Method for determining the positioning of a vehicle relative to a parking space and the width of the parking space

Abstract

Method for determining the positioning of a vehicle relative to a box and the width of the box. The invention relates to a method for determining the positioning of a vehicle relative to a box and the width of the box, the vehicle having headlights each projecting light in the form of a light trail towards the box, the light trails intersecting at a first point, the method comprising: Generation by a camera of the vehicle, of a first image including the box illuminated by the vehicle; Obtaining a third image from the first image and a second image including the light trail of the headlights in the absence of an obstacle; Projection into a coordinate system attached to the ground, of the third image, of a line representing a distance between the vehicle and the nearest obstacle, and of a second point located in the middle of the vehicle, the longitudinal distance between the line and the second point giving a longitudinal offset;Adding a line through the first and second points to the projection; Extracting the contours of the third projected image to obtain the width of the box, the transverse distance between the second point and a third point located in the middle of the box opening giving a transverse offset, and the angle between the line and a longitudinal axis passing through the third point giving an angular offset.

Description

Title of the invention: Method for determining the positioning of a vehicle relative to a parking space and the width of the parking space. TECHNICAL FIELD OF THE INVENTION

[0001] The technical field of the invention is that of vehicles, and more particularly that of vehicles equipped with an autonomous parking function.

[0002] The present invention relates to a method for determining the positioning of a vehicle relative to a parking space, and in particular a method for determining the positioning of a vehicle relative to a parking space and the width of the parking space. The invention also relates to a parking method, a vehicle, and an associated computer program. TECHNOLOGICAL BACKGROUND OF THE INVENTION

[0003] The autonomous parking function, or Remote Parking (RPK), is a feature that allows the driver to park a vehicle in a parking space located, for example, in a garage, while outside the vehicle and using a connected device, such as a smartphone. This function is particularly advantageous when the parking space is narrow and does not allow the vehicle's occupants to open the doors wide enough to exit the vehicle once it is parked.

[0004] For the autonomous parking maneuver to succeed, it is essential that the vehicle be correctly positioned in front of the parking space and therefore in front of the garage if the parking space is located within a garage. To this end, suppliers provide recommendations regarding certain vehicle positioning parameters. As illustrated in [Fig. 1], the positioning parameters are, for example, an angular offset DA between a longitudinal axis of the vehicle 3011 passing through the middle of the vehicle 301 and a longitudinal axis 3001 of the garage 300 passing through the middle of the garage 300; a longitudinal offset DL; and a transverse offset DT between a point 3015 of the vehicle 301 located on the longitudinal axis of the vehicle 3011 and closest to the garage 300, and a point 3003 of the garage 300 located on the longitudinal axis 3001 of the garage 300 and closest to the vehicle 301.

[0005] However, positioning a vehicle according to such recommendations is far from easy for a non-expert driver. This difficulty can be further increased if the maneuvering area in front of the parking space is constrained in terms of dimensions, such as in a corridor serving two rows of parking spaces in the underground parking garage of a building.

[0006] Furthermore, the description of the parking space in terms of dimensions and the description of its positioning relative to the vehicle is imprecise, making it impossible to anticipate the parking maneuver with the help of the autonomous parking function. Indeed, the ultrasonic sensors cannot detect a sharp angle such as that found at the entrance to a parking space, and neither the ultrasonic sensors nor the aerial view of the vehicle can provide precise information about the parking space, such as its width or length.

[0007] There is therefore a need to determine more precisely the positioning of a vehicle relative to a parking space, as well as the width of the space, to ensure the success of a maneuver performed by the autonomous parking function. Summary of the invention

[0008] The invention offers a solution to the problems mentioned above, by making it possible to determine the positioning of a vehicle in relation to a parking space located in a box as well as the width of the box, in order to increase the chances of success of a maneuver carried out via an autonomous parking function.

[0009] A first aspect of the invention relates to a method for determining the positioning of a vehicle relative to a box and the width of the box, the box having an opening, the vehicle being located in front of the box and comprising at least a first headlight and a second headlight each projecting illumination in the form of a light trail towards the box, the light trail of the first headlight intersecting with the light trail of the second headlight at a first point, the method comprising the following steps: • Generation by a vehicle camera of a first image including the box illuminated by the first headlight and the second headlight of the vehicle; • Obtaining a third image from the first image and a second image, the second image containing the light trail of the first lighthouse and the light trail of the second lighthouse in the absence of an obstacle; • Projection into a ground-linked coordinate system, of the third image, of a line representing a distance between the vehicle and at least one obstacle closest to the vehicle, and of a second point located in the middle of the vehicle, the line corresponding to the opening of the box and the longitudinal distance between the line and the second point giving a longitudinal offset between the vehicle and the box; • Adding to the projection, a line passing through the first point and the second point; • Extracting the contours of the third projected image to obtain the width of the box, the transverse distance between the second point and a third point located in the middle of the opening of the box giving a transverse offset between the vehicle and the box and the angle between the line and a longitudinal axis passing through the third point giving an angular offset between the vehicle and the box.

[0010] Thanks to the invention, the light trace left by the vehicle's headlights in the box is projected into the ground reference frame to determine certain positioning parameters of the vehicle relative to the box and the width of the box, and thus facilitate a subsequent maneuver of parking the vehicle in the box, and in particular a maneuver carried out by an autonomous parking function.

[0011] In addition to the characteristics mentioned in the preceding paragraph, the process according to the first aspect of the invention may have one or more complementary characteristics from among the following, considered individually or according to all technically possible combinations.

[0012] According to one embodiment, the vehicle camera is a front camera if the first headlight and the second headlight are located at the front of the vehicle and a rear vehicle camera if the first headlight and the second headlight are located at the rear of the vehicle.

[0013] According to a sub-variant of the previous variant, the projection step is carried out using a virtual camera located in the center of the front camera or the rear camera of the vehicle.

[0014] According to an embodiment compatible with the previous embodiment, the second image is generated during the vehicle development phase.

[0015] According to an embodiment compatible with the previous embodiments, the distance is obtained via a plurality of ultrasonic sensors of the vehicle.

[0016] According to an embodiment compatible with the previous embodiments, the point located in the middle of the vehicle is the middle of the front of the vehicle if the first headlight and the second headlight are located at the front of the vehicle and the middle of the rear of the vehicle if the first headlight and the second headlight are located at the rear of the vehicle.

[0017] A second aspect of the invention relates to a method of parking a vehicle in a box, the parking method comprising the steps of the determination method according to the first aspect of the invention, and a step of parking the vehicle in the box using the positioning of the vehicle and the width of the box determined.

[0018] Thus, the determined information is used to park the vehicle inside the box, by the driver as driving aids, or by an autonomous parking function as calculation data for the maneuver to be performed.

[0019] According to one embodiment, the parking step is carried out using an autonomous parking function implemented on the vehicle.

[0020] A third aspect of the invention relates to a vehicle comprising means for implementing the method according to the first or second aspect of the invention.

[0021] A fourth aspect of the invention relates to a computer program product comprising instructions which, when the program is executed by a computer, lead the computer to carry out the steps of the process according to the first or second aspect of the invention.

[0022] The invention and its various applications will be better understood by reading the following description and examining the accompanying figures. BRIEF DESCRIPTION OF THE FIGURES

[0023] The figures are presented for illustrative purposes only and are not in any way limiting to the invention. • Fig. 1 illustrates a set of vehicle positioning parameters that influence the success of an autonomous parking maneuver in a parking space located in a box. • Figure 2 is a synoptic diagram illustrating the sequence of steps of a determination method according to the invention. • Fig. 3 is an example of the first image generated during a first step of the determination process according to the invention. • Fig. 4 is an example of a second image used during a second step of the determination process according to the invention. • Fig. 5 is an example of the third image obtained after the second step of the determination process according to the invention. • The [Fig.6] is an example of a projection obtained after a fifth step of the determination process according to the invention, in the absence of angular and transverse offset of the vehicle relative to the box. • Figure 7 is an example of a projection obtained after the fifth step of the method of determination according to the invention, with a transverse offset of the vehicle relative to the box. • Fig. 8 is an example of the projection obtained after the fifth step of the determination process according to the invention, with an angular offset of the vehicle relative to the box. • Figure 9 is a synoptic diagram illustrating the sequence of steps of a parking method according to the invention. • The [Fig. 10] is a schematic representation of a vehicle according to the invention. DETAILED DESCRIPTION

[0024] Unless otherwise specified, the same element appearing on different figures has a unique reference.

[0025] A first aspect of the invention relates to a method for determining, on the one hand, the positioning of a vehicle 301 in relation to a box 300 in which the driver of the vehicle 301 wishes to park his vehicle 301, and on the other hand, the width of the box 300. The sequence of steps of the method 100 is illustrated in [Fig.2].

[0026] The box 300 has an opening 3002 through which the vehicle 301 can be entered for parking inside the box 300.

[0027] As previously described with reference to the prior art and visible in [Fig. 1], the positioning of the vehicle 301 relative to the box 300 can for example be described by an angular offset DA, a longitudinal offset DL and a transverse offset DT between the vehicle 301 and the box 300.

[0028] The term "longitudinal" refers to the dimension along its length relative to a given reference frame, and "transverse" refers to the dimension along its width relative to the given reference frame and is therefore perpendicular to the longitudinal dimension. In the remainder of this description, and unless otherwise stated, the reference frame is that of box 300, and everything referred to as longitudinal is therefore along the length of box 300, and everything referred to as transverse DT is therefore along the width of box 300.

[0029] At the time of implementation of the method 100, the vehicle 301 is located in front of the box 300 and includes at least a first headlight 3012 and a second headlight 3012, each projecting light towards the box 300 in the form of a light trail. The light trail 3013 of the first headlight 3012 intersects with the light trail 3013 of the second headlight 3012 at a first point 3014, as can be seen in Figures 3 to 8 and in [Fig. 10].

[0030] Furthermore, there is no light in box 300 and the materials composing the walls of box 300 do not reflect light.

[0031] A first step 101 of the process 100 consists of generating a first image 1010 using a camera of the vehicle 301 so that the box 300 illuminated by the first headlight 3012 and the second headlight 3012 of the vehicle 301 is visible on the first image 1010 as illustrated in [Fig.3].

[0032] On [Fig.3], the back wall and the side walls of box 300 are visible.

[0033] The camera 3015 of the vehicle 301 is, for example, a front camera if the parking maneuver is carried out in forward gear, or a rear camera or Permanent rear view in English, if the parking maneuver is carried out in reverse gear.

[0034] A second step 102 of the process 100 consists of obtaining a third image 1021 as illustrated in [Fig.5], from the first image 1010 generated in the first step 101 and a second image 1020 as illustrated in [Fig.4].

[0035] The second image 1020 only includes the light trace 3013 of the first headlight 3012 and the light trace 3013 of the second headlight 3012 in the absence of an obstacle and was for example previously generated during a development phase of the vehicle 301.

[0036] The third image 1021 is obtained for example by subtracting the pixels of the second image 1020 from the pixels of the first image 1010, then by filtering, keeping only the pixels having the luminance characteristics of the headlights.

[0037] The third image 1021 only includes the light trace 3013 of the first lighthouse 3012 and the light trace 3013 of the second lighthouse 3012 visible in the first image 1010.

[0038] A third step 103 of the method 100 consists of projecting the third image 1021, a line 1031 representing a distance between the vehicle 301 and at least one obstacle closest to the vehicle 301, as well as a second point 1032 located in the middle of the vehicle 301, in a frame of reference attached to the ground, as illustrated in Figures 6 to 8.

[0039] Line 1031 then coincides with the opening 3002 of box 300.

[0040] The third step 103 is for example carried out using a virtual camera located in the center of the camera 3015 of the vehicle 301.

[0041] Line 1031 is obtained for example via a plurality of ultrasonic sensors 3016 of the vehicle 301 located at the front or rear of the vehicle 301, depending on whether the parking maneuver is carried out in forward or reverse.

[0042] The second point 1032 is for example the middle of the front of the vehicle 301 or the middle of the rear of the vehicle 301 depending on whether the parking maneuver is carried out in forward or reverse.

[0043] As illustrated in figures 6 to 8, the longitudinal distance between line 1031 and second point 1032 gives a longitudinal offset DL between vehicle 301 and box 300.

[0044] A fourth step 104 of the process 100 consists of adding to the projection 1033 obtained in the third step 103, a straight line 1041 passing through the first point 1014 and through the second point 1032 as illustrated in figures 6 to 8.

[0045] A fifth step 105 of the process 100 consists of extracting the contours 1051 of the third image 1021 in the projection 1033.

[0046] Contour extraction is for example carried out by applying a filter, such as a Prewitt, Sobel or Canny filter.

[0047] The extracted longitudinal contours 1051 correspond to the side walls of box 300 and the transverse distance between the extracted longitudinal contours 1051 then corresponds to the width 3004 of box 300.

[0048] The transverse distance between the second point 1042 and a third point 3003 located in the middle of the opening 3002 of the box 300 gives the transverse offset DT between the vehicle 301 and the box 300, as illustrated in [Fig.7].

[0049] The angle between the line 1051 and the longitudinal axis 3001 of the box 300 passing through the third point 3003 gives the angular offset DA between the vehicle 301 and the box 300, as illustrated in [Fig.8].

[0050] The coordinates of the third point 3003 can be determined using the line 1031 corresponding to the opening 3002 of box 300 and the extracted longitudinal contours 1051 corresponding to the side walls of box 300.

[0051] A second aspect of the invention relates to a method of parking the vehicle 301 in the box 300, the sequence of which is shown in [Fig.9].

[0052] The process 200 comprises the steps of the process 100 according to the first aspect of the invention and a step 201 of parking the vehicle 301 in the box 300.

[0053] The parking step 201 uses the vehicle positioning parameters 301 and the box width 300 determined by the determination method 100.

[0054] According to one embodiment, the parking step 201 is carried out using an autonomous parking function implemented on the vehicle 301.

[0055] The autonomous parking function allows the driver to park his vehicle in a parking space while outside the vehicle, without having to perform the parking maneuver himself.

[0056] Once the vehicle 301 is positioned in front of the box 300, the autonomous parking function is triggered by the driver using a connected object, for example a smartphone or tablet, and the vehicle 301 parks itself autonomously in the box 300.

[0057] According to another embodiment, the parking step 201 is carried out by the driver of the vehicle 301 using the positioning parameters of the vehicle 301 and the width of the box 300 determined as driving aids.

[0058] A third aspect of the invention relates to a vehicle 301 enabling the implementation of the method 100 according to the first aspect of the invention and / or of the method 200 according to the second aspect of the invention.

[0059] As illustrated in [Fig.10], the vehicle 301 comprises at least a first headlight 3012 and a second headlight 3012, a camera 3015, a plurality of ultrasonic sensors 3016 and an on-board computer 3017.

[0060] The vehicle 301 includes, for example, 6 ultrasonic sensors 3016 at the front and 6 ultrasonic sensors 3016 at the rear.

[0061] The vehicle 301 may also include a plurality of peripheral cameras, or surround view in English, to implement the autonomous parking function, using the plurality of ultrasonic sensors 3016.

Claims

1. Demands Method (100) for determining the positioning of a vehicle (301) relative to a box (300) and of a width (3004) of the box (300), the box (300) having an opening (3002), the vehicle (301) being located in front of the box (300) and having at least a first headlight (3012) and a second headlight (3012) each projecting illumination in the form of a light trail (3013) towards the box (300), the light trail (3013) of the first headlight (3012) intersecting with the light trail (3013) of the second headlight (3012) at a first point (3014), the method (100) comprising the following steps: - Generation (101) by a camera (3015) of the vehicle (301), of a first image (1010) comprising the box (300) lit by the first headlight (3012) and the second headlight (3012) of the vehicle (301); - Obtaining (102) a third image (1021) from the first image (1010) and a second image (1020), the second image (1020) containing the light trail (3013) of the first lighthouse (3012) and the light trail (3013) of the second lighthouse (3012) in the absence of an obstacle; - Projection (103, 1033) in a ground-linked frame of reference, of the third image (1021), of a line (1031) representing a distance between the vehicle (301) and at least one obstacle closest to the vehicle (301), and of a second point (1032) located in the middle of the vehicle (301), the line (1031) corresponding to the opening (3002) of the box (300) and the longitudinal distance between the line (1031) and the second point (1032) giving a longitudinal offset (DL) between the vehicle (301) and the box (300); - Addition (104) in the projection (1033), of a line (1041) passing through the first point (1014) and the second point (1032); - Extraction (105) of the contours (1051) of the third image (1021) projected to obtain a width of the box (300), the transverse distance between the second point (1042) and a third point (3003) located in the middle of the opening (3002) of the box (300) giving a transverse offset (DT) between the vehicle (301) and the box (300) and the angle between the line (1051) and a longitudinal axis (3001) passing through the third point (3003) giving an angular offset (DA) between the vehicle (301) and the box (300).

2. Method (100) according to claim 1, wherein the camera (3015) of the vehicle (301) is a front camera if the first headlight (3012) and the second headlight (3012) are located at the front of the vehicle (301) and a rear camera of the vehicle (301) if the first headlight (3012) and the second headlight (3012) are located at the rear of the vehicle (301).

3. Method (100) according to claim 2, wherein the projection step (103) is carried out using a virtual camera located in the center of the front camera or the rear camera of the vehicle (301).

4. Method (100) according to any one of the preceding claims, wherein the second image is generated during the vehicle development phase (301).

5. Method (100) according to any one of the preceding claims, wherein the distance is obtained via a plurality of ultrasonic sensors (3016) of the vehicle (301).

6. Method (100) according to any one of the preceding claims, wherein the point located in the middle of the vehicle (301) is the middle of the front of the vehicle (301) if the first headlight (3012) and the second headlight (3012) are located at the front of the vehicle (301) and the middle of the rear of the vehicle (301) if the first headlight (3012) and the second headlight (3012) are located at the rear of the vehicle (301).

7. Method (200) of parking a vehicle (301) in a box (300), the parking method (200) comprising the steps of the determination method (100) according to any one of the preceding claims, and a step (201) of parking the vehicle (301) in the box (300) using the determined positioning of the vehicle (301) and the width of the box (300).

8. Method (200) according to claim 7, wherein the parking step (201) is carried out using an autonomous parking function implemented on the vehicle (301).

9. Vehicle (301) comprising means for implementing the method (100, 200) according to any one of the preceding claims.

10. Product computer program comprising instructions which, when the program is executed by a computer, cause the computer to carry out the steps of the process (100, 200) according to any one of claims 1 to 8.