Vehicle positioning error correction method and device in automatic parking, medium and equipment

Abstract

The invention relates to the field of computer models, and discloses a vehicle positioning error correction method and device in automatic parking, a medium and equipment. The method comprises the following steps: entering a parking space searching stage when an automatic parking request is received, and determining a global coordinate system; environment sensing data under a global coordinate system are tracked and recorded in real time in the parking space searching stage, all parking spaces in the searching environment are determined, and all the parking spaces are displayed on a vehicle screen; after a user selects a target parking space, entering a vehicle parking stage, and determining a parking coordinate system; converting the environmental perception data under the global coordinate system tracked and recorded in the parking space searching stage into a parking coordinate system, and planning a parking path; automatic parking is carried out, and environment sensing data recorded in a parking coordinate system are tracked in real time in the parking process; and determining the real-time deviation of the reference object, and carrying out error correction on the real-time positioning of the vehicle. According to the invention, the number of times of garage rolling is reduced, and the problems of skew parking posture and collision of the vehicle are avoided.

Description

Technical Field

[0001] The present invention relates to the field of intelligent parking technology, and in particular to a method and device, medium and equipment for correcting vehicle positioning errors in automatic parking. Background Art

[0002] The automatic assisted parking system achieves precise parking of the vehicle by sensing the vehicle's surrounding environment and locating the vehicle's position. Therefore, it is very important to accurately locate the vehicle, target parking space and surrounding environment during the parking process.

[0003] During the parking process, the position of the vehicle changes, and the conventional vehicle positioning method is based on the estimation of chassis signals. The estimation process cannot guarantee its accuracy. Therefore, the error in the change of vehicle position will cause the position of the target parking space and obstacles in the coordinate system to shift, causing the parking posture of the vehicle to be skewed, and even a collision in severe cases. Summary of the invention

[0004] In order to solve at least one of the above technical problems, the present invention provides a method, device, medium and equipment for correcting vehicle positioning errors in automatic parking.

[0005] According to a first aspect, a method for correcting vehicle positioning errors in automatic parking provided by an embodiment of the present invention includes:

[0006] Entering a parking space search phase upon receiving an automatic parking request, and determining a global coordinate system according to an initial positioning of the vehicle when entering the parking space search phase;

[0007] In the parking space search phase, the environmental perception data in the global coordinate system is tracked and recorded in real time, each parking space in the search environment is determined according to the environmental perception data in the global coordinate system, and each parking space is displayed on the vehicle screen;

[0008] When the user selects a target parking space from among the parking spaces displayed on the vehicle screen, the vehicle enters the parking phase, and a parking coordinate system is determined according to the target parking space;

[0009] Converting the environmental perception data in the global coordinate system tracked and recorded in the parking space search phase to the parking coordinate system, and planning a parking path according to the environmental perception data converted to the parking coordinate system;

[0010] Automatically parking the vehicle according to the parking path, and tracking and recording the environmental perception data in the parking coordinate system in real time during the parking process;

[0011] A real-time deviation of at least one reference object is determined according to the environmental perception data in the parking coordinate system, and an error correction is performed on the real-time positioning of the vehicle according to the real-time deviation of the at least one reference object.

[0012] In one embodiment, determining the global coordinate system according to the initial positioning of the vehicle when entering the parking space search phase includes:

[0013] The center position of the rear axle of the vehicle when the vehicle enters the parking space search phase is taken as the origin; the direction of the front of the vehicle when the vehicle enters the parking space search phase is taken as the positive direction of the horizontal coordinate axis, and the direction perpendicular to the horizontal coordinate axis and pointing to one side of the vehicle body is taken as the vertical coordinate axis; the heading angle of the vehicle when entering the parking space search phase is set to 0°;

[0014] And / or, determining a parking coordinate system according to the target parking space includes:

[0015] The preset position in the target parking space is taken as the origin; the direction perpendicular to the line between the two entrance corner points of the target parking space and pointing to the entrance of the target parking space is the positive direction of the longitudinal axis, and the direction perpendicular to the longitudinal axis and pointing to one side of the vehicle body is the horizontal axis; the heading angle of the positive direction of the horizontal axis is set to 0°.

[0016] In one embodiment, the environmental perception data tracked and recorded in the parking space search phase includes: a two-dimensional map corresponding to the vehicle communicable area determined based on each obstacle in the search environment;

[0017] Correspondingly, planning the parking path according to the environmental perception data converted into the parking coordinate system includes: planning the parking path according to the two-dimensional map in the environmental perception data converted into the parking coordinate system.

[0018] In one embodiment, the environment perception data tracked and recorded in the parking space search phase includes lane line information;

[0019] Correspondingly, determining the real-time deviation of at least one reference object according to the environmental perception data in the parking coordinate system, and performing error correction on the real-time positioning of the vehicle according to the real-time deviation of the at least one reference object, includes:

[0020] Calculate the real-time confidence of the lane line based on the lane line information currently tracked and recorded;

[0021] When the real-time confidence level of the lane line is higher than a first preset confidence level, selecting the lane line as the first reference object;

[0022] Comparing the lane line heading currently tracked and recorded for the first reference object with the lane line heading tracked and recorded during the parking space search phase in real time;

[0023] Determine the real-time deviation of the lane heading of the first reference object according to the real-time comparison result;

[0024] The real-time rotation amount of the parking coordinate system is determined according to the real-time deviation of the lane line heading of the first reference object, and the real-time error correction of the heading of the vehicle is performed according to the real-time rotation amount of the parking coordinate system.

[0025] In one embodiment, the environmental perception data tracked and recorded in the parking space search stage also includes the coordinates of the corner points of the parking space and the coordinates of the obstacle closest to the entrance corner point of the parking space;

[0026] Correspondingly, determining the real-time deviation of at least one reference object according to the environmental perception data in the parking coordinate system, and performing error correction on the real-time positioning of the vehicle according to the real-time deviation of the at least one reference object, further includes:

[0027] Calculate the distance from each parking space corner point to the corresponding side fisheye, and calculate the distance from the obstacle closest to the parking space entrance corner point to the corresponding side fisheye;

[0028] According to the distances from each parking space corner point to the corresponding side fisheye, the real-time confidence of each parking space corner point is calculated respectively; according to the distance from the obstacle closest to the entrance corner point of the parking space to the corresponding side fisheye, the real-time confidence of the obstacle closest to the entrance corner point of the parking space is calculated;

[0029] When there is a real-time confidence higher than the second preset confidence among the real-time confidences of the corner points of the parking spaces and the obstacles closest to the entrance corner points of the parking spaces, the corner points of the parking spaces or the obstacles corresponding to the real-time confidence higher than the second preset confidence are selected as the second reference objects;

[0030] Comparing the reference object coordinates currently tracked and recorded for the second reference object with the reference object coordinates tracked and recorded during the parking space search phase in real time;

[0031] According to the real-time comparison result, the real-time positioning deviation of the second reference object is determined, and the horizontal and vertical coordinates of the vehicle are corrected in real time according to the real-time positioning deviation of the second reference object.

[0032] In one embodiment, the method further comprises:

[0033] After planning the parking path, a first sub-stage, a second sub-stage and a third sub-stage of the parking process are defined; wherein the first sub-stage is a stage in which the vehicle backs up for the first time, the second sub-stage is a stage in which the vehicle moves forward to straighten the vehicle body, and the third sub-stage is a stage in which the vehicle backs up into the garage;

[0034] Determining the sub-stage of the parking process according to the vehicle action information during the parking process;

[0035] Correspondingly, before determining the real-time rotation amount of the parking coordinate system according to the real-time lane heading deviation of the first reference object, the method further includes:

[0036] If the parking process is in the first sub-stage or the second sub-stage, determining whether the lane line heading real-time deviation is higher than a first direction deviation threshold; if the lane line heading real-time deviation is higher than the first direction deviation threshold, determining the real-time rotation amount of the parking coordinate system according to the lane line heading real-time deviation of the first reference object;

[0037] If the parking process is in the third sub-stage, determining whether the lane line heading real-time deviation exceeds a second direction deviation threshold; if the lane line heading real-time deviation exceeds the second direction deviation threshold, determining the real-time rotation amount of the parking coordinate system according to the lane line heading real-time deviation of the first reference object; wherein the second direction deviation threshold is less than the first direction deviation threshold;

[0038] And / or, correspondingly, before performing real-time error correction on the horizontal and vertical coordinates of the vehicle according to the real-time positioning deviation of the second reference object, the method further includes:

[0039] If the parking process is in the first sub-stage or the second sub-stage, determining whether the real-time positioning deviation of the second reference object is higher than a first position deviation threshold; if the real-time positioning deviation of the second reference object is higher than the first position deviation threshold, correcting the horizontal and vertical coordinates of the vehicle according to the real-time positioning deviation of the second reference object;

[0040] If the parking process is in the third sub-stage, determine whether the real-time positioning deviation of the second reference object is higher than a second position deviation threshold; if the real-time positioning deviation of the second reference object is higher than the second position deviation threshold, perform the correction of the horizontal and vertical coordinates of the vehicle according to the real-time positioning deviation of the second reference object; wherein, the second position deviation threshold is lower than the first position deviation threshold.

[0041] In one embodiment, the relationship between the real-time confidence of the lane line and the first distance conforms to a linear relationship or a normal distribution relationship; the smaller the first distance is, the greater the real-time confidence corresponding to the lane line is; the first distance is the distance between the lane line and the line connecting the left and right fisheyes;

[0042] And / or, the relationship between the real-time confidence of each parking space corner point and the second distance conforms to a linear relationship or a normal distribution relationship; the smaller the second distance is, the greater the real-time confidence corresponding to the parking space corner point is; the second distance is the distance between the parking space corner point and the corresponding side fisheye;

[0043] And / or, the relationship between the real-time confidence of the obstacle closest to the entrance corner point of the parking space and the third distance conforms to a linear relationship or a normal distribution relationship; the smaller the third distance is, the greater the real-time confidence corresponding to the obstacle closest to the entrance corner point of the parking space is; the third distance is the distance between the obstacle closest to the entrance corner point of the parking space and the corresponding side fisheye.

[0044] According to a second aspect, a vehicle positioning error correction device in automatic parking provided by an embodiment of the present invention comprises:

[0045] A first determination module, configured to enter a parking space search phase upon receiving an automatic parking request, and determine a global coordinate system according to an initial positioning of the vehicle when entering the parking space search phase;

[0046] A parking space search module, used for tracking and recording the environmental perception data in the global coordinate system in real time during the parking space search phase, determining each parking space in the search environment according to the environmental perception data in the global coordinate system, and displaying each parking space on the vehicle screen;

[0047] A second determination module, configured to enter a vehicle parking phase after the user selects a target parking space from among the parking spaces displayed on the vehicle screen, and determine a parking coordinate system according to the target parking space;

[0048] a path planning module, used to convert the environmental perception data in the global coordinate system tracked and recorded in the parking space search phase into the parking coordinate system, and plan a parking path according to the environmental perception data converted into the parking coordinate system;

[0049] An automatic parking module, used to perform automatic parking according to the parking path, and to track and record the environmental perception data in the parking coordinate system in real time during the parking process;

[0050] The positioning correction module is used to determine the real-time deviation of at least one reference object according to the environmental perception data in the parking coordinate system, and perform error correction on the real-time positioning of the vehicle according to the real-time deviation of the at least one reference object.

[0051] According to the third aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed in a computer, the computer is caused to execute the method provided in the first aspect.

[0052] According to a fourth aspect, a computing device provided by an embodiment of the present invention includes a memory and a processor, wherein the memory stores executable code, and when the processor executes the executable code, the method provided by the first aspect is implemented.

[0053] The embodiment of the present invention has the following technical effects: a global coordinate system is determined according to the initial positioning of the vehicle when entering the parking space search stage, environmental perception data under the global coordinate system is tracked and recorded in real time during the parking space search stage, each parking space in the search environment is determined according to the environmental perception data under the global coordinate system, and each parking space is displayed on the vehicle screen. When the user selects a target parking space from the parking spaces displayed on the vehicle screen, the vehicle enters the parking stage, and a parking coordinate system is determined according to the target parking space, and then the environmental perception data under the global coordinate system tracked and recorded during the parking space search stage is converted to the parking coordinate system, and then a parking path is planned based on the environmental perception data converted to the parking coordinate system, and then automatic parking is performed according to the parking path, and the environmental perception data under the parking coordinate system is tracked and recorded in real time during the parking process, so as to determine the real-time deviation of at least one reference object, and the real-time positioning of the vehicle is corrected according to the real-time deviation of at least one reference object. During the parking process, the system keeps tracking and memorizing environmental perception data, determines the real-time deviation of the reference object based on the environmental perception data, and then corrects the positioning error of the vehicle according to the real-time deviation of the reference object. In this way, the positioning error of the vehicle is compensated. This not only reduces the number of times the parking lot is maneuvered and thus improves the parking efficiency, but also ensures the parking accuracy and avoids the problem of skew parking posture and collision of the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0055] Figure 1 is a flow chart of a method for correcting vehicle positioning errors in automatic parking according to an embodiment of the present invention;

[0056] Figure 2 is a schematic diagram of a parking space in one embodiment of the present invention;

[0057] Figure 3 is a schematic diagram of an actual scene in an embodiment of the present invention;

[0058] Figure 4 It is a structural block diagram of a vehicle positioning error correction device in automatic parking in one embodiment of the present invention. DETAILED DESCRIPTION

[0059] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.

[0060] In a first aspect, an embodiment of the present invention provides a method for correcting vehicle positioning errors in automatic parking, see Figure 1 The method comprises the following steps S110 to S160:

[0061] S110, entering a parking space search phase upon receiving an automatic parking request, and determining a global coordinate system according to an initial positioning of the vehicle when entering the parking space search phase;

[0062] For example, when a user applies for automatic parking on a vehicle, the vehicle's automatic parking system will receive the automatic parking request and then enter the first stage: the parking space search stage. The so-called parking space search stage refers to the process of searching for various parking spaces in the surrounding environment and displaying the searched parking spaces for the user to select.

[0063] It can be seen that the position of the vehicle when entering the parking space search phase is the initial position, and the global coordinate system is determined based on the initial position.

[0064] In one embodiment, the global coordinate system is determined according to the initial positioning of the vehicle when entering the parking space search stage in S110, including: taking the center position of the rear axle of the vehicle when entering the parking space search stage as the origin; the front direction of the vehicle when entering the parking space search stage as the positive direction of the horizontal coordinate axis, and the direction perpendicular to the horizontal coordinate axis and pointing to one side of the vehicle body as the vertical coordinate axis; the heading angle of the vehicle when entering the parking space search stage is set to 0°.

[0065] It can be seen that when the vehicle enters the parking space search phase, the horizontal coordinate of the center position of the rear axle of the vehicle in the global coordinate system is 0, the vertical coordinate is 0, and the heading angle is 0.

[0066] For example, see Figure 3 , the center position of the rear axle of the vehicle in the initial position is taken as the origin of the global coordinate system, the direction of the front of the vehicle is taken as the positive direction of the horizontal coordinate axis, and the direction perpendicular to the horizontal coordinate axis and pointing to the left side of the vehicle body is taken as the positive direction of the vertical coordinate, that is, the global coordinate system adopts a left-handed coordinate system.

[0067] S120, in the parking space search phase, tracking and recording the environmental perception data in the global coordinate system in real time, determining each parking space in the search environment according to the environmental perception data in the global coordinate system, and displaying each parking space on the vehicle screen;

[0068] The environmental perception data can be collected by a perception system, which refers to a hardware perception system based on vision and / or ultrasound. The hardware perception system based on vision, for example, is a fisheye installed on the left and right sides of the vehicle, through which images of the surrounding environment can be collected.

[0069] That is to say, the perception system on the vehicle collects environmental perception data in real time, and then projects the collected environmental perception data into the global coordinate system to obtain the environmental perception data in the global coordinate system, thereby realizing the tracking and recording of the environmental perception data in the global coordinate system, which can also be called tracking memory.

[0070] The environmental perception data tracked and recorded in the parking space search phase may include: a two-dimensional map corresponding to the vehicle communicable area determined based on various obstacles in the search environment, lane line information, coordinates of parking space corner points, coordinates of obstacles closest to the entrance corner points of the parking space, etc. Lane lines refer to lines in the middle of the road used to indicate the direction of the road.

[0071] It can be seen that the two-dimensional map corresponding to the vehicle communicable area is a two-dimensional map of the area where the vehicle can pass surrounded by various obstacles in the surrounding environment. The lane line information is the curve equation of the lane line, and the lane line is represented by a cubic equation. Because the lane line is not a point, it cannot be represented by coordinates. Usually a parking space has four corner points, so the coordinates of the parking space corner points can be the coordinates of the four corner points of the parking space, and the area surrounded by the four corner points constitutes a parking space. There are two entry corner points among the four corner points, for example, entry corner points A and B. The obstacles closest to the entry corner points of the parking space may include the obstacle closest to the entry corner point A of the parking space and the obstacle closest to the entry corner point B of the parking space.

[0072] It is understandable that based on the environmental perception data in the global coordinate system, each parking space in the search environment can be determined and displayed on the vehicle screen, so that the user can see each searched parking space, each parking space has its own number.

[0073] Among them, the coordinate system used in the vehicle search stage is the global coordinate system, because the global coordinate system is more convenient for searching parking spaces.

[0074] S130, when the user selects a target parking space from among the parking spaces displayed on the vehicle screen, the vehicle enters a parking phase, and a parking coordinate system is determined according to the target parking space;

[0075] It can be seen that after the user sees various parking spaces on the vehicle screen, he can select a parking space as the target parking space according to his own habits or preferences. After the user selects the target parking space, the automatic parking system switches from the parking space search stage to the vehicle parking stage, that is, enters the second stage. The coordinate system used in the vehicle parking stage is the parking coordinate system.

[0076] In one embodiment, determining the parking coordinate system according to the target parking space in S130 may include: taking a preset position in the target parking space as the origin; a direction perpendicular to a line between two entrance corner points of the target parking space and pointing to the entrance of the target parking space as the positive direction of the longitudinal axis, and a direction perpendicular to the longitudinal axis and pointing to one side of the vehicle body as the transverse axis; and setting the heading angle of the positive direction of the transverse axis to 0°.

[0077] For example, see Figure 2 , the center point of the line between the two corner points C and D at the rear of the target parking space is taken as the origin of the parking coordinate system, the direction perpendicular to the line between the two entrance corner points A and B of the target parking space and pointing to the entrance of the target parking space is the positive direction of the longitudinal axis, the direction perpendicular to the longitudinal axis and pointing to the right side of the vehicle body is the positive direction of the horizontal axis, and the heading angle of the positive direction of the horizontal axis is set to 0°.

[0078] For another example, see Figure 3 The origin of the parking coordinate system is the position where the center of the rear axle of the vehicle should be after it is parked in the target parking space. The x-axis points to the right side of the vehicle body, and the y-axis points to the entrance of the target parking space.

[0079] S140, converting the environmental perception data in the global coordinate system tracked and recorded in the parking space search phase into the parking coordinate system, and planning a parking path according to the environmental perception data converted into the parking coordinate system;

[0080] That is, the environmental perception data in the global coordinate system previously tracked and recorded in the parking space search phase is converted to the parking coordinate system to obtain the environmental perception data in the parking coordinate system, and then the parking path is planned based on the environmental perception data in the parking coordinate system.

[0081] Among them, the path planning is carried out in the parking coordinate system, because it is more convenient to carry out path planning in the parking coordinate system.

[0082] In one embodiment, when the environmental perception data tracked and recorded in the parking space search stage includes a two-dimensional map corresponding to the vehicle communicable area determined based on various obstacles in the search environment, planning the parking path according to the environmental perception data converted to the parking coordinate system in S140 may include: planning the parking path according to the two-dimensional map in the environmental perception data converted to the parking coordinate system.

[0083] It can be seen that based on the two-dimensional map corresponding to the vehicle communication area, the parking path can be planned to obtain a planned parking path.

[0084] S150, performing automatic parking according to the parking path, and tracking and recording the environmental perception data in the parking coordinate system in real time during the parking process;

[0085] It can be seen that after the parking path is obtained, the automatic parking operation is performed according to the parking path, and during the parking process, the environmental perception data collected by the perception system in real time is projected to the parking coordinate system to obtain the environmental perception data in the parking coordinate system, thereby realizing the tracking and recording of the environmental perception data in the parking coordinate system, that is, tracking memory.

[0086] S160: Determine a real-time deviation of at least one reference object according to the environmental perception data in the parking coordinate system, and perform error correction on the real-time positioning of the vehicle according to the real-time deviation of the at least one reference object.

[0087] It can be seen that based on the environmental perception data in the parking coordinate system, at least one reference object can be determined, and then the real-time deviation of the at least one reference object can be determined. Based on the real-time deviation of the at least one reference object, the real-time positioning of the vehicle can be corrected, thereby obtaining relatively accurate vehicle positioning information, and then automatic parking can be performed based on the relatively accurate vehicle positioning information, which can improve the parking effect. That is, since the environmental perception data is tracked and memorized during the parking process, the real-time deviation of the reference object is determined based on the environmental perception data, and then the positioning error of the vehicle is corrected according to the real-time deviation of the reference object, the positioning error of the vehicle is compensated, which not only reduces the number of times of rubbing the parking lot and thus improves the parking efficiency, but also ensures the parking accuracy and thus avoids the problem of skewed parking posture and collision of the vehicle.

[0088] In one embodiment, when the environmental perception data tracked and recorded in the parking space search phase includes lane line information, the step of determining the real-time deviation of at least one reference object according to the environmental perception data in the parking coordinate system in S160, and performing error correction on the real-time positioning of the vehicle according to the real-time deviation of the at least one reference object, may include the following steps S161a to S165a:

[0089] S161 a. Calculate the real-time confidence of the lane line according to the lane line information currently tracked and recorded;

[0090] Specifically, the real-time confidence of the lane line can be determined based on the distance between the lane line and the line connecting the left and right fisheyes. For the convenience of description, the distance between the lane line and the line connecting the left and right fisheyes is defined as the first distance. The smaller the first distance, the greater the real-time confidence corresponding to the lane line. In other words, the closer the lane line is to the line connecting the left and right fisheyes, the greater the real-time confidence of the lane line. The relationship between the real-time confidence of the lane line and the first distance can conform to a linear relationship or a normal distribution relationship.

[0091] That is, in one embodiment, the relationship between the real-time confidence of the lane line and the first distance conforms to a linear relationship or a normal distribution relationship; the smaller the first distance is, the greater the real-time confidence corresponding to the lane line is; the first distance is the distance between the lane line and the line connecting the left and right fisheyes.

[0092] For example, the real-time confidence of the lane line is determined according to the normal distribution model, which is expressed as follows:

[0093] Among them, x is a variable, which represents the first distance, and of course it can also represent the second distance and the third distance mentioned below. The closer the distance, the higher the confidence. u and σ are constants. The choice of σ is very important, which controls the speed at which the confidence decreases with the degree of deviation. A smaller σ means that the confidence decreases rapidly. A larger σ means that even if the vehicle is not exactly parked in the center, a relatively high confidence can be obtained. In order to achieve correction only when the confidence is relatively high, a relatively small σ can be selected.

[0094] S162 a. Selecting the lane line as the first reference object when the real-time confidence level of the lane line is higher than a first preset confidence level;

[0095] That is, the real-time confidence of the lane line changes as the vehicle parks. The closer the vehicle is to the lane line, the higher the real-time confidence of the lane line. When the real-time confidence of the lane line is greater than the first preset confidence, the lane line is used as the first reference object.

[0096] In fact, during the entire process of the vehicle parking phase, no matter it is the first sub-stage, the second sub-stage or the third sub-stage mentioned later, the lane lines are next to or below the vehicle and the distance is very close. Therefore, in these three sub-stages, the real-time confidence of the lane lines is actually very high. Therefore, in these three sub-stages, the heading of the lane lines can be used to correct the heading of the vehicle.

[0097] S163 a. Performing a real-time comparison between the lane line heading currently tracked and recorded for the first reference object and the lane line heading tracked and recorded during the parking space search phase;

[0098] It is understandable that the lane heading will not change during the parking phase or the vehicle search phase, so the lane heading of the vehicle during the parking process and the lane heading tracked and recorded during the parking search phase should be consistent in theory. However, during the parking process, the error in vehicle positioning will cause the lane heading tracked and recorded by the perception system to deviate from the lane heading previously tracked and recorded. In fact, not only the lane heading will change due to the error in vehicle positioning, but also the coordinates of the corner points of the parking space and the coordinates of the nearest obstacle will change due to the error in vehicle positioning.

[0099] S164 a. Determine the real-time deviation of the lane heading of the first reference object according to the real-time comparison result;

[0100] S165 a. Determine a real-time rotation amount of the parking coordinate system according to the real-time deviation of the lane heading of the first reference object, and perform real-time error correction on the heading of the vehicle according to the real-time rotation amount of the parking coordinate system.

[0101] It can be seen that the real-time comparison result of the lane line heading can be used to know the real-time deviation of the lane line heading, and then the real-time rotation amount of the parking coordinate system can be known based on the real-time deviation of the lane line heading, that is, the angle at which the horizontal axis of the parking coordinate system has rotated compared to the original horizontal axis, and then the real-time error correction of the vehicle's heading can be performed based on the real-time rotation amount of the parking coordinate system.

[0102] Among them, the role of the lane line is to correct the vehicle's heading.

[0103] At this point, the error in the vehicle's heading can be corrected.

[0104] In one embodiment, when the environmental perception data tracked and recorded in the parking space search stage also includes the coordinates of the corner points of the parking space and the coordinates of the obstacle closest to the entrance corner point of the parking space, the step of determining the real-time deviation of at least one reference object according to the environmental perception data in the parking coordinate system in S160, and performing error correction on the real-time positioning of the vehicle according to the real-time deviation of the at least one reference object, may further include the following steps S161b to S165b:

[0105] S161b, calculating the distance from each parking space corner point to the corresponding side fisheye, and calculating the distance from the obstacle closest to the entrance corner point of the parking space to the corresponding side fisheye;

[0106] For example, see Figure 2, parking corner points A and C correspond to the fisheye on the left side of the vehicle, and parking corner points B and D correspond to the fisheye on the right side of the vehicle. Therefore, the distances from parking corner points A and C to the left fisheye are calculated, and the distances from parking corner points B and D to the right fisheye are calculated.

[0107] For another example, the closest obstacle to the entrance corner point A is the vehicle p parked in the parking space to the left of the target parking space, and the closest obstacle to the entrance corner point B is the vehicle q parked in the parking space to the right of the target parking space. Therefore, the distance between vehicle p and the left fisheye of vehicle k and the distance between vehicle q and the right fisheye of vehicle k are calculated.

[0108] S162b, calculating the real-time confidence of each parking space corner point according to the distance from each parking space corner point to the corresponding side fisheye; calculating the real-time confidence of the obstacle closest to the entrance corner point of the parking space according to the distance from the obstacle closest to the entrance corner point of the parking space to the corresponding side fisheye;

[0109] In one embodiment, the relationship between the real-time confidence of each parking space corner point and the second distance conforms to a linear relationship or a normal distribution relationship; the smaller the second distance is, the greater the real-time confidence corresponding to the parking space corner point is; the second distance is the distance between the parking space corner point and the corresponding side fisheye;

[0110] In one embodiment, the relationship between the real-time confidence of the obstacle closest to the entrance corner point of the parking space and the third distance conforms to a linear relationship or a normal distribution relationship; the smaller the third distance is, the greater the real-time confidence corresponding to the obstacle closest to the entrance corner point of the parking space is; the third distance is the distance between the obstacle closest to the entrance corner point of the parking space and the corresponding side fisheye.

[0111] It can be seen that the calculation process of the real-time confidence of the parking space corner point and the real-time confidence of the obstacle closest to the parking space entrance corner point is the same as the real-time confidence of the lane line mentioned above. The smaller the distance, the higher the real-time confidence, and the relationship between the distance and the real-time confidence can be a linear relationship or a normal distribution relationship.

[0112] S163b, when there is a real-time confidence level higher than the second preset confidence level among the real-time confidence levels of the parking space corner points and the obstacles closest to the entrance corner points of the parking spaces, selecting the parking space corner points or obstacles corresponding to the real-time confidence level higher than the second preset confidence level as the second reference object;

[0113] S164b, performing real-time comparison of the reference object coordinates currently tracked and recorded for the second reference object with the reference object coordinates tracked and recorded during the parking space search phase;

[0114] S165b. Determine the real-time positioning deviation of the second reference object according to the real-time comparison result, and perform real-time error correction on the horizontal and vertical coordinates of the vehicle according to the real-time positioning deviation of the second reference object.

[0115] It is understandable that at the beginning of parking, the second distance and the third distance are relatively large, and the real-time confidences are relatively small at this time, and the confidence is relatively low, so the reference object is not determined, and the horizontal and vertical coordinates of the vehicle are not corrected in real time. When the vehicle is parked, the second distance and the third distance are getting smaller and smaller, so the real-time confidence increases. When the real-time confidence increases to higher than the second preset confidence, the parking space corner point or obstacle corresponding to the real-time confidence higher than the second preset confidence is used as the second reference object, and then the coordinates of the second reference object at this time are determined and compared with the previous coordinates in real time, so as to obtain a real-time comparison result. According to the real-time comparison result, it can be known how much the coordinates of the second reference object have shifted, and then the shift is used to perform real-time error correction on the horizontal and vertical coordinates of the vehicle.

[0116] At this point, the error correction of the vehicle's horizontal and vertical coordinates is achieved.

[0117] In order to further refine the parking process, the method provided in the embodiment of the present invention may further include the following steps 1 and 2:

[0118] 1. After planning the parking path, define the first sub-stage, the second sub-stage and the third sub-stage of the parking process; wherein the first sub-stage is the stage in which the vehicle backs up for the first time, the second sub-stage is the stage in which the vehicle moves forward to straighten the vehicle body, and the third sub-stage is the stage in which the vehicle backs up into the garage;

[0119] 2. Determine the sub-stage of the parking process based on the vehicle motion information during the parking process;

[0120] It can be seen that the parking process is divided into three sub-stages, first the first backing up, then the forward movement to straighten the vehicle body, and finally the vehicle backing up into the parking garage. As for which sub-stage it is in, it can be determined based on the vehicle action information, such as steering, reversing or forward movement, etc. Figure 3 , we can see the status of the vehicle in different sub-stages.

[0121] Furthermore, before determining the real-time rotation amount of the parking coordinate system according to the real-time lane heading deviation of the first reference object in S165a, the method may further include:

[0122] (1) if the parking process is in the first sub-stage or the second sub-stage, determining whether the lane line heading real-time deviation is higher than a first direction deviation threshold; if the lane line heading real-time deviation is higher than the first direction deviation threshold, determining the real-time rotation amount of the parking coordinate system according to the lane line heading real-time deviation of the first reference object;

[0123] (2) If the parking process is in the third sub-stage, determine whether the real-time lane line heading deviation exceeds a second direction deviation threshold; if the real-time lane line heading deviation exceeds the second direction deviation threshold, determine the real-time rotation amount of the parking coordinate system according to the real-time lane line heading deviation of the first reference object; wherein the second direction deviation threshold is less than the first direction deviation threshold.

[0124] It can be seen that after calculating the real-time deviation of the lane line heading, it is first determined whether the real-time deviation of the lane line heading is greater than the direction deviation threshold. If it is greater than the direction deviation threshold, the real-time rotation amount of the parking coordinate system is determined according to the real-time deviation of the lane line heading of the first reference object. It can be seen that when the real-time deviation of the lane line heading is relatively small, the vehicle heading does not need to be corrected for error. Only when the real-time deviation of the lane line heading is relatively large, the vehicle heading is corrected for error. Moreover, the second direction deviation threshold is smaller than the first direction deviation threshold. It can be seen that the heading error correction range of the first sub-stage and the second sub-stage is relatively wide, because when the real-time deviation of the lane line heading in these two sub-stages is relatively large, it can be corrected in the third sub-stage. The heading error correction range of the third sub-stage is relatively narrow, so the heading error can be corrected in time in the third sub-stage. In this way, the continuity of the parking trajectory of the vehicle can be ensured, and the number of rubbing the parking garage is relatively small.

[0125] In other words, in order to ensure the continuity of the parking trajectory and reduce the number of parking mistakes, a correction trigger mechanism is set. The correction operation will only be triggered when the real-time deviation of the lane line heading exceeds a certain error. Otherwise, the continuity of the parking trajectory cannot be guaranteed, and it will also cause a large number of parking mistakes. Moreover, the heading deviations that occur in the first and second sub-stages can be corrected during the re-planning of the third sub-stage. Therefore, the correction trigger range in the first and second sub-stages can be relaxed. Of course, when the heading deviation is large, it must be corrected, otherwise it will be difficult to enter the parking lot in the third sub-stage, and even a collision problem will occur in severe cases.

[0126] Furthermore, before performing real-time error correction on the horizontal and vertical coordinates of the vehicle according to the real-time positioning deviation of the second reference object in S165b, the method may further include:

[0127] (1) if the parking process is in the first sub-stage or the second sub-stage, determining whether the real-time positioning deviation of the second reference object is higher than a first position deviation threshold; if the real-time positioning deviation of the second reference object is higher than the first position deviation threshold, correcting the horizontal and vertical coordinates of the vehicle according to the real-time positioning deviation of the second reference object;

[0128] (2) If the parking process is in the third sub-stage, determine whether the real-time positioning deviation of the second reference object is higher than a second position deviation threshold; if the real-time positioning deviation of the second reference object is higher than the second position deviation threshold, correct the horizontal and vertical coordinates of the vehicle according to the real-time positioning deviation of the second reference object; wherein the second position deviation threshold is lower than the first position deviation threshold.

[0129] It can be seen that after calculating the real-time positioning deviation of the second reference object, it is first determined whether the real-time positioning deviation of the second reference object is greater than the position deviation threshold. If it is greater than the position deviation threshold, the horizontal and vertical coordinates of the vehicle are corrected according to the real-time positioning deviation of the second reference object. It can be seen that when the position deviation threshold is relatively small, the horizontal and vertical coordinates of the vehicle can be corrected for errors. Only when the real-time positioning deviation of the second reference object is relatively large, the horizontal and vertical coordinate errors of the vehicle are corrected. Moreover, the second position deviation threshold is smaller than the first position deviation threshold. It can be seen that the position deviation thresholds of the first sub-stage and the second sub-stage are relatively high, because when the real-time positioning deviation of the second reference object in these two sub-stages is relatively large, it can be corrected in the third sub-stage. The position deviation threshold of the third sub-stage is relatively low, so the horizontal and vertical coordinate errors of the vehicle can be corrected in time in the third sub-stage. In this way, the continuity of the parking trajectory of the vehicle can be ensured, and the number of rubbing the warehouse is relatively small.

[0130] It is understandable that the chassis system needs to be used to position the vehicle during the entire automatic parking process. Specifically, the vehicle steering wheel angle and wheel speed pulse can be used to calculate the turning radius corresponding to each steering wheel angle according to the vehicle angle mapping relationship under Ackerman steering and steering wheel angle, and then the arc length of the travel is calculated according to the turning radius and wheel speed pulse, and then the lateral and longitudinal coordinate components of the vehicle are calculated based on the arc length.

[0131] In actual scenarios, in order to eliminate the time difference between the data collected by the perception system during the vehicle's movement and the vehicle positioning information determined by the chassis system, it is necessary to synchronize the perception system and the chassis. That is, in order to compensate for the different frequencies of the chassis system and the perception system, each frame of information needs to be synchronized before coordinate transformation. In addition, in order to eliminate single-frame errors, multi-frame fusion error processing can also be performed.

[0132] In some special scenarios, such as parking spaces with pillars, it is necessary to remember the boundary points of the pillars. Therefore, in addition to collecting parking space information, the perception system can also mark some attributes of the parking space, such as whether there are pillars.

[0133] In theory, whether in the vehicle search phase or in the vehicle parking phase, the target parking space, obstacles in the surrounding environment, and lane line heading will not change. However, as the positioning of the vehicle during the parking process is erroneous, the information such as the target parking space, obstacles in the surrounding environment, and lane line heading collected by the perception system in real time will deviate from the previously recorded information. This deviation is caused by the positioning error of the vehicle. Therefore, during the entire parking process, the coordinates of the corner points of the target parking space, the coordinates of the nearest obstacle, the lane line heading, and other information with obvious characteristics will be monitored in real time. Then, based on the offset of this information, the heading, horizontal and vertical coordinates of the vehicle in the parking coordinate system will be corrected, thereby correcting the relative relationship between the vehicle positioning and the corner points of the target parking space and the lane line heading of the nearest obstacle, so as to achieve accurate obstacle avoidance and parking of the vehicle. Since the positioning error of the vehicle is tracked, memorized, and corrected during the parking process, the positioning error of the previous process can generally be compensated before the entry planning, which not only ensures the parking efficiency, but also ensures the parking accuracy.

[0134] In a second aspect, an embodiment of the present invention provides a vehicle positioning error correction device in automatic parking, see Figure 4 , the device 100 comprises:

[0135] A first determination module 110, configured to enter a parking space search phase upon receiving an automatic parking request, and determine a global coordinate system according to an initial positioning of the vehicle when entering the parking space search phase;

[0136] A parking space search module 120, configured to track and record the environmental perception data in the global coordinate system in real time during the parking space search phase, determine each parking space in the search environment according to the environmental perception data in the global coordinate system, and display each parking space on the vehicle screen;

[0137] A second determination module 130, configured to enter a vehicle parking phase after the user selects a target parking space from among the parking spaces displayed on the vehicle screen, and determine a parking coordinate system according to the target parking space;

[0138] A path planning module 140, configured to convert the environmental perception data in the global coordinate system tracked and recorded during the parking space search phase into the parking coordinate system, and plan a parking path according to the environmental perception data converted into the parking coordinate system;

[0139] An automatic parking module 150, configured to perform automatic parking according to the parking path, and to track and record the environmental perception data in the parking coordinate system in real time during the parking process;

[0140] The positioning correction module 160 is used to determine the real-time deviation of at least one reference object according to the environmental perception data in the parking coordinate system, and perform error correction on the real-time positioning of the vehicle according to the real-time deviation of the at least one reference object.

[0141] In one embodiment, the first determination module is specifically used to: take the center position of the rear axle of the vehicle when the vehicle enters the parking space search stage as the origin; the direction of the front of the vehicle when the vehicle enters the parking space search stage as the positive direction of the horizontal coordinate axis, and the direction perpendicular to the horizontal coordinate axis and pointing to one side of the vehicle body as the vertical coordinate axis; set the heading angle of the vehicle when entering the parking space search stage to 0°; and / or, the second determination module is specifically used to: take the preset position in the target parking space as the origin; the direction perpendicular to the line between the two entrance corner points of the target parking space and pointing to the entrance of the target parking space is the positive direction of the vertical coordinate axis, and the direction perpendicular to the vertical coordinate axis and pointing to one side of the vehicle body is the horizontal coordinate axis; set the heading angle of the positive direction of the horizontal coordinate axis to 0°.

[0142] In one embodiment, the environmental perception data tracked and recorded in the parking space search stage includes: a two-dimensional map corresponding to the vehicle communicable area determined based on various obstacles in the search environment; correspondingly, the parking path planning module plans the parking path according to the environmental perception data converted to the parking coordinate system, including: planning the parking path according to the two-dimensional map in the environmental perception data converted to the parking coordinate system.

[0143] In one embodiment, the environment perception data tracked and recorded in the parking space search phase includes lane line information;

[0144] Correspondingly, the positioning correction module includes:

[0145] A first calculation unit, used to calculate the real-time confidence of the lane line according to the lane line information currently tracked and recorded;

[0146] A first selection unit, configured to select a lane line as a first reference object when the real-time confidence level of the lane line is higher than a first preset confidence level;

[0147] A first comparison unit is used to compare in real time the lane line heading currently tracked and recorded for the first reference object with the lane line heading tracked and recorded during the parking space search phase;

[0148] A first determining unit, configured to determine a real-time deviation of a lane line heading of the first reference object according to a real-time comparison result;

[0149] The first correction unit is used to determine the real-time rotation amount of the parking coordinate system according to the real-time deviation of the lane line heading of the first reference object, and perform real-time error correction on the heading of the vehicle according to the real-time rotation amount of the parking coordinate system.

[0150] In one embodiment, the environmental perception data tracked and recorded in the parking space search stage also includes the coordinates of the corner points of the parking space and the coordinates of the obstacle closest to the entrance corner point of the parking space;

[0151] Correspondingly, the positioning correction module also includes:

[0152] The second calculation unit is used to calculate the distance from each parking space corner point to the corresponding side fisheye, and calculate the distance from the obstacle closest to the entrance corner point of the parking space to the corresponding side fisheye;

[0153] The third calculation unit is used to calculate the real-time confidence of each parking space corner point according to the distance from each parking space corner point to the corresponding side fisheye; calculate the real-time confidence of the obstacle closest to the entrance corner point of the parking space according to the distance from the obstacle closest to the entrance corner point of the parking space to the corresponding side fisheye;

[0154] A second selection unit is used to select the parking space corner point or obstacle corresponding to the real-time confidence higher than the second preset confidence as the second reference object when there is a real-time confidence higher than the second preset confidence among the real-time confidences of the parking space corner points and the obstacles closest to the entrance corner point of the parking space;

[0155] A second comparison unit is used to compare in real time the reference object coordinates currently tracked and recorded for the second reference object with the reference object coordinates tracked and recorded during the parking space search phase;

[0156] The second correction unit is used to determine the real-time positioning deviation of the second reference object according to the real-time comparison result, and perform real-time error correction on the horizontal and vertical coordinates of the vehicle according to the real-time positioning deviation of the second reference object.

[0157] In one embodiment, the device may further include:

[0158] a stage definition module, for defining the first sub-stage, the second sub-stage and the third sub-stage of the parking process after planning the parking path; wherein the first sub-stage is the stage in which the vehicle backs up for the first time, the second sub-stage is the stage in which the vehicle moves forward to straighten the vehicle body, and the third sub-stage is the stage in which the vehicle backs up into the garage;

[0159] A stage determination module, used to determine the sub-stage of the parking process according to the vehicle action information during the parking process;

[0160] Correspondingly, the device further includes:

[0161] a first judgment module, configured to, before the first correction unit determines the real-time rotation amount of the parking coordinate system according to the real-time deviation of the lane line heading of the first reference object, if the parking process is in the first sub-stage or the second sub-stage, determine whether the real-time deviation of the lane line heading is higher than a first direction deviation threshold; if the real-time deviation of the lane line heading is higher than the first direction deviation threshold, the first correction unit determines the real-time rotation amount of the parking coordinate system according to the real-time deviation of the lane line heading of the first reference object;

[0162] a second judgment module, configured to, before the first correction unit determines the real-time rotation amount of the parking coordinate system according to the real-time deviation of the lane line heading of the first reference object, if the parking process is in the third sub-stage, determine whether the real-time deviation of the lane line heading exceeds a second direction deviation threshold; if the real-time deviation of the lane line heading exceeds the second direction deviation threshold, the first correction unit determines the real-time rotation amount of the parking coordinate system according to the real-time deviation of the lane line heading of the first reference object; wherein the second direction deviation threshold is less than the first direction deviation threshold;

[0163] And / or, correspondingly, the device further includes:

[0164] a third judgment module, used for judging whether the real-time positioning deviation of the second reference object is higher than a first position deviation threshold before the second correction unit performs real-time error correction on the horizontal and vertical coordinates of the vehicle according to the real-time positioning deviation of the second reference object if the parking process is in the first sub-stage or the second sub-stage; if the real-time positioning deviation of the second reference object is higher than the first position deviation threshold, the second correction unit performs real-time error correction on the horizontal and vertical coordinates of the vehicle according to the real-time positioning deviation of the second reference object;

[0165] a fourth judgment module, for judging whether the real-time positioning deviation of the second reference object is higher than a second position deviation threshold before the second correction unit performs real-time error correction on the horizontal and vertical coordinates of the vehicle according to the real-time positioning deviation of the second reference object if the parking process is in the third sub-stage; if the real-time positioning deviation of the second reference object is higher than the second position deviation threshold, the second correction unit performs real-time error correction on the horizontal and vertical coordinates of the vehicle according to the real-time positioning deviation of the second reference object; wherein the second position deviation threshold is lower than the first position deviation threshold.

[0166] In one embodiment, the relationship between the real-time confidence of the lane line and the first distance conforms to a linear relationship or a normal distribution relationship; the smaller the first distance is, the greater the real-time confidence corresponding to the lane line is; the first distance is the distance between the lane line and the line connecting the left and right fisheyes; and / or, the relationship between the real-time confidence of each parking space corner point and the second distance conforms to a linear relationship or a normal distribution relationship; the smaller the second distance is, the greater the real-time confidence corresponding to the parking space corner point is; the second distance is the distance between the parking space corner point and the corresponding side fisheye; and / or, the relationship between the real-time confidence of the obstacle closest to the entrance corner point of the parking space and the third distance conforms to a linear relationship or a normal distribution relationship; the smaller the third distance is, the greater the real-time confidence corresponding to the obstacle closest to the entrance corner point of the parking space is; the third distance is the distance between the obstacle closest to the entrance corner point of the parking space and the corresponding side fisheye.

[0167] It is understandable that the explanation, specific implementation, beneficial effects, examples, etc. of the relevant contents in the device provided in the embodiment of the present invention can be found in the corresponding parts of the method provided in the first aspect, and will not be repeated here.

[0168] In a third aspect, an embodiment of the present invention provides a computer-readable medium, on which computer instructions are stored. When the computer instructions are executed by a processor, the processor executes the method provided in the first aspect.

[0169] Specifically, a system or device equipped with a storage medium can be provided, on which software program code that implements the functions of any of the above-mentioned embodiments is stored, and a computer (or CPU or MPU) of the system or device can be enabled to read and execute the program code stored in the storage medium.

[0170] In this case, the program code itself read from the storage medium can realize the function of any one of the above-mentioned embodiments, and thus the program code and the storage medium storing the program code constitute a part of the present invention.

[0171] The storage medium embodiments for providing the program code include a floppy disk, a hard disk, a magneto-optical disk, an optical disk (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), a magnetic tape, a non-volatile memory card, and a ROM. Alternatively, the program code can be downloaded from a server computer by a communication network.

[0172] In addition, it should be clear that the functions of any of the above embodiments can be implemented not only by executing the program code read by the computer, but also by enabling an operating system operating on the computer to complete part or all of the actual operations based on instructions from the program code.

[0173] In addition, it can be understood that the program code read from the storage medium is written to a memory provided in an expansion board inserted into the computer or to a memory provided in an expansion module connected to the computer, and then based on the instructions of the program code, a CPU installed on the expansion board or expansion module is enabled to perform part or all of the actual operations, thereby realizing the functions of any of the above-mentioned embodiments.

[0174] It is understandable that the explanation, specific implementation methods, beneficial effects, examples, etc. of the relevant contents in the computer-readable medium provided in the embodiment of the present invention can be found in the corresponding parts of the method provided in the first aspect, and will not be repeated here.

[0175] In a fourth aspect, an embodiment of the present specification provides a computing device, including a memory and a processor, wherein the memory stores executable code, and when the processor executes the executable code, it implements the method in any embodiment of the specification.

[0176] It is understandable that the explanation, specific implementation, beneficial effects, examples, etc. of the relevant contents in the computing device provided in the embodiment of the present invention can be found in the corresponding parts of the method provided in the first aspect, and will not be repeated here.

[0177] It should be noted that the terms used in the present invention are only for describing specific embodiments, rather than limiting the scope of the present application. As shown in the present specification and claims, unless the context clearly indicates an exception, the words "one", "a", "a kind of" and / or "the" do not specifically refer to the singular, but may also include the plural. The terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such a process, method or device. In the absence of further restrictions, the elements defined by the sentence "include one..." do not exclude the presence of other identical elements in the process, method or device including the elements.

[0178] It should also be noted that the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. Unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", etc. should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be a connection between the two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0179] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present invention.

Claims

1. A method for correcting vehicle positioning error in automatic parking, characterized in that: include: Entering a parking space search phase upon receiving an automatic parking request, and determining a global coordinate system according to an initial positioning of the vehicle when entering the parking space search phase; In the parking space search phase, the environmental perception data in the global coordinate system is tracked and recorded in real time, each parking space in the search environment is determined according to the environmental perception data in the global coordinate system, and each parking space is displayed on the vehicle screen; When the user selects a target parking space from among the parking spaces displayed on the vehicle screen, the vehicle enters the parking phase, and a parking coordinate system is determined according to the target parking space; Converting the environmental perception data in the global coordinate system tracked and recorded in the parking space search phase to the parking coordinate system, and planning a parking path according to the environmental perception data converted to the parking coordinate system; Automatically parking the vehicle according to the parking path, and tracking and recording the environmental perception data in the parking coordinate system in real time during the parking process; A real-time deviation of at least one reference object is determined according to the environmental perception data in the parking coordinate system, and an error correction is performed on the real-time positioning of the vehicle according to the real-time deviation of the at least one reference object.

2. The method according to claim 1, characterized in that The determining of the global coordinate system according to the initial positioning of the vehicle when entering the parking space search phase includes: The center position of the rear axle of the vehicle when the vehicle enters the parking space search phase is taken as the origin; the direction of the front of the vehicle when the vehicle enters the parking space search phase is taken as the positive direction of the horizontal coordinate axis, and the direction perpendicular to the horizontal coordinate axis and pointing to one side of the vehicle body is taken as the vertical coordinate axis; the heading angle of the vehicle when entering the parking space search phase is set to 0°; And / or, determining a parking coordinate system according to the target parking space includes: The preset position in the target parking space is taken as the origin; the direction perpendicular to the line between the two entrance corner points of the target parking space and pointing to the entrance of the target parking space is the positive direction of the longitudinal axis, and the direction perpendicular to the longitudinal axis and pointing to one side of the vehicle body is the horizontal axis; the heading angle of the positive direction of the horizontal axis is set to 0°.

3. The method according to claim 1, characterized in that The environmental perception data tracked and recorded in the parking space search phase includes: a two-dimensional map corresponding to the vehicle communicable area determined based on each obstacle in the search environment; Correspondingly, planning the parking path according to the environmental perception data converted into the parking coordinate system includes: planning the parking path according to the two-dimensional map in the environmental perception data converted into the parking coordinate system.

4. The method according to claim 1, characterized in that The environmental perception data tracked and recorded in the parking space search phase includes lane line information; Correspondingly, determining the real-time deviation of at least one reference object according to the environmental perception data in the parking coordinate system, and performing error correction on the real-time positioning of the vehicle according to the real-time deviation of the at least one reference object, includes: Calculate the real-time confidence of the lane line based on the lane line information currently tracked and recorded; When the real-time confidence level of the lane line is higher than a first preset confidence level, selecting the lane line as the first reference object; Comparing the lane line heading currently tracked and recorded for the first reference object with the lane line heading tracked and recorded during the parking space search phase in real time; Determine the real-time deviation of the lane heading of the first reference object according to the real-time comparison result; The real-time rotation amount of the parking coordinate system is determined according to the real-time deviation of the lane line heading of the first reference object, and the real-time error correction of the heading of the vehicle is performed according to the real-time rotation amount of the parking coordinate system.

5. The method according to claim 4, characterized in that The environmental perception data tracked and recorded in the parking space search phase also includes the coordinates of the corner points of the parking space and the coordinates of the obstacle closest to the entrance corner point of the parking space; Correspondingly, determining the real-time deviation of at least one reference object according to the environmental perception data in the parking coordinate system, and performing error correction on the real-time positioning of the vehicle according to the real-time deviation of the at least one reference object, further includes: Calculate the distance from each parking space corner point to the corresponding side fisheye, and calculate the distance from the obstacle closest to the parking space entrance corner point to the corresponding side fisheye; According to the distances from each parking space corner point to the corresponding side fisheye, the real-time confidence of each parking space corner point is calculated respectively; according to the distance from the obstacle closest to the entrance corner point of the parking space to the corresponding side fisheye, the real-time confidence of the obstacle closest to the entrance corner point of the parking space is calculated; When there is a real-time confidence higher than the second preset confidence among the real-time confidences of the corner points of the parking spaces and the obstacles closest to the entrance corner points of the parking spaces, the corner points of the parking spaces or the obstacles corresponding to the real-time confidence higher than the second preset confidence are selected as the second reference objects; Comparing the reference object coordinates currently tracked and recorded for the second reference object with the reference object coordinates tracked and recorded during the parking space search phase in real time; According to the real-time comparison result, the real-time positioning deviation of the second reference object is determined, and the horizontal and vertical coordinates of the vehicle are corrected in real time according to the real-time positioning deviation of the second reference object.

6. The method according to claim 5, characterized in that Also includes: After planning the parking path, a first sub-stage, a second sub-stage and a third sub-stage of the parking process are defined; wherein the first sub-stage is a stage in which the vehicle backs up for the first time, the second sub-stage is a stage in which the vehicle moves forward to straighten the vehicle body, and the third sub-stage is a stage in which the vehicle backs up into the garage; Determining the sub-stage of the parking process according to the vehicle action information during the parking process; Correspondingly, before determining the real-time rotation amount of the parking coordinate system according to the real-time lane heading deviation of the first reference object, the method further includes: If the parking process is in the first sub-stage or the second sub-stage, determining whether the lane line heading real-time deviation is higher than a first direction deviation threshold; if the lane line heading real-time deviation is higher than the first direction deviation threshold, determining the real-time rotation amount of the parking coordinate system according to the lane line heading real-time deviation of the first reference object; If the parking process is in the third sub-stage, determining whether the lane line heading real-time deviation exceeds a second direction deviation threshold; if the lane line heading real-time deviation exceeds the second direction deviation threshold, determining the real-time rotation amount of the parking coordinate system according to the lane line heading real-time deviation of the first reference object; wherein the second direction deviation threshold is less than the first direction deviation threshold; And / or, correspondingly, before performing real-time error correction on the horizontal and vertical coordinates of the vehicle according to the real-time positioning deviation of the second reference object, the method further includes: If the parking process is in the first sub-stage or the second sub-stage, determining whether the real-time positioning deviation of the second reference object is higher than a first position deviation threshold; if the real-time positioning deviation of the second reference object is higher than the first position deviation threshold, correcting the horizontal and vertical coordinates of the vehicle according to the real-time positioning deviation of the second reference object; If the parking process is in the third sub-stage, determine whether the real-time positioning deviation of the second reference object is higher than a second position deviation threshold; if the real-time positioning deviation of the second reference object is higher than the second position deviation threshold, perform the correction of the horizontal and vertical coordinates of the vehicle according to the real-time positioning deviation of the second reference object; wherein, the second position deviation threshold is lower than the first position deviation threshold.

7. The method according to claim 5, characterized in that The relationship between the real-time confidence of the lane line and the first distance conforms to a linear relationship or a normal distribution relationship; the smaller the first distance is, the greater the real-time confidence corresponding to the lane line is; the first distance is the distance between the lane line and the line connecting the left and right fisheyes; And / or, the relationship between the real-time confidence of each parking space corner point and the second distance conforms to a linear relationship or a normal distribution relationship; the smaller the second distance is, the greater the real-time confidence corresponding to the parking space corner point is; the second distance is the distance between the parking space corner point and the corresponding side fisheye; And / or, the relationship between the real-time confidence of the obstacle closest to the entrance corner point of the parking space and the third distance conforms to a linear relationship or a normal distribution relationship; the smaller the third distance is, the greater the real-time confidence corresponding to the obstacle closest to the entrance corner point of the parking space is; the third distance is the distance between the obstacle closest to the entrance corner point of the parking space and the corresponding side fisheye.

8. A vehicle positioning error correction device in automatic parking, characterized in that: include: A first determination module, configured to enter a parking space search phase upon receiving an automatic parking request, and determine a global coordinate system according to an initial positioning of the vehicle when entering the parking space search phase; A parking space search module, used for tracking and recording the environmental perception data in the global coordinate system in real time during the parking space search phase, determining each parking space in the search environment according to the environmental perception data in the global coordinate system, and displaying each parking space on the vehicle screen; A second determination module, configured to enter a vehicle parking phase after the user selects a target parking space from among the parking spaces displayed on the vehicle screen, and determine a parking coordinate system according to the target parking space; a path planning module, used to convert the environmental perception data in the global coordinate system tracked and recorded in the parking space search phase into the parking coordinate system, and plan a parking path according to the environmental perception data converted into the parking coordinate system; An automatic parking module, used to perform automatic parking according to the parking path, and to track and record the environmental perception data in the parking coordinate system in real time during the parking process; The positioning correction module is used to determine the real-time deviation of at least one reference object according to the environmental perception data in the parking coordinate system, and perform error correction on the real-time positioning of the vehicle according to the real-time deviation of the at least one reference object.

9. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed in a computer, the computer is caused to execute the method according to any one of claims 1 to 7.

10. A computing device, characterized in that The method comprises a memory and a processor, wherein the memory stores executable codes, and when the processor executes the executable codes, the method according to any one of claims 1 to 7 is implemented.

Images