Parking path determination method and device

Abstract

The invention relates to a parking path determination method and device. The method comprises the steps of obtaining current vehicle pose information of a target vehicle, obstacle information in an area where the target vehicle is located and parking space pose information of each parking space, and determining a target virtual boundary associated with the target parking space according to the current vehicle pose information, the obstacle information and the parking space pose information of each parking space, the target parking space is an idle parking space where the target vehicle is parked, and determining an automatic parking path of the target vehicle parked in the target parking space according to the target virtual boundary, the current vehicle pose information and the parking space pose information of the target parking space. By adopting the method, the safety of automatic parking can be improved.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving technology, and in particular to a parking path determination method and device. Background Art

[0002] With the continuous development of the field of autonomous driving technology, automatic parking operations have gradually matured. Existing automatic parking operations generally use vehicle cameras to obtain image information of the target parking space, and plan an automatic parking path based on the image information of the target parking space so that the vehicle can automatically park in the target parking space.

[0003] However, with the existing automatic parking operation, the automatic parking path is planned only based on the image information of the target parking space, which easily plans the parking path into an untrustworthy area, thereby reducing the safety of the automatic parking operation. Summary of the invention

[0004] Based on this, it is necessary to provide a parking path determination method and device that can improve the safety of automatic parking in response to the above technical problems.

[0005] In a first aspect, the present application provides a parking path determination method, comprising:

[0006] Obtain the current vehicle position information of the target vehicle, as well as the obstacle information in the area where the target vehicle is located and the parking position information of each parking space;

[0007] Determine a target virtual boundary associated with a target parking space according to the current vehicle position information, obstacle information, and parking space position information of each parking space; wherein the target parking space is an empty parking space to be parked by the target vehicle;

[0008] An automatic parking path for the target vehicle to park in the target parking space is determined according to the target virtual boundary, the current vehicle posture information and the parking space posture information of the target parking space.

[0009] In one embodiment, determining a target virtual boundary associated with a target parking space according to current vehicle posture information, obstacle information, and parking space posture information of each parking space includes:

[0010] According to the parking position information of each parking space, an initial virtual boundary corresponding to the target parking space is constructed; according to the current vehicle position information and obstacle information, the initial virtual boundary is processed to obtain a target virtual boundary associated with the target parking space.

[0011] In one embodiment, constructing an initial virtual boundary corresponding to a target parking space according to the parking space posture information of each parking space includes:

[0012] According to the parking position information of each parking space, the entry edge position relationship between adjacent parking spaces is determined; according to the entry edge position relationship and the parking position information of the target parking space, the initial virtual boundary corresponding to the target parking space is constructed.

[0013] In one embodiment, constructing an initial virtual boundary corresponding to the target parking space according to the position relationship of the entry edge and the parking space posture information of the target parking space includes:

[0014] When the entry edge position relationship is located on the same straight line, the parking space boundary in the parking space posture information of the target parking space is used as the initial virtual boundary corresponding to the target parking space; when the entry edge position relationship is located on different straight lines, the line segment between the parking space vertex in the parking space posture information of the target parking space and the parking space vertex in the parking space posture information of the adjacent parking space of the target parking space, as well as the parking space boundary in the parking space posture information of the target parking space, are used as the initial virtual boundary corresponding to the target parking space.

[0015] In one embodiment, the initial virtual boundary includes an initial virtual lower boundary, which is the bottom edge of the target parking space, and the target virtual boundary includes a target virtual lower boundary; the initial virtual boundary is processed according to the current vehicle posture information and obstacle information to obtain the target virtual boundary associated with the target parking space, including:

[0016] According to preset lower boundary extension parameters, the initial virtual lower boundary is laterally extended to obtain a candidate virtual lower boundary; according to preset lower boundary translation parameters, the candidate virtual lower boundary is translated in the vehicle parking direction, and during the translation process, according to current vehicle posture information and obstacle information, the candidate virtual lower boundary is subjected to collision detection processing to obtain a first collision detection result; according to the first collision detection result and the candidate virtual lower boundary, a target virtual lower boundary associated with the target parking space is determined.

[0017] In one embodiment, determining a target virtual lower boundary associated with the target parking space according to the first collision detection result and the candidate virtual lower boundary includes:

[0018] When the first collision detection result is that there is a collision between the candidate virtual lower boundary and the target vehicle and / or obstacle, the candidate virtual lower boundary is translated to the previous translation position corresponding to the boundary collision position to obtain the target virtual lower boundary; when the first collision detection result is that there is no collision between the candidate virtual lower boundary and the target vehicle and / or obstacle, the candidate virtual lower boundary translated to the maximum lower boundary translation position is used as the target virtual lower boundary.

[0019] In one embodiment, the initial virtual boundary also includes an initial virtual side boundary, and the target virtual boundary includes a target virtual side boundary; the initial virtual boundary is processed according to the current vehicle posture information and obstacle information to obtain a target virtual boundary associated with the target parking space, including:

[0020] Determine a longitudinal adjustment parameter of an initial virtual side boundary; use the longitudinal adjustment parameter of the initial virtual side boundary to longitudinally adjust the initial virtual side boundary to obtain a candidate virtual side boundary; according to a preset side boundary translation parameter, move the candidate virtual side boundary to the corresponding side, and during the movement, perform collision detection processing on the candidate virtual side boundary according to current vehicle posture information and obstacle information to obtain a second collision detection result; determine a target virtual side boundary associated with the target parking space according to the second collision detection result and the candidate virtual side boundary.

[0021] In one embodiment, determining the longitudinal adjustment parameter of the initial virtual lateral boundary includes:

[0022] According to the parking space posture information of the target parking space, a longitudinal obstacle detection frame corresponding to the initial virtual side boundary is constructed; along the parking direction of the vehicle, the longitudinal obstacle detection frame is moved on the straight line where the initial virtual side boundary is located, and during the movement, according to the current vehicle posture information and obstacle information, a collision detection process is performed on the longitudinal obstacle detection frame to obtain a third collision detection result; according to the third collision detection result, a longitudinal adjustment parameter of the initial virtual side boundary is determined.

[0023] In one embodiment, the initial virtual boundary also includes an initial virtual upper boundary, the initial virtual upper boundary is the entry boundary of the target parking space, and the target virtual boundary includes a target virtual upper boundary; the initial virtual boundary is processed according to the current vehicle posture information and obstacle information to obtain the target virtual boundary associated with the target parking space, including:

[0024] According to preset upper boundary extension parameters, the initial virtual upper boundary is laterally extended to obtain a candidate virtual upper boundary; according to the parking method of the target vehicle, the maximum upper boundary translation position is determined; according to the maximum upper boundary translation position, the candidate virtual upper boundary is moved in the vertical direction of the initial virtual upper boundary, and during the movement, according to the current vehicle posture information and obstacle information, the candidate virtual upper boundary is subjected to collision detection processing to obtain a fourth collision detection result; according to the fourth collision detection result and the candidate virtual upper boundary, a target virtual upper boundary associated with the target parking space is determined.

[0025] In one embodiment, the target virtual boundary includes a target virtual extension boundary; the initial virtual boundary is processed according to the current vehicle posture information and the obstacle information to obtain the target virtual boundary associated with the target parking space, including:

[0026] The initial virtual upper boundary is translated in the straight line direction of the initial virtual upper boundary, and during the translation process, the initial virtual upper boundary is subjected to collision detection processing according to the current vehicle posture information and obstacle information to obtain a fifth collision detection result; based on the fifth collision detection result and the initial virtual upper boundary, a target virtual extension boundary associated with the target parking space is determined.

[0027] In a second aspect, the present application further provides a parking path determination device, comprising:

[0028] An information acquisition module is used to obtain the current vehicle position information of the target vehicle, as well as the obstacle information in the area where the target vehicle is located and the parking position information of each parking space;

[0029] A boundary determination module is used to determine a target virtual boundary associated with a target parking space according to the current vehicle posture information, obstacle information and parking space posture information of each parking space; wherein the target parking space is an empty parking space to be parked by the target vehicle;

[0030] The path determination module is used to determine the automatic parking path for the target vehicle to park in the target parking space according to the target virtual boundary, the current vehicle posture information and the parking space posture information of the target parking space.

[0031] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0032] Obtain the current vehicle position information of the target vehicle, as well as the obstacle information in the area where the target vehicle is located and the parking position information of each parking space;

[0033] Determine a target virtual boundary associated with a target parking space according to the current vehicle position information, obstacle information, and parking space position information of each parking space; wherein the target parking space is an empty parking space to be parked by the target vehicle;

[0034] An automatic parking path for the target vehicle to park in the target parking space is determined according to the target virtual boundary, the current vehicle posture information and the parking space posture information of the target parking space.

[0035] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0036] Obtain the current vehicle position information of the target vehicle, as well as the obstacle information in the area where the target vehicle is located and the parking position information of each parking space;

[0037] Determine a target virtual boundary associated with a target parking space according to the current vehicle position information, obstacle information, and parking space position information of each parking space; wherein the target parking space is an empty parking space to be parked by the target vehicle;

[0038] An automatic parking path for the target vehicle to park in the target parking space is determined according to the target virtual boundary, the current vehicle posture information and the parking space posture information of the target parking space.

[0039] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0040] Obtain the current vehicle position information of the target vehicle, as well as the obstacle information in the area where the target vehicle is located and the parking position information of each parking space;

[0041] Determine a target virtual boundary associated with a target parking space according to the current vehicle position information, obstacle information, and parking space position information of each parking space; wherein the target parking space is an empty parking space to be parked by the target vehicle;

[0042] An automatic parking path for the target vehicle to park in the target parking space is determined according to the target virtual boundary, the current vehicle posture information and the parking space posture information of the target parking space.

[0043] The above parking path determination method and device determine the target virtual boundary associated with the target parking space to be parked by the target vehicle according to the current vehicle posture information of the target vehicle, the obstacle information in the area where the target vehicle is located, and the parking space posture information of each parking space, and then determine the automatic parking path for the target vehicle to park in the target parking space according to the target virtual boundary, the current vehicle posture information, and the parking space posture information of the target parking space. Compared with the related art, in which the automatic parking path is planned only according to the image information of the target parking space, the above method can ensure the reliability of the automatic parking path by using the target virtual boundary associated with the target parking space to constrain the automatic parking path, thereby ensuring the safety of the automatic parking operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0045] Figure 1 A diagram showing an application environment of a parking path determination method in one embodiment;

[0046] Figure 2is a flow chart of a method for determining a parking path in one embodiment;

[0047] Figure 3A A schematic diagram of a process of constructing an initial virtual boundary in one embodiment;

[0048] Figure 3B A schematic diagram of a common vertical parking space / oblique diamond parking space in one embodiment;

[0049] Figure 3C is a schematic diagram of an oblique rectangular parking space in one embodiment;

[0050] Figure 3D is a schematic diagram of an initial virtual upper boundary in one embodiment;

[0051] Figure 3E is a schematic diagram of an initial virtual upper boundary in another embodiment;

[0052] Figure 4A A schematic diagram of a process for determining a target virtual lower boundary in one embodiment;

[0053] Figure 4B is a schematic diagram of a candidate virtual lower boundary moving process in one embodiment;

[0054] Figure 5A A schematic diagram of a process for determining a target virtual side boundary in one embodiment;

[0055] Figure 5B is a schematic diagram of a candidate virtual side boundary moving process in one embodiment;

[0056] Fig. 6A A schematic diagram of a flow chart for determining a longitudinal adjustment parameter in one embodiment;

[0057] Figure 6B is a schematic diagram of a longitudinal obstacle detection frame in one embodiment;

[0058] Figure 7 A schematic diagram of a process for determining a target virtual upper boundary in one embodiment;

[0059] Fig. 8A A schematic diagram of a process for determining a target virtual extension boundary in one embodiment;

[0060] Figure 8B is a schematic diagram of a target virtual extension boundary moving process in one embodiment;

[0061] Fig. 9 is a flow chart of a parking path determination method according to another embodiment;

[0062] Fig.10 is a structural block diagram of a parking path determination device in one embodiment;

[0063] Fig.11 is a structural block diagram of a parking path determination device in another embodiment;

[0064] Fig.12 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0065] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0066] The parking path determination method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown in the figure, the sensing device 102 communicates with the vehicle terminal 104 through the network. For example, the vehicle terminal 104 obtains the current vehicle posture information of the target vehicle, the obstacle information in the area where the target vehicle is located, and the parking posture information of each parking space through the sensing device 102, and determines the target virtual boundary associated with the target parking space according to the current vehicle posture information, the obstacle information, and the parking posture information of each parking space, and then determines the automatic parking path for the target vehicle to park in the target parking space according to the target virtual boundary, the current vehicle posture information, and the parking posture information of the target parking space.

[0067] With the continuous development of the field of autonomous driving technology, automatic parking operations have gradually matured. Existing automatic parking operations generally use vehicle cameras to obtain image information of the target parking space, and plan an automatic parking path based on the image information of the target parking space so that the vehicle can automatically park in the target parking space.

[0068] However, with the existing automatic parking operation, the automatic parking path is planned only based on the image information of the target parking space, which easily plans the parking path into an untrustworthy area, thereby reducing the safety of the automatic parking operation.

[0069] Based on this, in an exemplary embodiment, Figure 2 As shown, a parking path determination method is provided, and the method is applied to Figure 1 The vehicle-mounted terminal in the example is used for explanation, and the specific steps include:

[0070] S201, obtaining the current vehicle position information of the target vehicle, the obstacle information in the area where the target vehicle is located, and the parking position information of each parking space.

[0071] Among them, the target vehicle is an intelligent vehicle with automatic parking requirements; the current vehicle posture information refers to the position information and posture information of the target vehicle at the current moment, for example, the posture information may include the vehicle's driving direction and vehicle orientation, etc.; the obstacle information is the information of obstacle objects that hinder the vehicle's driving in the area where the target vehicle is located; the parking space posture information is the position information and posture information of the parking space, for example, the parking space vertex coordinates and parking space orientation, etc.

[0072] Optionally, after the target vehicle enters the parking area, the sensor equipment deployed on the target vehicle can be used to obtain obstacle information and parking position information of each parking space in the area where the target vehicle is located. At the same time, the current vehicle position information of the target vehicle is obtained through the control equipment in the target vehicle.

[0073] For example, a camera and a lidar deployed on the target vehicle can be used to obtain obstacle information and parking position information of each parking space in the area where the target vehicle is located; then, the GPS (Global Positioning System) inside the vehicle is used to obtain the current vehicle position information of the target vehicle.

[0074] S202: Determine a target virtual boundary associated with a target parking space according to the current vehicle position information, obstacle information, and parking position information of each parking space.

[0075] Among them, the target parking space is an empty parking space where the target vehicle is to be parked; the target virtual boundary is an auxiliary line used to optimize the automatic parking path, and the area included in the target virtual boundary is a trusted area, that is, there are no obstacles in the area included in the target virtual boundary.

[0076] Optionally, in order to ensure the rationality of the determination of the target parking space, the vacancy status of each parking space in the area where the target vehicle is located can be displayed through a display device in the target vehicle; then, the target parking space can be determined based on the driver's click operation on a certain vacant parking space, or the most suitable target parking space can be determined based on the current vehicle's position information and the parking position information of each vacant parking space.

[0077] Furthermore, the current vehicle posture information, obstacle information, and parking posture information of each parking space can be simultaneously input into the trained virtual boundary construction model, and the virtual boundary construction model outputs the target virtual boundary associated with the target parking space according to the current vehicle posture information, obstacle information, parking posture information of each parking space, and model parameters. It can be understood that since the current vehicle posture information changes in real time, the target virtual boundary corresponding to the target vehicle also changes in fact.

[0078] Alternatively, an initial virtual boundary corresponding to the target parking space is constructed according to the parking space posture information of each parking space, and the initial virtual boundary is processed according to the current vehicle posture information and obstacle information to obtain a target virtual boundary associated with the target parking space. The initial virtual boundary is a virtual boundary initially constructed.

[0079] Optionally, an initial virtual boundary can be constructed based on the parking position information of each parking space and the target parking space; then, the initial virtual boundary is expanded outward, and during the outward expansion process, collision detection is performed on the outward expanded initial virtual boundary based on the current vehicle position information and obstacle information.

[0080] Furthermore, in the absence of a collision, the initial virtual boundary at the end of the expansion is used as the target virtual boundary associated with the target parking space; in the event of a collision, based on the collision position, the initial virtual boundary corresponding to the area that does not include obstacles and the vehicle outline of the target vehicle during the expansion process is used as the target virtual boundary associated with the target parking space.

[0081] S203, determining an automatic parking path for the target vehicle to park in the target parking space according to the target virtual boundary, the current vehicle position information and the parking space position information of the target parking space.

[0082] The automatic parking path is the automatic driving path of the target vehicle when performing automatic parking processing.

[0083] Optionally, after determining the target virtual boundary, the target virtual boundary, the current vehicle posture information and the parking posture information of the target parking space can be simultaneously input into a trained path planning model, and the path planning model outputs an automatic parking path for the target vehicle to park in the target parking space based on the target virtual boundary, the current vehicle posture information, the parking posture information of the target parking space and the model parameters.

[0084] Alternatively, the current vehicle posture information and the parking space posture information of the target parking space may be used to predetermine an initial parking path; then, the initial parking path is constrained using the target virtual boundary to obtain an automatic parking path for the target vehicle to park in the target parking space.

[0085] Furthermore, the automatic parking path for the target vehicle to park in the target parking space can be displayed to the driver through a display device in the target vehicle.

[0086] In the above parking path determination method, the target virtual boundary associated with the target parking space to be parked by the target vehicle is determined based on the current vehicle posture information of the target vehicle, the obstacle information in the area where the target vehicle is located, and the parking space posture information of each parking space, and then the automatic parking path for the target vehicle to park in the target parking space is determined based on the target virtual boundary, the current vehicle posture information, and the parking space posture information of the target parking space. Compared with the related art, in which the automatic parking path is planned only based on the image information of the target parking space, the above method uses the target virtual boundary associated with the target parking space to constrain the automatic parking path, which can ensure the reliability of the automatic parking path and thus ensure the safety of the automatic parking operation.

[0087] In order to ensure the reliability of the initial virtual boundary construction, based on the above embodiment, in this embodiment, an optional method for constructing the initial virtual boundary is provided, such as Figure 3A As shown, the specific steps include:

[0088] S301, determining the parking edge position relationship between adjacent parking spaces according to the parking position information of each parking space.

[0089] Among them, the entry edge is the boundary that the vehicle passes through when entering the parking space; the entry edge position relationship is the position relationship between the entry edges of adjacent parking spaces.

[0090] Optionally, the position information corresponding to the entry edge of each parking space can be extracted from the parking position information of each parking space; then, the position relationship of the entry edges between adjacent parking spaces is determined based on the position information corresponding to the entry edge of each parking space.

[0091] Alternatively, the target parking space vertex of the target parking space and the adjacent parking space vertices of the adjacent parking spaces of the target parking space may be extracted from the parking space posture information of each parking space; then, based on the positional relationship between the target parking space vertex and the adjacent parking space vertices, the positional relationship of the parking edge between the adjacent parking spaces may be determined.

[0092] S302: construct an initial virtual boundary corresponding to the target parking space according to the position relationship of the entry edge and the parking space posture information of the target parking space.

[0093] Optionally, after determining the in-and-out edge position relationship, the parking space boundary of the target parking space can be extracted from the parking space posture information of the target parking space; then, based on the in-and-out edge position relationship, the parking space boundary of the target parking space is processed to obtain an initial virtual boundary corresponding to the target parking space.

[0094] Exemplarily, when the position relationship of the parking edges is located on the same straight line, the parking space boundary in the parking space posture information of the target parking space is used as the initial virtual boundary corresponding to the target parking space.

[0095] Optionally, when the position relationship of the entry edge is located on the same straight line, the target parking space may be as follows Figure 3B For the ordinary vertical parking space / oblique diamond parking space shown, at this time, the parking space boundary in the parking space posture information of the target parking space can be directly used as the initial virtual boundary corresponding to the target parking space.

[0096] Alternatively, when the position relationship of the entry edge is on different straight lines, the line segment between the parking space vertex in the parking space posture information of the target parking space and the parking space vertex in the parking space posture information of the adjacent parking space of the target parking space, as well as the parking space boundary in the parking space posture information of the target parking space, are used as the initial virtual boundary corresponding to the target parking space.

[0097] Optionally, when the position relationship of the entry edge is located on different straight lines, the target parking space may be as follows Figure 3C For the oblique rectangular parking space shown in the figure, at this time, based on the orientation of the parking space, the line segment between the parking space vertex in the parking space posture information of the target parking space and the parking space vertex in the parking space posture information of the adjacent parking space of the target parking space can be used as the virtual upper boundary; then, the side edge and the bottom edge of the parking space boundary of the target parking space can also be used as the initial virtual boundary corresponding to the target parking space.

[0098] In the embodiment of the present application, by constructing an initial virtual boundary corresponding to the target parking space according to the position relationship of the entry edge and the parking space posture information of the target parking space, the reliability of the construction of the initial virtual boundary can be ensured.

[0099] It can be understood that due to the staggered distribution of the oblique rectangular parking spaces, on the basis of the above embodiment, in this embodiment, for the oblique rectangular parking spaces, another optional method of constructing the initial virtual upper boundary is provided, specifically, according to the relationship between the side of the target parking space and the side of the adjacent parking space, it is determined whether the target parking space is an oblique rectangular parking space; if so, the initial virtual upper boundary corresponding to the target parking space is constructed according to the parking position information of each parking space.

[0100] Optionally, the vertex A1B1 of any side line L1 of the target vehicle and the vertex A2B2 of the other side line L2 of the adjacent parking space may be obtained first, wherein L1 and L2 cannot be both left side lines or both right side lines.

[0101] Then, the following three conditions are used to determine the shape of the parking space: 1) the cross product of vector A1B1 and vector A2B2 is close to 0 within a certain error range; 2) the distance S1 between vertex A2 and L1 and the distance S2 between vertex B2 and L1 are determined, and the sum of S1 and S2 is less than the first preset threshold; 3) the distance S from vertex A2 to the entrance line of the target parking space is determined A2 , and the distance S from vertex B2 to the bottom line of the target parking space B2 , and S A2 With SB2 The sum is greater than a second preset threshold.

[0102] If the target parking space and the adjacent parking space meet the above three conditions, it proves that the parking space shape of the target parking space is an oblique rectangular parking space; if the target parking space and the adjacent parking space do not meet any of the above conditions, it proves that the parking space shape of the target parking space is an ordinary vertical parking space / oblique diamond parking space.

[0103] After determining that the target parking space is an oblique rectangular parking space, the dot product dot between vector A1B1 and vector A1A2 can be calculated. A2 , and the dot product between vector A1B1 and vector A1B2 B2 ; Then, if vector A1B1 and vector A1B2 satisfy 0 <dot A2 <∣A1B1∣ 2 , or 0 <dot B2 <∣A1B1∣ 2 If any of the conditions in the above table is met, it is determined that there is only one vertex between vertex A2 and vertex B2 that is located between vectors A1B1.

[0104] If the vector A1B1 that satisfies the above conditions is the left side of the target parking space (the left side in the driving direction), and the vertex A2 is within the range of the vector A1B1, and the vertex B2 is not within the range of the vector A1B1, then it is proved that the target parking space is Figure 3D At this time, the vector P1 is constructed based on the left front vertex of the target parking space and the parking space on its left. However, since the initial upper boundary constructed currently offsets the target parking space, it is necessary to withdraw the distance of P1 module to the right based on the vector P1, and then obtain the initial virtual upper boundary P1 corresponding to the target parking space. ’ .

[0105] If the vector A1B1 that satisfies the above conditions is the left side of the target parking space, and the vertex B2 is within the range of the vector A1B1, and the vertex A2 is not within the range of the vector A1B1, then it is proved that the target parking space is Figure 3D At this time, the initial upper boundary P2 is constructed based on the right front vertex of the target parking space and the parking space on its left. It is worth noting that the initial upper boundary constructed at this time does not deviate from the target parking space, so no additional retracement is required.

[0106] If the vector A1B1 that satisfies the above conditions is the right side of the target parking space (the right side in the direction of entry), and the vertex A2 is within the range of the vector A1B1, and the vertex B2 is not within the range of the vector A1B1, then it is proved that the target parking space is Figure 3EAt this time, vector P3 is constructed based on the right front vertex of the target parking space and the parking space on its right. However, since the initial upper boundary currently constructed is offset from the target parking space, it is also necessary to use vector P3 as the basis and withdraw the distance of P3 module to the left to obtain the initial virtual upper boundary P3 corresponding to the target parking space. ’ .

[0107] If the vector A1B1 that meets the above conditions is the right side of the target parking space, and the vertex B2 is within the range of the vector A1B1, and the vertex A2 is not within the range of the vector A1B1, then the initial upper boundary P4 is constructed based on the target parking space and the left front vertex of the parking space to its right. It is worth noting that the initial upper boundary constructed at this time does not deviate from the target parking space, so no additional retracement is required.

[0108] In the embodiment of the present application, by constructing an initial virtual upper boundary corresponding to the target parking space according to the parking space shape of the target parking space and the parking space posture information of each parking space, the reliability of the construction of the initial virtual upper boundary can be ensured.

[0109] In order to ensure the reliability of the construction of the target virtual lower boundary, based on the above embodiment, in this embodiment, the initial virtual boundary includes the initial virtual lower boundary, the initial virtual lower boundary is the bottom edge of the target parking space, and the target virtual boundary includes the target virtual lower boundary; further, an optional method for determining the target virtual lower boundary is provided, such as Figure 4A As shown, the specific steps include:

[0110] S401, according to a preset lower boundary extension parameter, the initial virtual lower boundary is extended horizontally to obtain a candidate virtual lower boundary.

[0111] The lower boundary extension parameter is a length parameter for horizontally extending the initial virtual lower boundary; and the candidate virtual lower boundary is the virtual lower boundary after the horizontal extension process.

[0112] Optionally, a lower boundary extension parameter may be preset based on historical parking data; accordingly, the lower boundary extension parameter may be used to extend the initial virtual lower boundary laterally to the left and right sides to obtain a candidate virtual lower boundary.

[0113] It can be understood that during the lateral extension process, it is possible to determine whether the initial virtual lower boundary collides with the target vehicle or obstacle during the extension process based on the current vehicle posture information and obstacle information. If no collision occurs, the initial virtual lower boundary after the lateral extension process is directly used as the candidate virtual lower boundary; if there is a collision, it is necessary to withdraw an extension action based on the collision position to obtain a candidate virtual lower boundary.

[0114] S402, according to the preset lower boundary translation parameters, the candidate virtual lower boundary is translated in the vehicle parking direction, and during the translation process, according to the current vehicle posture information and obstacle information, the candidate virtual lower boundary is subjected to collision detection processing to obtain a first collision detection result.

[0115] Among them, the lower boundary translation parameter is the single translation distance of the candidate virtual lower boundary during translation; the vehicle parking direction is the driving direction of the target vehicle when it parks in the target parking space; and the first collision detection result is the collision result between the candidate virtual lower boundary and the target vehicle / obstacle.

[0116] Optionally, the lower boundary translation parameter can be preset based on historical parking data; accordingly, the lower boundary translation parameter can be used to perform multiple translation processing on the candidate virtual lower boundary in the parking direction of the vehicle. Figure 4B , using the lower boundary translation parameter, the candidate virtual lower boundary H1 is translated multiple times in the vehicle parking direction.

[0117] Furthermore, a contour extraction function can be used to extract the vehicle rectangular frame where the target vehicle is located according to the current vehicle posture information; then, during the translation process, it can be determined whether the candidate virtual lower boundary collides with the target vehicle / obstacle based on the position information of the vehicle rectangular frame and the obstacle, thereby obtaining the first collision detection result.

[0118] S403: Determine a target virtual lower boundary associated with the target parking space according to the first collision detection result and the candidate virtual lower boundary.

[0119] Optionally, based on the first collision detection result, a target virtual lower boundary associated with the target parking space may be determined from candidate virtual lower boundaries at various positions during the translation process.

[0120] Exemplarily, when the first collision detection result is that there is a collision between the candidate virtual lower boundary and the target vehicle and / or obstacle, the candidate virtual lower boundary is translated to the previous translation position corresponding to the boundary collision position to obtain the target virtual lower boundary; when the first collision detection result is that there is no collision between the candidate virtual lower boundary and the target vehicle and / or obstacle, the candidate virtual lower boundary translated to the maximum lower boundary translation position is used as the target virtual lower boundary.

[0121] The boundary collision position is the position where the candidate virtual lower boundary collides with the target vehicle and / or obstacle during the translation process; the maximum lower boundary translation position is the maximum position to which the preset candidate virtual lower boundary can translate.

[0122] Optionally, if the candidate virtual lower boundary collides with the target vehicle and / or obstacle during the translation process, in order to ensure the reliability of the target virtual lower boundary, it is necessary to withdraw the candidate virtual lower boundary to the previous translation position corresponding to the boundary collision position when a collision occurs, so as to obtain the target virtual lower boundary.

[0123] If there is no collision between the candidate virtual lower boundary and the target vehicle and / or obstacle during the translation process, the candidate virtual lower boundary translated to the maximum lower boundary translation position can be directly used as the target virtual lower boundary.

[0124] In an embodiment of the present application, the initial virtual lower boundary is laterally extended, translated, and collision detected based on the current vehicle posture information, obstacle information, lower boundary extension parameters, and lower boundary translation parameters to obtain the target virtual lower boundary, thereby ensuring the reliability of the target virtual lower boundary.

[0125] In order to ensure the reliability of the construction of the target virtual side boundary, based on the above embodiment, in this embodiment, the initial virtual boundary also includes the initial virtual side boundary, and the target virtual boundary includes the target virtual side boundary; further, an optional method for determining the target virtual side boundary is provided, such as Figure 5A As shown, the specific steps include:

[0126] S501, determining a longitudinal adjustment parameter of an initial virtual side boundary.

[0127] Among them, the initial virtual side boundary includes an initial virtual left boundary and an initial virtual right boundary; the longitudinal adjustment parameter is an adjustment parameter of the initial virtual side boundary in the longitudinal direction, and further, the longitudinal adjustment parameter of the initial virtual left boundary and the longitudinal adjustment parameter of the initial virtual right boundary can be the same or different.

[0128] Optionally, the longitudinal adjustment parameters of the initial virtual side boundary may be determined based on the parking space posture information of the target parking space, the current vehicle posture information and the obstacle information.

[0129] Exemplarily, the parking space posture information of the target parking space, the current vehicle posture information and the obstacle information can be simultaneously input into a trained parameter determination model, and the parameter determination model outputs the longitudinal adjustment parameters of the initial virtual side boundary according to the parking space posture information of the target parking space, the current vehicle posture information, the obstacle information and the model parameters.

[0130] S502 : Using the longitudinal adjustment parameters of the initial virtual side boundary, longitudinally adjust the initial virtual side boundary to obtain a candidate virtual side boundary.

[0131] The candidate virtual side boundary is a virtual side boundary that has been longitudinally extended.

[0132] Optionally, the longitudinal adjustment parameter of the initial virtual left boundary may be used to longitudinally adjust the initial virtual left boundary in the vehicle parking direction to obtain a candidate virtual left boundary; correspondingly, the longitudinal adjustment parameter of the initial virtual right boundary may be used to longitudinally adjust the initial virtual right boundary in the vehicle parking direction to obtain a candidate virtual right boundary.

[0133] S503, according to the preset side boundary translation parameters, the candidate virtual side boundary is moved to the corresponding side edge, and during the movement, according to the current vehicle posture information and obstacle information, the candidate virtual side boundary is subjected to collision detection processing to obtain a second collision detection result.

[0134] Among them, the side boundary translation parameter is the single translation distance of the candidate virtual side boundary during translation; the second collision detection result is the collision result between the candidate virtual side boundary and the target vehicle / obstacle, and the second collision detection result includes the collision result between the candidate virtual left boundary and the target vehicle / obstacle and the collision result between the candidate virtual right boundary and the target vehicle / obstacle.

[0135] Optionally, the side boundary translation parameter can be preset according to the historical parking data; accordingly, the side boundary translation parameter can be used to move the candidate virtual side boundary to the corresponding side. Figure 5B , using the left boundary translation parameter, the candidate virtual left boundary H2 is translated to the left for multiple times; using the right boundary translation parameter, the candidate virtual right boundary H3 is translated to the right for multiple times.

[0136] Furthermore, a contour extraction function can be used to extract the vehicle rectangular frame where the target vehicle is located according to the current vehicle posture information; then, during the translation process, it can be determined whether the candidate virtual side boundary collides with the target vehicle / obstacle based on the position information of the vehicle rectangular frame and the obstacle, thereby obtaining a second collision detection result.

[0137] S504: Determine a target virtual side boundary associated with the target parking space according to the second collision detection result and the candidate virtual side boundaries.

[0138] The target virtual side boundary includes a target virtual left boundary and a target virtual right boundary.

[0139] Optionally, if the candidate virtual left boundary collides with the target vehicle and / or obstacle during translation, it is necessary to withdraw the candidate virtual left boundary to the previous translation position corresponding to the boundary collision position when a collision occurs, thereby obtaining the target virtual left boundary.

[0140] If there is no collision between the candidate virtual left boundary and the target vehicle and / or obstacle during the translation process, the candidate virtual left boundary translated to the maximum left boundary translation position can be directly used as the target virtual left boundary.

[0141] It is understandable that the method for determining the target virtual right boundary is the same as the method for determining the target virtual left boundary, and thus will not be described in detail.

[0142] In an embodiment of the present application, the initial virtual side boundary is extended, translated, and collision detected based on the current vehicle posture information, obstacle information, longitudinal adjustment parameters, and side boundary translation parameters to obtain the target virtual side boundary, thereby ensuring the reliability of the target virtual side boundary.

[0143] In order to ensure the accuracy of the longitudinal adjustment parameters, based on the above embodiments, in this embodiment, an optional method for determining the longitudinal adjustment parameters is provided, such as Fig. 6A As shown, the specific steps include:

[0144] S601: construct a longitudinal obstacle detection frame corresponding to an initial virtual side boundary according to the parking space posture information of the target parking space.

[0145] The longitudinal obstacle detection frame is a virtual boundary frame used for collision detection. Furthermore, the initial virtual left boundary and the initial virtual right boundary correspond to different longitudinal obstacle detection frames.

[0146] Optionally, based on the parking space posture information of the target parking space, a longitudinal obstacle detection frame corresponding to the initial virtual left boundary can be constructed at the position corresponding to the initial virtual left boundary, and a longitudinal obstacle detection frame corresponding to the initial virtual right boundary can be constructed at the position corresponding to the initial virtual right boundary.

[0147] For example, refer to Figure 6B For the initial virtual left boundary, a longitudinal obstacle detection frame can be constructed at the upper left of the target parking space based on the length of the parking side of the target parking space and half of the depth of the parking space. Among them, X1 is 1 meter and X2 is 0.3 meters.

[0148] S602, moving the longitudinal obstacle detection frame on the straight line where the initial virtual side boundary is located along the vehicle parking direction, and performing collision detection processing on the longitudinal obstacle detection frame according to the current vehicle posture information and obstacle information during the movement to obtain a third collision detection result.

[0149] Optionally, the longitudinal obstacle detection frame can be moved along the parking direction of the vehicle so that the longitudinal obstacle detection frame moves on the straight line where the initial virtual side boundary is located; then, a contour extraction function can be used to extract the vehicle rectangular frame where the target vehicle is located according to the current vehicle posture information, and then according to the position information of the vehicle rectangular frame and the obstacle, it is determined whether the initial virtual side boundary collides with the target vehicle / obstacle, thereby obtaining a third collision detection result.

[0150] S603: Determine a longitudinal adjustment parameter of the initial virtual side boundary according to the third collision detection result.

[0151] Optionally, if the candidate virtual side boundary collides with the target vehicle and / or obstacle during the translation process, in order to ensure the reliability of the candidate virtual side boundary, the parking space vertex needs to be adjusted according to the collision position to obtain the longitudinal adjustment parameter.

[0152] For example, for the initial virtual left boundary, in the event of a collision, the vertex of the entry edge can be kept stationary, and at the same time, based on the collision position, the vertex of the bottom edge of the left boundary can be moved to a fixed safe position outside the parking space on the straight line where the initial virtual left boundary is located; then, the longitudinal adjustment parameters can be determined based on the moving distance.

[0153] If there is no collision between the candidate virtual lower boundary and the target vehicle and / or obstacle during the translation process, the parking space vertex can be adjusted according to a preset safe moving distance to obtain a longitudinal adjustment parameter.

[0154] For example, for the initial virtual left boundary, in the event of a collision, the position of the vertex of the bottom edge of the garage can be kept unchanged, and the vertex of the entrance edge can be moved to a fixed safe position outside the parking space on the straight line where the initial virtual left boundary is located; then, the longitudinal adjustment parameters can be determined based on the moving distance.

[0155] In the embodiment of the present application, the longitudinal adjustment parameter is determined by using the longitudinal obstacle detection frame corresponding to the initial virtual side boundary, so that the accuracy of the longitudinal adjustment parameter can be guaranteed.

[0156] In order to ensure the reliability of the target virtual upper boundary, based on the above embodiment, in this embodiment, the initial virtual boundary also includes an initial virtual upper boundary, the initial virtual upper boundary is the parking boundary of the target parking space, and the target virtual boundary includes the target virtual upper boundary; further, an optional method for determining the target virtual upper boundary is provided, such as Figure 7 As shown, the specific steps include:

[0157] S701 , according to a preset upper boundary extension parameter, the initial virtual upper boundary is extended horizontally to obtain a candidate virtual upper boundary.

[0158] The upper boundary extension parameter is a length parameter for horizontally extending the initial virtual upper boundary; and the candidate virtual upper boundary is the virtual upper boundary after the horizontal extension process.

[0159] Optionally, an upper boundary extension parameter may be preset based on historical parking data; accordingly, the upper boundary extension parameter may be used to extend the initial virtual upper boundary laterally to the left and right sides with the midpoint of the initial virtual upper boundary as the starting point to obtain a candidate virtual upper boundary.

[0160] It can be understood that during the lateral extension process, it is possible to determine whether the initial virtual upper boundary collides with the target vehicle or obstacle during the extension process based on the current vehicle posture information and obstacle information. If no collision occurs, the initial virtual upper boundary after the lateral extension process is directly used as the candidate virtual upper boundary; if there is a collision, it is necessary to withdraw an extension action based on the collision position to obtain a candidate virtual upper boundary.

[0161] S702: Determine a maximum upper boundary translation position according to the parking mode of the target vehicle.

[0162] The vehicle parking mode refers to the mode in which the target vehicle parks in the target parking space, which may include front parking and parking space parking; the maximum upper boundary translation position refers to the maximum position to which the preset candidate virtual upper boundary can be translated.

[0163] Optionally, since different parking methods require different moving spaces, the maximum upper boundary translation position can be determined according to the parking method of the target vehicle. For example, the maximum upper boundary translation position corresponding to the front parking method is 7 meters, and the maximum upper boundary translation position corresponding to the rear parking method is 5.5 meters.

[0164] S703, according to the maximum upper boundary translation position, the candidate virtual upper boundary is moved in the vertical direction of the initial virtual upper boundary, and during the movement, according to the current vehicle posture information and obstacle information, the candidate virtual upper boundary is subjected to collision detection processing to obtain a fourth collision detection result.

[0165] The fourth collision detection result is a collision result between the candidate virtual upper boundary and the target vehicle / obstacle.

[0166] Optionally, the candidate virtual upper boundary can be moved in the vertical direction of the initial virtual upper boundary within the maximum upper boundary translation position; then, a contour extraction function can be used to extract the vehicle rectangular frame where the target vehicle is located according to the current vehicle posture information, and during the translation process, it can be determined whether the candidate virtual upper boundary collides with the target vehicle / obstacle based on the position information of the vehicle rectangular frame and the obstacle, thereby obtaining a fourth collision detection result.

[0167] S704: Determine a target virtual upper boundary associated with the target parking space according to the fourth collision detection result and the candidate virtual upper boundary.

[0168] Optionally, if the candidate virtual upper boundary collides with the target vehicle and / or obstacle during the translation process, in order to ensure the reliability of the target virtual upper boundary, it is necessary to withdraw the candidate virtual upper boundary to the previous translation position corresponding to the boundary collision position when a collision occurs, so as to obtain the target virtual upper boundary.

[0169] If there is no collision between the candidate virtual upper boundary and the target vehicle and / or obstacle during the translation process, the candidate virtual upper boundary translated to the maximum lower upper boundary translation position can be directly used as the target virtual upper boundary.

[0170] In an embodiment of the present application, the initial virtual upper boundary is laterally extended, translated, and collision detected based on the current vehicle posture information, obstacle information, and upper boundary extension parameters to obtain the target virtual upper boundary, thereby ensuring the reliability of the target virtual upper boundary.

[0171] In order to ensure the reliability of the target virtual extension boundary, based on the above embodiment, in this embodiment, the target virtual boundary includes a target virtual extension boundary; further, an optional method for determining the target virtual extension boundary is provided, as shown in FIG8, specifically including the following steps:

[0172] S801, the initial virtual upper boundary is translated in the straight line direction where the initial virtual upper boundary is located, and during the translation process, collision detection processing is performed on the initial virtual upper boundary according to the current vehicle posture information and obstacle information to obtain a fifth collision detection result.

[0173] The fifth collision detection result is a collision result between the initial virtual extension boundary and the target vehicle / obstacle.

[0174] Optionally, the initial virtual upper boundary is extended laterally using an extended boundary extension parameter to obtain an initial virtual extended boundary; subsequently, the initial virtual extended boundary can be translated to the left and right in the straight line direction where the initial virtual upper boundary is located.

[0175] Furthermore, during the translation process, a contour extraction function can be used to extract the vehicle rectangular frame where the target vehicle is located based on the current vehicle posture information, and then based on the position information of the vehicle rectangular frame and the obstacle, it is determined whether the initial virtual extended boundaries on the left and right sides collide with the target vehicle / obstacle, thereby obtaining the fifth collision detection result.

[0176] For example, you can refer to Figure 8B, the initial virtual extended left boundary H4 is translated to the left for multiple times; the initial virtual extended right boundary H5 is translated to the right for multiple times.

[0177] S802: Determine a target virtual extension boundary associated with the target parking space according to the fifth collision detection result and the initial virtual upper boundary.

[0178] Optionally, if the initial virtual extended boundary collides with the target vehicle and / or obstacle during the translation process, in order to ensure the reliability of the target virtual extended boundary, it is necessary to withdraw the initial virtual extended boundary to the previous translation position corresponding to the boundary collision position when a collision occurs, so as to obtain the target virtual extended boundary.

[0179] If there is no collision between the initial virtual extended boundary and the target vehicle and / or obstacle during the translation process, the initial virtual extended boundary translated to the maximum boundary translation position may be directly used as the target virtual extended boundary.

[0180] In the embodiment of the present application, the target virtual extension boundary is obtained by translating the initial virtual upper boundary in the straight line direction where the initial virtual upper boundary is located, so that the reliability of the target virtual extension boundary can be guaranteed.

[0181] Fig. 9 FIG. 2 is a flow chart of a method for determining a parking path in another embodiment. Based on the above embodiment, this embodiment provides an optional example of a method for determining a parking path. Fig. 9 The specific implementation process is as follows:

[0182] S901, obtaining the current vehicle position information of the target vehicle, as well as the obstacle information in the area where the target vehicle is located and the parking position information of each parking space.

[0183] S902: Determine the parking edge position relationship between adjacent parking spaces based on the parking position information of each parking space.

[0184] S903: construct an initial virtual boundary corresponding to the target parking space according to the position relationship of the parking edge and the parking space posture information of the target parking space where the target vehicle is to be parked.

[0185] Exemplarily, when the entry edge position relationship is located on the same straight line, the parking space boundary in the parking space posture information of the target parking space is used as the initial virtual boundary corresponding to the target parking space; when the entry edge position relationship is located on different straight lines, the line segment between the parking space vertex in the parking space posture information of the target parking space and the parking space vertex in the parking space posture information of the adjacent parking space of the target parking space, as well as the parking space boundary in the parking space posture information of the target parking space, are used as the initial virtual boundary corresponding to the target parking space.

[0186] S904: Process the initial virtual boundary according to the current vehicle posture information and obstacle information to obtain a target virtual boundary associated with the target parking space.

[0187] Exemplarily, according to preset lower boundary extension parameters, the initial virtual lower boundary is laterally extended to obtain a candidate virtual lower boundary; according to preset lower boundary translation parameters, the candidate virtual lower boundary is translated in the vehicle parking direction, and during the translation process, according to current vehicle posture information and obstacle information, the candidate virtual lower boundary is subjected to collision detection processing to obtain a first collision detection result; according to the first collision detection result and the candidate virtual lower boundary, the target virtual lower boundary associated with the target parking space is determined.

[0188] When the first collision detection result is that there is a collision between the candidate virtual lower boundary and the target vehicle and / or obstacle, the candidate virtual lower boundary is translated to the previous translation position corresponding to the boundary collision position to obtain the target virtual lower boundary; when the first collision detection result is that there is no collision between the candidate virtual lower boundary and the target vehicle and / or obstacle, the candidate virtual lower boundary translated to the maximum lower boundary translation position is used as the target virtual lower boundary.

[0189] Determine a longitudinal adjustment parameter of an initial virtual side boundary; use the longitudinal adjustment parameter of the initial virtual side boundary to longitudinally adjust the initial virtual side boundary to obtain a candidate virtual side boundary; according to a preset side boundary translation parameter, move the candidate virtual side boundary to the corresponding side, and during the movement, perform collision detection processing on the candidate virtual side boundary according to current vehicle posture information and obstacle information to obtain a second collision detection result; determine a target virtual side boundary associated with the target parking space according to the second collision detection result and the candidate virtual side boundary.

[0190] Exemplarily, based on the parking space posture information of the target parking space, a longitudinal obstacle detection frame corresponding to the initial virtual side boundary is constructed; along the parking direction of the vehicle, the longitudinal obstacle detection frame is moved on the straight line where the initial virtual side boundary is located, and during the movement, according to the current vehicle posture information and obstacle information, the longitudinal obstacle detection frame is subjected to collision detection processing to obtain a third collision detection result; based on the third collision detection result, the longitudinal adjustment parameter of the initial virtual side boundary is determined.

[0191] According to preset upper boundary extension parameters, the initial virtual upper boundary is laterally extended to obtain a candidate virtual upper boundary; according to the parking method of the target vehicle, the maximum upper boundary translation position is determined; according to the maximum upper boundary translation position, the candidate virtual upper boundary is moved in the vertical direction of the initial virtual upper boundary, and during the movement, according to the current vehicle posture information and obstacle information, the candidate virtual upper boundary is subjected to collision detection processing to obtain a fourth collision detection result; according to the fourth collision detection result and the candidate virtual upper boundary, a target virtual upper boundary associated with the target parking space is determined.

[0192] The initial virtual upper boundary is translated in the straight line direction of the initial virtual upper boundary, and during the translation process, the initial virtual upper boundary is subjected to collision detection processing according to the current vehicle posture information and obstacle information to obtain a fifth collision detection result; based on the fifth collision detection result and the initial virtual upper boundary, a target virtual extension boundary associated with the target parking space is determined.

[0193] S905 , determining an automatic parking path for the target vehicle to park in the target parking space according to the target virtual boundary, the current vehicle position information, and the parking space position information of the target parking space.

[0194] The specific process of S901-S905 above can refer to the description of the above method embodiment, and its implementation principle and technical effect are similar, which will not be repeated here.

[0195] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0196] Based on the same inventive concept, the embodiment of the present application also provides a parking path determination device for implementing the parking path determination method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations of one or more parking path determination device embodiments provided below can refer to the limitations of the parking path determination method above, and will not be repeated here.

[0197] In an exemplary embodiment, Fig.10As shown, a parking path determination device 1 is provided, comprising: an information acquisition module 10, a boundary determination module 20 and a path determination module 30, wherein:

[0198] The information acquisition module 10 is used to acquire the current vehicle position information of the target vehicle, as well as the obstacle information in the area where the target vehicle is located and the parking position information of each parking space;

[0199] The boundary determination module 20 is used to determine the target virtual boundary associated with the target parking space according to the current vehicle posture information, obstacle information and parking space posture information of each parking space; wherein the target parking space is an empty parking space to be parked by the target vehicle;

[0200] The path determination module 30 is used to determine an automatic parking path for the target vehicle to park in the target parking space according to the target virtual boundary, the current vehicle posture information and the parking space posture information of the target parking space.

[0201] In an exemplary embodiment, Fig.11 As shown, the boundary determination module 20 includes:

[0202] An initial boundary construction unit 21 is used to construct an initial virtual boundary corresponding to a target parking space according to the parking space position information of each parking space;

[0203] The target boundary determination unit 22 is used to process the initial virtual boundary according to the current vehicle posture information and obstacle information to obtain a target virtual boundary associated with the target parking space.

[0204] In an exemplary embodiment, the initial boundary construction unit 21 is specifically used for:

[0205] According to the parking position information of each parking space, the entry edge position relationship between adjacent parking spaces is determined; according to the entry edge position relationship and the parking position information of the target parking space, the initial virtual boundary corresponding to the target parking space is constructed.

[0206] In an exemplary embodiment, the initial boundary construction unit 21 is further configured to:

[0207] When the entry edge position relationship is located on the same straight line, the parking space boundary in the parking space posture information of the target parking space is used as the initial virtual boundary corresponding to the target parking space; when the entry edge position relationship is located on different straight lines, the line segment between the parking space vertex in the parking space posture information of the target parking space and the parking space vertex in the parking space posture information of the adjacent parking space of the target parking space, as well as the parking space boundary in the parking space posture information of the target parking space, are used as the initial virtual boundary corresponding to the target parking space.

[0208] In an exemplary embodiment, the initial virtual boundary includes an initial virtual lower boundary, the initial virtual lower boundary is the bottom edge of the target parking space, and the target virtual boundary includes a target virtual lower boundary; further, the target boundary determination unit 22 is specifically used to:

[0209] According to preset lower boundary extension parameters, the initial virtual lower boundary is laterally extended to obtain a candidate virtual lower boundary; according to preset lower boundary translation parameters, the candidate virtual lower boundary is translated in the vehicle parking direction, and during the translation process, according to current vehicle posture information and obstacle information, the candidate virtual lower boundary is subjected to collision detection processing to obtain a first collision detection result; according to the first collision detection result and the candidate virtual lower boundary, a target virtual lower boundary associated with the target parking space is determined.

[0210] In an exemplary embodiment, the target boundary determination unit 22 is further configured to:

[0211] When the first collision detection result is that there is a collision between the candidate virtual lower boundary and the target vehicle and / or obstacle, the candidate virtual lower boundary is translated to the previous translation position corresponding to the boundary collision position to obtain the target virtual lower boundary; when the first collision detection result is that there is no collision between the candidate virtual lower boundary and the target vehicle and / or obstacle, the candidate virtual lower boundary translated to the maximum lower boundary translation position is used as the target virtual lower boundary.

[0212] In an exemplary embodiment, the initial virtual boundary further includes an initial virtual side boundary, and the target virtual boundary includes a target virtual side boundary; further, the target boundary determination unit 22 is further configured to:

[0213] Determine a longitudinal adjustment parameter of an initial virtual side boundary; use the longitudinal adjustment parameter of the initial virtual side boundary to longitudinally adjust the initial virtual side boundary to obtain a candidate virtual side boundary; according to a preset side boundary translation parameter, move the candidate virtual side boundary to the corresponding side, and during the movement, perform collision detection processing on the candidate virtual side boundary according to current vehicle posture information and obstacle information to obtain a second collision detection result; determine a target virtual side boundary associated with the target parking space according to the second collision detection result and the candidate virtual side boundary.

[0214] In an exemplary embodiment, the target boundary determination unit 22 is further configured to:

[0215] According to the parking space posture information of the target parking space, a longitudinal obstacle detection frame corresponding to the initial virtual side boundary is constructed; along the parking direction of the vehicle, the longitudinal obstacle detection frame is moved on the straight line where the initial virtual side boundary is located, and during the movement, according to the current vehicle posture information and the obstacle information, a collision detection process is performed on the longitudinal obstacle detection frame to obtain a third collision detection result; according to the third collision detection result, a longitudinal adjustment parameter of the initial virtual side boundary is determined.

[0216] In an exemplary embodiment, the initial virtual boundary also includes an initial virtual upper boundary, the initial virtual upper boundary is the parking boundary of the target parking space, and the target virtual boundary includes a target virtual upper boundary; further, the target boundary determination unit 22 is further used to:

[0217] According to preset upper boundary extension parameters, the initial virtual upper boundary is laterally extended to obtain a candidate virtual upper boundary; according to the parking method of the target vehicle, the maximum upper boundary translation position is determined; according to the maximum upper boundary translation position, the candidate virtual upper boundary is moved in the vertical direction of the initial virtual upper boundary, and during the movement, according to the current vehicle posture information and obstacle information, the candidate virtual upper boundary is subjected to collision detection processing to obtain a fourth collision detection result; according to the fourth collision detection result and the candidate virtual upper boundary, a target virtual upper boundary associated with the target parking space is determined.

[0218] In an exemplary embodiment, the target virtual boundary includes a target virtual extension boundary; further, the target boundary determination unit 22 is further configured to:

[0219] The initial virtual upper boundary is translated in the straight line direction of the initial virtual upper boundary, and during the translation process, the initial virtual upper boundary is subjected to collision detection processing according to the current vehicle posture information and obstacle information to obtain a fifth collision detection result; based on the fifth collision detection result and the initial virtual upper boundary, a target virtual extension boundary associated with the target parking space is determined.

[0220] Each module in the above parking path determination device can be implemented in whole or in part by software, hardware or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0221] In an exemplary embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. Fig.12As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, near field communication (NFC) or other technologies. When the computer program is executed by the processor, a parking path determination method is implemented. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0222] Those skilled in the art will understand that Fig.12 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0223] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0224] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0225] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0226] It should be noted that the data involved in this application (including but not limited to current vehicle posture information HE parking space posture information, etc.) are all data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0227] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0228] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0229] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A parking path determination method, characterized in that: The method comprises: Obtaining the current vehicle position information of the target vehicle, as well as the obstacle information in the area where the target vehicle is located and the parking position information of each parking space; Determine a target virtual boundary associated with a target parking space according to the current vehicle posture information, the obstacle information and the parking space posture information of each parking space; wherein the target parking space is an empty parking space to be parked by the target vehicle; An automatic parking path for the target vehicle to park in the target parking space is determined according to the target virtual boundary, the current vehicle position information and the parking space position information of the target parking space.

2. The method according to claim 1, characterized in that The step of determining a target virtual boundary associated with a target parking space according to the current vehicle posture information, the obstacle information, and the parking posture information of each parking space includes: Constructing an initial virtual boundary corresponding to the target parking space according to the parking space posture information of each parking space; The initial virtual boundary is processed according to the current vehicle posture information and the obstacle information to obtain a target virtual boundary associated with the target parking space.

3. The method according to claim 2, characterized in that The step of constructing an initial virtual boundary corresponding to the target parking space according to the parking space posture information of each parking space includes: According to the parking position information of each parking space, determine the parking edge position relationship between adjacent parking spaces; An initial virtual boundary corresponding to the target parking space is constructed according to the position relationship of the entry edge and the parking space posture information of the target parking space.

4. The method according to claim 3, characterized in that: The step of constructing an initial virtual boundary corresponding to the target parking space according to the position relationship of the entry edge and the parking space posture information of the target parking space includes: In the case where the position relationship of the parking edges is located on the same straight line, the parking space boundary in the parking space posture information of the target parking space is used as the initial virtual boundary corresponding to the target parking space; When the position relationship of the entry edge is located on different straight lines, the line segment between the parking space vertex in the parking space posture information of the target parking space and the parking space vertex in the parking space posture information of the adjacent parking space of the target parking space, as well as the parking space boundary in the parking space posture information of the target parking space, are used as the initial virtual boundary corresponding to the target parking space.

5. The method according to claim 2, characterized in that: The initial virtual boundary includes an initial virtual lower boundary, the initial virtual lower boundary is the bottom edge of the target parking space, and the target virtual boundary includes a target virtual lower boundary; The processing of the initial virtual boundary according to the current vehicle posture information and the obstacle information to obtain a target virtual boundary associated with the target parking space includes: According to a preset lower boundary extension parameter, the initial virtual lower boundary is laterally extended to obtain a candidate virtual lower boundary; According to a preset lower boundary translation parameter, the candidate virtual lower boundary is translated in the vehicle parking direction, and during the translation process, a collision detection process is performed on the candidate virtual lower boundary according to the current vehicle posture information and the obstacle information to obtain a first collision detection result; A target virtual lower boundary associated with the target parking space is determined according to the first collision detection result and the candidate virtual lower boundary.

6. The method according to claim 5, characterized in that The step of determining a target virtual lower boundary associated with the target parking space according to the first collision detection result and the candidate virtual lower boundary includes: When the first collision detection result indicates that there is a collision between the candidate virtual lower boundary and the target vehicle and / or obstacle, the candidate virtual lower boundary is translated to a previous translation position corresponding to a boundary collision position to obtain a target virtual lower boundary; When the first collision detection result is that there is no collision between the candidate virtual lower boundary and the target vehicle and / or obstacle, the candidate virtual lower boundary translated to the maximum lower boundary translation position is used as the target virtual lower boundary.

7. The method according to claim 2, characterized in that The initial virtual boundary also includes an initial virtual side boundary, and the target virtual boundary includes a target virtual side boundary; The processing of the initial virtual boundary according to the current vehicle posture information and the obstacle information to obtain a target virtual boundary associated with the target parking space includes: determining a longitudinal adjustment parameter of the initial virtual lateral boundary; Using the longitudinal adjustment parameter of the initial virtual side boundary, longitudinally adjusting the initial virtual side boundary to obtain a candidate virtual side boundary; According to a preset side boundary translation parameter, the candidate virtual side boundary is moved toward a corresponding side edge, and during the movement, a collision detection process is performed on the candidate virtual side boundary according to the current vehicle posture information and the obstacle information to obtain a second collision detection result; A target virtual side boundary associated with the target parking space is determined according to the second collision detection result and the candidate virtual side boundary.

8. The method according to claim 7, characterized in that The determining of the longitudinal adjustment parameter of the initial virtual side boundary includes: Constructing a longitudinal obstacle detection frame corresponding to the initial virtual side boundary according to the parking space posture information of the target parking space; Moving the longitudinal obstacle detection frame on the straight line where the initial virtual side boundary is located along the vehicle parking direction, and performing collision detection processing on the longitudinal obstacle detection frame according to the current vehicle posture information and the obstacle information during the movement to obtain a third collision detection result; A longitudinal adjustment parameter of the initial virtual side boundary is determined according to the third collision detection result.

9. The method according to claim 2, characterized in that: The initial virtual boundary also includes an initial virtual upper boundary, the initial virtual upper boundary is the parking boundary of the target parking space, and the target virtual boundary includes a target virtual upper boundary; The processing of the initial virtual boundary according to the current vehicle posture information and the obstacle information to obtain a target virtual boundary associated with the target parking space includes: According to a preset upper boundary extension parameter, the initial virtual upper boundary is laterally extended to obtain a candidate virtual upper boundary; Determining a maximum upper boundary translation position according to a parking mode of the target vehicle; According to the maximum upper boundary translation position, the candidate virtual upper boundary is moved in a vertical direction of the initial virtual upper boundary, and during the movement, according to the current vehicle posture information and the obstacle information, a collision detection process is performed on the candidate virtual upper boundary to obtain a fourth collision detection result; A target virtual upper boundary associated with the target parking space is determined according to the fourth collision detection result and the candidate virtual upper boundary.

10. The method according to claim 9, characterized in that The target virtual boundary includes a target virtual extension boundary; The processing of the initial virtual boundary according to the current vehicle posture information and the obstacle information to obtain a target virtual boundary associated with the target parking space includes: The initial virtual upper boundary is translated in the straight line direction where the initial virtual upper boundary is located, and during the translation process, a collision detection process is performed on the initial virtual upper boundary according to the current vehicle posture information and the obstacle information to obtain a fifth collision detection result; A target virtual extension boundary associated with the target parking space is determined according to the fifth collision detection result and the initial virtual upper boundary.

11. A parking path determination device, characterized in that: The device comprises: An information acquisition module is used to acquire the current vehicle posture information of the target vehicle, as well as the obstacle information in the area where the target vehicle is located and the parking posture information of each parking space; A boundary determination module, used to determine a target virtual boundary associated with a target parking space according to the current vehicle posture information, the obstacle information and the parking space posture information of each parking space; wherein the target parking space is an empty parking space to be parked by the target vehicle; The path determination module is used to determine an automatic parking path for the target vehicle to park in the target parking space according to the target virtual boundary, the current vehicle posture information and the parking space posture information of the target parking space.

Images