 10 includes a control system 16 configured to move the mobile image acquisition means 12 on the support 14 of the infrastructure 100 so as to follow the vehicle 1 traveling in the infrastructure 100.

[0044] In addition, and according to an optional variant of the embodiment described above, the mobile image acquisition means 12 is also translationally mobile along a second direction on a support 14 of the infrastructure 100, the second direction being substantially perpendicular to the first direction. Thus, if the first direction of translation is horizontal, the second direction of translation is vertical. This additional degree of mobility allows the control system 16 of the mobile image acquisition means 12 to adjust the height of the mobile image acquisition means 12. In particular, the mobile image acquisition means 12 can be placed at the same distance from the ground as the image acquisition means 2 arranged on the front of the vehicle 1. The height at which the image acquisition means 2 of the vehicle 1 is located may differ from one vehicle to another.The additional degree of mobility allows the control system 16 of the mobile image acquisition means 12 to reproduce as faithfully as possible the position of the image acquisition means 2 of the vehicle 1.

[0045] Optionally, the moving image acquisition means 12 is also movable in rotation about a first axis of rotation so that the control system 16 can pivot the moving image acquisition means 12 about said first axis of rotation. This first axis of rotation is in particular a horizontal axis. This makes it possible to tilt the moving image acquisition means 12 so as to reproduce the angle of inclination of the image acquisition means 2 of the vehicle 1. Such an inclination of the mobile image acquisition means 12 then makes it possible to acquire images of the environment in front of the vehicle 1, and more particularly to acquire images of the ground just in front of the front face of the vehicle 1. The inclination of the image acquisition means 2 of the vehicle 1 can be different from one vehicle to another, which is why such mobility of the mobile image acquisition means 12 in rotation around a first axis of rotation is particularly appreciable.

[0046] Having the same angle of inclination for the image acquisition means 2 of the vehicle 1 and for the mobile image acquisition means 12 of the maneuvering assistance system 10 allows for similar images of the environment in front of the vehicle 1, which facilitates the processing of images acquired during the progression of the vehicle 1 along the predetermined trajectory.

[0047] In addition, according to an optional variant of the embodiment described above, the mobile image acquisition means 12 is also rotationally mobile about a second axis of rotation so that the control system 16 can pivot the mobile image acquisition means 12 about said second axis of rotation. This second axis of rotation is substantially perpendicular to the first axis of rotation. Thus, if the first axis of rotation is a horizontal axis, the second axis of rotation is a vertical axis. This additional degree of mobility allows the control system 16 of the mobile image acquisition means 12 to apply a yaw motion to the mobile image acquisition means 12.

[0048] According to a preferred embodiment, the support 14 mentioned above comprises a rail or a rail arrangement. In this preferred embodiment, the control system 16 is configured to move at least one mobile image acquisition means 12 along said rail or along said rails of the arrangement. The rail(s) extend, for example, parallel to the ground of the infrastructure 100, in particular in a substantially horizontal direction if the ground of the infrastructure 100 is generally flat, or in a direction inclined with respect to the horizontal if the ground is sloped.

[0049] Such a rail or rail arrangement allows the control system 16 of at least one mobile image acquisition means 12 to be guided reliably and repeatably. This rail or rail arrangement can also ensure the stability of the mobile image acquisition means 12 during its trajectory.

[0050] In a first embodiment, the rail or arrangement of several rails that forms the support 14 for the mobile image acquisition means 12 is, for example, arranged at the level of a wall of the infrastructure 100 in which the vehicle 1 travels. In this way, the control system 16 can move the camera along from the wall next to vehicle 1 in order to follow it while acquiring images of the environment in front of vehicle 1.

[0051] Thus, according to one embodiment, the positioning coordinates successively occupied by the mobile image acquisition means 12 coincide with the positioning coordinates successively occupied by the vehicle 1. The first direction in which the control system 16 moves the mobile image acquisition means 12 relative to the support 14 is therefore substantially the same direction as that taken by the motor vehicle 1 when it autonomously travels a predetermined trajectory in the infrastructure 100.

[0052] In other words, the mobile image acquisition means 12 and the at least partially automated vehicle 1 progress together, preferably at the same speed, along the route taken by said vehicle 1. During this route, the image acquisition means 2 arranged on the front face of the vehicle 1 and the at least one mobile image acquisition means 12 of the maneuvering assistance system 10 each acquire a series of images with viewpoints that are certainly different, but still very similar, as illustrated in Figures 2 and 3.

[0053] The control system 16 and the mobile image acquisition means 12 are also equipped with a second communication module 13. This second communication module 13 allows, on the one hand, the transfer of images acquired by the mobile image acquisition means 12 to a receiver external to the vehicle 1, and on the other hand, the receipt of movement instructions indicating to the control system 16 where and how the mobile image acquisition means 12 should be positioned. The second communication module 13 may, for example, include a transmitter at the level of the mobile image acquisition means 12 to send the acquired images and it may include a receiver arranged at the level of the control system 16 to receive the movement instructions which will then be relayed to the mobile image acquisition means 12.

[0054] The maneuvering assistance system 10 further includes a central unit 20 which also includes a third communication module 23. The third communication module 23 of the central unit 20 is configured to communicate on the one hand with the first communication module 3 of the vehicle 1, and configured on the other hand to communicate with the second communication module 13 of the steering system 16 and of the mobile image acquisition means 12.

[0055] More specifically, the third communication module 23 of the central unit 20 is configured to transmit positioning coordinates, including GPS coordinates of a trajectory, to the first communication module 3 of the vehicle 1 and / or positioning coordinates to the second communication module 13 of the steering system 16 and the image acquisition means Mobile 12. In this way, the central unit 20 can ensure consistency between the movements of the mobile image acquisition device 12 and the movements of the vehicle 1. In particular, the central unit 20 can adjust the movements of the mobile image acquisition device 12 according to the position of the vehicle 1 traveling within the infrastructure 100. The third communication module 23 of the central unit 20 sends the GPS coordinates to the first communication module 3 of the vehicle 1 at a given frequency, for example every 100 ms, thus dictating to the vehicle 1 the trajectory to follow as it travels within the infrastructure 100.

[0056] Furthermore, the third communication module 23 of the central unit 20 is configured to communicate with the first communication module 3 of the vehicle 1 to exchange data concerning said vehicle 1. This data includes, for example, the position of the image acquisition means 2 in the frame of reference of the vehicle 1, the dimensions of the vehicle 1, in particular its length, width and height, as well as the width and height of the sensor of the image acquisition means 2 arranged on the front face of the vehicle 1, the horizontal and vertical resolution of the acquired images and the polynomial coefficients describing the distortion due to the "fish-eye" type lens of said image acquisition means 2.

[0057] This third communication module 23 of the central unit 20 is also configured to receive images acquired by the image acquisition means 2 located at the front of said vehicle 1, via the first communication module 3, and images acquired by the mobile image acquisition means 12, via the second communication module 13.

[0058] The central unit 20 is also configured to apply image processing to said images. The objective of this processing is to identify a potential obstacle 40 present in the trajectory of the vehicle 1 traveling in the infrastructure 100. This image processing is part of a method for assisting the maneuvering and parking of a vehicle 1 using a system as described above. The method comprises a plurality of steps detailed below.

[0059] In a first step of the process, the first communication module 3 of the vehicle 1 sends data from said vehicle 1 to the third communication module 23 of the central unit 20. This step takes place, for example, when a user drops off their vehicle 1 at the entrance of the infrastructure 100, or shortly after this drop-off. In the next step, the mobile image acquisition means 12 is positioned at the same longitudinal coordinate as the image acquisition means 2 mounted on the vehicle 1. In other words, the mobile image acquisition means 12 aligns with the image acquisition means 2 arranged on vehicle 1 in order to follow said vehicle 1 during its predetermined trajectory in infrastructure 100. Next, the image acquisition means 2 arranged on the vehicle 1 acquires a first image of the environment in front of the vehicle. Figure 2 shows an example of such a first image. Then the first communication module 3 of vehicle 1 sends the first acquired image to the third communication module 23 of the central unit 20 of the infrastructure 100. Then the central unit 20 generates a distortion-free image from the first image sent by the first communication module 3 of vehicle 1. Figure 4 shows an example of such an image. When vehicle 1 moves autonomously along the predetermined path to its parking position, the position of the mobile image acquisition means 12 is updated, that is to say, the mobile image acquisition means 12 is positioned at the same longitudinal coordinate as the image acquisition means 2 arranged on the vehicle 1, and this throughout the path of the vehicle 1 traveling in the infrastructure 100. In other words, the mobile image acquisition means 12 and the at least partially automated vehicle 1 move together, preferably at the same speed, along the path taken by said vehicle 1. The mobile image acquisition means 12 acquires a second image of the environment in front of the vehicle 1. Figure 3 shows an example of such a second image. Then, the second communication module 13 of the mobile image acquisition means 12 sends the second acquired image to the central unit 20 of the infrastructure 100. The central unit 20 then generates a distortion-free image from the second received image. Figure 5 shows an example of such an image. Finally, the central unit 20 applies image processing to the distortion-free images to detect one or more potential obstacles 40 in the path of the vehicle 1.

[0060] Generally, the image acquisition means 2 mounted on the vehicle 1 acquires a series of first images of the environment in front of the vehicle 1 throughout the vehicle 1's journey.

[0061] Similarly, the mobile image acquisition means 12 acquires a series of second images of the environment in front of the vehicle 1 throughout the journey of the vehicle 1.

[0062] Optionally, the method may include an additional step in which the central unit 20 applies a change of focal length as well as a change of horizontal and vertical resolutions to the second images acquired by the second communication module 13 of the mobile image acquisition means 12, images in which the distortions due to the wide angle of the image acquisition means The 12 mobile images were corrected. Figure 6 shows an example of the result obtained from the image in Figure 5, to which a change in focal length and a change in horizontal and vertical resolution were applied. These changes notably modify the field of view in the resulting image.

[0063] To perform this step, and according to a particular embodiment of the vehicle 1 maneuvering and parking assistance system 10, the mobile image acquisition means 12 is configured to reproduce the image focal length and the horizontal and vertical resolutions of the image of the image acquisition means 2 of the vehicle 1. To do this, the first communication module 3 of the vehicle 1 sends to the third communication module 23 of the central unit 20 data specific to the vehicle 1 and data specific to the image acquisition means 2 of the vehicle 1, then the third communication module 23 of the central unit 20 transmits this data to the second communication module 13 of the control system 16 and the mobile image acquisition means 12.

[0064] These data are for example the position of the camera in the vehicle 1 frame, the length, width and height of the vehicle 1, the width and height of the sensor of the image acquisition means 2, the horizontal and vertical resolution of the image acquisition means 2 and the polynomial coefficients describing the distortion.

[0065] Once this data is received by the second communication module 13 of the control system 16 and the mobile image acquisition means 12, the parameters of the latter are modified so as to reproduce the same focal length of the image of the image acquisition means 2 of the vehicle 1 and so as to adapt the horizontal and vertical resolutions of the image so that they are identical or at least similar to those of the image acquisition means 2 of the vehicle 1. Such a reproduction of the parameters of the image acquisition means 2 of the vehicle 1 at the level of the mobile image acquisition means 12 of the system allows for fast and efficient image processing, so as to make the detection of an obstacle more reliable.

[0066] Regarding the image processing applied by the central unit 20, the central unit 20 compares the images received by the image acquisition means 2 of the vehicle 1 with images from a previously performed calibration phase. Figure 7 shows an example of such an image acquired during a calibration phase by an image acquisition means positioned on the front of a reference vehicle. In the example of the image shown in Figure 7, the distortions due to the wide angle of the image acquisition means have already been corrected; therefore, these distortions do not appear in Figure 7.

[0067] During the calibration phase, which is carried out for example during the installation of infrastructure 100, before its commissioning, a reference vehicle travels one or more example trajectories, for example from the entrance of infrastructure 100 to a parking space in this infrastructure 100. There are in particular as many trajectories as there are parking spaces in the infrastructure 100. Each predetermined trajectory then allows the at least partially automated automobile vehicle 1 to be guided to a precise parking space, specific to the trajectory, when the infrastructure 100 is functional and in service.

[0068] During this calibration phase, the trajectory(s) followed by the reference vehicle are free of any obstacles. The GPS coordinates of the reference vehicle traveling this trajectory(s) are recorded periodically so as to keep them in memory for use as a reference when the infrastructure 100 is put into service, that is to say, when a vehicle 1, at least partially automated, other than the reference vehicle, travels in the infrastructure 100.

[0069] Also during this calibration phase, the positions of at least one mobile image acquisition means 12, which follows the reference vehicle traveling along the trajectory(ies), are also recorded. At each positioning coordinate recorded for the at least one mobile image acquisition means 12, an image acquired by it is also recorded.

[0070] When a vehicle 1, at least partially automated, travels within the operational infrastructure 100, the central unit 20 compares the images received by the image acquisition means 2 of vehicle 1 with the images from the calibration phase. More specifically, the central unit 20 can be configured to subtract an image from the calibration phase associated with a specific position of the reference vehicle from an image acquired by the image acquisition means 2 of vehicle 1 for that same position along its route.

[0071] To enable a reliable comparison of these two images, the central unit 20 can apply image processing to the images from the calibration phase for all the positions occupied by the reference vehicle during a trajectory traveled during the calibration phase. In particular, the central unit 20 can apply a tilt angle to said images to reflect the tilt of the image acquisition means 2 of the vehicle 1 traveling in the infrastructure 100. In other words, the central unit 20 can apply a tilt angle to the images from the calibration phase to reproduce the tilt of the image acquisition means 2 of the vehicle 1. Figure 8 gives an example of an image resulting from such processing.

[0072] As long as the system 10 does not detect an obstacle 40 on the trajectory of the vehicle 1, the latter continues to advance autonomously along the predefined trajectory.

[0073] However, if a significant difference exists between the images from the post-processing calibration phase and the images acquired by the image acquisition means 2 of vehicle 1 traveling in the infrastructure 100, the presence of an obstacle 40 in the trajectory of vehicle 1 is very likely. In particular, if the processing applied to the images involves image subtraction, and if the resulting image is not empty, i.e., if it contains something other than black pixels, the presence of an obstacle 40 in the trajectory of vehicle 1 is very likely.

[0074] Figure 9 shows an example of a distortion-free image generated by the central unit 20 from the image acquired by the image acquisition means 2 of the vehicle 1 and transmitted to the central unit 20 via the first communication module 3. In this image, an obstacle 40, here a construction cone, is present on the intended path for the motor vehicle 1 traveling in the infrastructure 100.

[0075] Similarly, [Fig. 10] shows an example of a distortion-free image generated by the central unit 20 from the image acquired by the moving image acquisition means 12. On this image, the same obstacle 40 as on [Fig. 9] is visible.

[0076] Figure 11 shows the result of the image processing applied to the image of Figure 9 by the central processing unit 20: the subtraction of the image of Figure 9 from the image obtained during the calibration phase does not produce a matrix of black pixels; the obstacle 40 (the construction cone) is present in the resulting image. This scenario can be generalized to various types of obstacles 40.

[0077] In the event that the system 10 detects an obstacle 40 on the trajectory of the vehicle 1, the system 10 can send a stop command to the vehicle 1 to immobilize it in order to prevent said vehicle 1 from colliding with the detected obstacle 40.

[0078] The central unit 20 can also apply processing to images acquired, on the one hand, by the image acquisition means 2 of the vehicle 1 and transmitted to the central unit 20 via the first communication module 3, and on the other hand, by the mobile image acquisition means 12 which follows the vehicle 1 traveling within the infrastructure 100 along the predetermined trajectory; this image processing makes it possible to evaluate the distance between the obstacle 40 and the vehicle 1. This processing is then similar to a stereovision method, also known as "stereoscopic measurement". Thus, in the case where an obstacle 40 is indeed detected, the central unit 20 uses a stereovision method to determine the dimensions, shapes and / or positions of objects present in these images and to evaluate the distance between the vehicle 1 and the detected obstacle 40.

[0079] In the example described above, the central unit 20 applies a stereovision method using the images in Figures 9 and 10, among others, to evaluate the distance between the vehicle 1 and the detected obstacle 40.

[0080] Various stereovision methods are known. However, it is possible to improve the results obtained with certain stereovision methods by providing images with specific characteristics. In the case of the stereovision method applied by the central unit 20, it is more efficient, that is to say, more precise in estimating the distance between vehicle 1 and an obstacle 40 detected on the trajectory of vehicle 1, when said obstacle 40 is located at the center of the image used in the stereovision method.

[0081] Thus, when an obstacle 40 is detected on the trajectory of the vehicle 1, the method described above includes additional steps to allow an evaluation of the distance separating the vehicle 1 from the detected obstacle 40: - the third communication module 23 of the central unit 20 sends a command to the second communication module 13 of the control system 16 of the mobile image acquisition means 12 to rotate said mobile image acquisition means 12 around the second axis of rotation so as to center the images acquired by the mobile image acquisition means 12 on the detected obstacle 40; - the mobile image acquisition means 12 acquires a third image of the environment in front of the vehicle 1, an image in which the detected obstacle 40 is substantially in the center; - the second communication module 13 of the mobile image acquisition means 12 sends the third acquired image to the third communication module 23 of the central unit 20 of the infrastructure 100; - the central unit 20 generates a distortion-free image from the images received by the second communication module 13 of the mobile image acquisition means 12 and by the first communication module 3 of the image acquisition means 2 of the vehicle 1; - the central unit 20 applies a stereovision method to evaluate the distance between vehicle 1 and the detected obstacle 40.

[0082] Thus, in the case where the system 10 detects an obstacle 40 on the trajectory of the vehicle 1, the third communication module 23 of the central unit 20 can send a pivot command to the second communication module 13 of the control system 16 of the mobile image acquisition means 12 to pivot the mobile image acquisition means 12 in rotation around its second axis of rotation so as to center the images acquired by said means on the detected obstacle 40.

[0083] Once this rotation has been carried out, the automobile vehicle 1 appears in the images acquired by said mobile image acquisition means 12, as illustrated in the example in [Fig. 12]. The next step is to replace the vehicle 1 in these images with black pixels before or during the image processing applied by the central processing unit 20.To do this, the first communication module 3 of vehicle 1 can communicate to the third communication module 23 of the central unit 20 specific data of the vehicle 1, such as its dimensions and / or shape, or any other data enabling the central unit 20 to replace the pixels of the images on which a portion of the vehicle 1 appears with black pixels, to avoid a disruption of the processing enabling the detection of an obstacle 40 and / or to avoid a disruption of the stereovision method applied by the central unit 20 to determine the distance between the vehicle 1 and the obstacle 40 detected on the trajectory of the vehicle 1.

[0084] In order to be able to pivot the image acquisition means around the second axis of rotation so as to center the images acquired by the image acquisition means on the detected obstacle 40, at least one mobile image acquisition means 12 of the system is mobile in rotation around a first axis of rotation, generally horizontal, and around a second axis of rotation, generally vertical, as explained above in the description.

[0085] Figure 13 shows an example of the images obtained using a stereovision method applied by the central unit 20, which allows for the evaluation of the distance between the motor vehicle 1 and the detected obstacle 40. In these images, the pixels corresponding to the obstacle 40 are identified, and the corresponding 3D point is calculated. To do this, the central unit 20 draws the line connecting the pixels corresponding to the obstacle 40 in the image acquired by the image acquisition means 2, mounted on the front of the vehicle 1, and the center of said image acquisition means 2. Similarly, the central unit 20 draws the line connecting the pixels corresponding to the obstacle 40 in the image acquired by the mobile image acquisition means 12 and the center of said mobile image acquisition means 12. The desired 3D point corresponds to the intersection of the two lines.Calculating the 3D point of the detected obstacle 40 then allows us to evaluate the distance between the vehicle 1 and said obstacle 40.

[0086] Another embodiment can be envisaged for the maneuvering and parking assistance system 10. Such an embodiment of the system is illustrated in [Fig. 14].

[0087] In this second embodiment, the system comprises two means for acquiring moving images 121, 122. These two means for acquiring moving images 121, 122 may be identical, that is to say, have the same technical characteristics, or may be different.

[0088] The two moving image acquisition means 121, 122 are arranged on a support 14 located above the vehicle 1. The support 14 may be in the form of a rail or a lattice structure. If the infrastructure 100 includes a ceiling, the support 14 is, for example, arranged at the level of said ceiling. Alternatively, the support 14 may be in the form of an articulated crane on which the two image acquisition means 121, 122 are arranged. The two moving image acquisition means 121, 122 are oriented towards the ground of the infrastructure 100 so as to acquire images of an environment in front of the vehicle 1 traveling within the infrastructure 100.

[0089] In this second embodiment, a first mobile image acquisition means 121 is aligned vertically with the image acquisition means 2 arranged at the front of said vehicle 1 and a second mobile image acquisition means 122 is juxtaposed longitudinally in front of the first mobile image acquisition means 121. The first mobile image acquisition means 121 and the second mobile image acquisition means 122 are spaced a few hundred millimeters apart, for example 200mm.

[0090] In this second embodiment of the system 10, the control system 16 is configured to move the two mobile image acquisition means 121, 122 so as to acquire images of an environment in front of the vehicle 1 traveling autonomously within the infrastructure 100 along a predetermined trajectory. Figure 15 shows an example of the images acquired by the two mobile image acquisition means 121, 122. The two mobile image acquisition means 121, 122 of the system 10 thus each acquire a series of images from different viewpoints of the vehicle 1 traveling within the infrastructure 100 along a predetermined trajectory.

[0091] In this second embodiment of the system 10 for assisting the maneuvering and parking of a vehicle 1 in an infrastructure 100, the image acquisition means 2 of the vehicle 1 is used only for the initial detection of an obstacle 40 on the predetermined path intended to be followed by said vehicle 1. It is the images acquired by the two moving image acquisition means 121, 122 that are used by the central unit 20 to apply the stereovision method in order to evaluate the distance between the vehicle 1 and the obstacle 40.

[0092] Optionally, the control system 16 can be configured to move the second mobile image acquisition means 122 independently of the first mobile image acquisition means 121. In this option, the two mobile image acquisition means 121, 122 are not necessarily moved synchronously. Thus, the second mobile image acquisition means 122 can be moved away from the first mobile image acquisition means 121, for example to The second mobile image acquisition device 122 can be moved closer to the detected obstacle 40 in order to make stereovision more efficient and the distance calculations between vehicle 1 and obstacle 40 more precise. In other words, the second mobile image acquisition device 122 can be moved further away from the first mobile image acquisition device 121 in order to acquire images in which the detected obstacle 40 is located at the center of the images. This increases the reliability of the results from the stereovision method applied by the central unit 20 to the images acquired by the two mobile image acquisition devices 121 and 122, and thus allows for a more precise evaluation of the distance between vehicle 1 and the detected obstacle 40 along the trajectory of vehicle 1 traveling within the infrastructure 100.

Claims

1. Demands System (10) for assisting in the maneuvering and parking of a vehicle (1) in an infrastructure (100), the vehicle (1) being equipped with an image acquisition means (2) disposed at the front of said vehicle (1), and a first communication module (3), the system (10) being characterized in that it comprises: - at least one mobile image acquisition means (12) in translation along a first direction on a support (14) of the infrastructure (100), the mobile image acquisition means (12) being configured to acquire images of an environment in front of the vehicle (1) traveling autonomously in the infrastructure (100) following a predetermined trajectory; - a control system (16) configured to move the mobile image acquisition means (12) on the support (14) of the infrastructure (100) so as to follow the vehicle (1) circulating in the infrastructure (100), the control system (16) and at least one mobile image acquisition means (12) being equipped with a second communication module (13); - a central unit (20) comprising a third communication module (23) configured to receive images acquired by the image acquisition means (2) located at the front of said vehicle (1) and images acquired by the mobile image acquisition means (12), the central unit (20) being configured to apply image processing to said images enabling the identification of an obstacle (40) on the trajectory of the vehicle (1) traveling in the infrastructure (100), the third communication module (23) of the central unit (20) also being configured to communicate with the first communication module (3) of the vehicle (1) to exchange data concerning said vehicle (1), and configured to communicate with the second communication module (13) of the steering system (16) and the mobile image acquisition means (12), the third communication module (23) of the central unit (20) being configured to transmit positioning coordinates,including GPS coordinates of a trajectory, to the first communication module (3) of the vehicle (1) and / or positioning coordinates to the second module of, communication (13) of the control system (16) and of the mobile image acquisition means (12).

2. System according to any one of the preceding claims, characterized in that the mobile image acquisition means (12) is configured to reproduce the image focal length and the horizontal and vertical image resolutions of the image acquisition means (2) of the vehicle (1).

3. System according to any one of the preceding claims, characterized in that the translationally mobile image acquisition means (12) is also rotationally mobile about a first axis of rotation so that the control system (16) can pivot the mobile image acquisition means (12) about said first axis of rotation.

4. System according to the preceding claim, characterized in that the mobile image acquisition means (12) is also mobile in rotation about a second axis of rotation so that the control system (16) can pivot the mobile image acquisition means (12) about said second axis of rotation and in that this second axis of rotation is substantially perpendicular to the first axis of rotation.

5. System according to any one of the preceding claims, characterized in that the mobile image acquisition means (12) is also mobile in translation along a second direction on a support (14) of the infrastructure (100), the second direction being substantially perpendicular to the first direction.

6. System (10) according to claim 1, characterized in that it comprises two mobile image acquisition means (121, 122), the two image acquisition means being arranged on a support (14) located above the vehicle (1), the two mobile image acquisition means (121, 122) being oriented towards the ground of the infrastructure (100) so as to acquire images of an environment in front of the vehicle (1) traveling in the infrastructure (100), a first mobile image acquisition means (121) being vertically aligned with the image acquisition means (2) disposed in front of said vehicle (1), a second mobile image acquisition means (122) being longitudinally juxtaposed in front of the first mobile image acquisition means (121).

7. System according to the preceding claim, characterized in that the control system (16) is configured to move the second mobile image acquisition means (122) independently of the first mobile image acquisition means (121).

8. System according to any one of the preceding claims, characterized in that the support (14) comprises a rail or arrangement of rails and in that the control system (16) is configured to move at least one mobile image acquisition means (12) along said rail or along said rails of the arrangement.

9. A method for assisting in the maneuvering and parking of a vehicle (1) from a system according to any one of the preceding claims, characterized in that it comprises the following steps: - the first communication module (3) of the vehicle (1) sends data from said vehicle (1) to the third communication module (23) of the central unit (20); - the mobile image acquisition means (12) is positioned at the same longitudinal coordinate as the image acquisition means (2) arranged on the vehicle (1); - the image acquisition means (2) arranged on the vehicle (1) acquires a first image of the environment in front of the vehicle (1); - the first communication module (3) of the vehicle (1) sends the first acquired image to the third communication module (23) of the central unit (20) of the infrastructure (100);- The central unit (20) generates a distortion-free image from the image sent by the first communication module (3) of the vehicle (1); - The position of the mobile image acquisition means (12) is updated; - The mobile image acquisition means (12) acquires a second image of the environment in front of the vehicle (1); - The second communication module (23) of the mobile image acquisition means (12) sends the second acquired image to the third communication module (23) of the central unit (20) of the infrastructure (100); - The central unit (20) generates a distortion-free image from the received image;

10. - the central unit (20) applies image processing to the distortion-free images to detect one or more potential obstacles (40) on the trajectory of the vehicle (1). A method according to the preceding claim based on a system according to claim 4, characterized in that it also comprises the following steps when an obstacle (40) is detected on the trajectory of the vehicle (1): - the third communication module (23) of the central unit (20) sends a command to the second communication module (13) of the control system (16) of the mobile image acquisition means (12) to rotate said mobile image acquisition means (12) around the second axis of rotation so as to center the images acquired by the mobile image acquisition means (12) on the detected obstacle (40); - the mobile image acquisition means (12) acquires a third image of the environment in front of the vehicle (1); - the second communication module (23) of the mobile image acquisition means (12) sends the third acquired image to the third communication module (23) of the central unit (20) of the infrastructure (100); - the central unit (20) generates a distortion-free image from the images received by the second communication module (13) of the mobile image acquisition means (12) and by the first communication module (3) of the image acquisition means (2) of the vehicle (1); - the central unit (20) applies a stereovision method to evaluate the distance between the vehicle (1) and the detected obstacle (40).

Images