An automatic parking reverse path planning method, device, equipment and medium

Abstract

The application discloses an automatic parking reverse path planning method, device, equipment and medium, and the method comprises the steps of determining the maximum trajectory range of a current vehicle when the vehicle travels in each direction based on a given starting pose according to the performance of the vehicle; acquiring an obstacle grid map, and determining a target parking space position and a target vehicle pose in the obstacle grid map; taking a set position when a vehicle head drives out of the parking space as an intermediate vehicle pose, taking the intermediate vehicle pose as a starting pose, superimposing the maximum trajectory range in the direction of the target vehicle pose on the obstacle grid map, and analyzing to obtain a trajectory range capable of avoiding obstacles; randomly sampling in the trajectory range, selecting path points from the intermediate vehicle pose, obtaining a reverse path planning sequence, and inverting to obtain a forward path planning sequence. The application can process the path planning problem of parking demand in a complex environment.

Description

Technical Field

[0001] The present invention belongs to the technical field of automatic parking planning, and in particular relates to a method, device, equipment and medium for reverse path planning of automatic parking. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] With the acceleration of urbanization, the problem of parking difficulties has become increasingly prominent, especially in narrow or complex environments. Automatic parking systems face challenges in path planning efficiency and safety. Traditional automatic parking path planning methods mainly use forward path planning strategies, that is, starting from the current position of the vehicle, looking for a path to the target parking space. This method has significant limitations when dealing with complex environments (such as narrow alleys and areas with multiple obstacles). Forward path planning often requires real-time calculation and adjustment of the path, which requires high computing resources. In an environment with many dynamic obstacles, the efficiency and reliability of path planning will be significantly reduced.

[0004] In recent years, reverse optimization path planning for automated parking has gained increasing attention. This approach essentially constructs a path from the target parking space to the vehicle's current location. Compared to forward planning, reverse planning offers advantages for complex parking environments. However, current path planning methods lack consideration of vehicle dynamics, and their efficiency needs to be improved. Summary of the Invention

[0005] To overcome the above-mentioned deficiencies of the prior art, the present invention provides a method, device, equipment and medium for reverse path planning for automatic parking, which can more efficiently and intuitively handle the path planning problem of parking requirements in complex environments.

[0006] To achieve the above objectives, a first aspect of the present invention provides a reverse path planning method for automatic parking, comprising the following steps:

[0007] Determine the maximum trajectory range of the current vehicle when traveling in each direction based on a given starting position according to vehicle performance;

[0008] Obtain an obstacle grid map and determine the target parking space position and target vehicle posture in the obstacle grid map;

[0009] The set position of the vehicle's front end when it leaves the parking space is used as the intermediate vehicle posture, the intermediate vehicle posture is used as the starting posture, and the maximum trajectory range in the direction of the target vehicle posture is superimposed on the obstacle grid map to analyze and obtain the trajectory range that can avoid obstacles;

[0010] Random sampling is performed within the trajectory range, and the distance constraints to obstacles and the shortest overall path are used as constraints. Path points are selected starting from the intermediate vehicle posture to obtain a reverse path planning sequence, which is then flipped to obtain a forward path planning sequence.

[0011] Furthermore, the vehicle performance includes an optional range of the front wheel steering angle and the driving step length, and sampling is performed within the optional range of the front wheel steering angle and the driving step length of the current vehicle to obtain multiple combinations of front wheel steering angle and driving step length; based on each group of front wheel steering angle and driving step length, the vehicle running trajectory is generated in different driving directions to obtain the trajectory range in different driving directions.

[0012] Furthermore, based on the size of the vehicle and the maximum trajectory range of the vehicle when moving forward, the initial position of the adjustment direction that can drive out of the parking space without crossing the line is determined, and this is used as the setting position of the intermediate vehicle posture.

[0013] Furthermore, after obtaining the trajectory range that can avoid obstacles, it is also determined whether the trajectory range can cover the vehicle in the initial vehicle posture. If not, on the obstacle grid map, based on the last superimposed trajectory range, the maximum trajectory range of the vehicle is superimposed again, and it is ensured that: the trajectory range obtained after the second superposition and obstacle avoidance analysis can cover the vehicle in the initial vehicle posture, can avoid obstacles, intersects with the trajectory range obtained by the previous superposition, and the intersection is larger than the set size of the vehicle; if a trajectory range that meets the above conditions is found, the trajectory ranges obtained by the two superpositions are combined as the random sampling range.

[0014] Furthermore, if a trajectory range that meets the above conditions cannot be found, the intermediate vehicle posture is reset; if a trajectory range that meets the above conditions is still not found after multiple resets, it is considered that a direction switch is required.

[0015] Furthermore, the path point selection process is as follows: starting from the position corresponding to the intermediate vehicle posture and ending at the position corresponding to the initial vehicle posture, a search random tree is constructed based on the sampling points, and a search box is set. Each time, the nodes in a search box are searched, and the node that minimizes the cost function is selected as the next path point; the path point p i The cost function formula is:

[0016]

[0017] Among them, d(start,p i ) is the distance from the target pose to the path point p i The distance, d obs (p i ) is the path point p i The inverse of the distance to the nearest obstacle, is the vehicle's arrival point pi The deviation of the front wheel angle change between the time when the path point is reached and the time when the path point is reached, d θ (p i ) is the vehicle's arrival point p i The heading angle deviation from the target posture at time d xy (p i ) is the path point p i The distance deviation from the target pose, λ, ω θ 、ω xy are all weight coefficients.

[0018] A second aspect of the present invention provides a reverse path planning device for automatic parking, comprising:

[0019] The vehicle trajectory range calculation module is configured to: determine the maximum trajectory range of the current vehicle when traveling in each direction based on a given starting position according to vehicle performance;

[0020] The target vehicle posture determination module is configured to: obtain an obstacle grid map and determine the target parking space position and the target vehicle posture in the obstacle grid map;

[0021] The sampling range determination module is configured to: use the set position of the vehicle's front end when it exits the parking space as the intermediate vehicle posture, use the intermediate vehicle posture as the starting posture, superimpose the maximum trajectory range in the direction of the target vehicle posture on the obstacle grid map, and analyze to obtain a trajectory range that can avoid obstacles;

[0022] The reverse path point optimization module is configured to perform random sampling within the trajectory range, using the distance constraints to obstacles and the shortest overall path as constraints, select path points starting from the intermediate vehicle posture, obtain a reverse path planning sequence, and flip it to obtain a forward path planning sequence.

[0023] A third aspect of the present invention provides an electronic device, comprising a processor and a memory, wherein the memory stores computer instructions, and when the computer instructions are executed by the processor, the electronic device executes the method described.

[0024] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which implements the method described above when the program is executed by a processor.

[0025] A fifth aspect of the present invention provides a computer program product, comprising a computer program, and when the computer program is executed by a processor, the method described is implemented.

[0026] One or more of the above technical solutions determine the possible driving range of the vehicle based on the current vehicle performance, and then obtain the trajectory range that the vehicle can reach when driving out of the parking space under the premise of avoiding obstacles. Using the trajectory range as the sampling range, starting from the target posture, sampling points are selected point by point based on the set constraints, and the optimal path is selected to expand and construct the path until the path reaches the initial posture of the vehicle, realizing reverse path construction. It is particularly suitable for occasions where the space around the parking space is narrow. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The accompanying drawings, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0028] Figure 1 This is a flow chart of a reverse path planning method for automatic parking according to an embodiment of the present invention;

[0029] Figure 2 Schematic diagram of a portion of the vehicle path range in an embodiment of the present invention. DETAILED DESCRIPTION

[0030] The following describes embodiments of the present application in more detail with reference to the accompanying drawings. Although certain embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be construed as limited to the embodiments described herein. Instead, these embodiments are provided to provide a more thorough and complete understanding of the present application. It should be understood that the drawings and embodiments of the present application are for illustrative purposes only and are not intended to limit the scope of protection of the present application.

[0031] In the description of the embodiments of the present application, the term “including” and similar terms should be understood as open inclusion, that is, “including but not limited to.” The term “based on” should be understood as “at least partially based on.”

[0032] One or more embodiments of the present invention provide a reverse path planning method for automatic parking. This method effectively improves the efficiency and safety of automatic parking by combining reverse path construction and optimization with iterative sampling and the RRT* algorithm, and provides an innovative solution for solving parking needs in narrow or complex environments. Specifically, Figure 1 and 2 As shown, the method includes:

[0033] Step 1: Determine the maximum trajectory range of the current vehicle when it travels in each direction based on a given starting position according to vehicle performance;

[0034] Step 2: Obtain the obstacle grid map and determine the target parking space position and target vehicle pose in the obstacle grid map;

[0035] Step 3: Using the set position of the vehicle's front end when it leaves the parking space as the intermediate vehicle posture and the intermediate vehicle posture as the starting posture, the maximum trajectory range in the direction of the target vehicle posture is superimposed on the obstacle grid map to analyze and obtain the trajectory range that can avoid obstacles;

[0036] Step 4: Perform random sampling within the trajectory range, using the distance constraint to the obstacle and the shortest overall path as constraints, select path points starting from the middle vehicle posture, and obtain a reverse path planning sequence. Flip it to obtain a forward path planning sequence.

[0037] By determining the vehicle's possible driving range based on its current performance, the system then derives the trajectory range that the vehicle can reach when exiting the parking space while avoiding obstacles. Using this trajectory range as the sampling range, the system selects and constructs a path based on sampling points, starting from the target position. Through iterative optimization, the optimal path is selected and expanded each time until the path reaches or approaches the vehicle's current position, thus achieving reverse path construction. This process is particularly suitable for situations where the space around parking spaces is limited. This process fully utilizes the intuitiveness and efficiency of reverse planning, significantly improving the flexibility and adaptability of path planning.

[0038] In step 1, the vehicle performance includes the front wheel angle and the optional range of driving step length η. The current vehicle's front wheel steering angle and driving step length are sampled within the optional range to obtain multiple front wheel steering angle and driving step length combinations. Based on each combination of front wheel steering angle and driving step length, vehicle trajectories are generated in different driving directions to obtain trajectory ranges in different driving directions.

[0039] Specifically, when the driving direction D is fixed, the front wheel turning angle of the vehicle and driving step length η are sampled within the optional range to form a set of control variables; different control variables are generated The corresponding vehicle running trajectory; the generation of different control variables The corresponding vehicle trajectory is: Using the Ackerman steering vehicle kinematic model and the second-order Runge-Kutta method to construct the vehicle state transfer equation is:

[0040]

[0041] In the formula, a set of control variables input into the vehicle state transfer equation in the i-th iteration is recorded as

[0042]

[0043] p i+1 To use the control variable u iThe vehicle posture after the state transfer under control, including the lateral position x after the state transfer i+1 , vertical position y i+1 , heading angle θ i+1 ;p i is the vehicle posture before state transfer, including the lateral position x before state transfer i , vertical position y i , heading angle θ i ; D is the direction of travel; L is the vehicle wheelbase.

[0044] Based on the vehicle's front wheel angle The maximum and minimum values ​​of the driving step length η can be used to obtain the trajectory boundaries of the vehicle under different driving directions, and then the trajectory range under different driving directions. Among them, different driving directions include left front, right front, left rear, and right rear.

[0045] It can be understood that when "trajectory" is mentioned in this article, it refers to a center point of the vehicle, such as the center of the rear bumper of the vehicle model, the center of the ellipse of the vehicle model, etc.

[0046] In step 2, the obstacle grid map may be determined based on the surrounding environment data. In fact, the surrounding environment data may be obtained based on point cloud data or in other ways, which are not specifically limited here.

[0047] After obtaining the obstacle grid map and determining the target parking space within it, the system then determines, based on the current vehicle's size, whether sufficient space is left and right when the vehicle is parked in the target space. Based on this, the target vehicle position in the space is determined. Generally, the target vehicle position is when the vehicle is parked in the middle of the space, parallel to the side parking space lines. However, in special cases, if an obstacle (such as another vehicle) is pressing against the line, the vehicle can be positioned off-center, provided sufficient space is left and right.

[0048] The target pose is used as the planning starting point and is represented by coordinates (x, y) and angle θ to provide a clear starting point for subsequent reverse path planning.

[0049] In step 3, it is assumed that the steering wheel state is recorded as the initial state, and three states, left and right, based on the initial state are recorded. Generally speaking, whether reversing into a parking space or parallel parking, the steering wheel state must be changed at least once. For example, when reversing into a parking space, the steering wheel must be straightened after the rear of the vehicle enters the parking space; when parallel parking, the steering wheel must be reversed after entering the parking space. Based on this, in this embodiment, the position and posture of the vehicle at which the steering wheel state is last adjusted after the rear of the vehicle enters the parking space is recorded as the intermediate vehicle posture.

[0050] For vertical and diagonal parking spaces, the intermediate vehicle posture is the posture when the front of the vehicle is in the initial steering wheel position and moves out of the parking space. For side parking spaces, the intermediate vehicle posture is the posture when the front of the vehicle moves out of the parking space and forms a certain angle with the target vehicle posture and points outward from the parking space.

[0051] As a specific implementation method, the initial position of the adjustment direction that can be used to drive out of the parking space without crossing the line can be determined based on the vehicle size and the maximum trajectory range of the vehicle when moving forward, and this can be used as the setting position of the intermediate vehicle posture.

[0052] As mentioned above, whether reversing into a garage or parallel parking, the steering wheel state needs to be changed at least once, and the driving direction may also need to be adjusted. Based on this, in step 4, it is first determined whether the trajectory range can cover the vehicle in the initial vehicle posture.

[0053] If it can be covered, it is considered that the initial vehicle posture can be reached within the current trajectory range. The trajectory range is randomly sampled, and the distance constraint to the obstacle and the shortest overall path are used as constraints. Path points are selected from the middle vehicle posture to obtain a reverse path planning sequence, which is flipped to obtain a forward path planning sequence.

[0054] If it cannot be covered, it is considered that the initial vehicle posture cannot be reached within the current trajectory range. In this case, as a specific implementation method, on the obstacle grid map, based on the last superimposed trajectory range, the maximum trajectory range of the vehicle is superimposed again, and it is guaranteed that: (1) the trajectory range obtained after superimposition and obstacle avoidance analysis can cover the vehicle under the initial vehicle posture, (2) it can avoid obstacles, and (3) it intersects with the trajectory range obtained by the previous superposition, and the intersection is larger than the vehicle set size (to ensure that the vehicle can pass through the intersection). If a trajectory range that meets the above conditions is found, the trajectory ranges obtained by the two superpositions are merged, and random sampling is performed within the range. The distance constraint with the obstacle and the shortest overall path are used as constraints. Path points are selected starting from the middle vehicle posture to obtain a reverse path planning sequence, which is then flipped to obtain a forward path planning sequence.

[0055] If there is no trajectory range that meets the above conditions, return to step 3 and reset the intermediate vehicle posture. If there is still no trajectory range that meets the above conditions after resetting multiple times, it is considered that direction switching is required. Since the space near the parking space is often narrow, not all positions meet the requirements for adjusting the direction. Therefore, first, based on the size of the vehicle and the maximum trajectory range of the vehicle when reversing, determine the adjustable space closest to the parking space and the vehicle's starting posture when adjusting in the adjustable space. The vehicle's starting posture is used as the second intermediate vehicle posture. On the obstacle grid map, the maximum trajectory range of the vehicle is superimposed to meet the following requirements: (1) The trajectory range obtained after superposition and obstacle avoidance analysis can cover the vehicle in the initial vehicle posture, and (2) it can avoid obstacles; based on the obtained trajectory range, determine the target direction that the vehicle needs to adjust, and record the vehicle posture after adjusting the direction as the third intermediate vehicle posture. Random sampling is performed within the trajectory range between the first intermediate vehicle posture and the second intermediate vehicle posture, and between the third intermediate vehicle posture and the target vehicle posture, and path planning is performed based on the above method.

[0056] Understandably, to ensure the accuracy and success rate of the above solution and avoid excessive algorithmic complexity, a reference distance between the vehicle and the target parking space can be set to a minimum. By limiting the path sampling points to the reachable range of the vehicle's trajectory, subsequent sampling points are guaranteed to be accessible to the vehicle, improving the success rate of path planning.

[0057] The above random sampling can be achieved by constructing a search tree or graph. Since these points are restricted to the possible driving trajectory of the vehicle, the accuracy of the path planning is guaranteed.

[0058] In step 4, the distance constraint to the obstacle and the shortest overall path are used as constraints, and the path point selection is performed starting from the intermediate vehicle posture. Specifically, the RRT* algorithm can be used to ensure the efficiency and safety of the path. Specifically, the position corresponding to the intermediate vehicle posture is used as the starting point, and the position corresponding to the initial vehicle posture is used as the end point. A search random tree is constructed based on the sampling points, and a search box is set. Each node within the search box is searched, and the node that minimizes the cost function is selected as the next path point. Specifically, the cost function of each point is defined as:

[0059]

[0060] Among them, C(p i ) is the path point p i The cost function, d(start,p i ) is the distance from the target pose to the path point p i distance, λ is the obstacle penalty coefficient, which is used to constrain path planning to ensure the shortest path; d obs (pi ) is the path point p i The reciprocal of the distance to the nearest obstacle is used to constrain the distance to the obstacle and keep the vehicle as far away from the obstacle as possible; is the weight of the front wheel angle change term, is the vehicle's arrival point p i The deviation of the front wheel angle change between the time when the target point is reached and the time when the target point is reached; ω θ is the heading angle deviation weight, dθ(p i ) is the vehicle's arrival point p i The heading angle deviation from the target posture at time ω is used to make the vehicle's posture closer to the initial posture; xy is the vehicle position deviation weight, d xy (p i ) is the path point p i By optimizing the cost function, the present invention can generate a parking path that is both efficient and safe.

[0061] By adjusting the weight ω θ 、ω xy , the cost function of the path can be optimized. Specifically, the closer to the initial vehicle posture, the more the heading angle deviation weight ω is increased. θ The cost function considers multiple factors simultaneously, eliminating the need for separate collision detection and simultaneously approaching the initial vehicle position in terms of position and attitude, thus improving path planning efficiency.

[0062] During the path planning process, the vehicle posture gradually approaches the initial posture by adjusting each path point based on the shortest distance to the target position, getting closer to the initial position, minimizing the front wheel angle adjustment, and the heading angle getting closer to the initial posture.

[0063] The optimized path is output in reverse, and combined with the target vehicle posture and the intermediate vehicle posture to obtain the parking trajectory from the current vehicle posture to the target parking posture, thus realizing automatic parking.

[0064] It is understandable that further verification can be performed based on the obtained path, and smooth operation can be performed within a reasonable range to avoid discomfort caused by frequent adjustments of direction in a short period of time.

[0065] Based on the above method, one or more embodiments of the present invention further provide an automatic parking reverse path planning device, comprising:

[0066] The vehicle trajectory range calculation module is configured to: determine the maximum trajectory range of the current vehicle when traveling in each direction based on a given starting position according to vehicle performance;

[0067] The target vehicle posture determination module is configured to: obtain an obstacle grid map and determine the target parking space position and the target vehicle posture in the obstacle grid map;

[0068] The sampling range determination module is configured to: use the set position of the vehicle's front end when it exits the parking space as the intermediate vehicle posture, use the intermediate vehicle posture as the starting posture, superimpose the maximum trajectory range in the direction of the target vehicle posture on the obstacle grid map, and analyze to obtain a trajectory range that can avoid obstacles;

[0069] The reverse path point optimization module is configured to perform random sampling within the trajectory range, using the distance constraints to obstacles and the shortest overall path as constraints, select path points starting from the intermediate vehicle posture, obtain a reverse path planning sequence, and flip it to obtain a forward path planning sequence.

[0070] One or more embodiments of the present invention further provide an electronic device that can be used to implement the automatic parking reverse path planning method in the above embodiments. The electronic device includes one or more processors, one or more memories coupled to the processors, and a communication module coupled to the processors.

[0071] The memory may include one or more non-volatile memories and one or more volatile memories. Examples of non-volatile memories include, but are not limited to, at least one of the following: read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, hard disk, compact disc (CD), digital video disc (DVD), or other magnetic storage and / or optical storage. Examples of volatile memories include, but are not limited to, at least one of the following: random access memory (RAM), or other volatile memories that do not persist during a power outage. The computer program may be stored in the ROM. When the processor executes the computer program, the above-mentioned automatic parking reverse path planning method is implemented.

[0072] In some embodiments, the program may be tangibly contained in a computer-readable medium, which may be included in a device (such as a memory) or other storage device accessible by the device. The program may be loaded from the computer-readable medium into RAM for execution. The computer-readable medium may include any type of tangible non-volatile memory, such as ROM, EPROM, flash memory, or a hard disk. The computer-readable storage medium stores a computer program that, when executed by a processor, implements the above-described automatic parking reverse path planning method.

[0073] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a server or terminal, the process or function described in the embodiment of the present application is generated in whole or in part. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that can be accessed by a server or terminal or a data storage device such as a server or data center that includes one or more available media. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, and a tape, etc.), an optical medium (e.g., a digital video disk (DVD), etc.), or a semiconductor medium (e.g., a solid-state drive, etc.).

[0074] In addition, although adopting specific order to describe each operation, this should be understood as requiring such operation to be carried out with shown specific order or with sequential order, or requiring all illustrated operations to be carried out to obtain desired result.Under certain environment, multitasking and parallel processing may be advantageous.Similarly, although comprising some specific implementation details in the above discussion, these should not be interpreted as limiting the scope of the application.Some features described in the context of independent embodiment can also be implemented in a single implementation in combination.On the contrary, the various features described in the context of independent implementation also can be implemented in a plurality of implementations individually or in the mode of any suitable subcombination.

[0075] Although the subject matter has been described in language specific to structural features and / or methodological logical acts, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are merely example forms of implementing the claims.

Claims

1. A reverse path planning method for automatic parking, characterized in that: The following steps are involved: Determine the maximum trajectory range of the current vehicle when traveling in each direction based on a given starting position according to vehicle performance; Obtain an obstacle grid map and determine the target parking space position and target vehicle posture in the obstacle grid map; The set position of the vehicle's front end when it leaves the parking space is used as the intermediate vehicle posture, the intermediate vehicle posture is used as the starting posture, and the maximum trajectory range in the direction of the target vehicle posture is superimposed on the obstacle grid map to analyze and obtain the trajectory range that can avoid obstacles; After obtaining the trajectory range that can avoid obstacles, it is also determined whether the trajectory range can cover the vehicle in the initial vehicle posture. If not, the maximum trajectory range of the vehicle is superimposed again on the obstacle grid map based on the last superimposed trajectory range, and it is ensured that: the trajectory range obtained after the second superposition and obstacle avoidance analysis can cover the vehicle in the initial vehicle posture, can avoid obstacles, intersects with the trajectory range obtained by the previous superposition, and the intersection is larger than the set vehicle size; if a trajectory range that meets the above conditions is found, the trajectory ranges obtained by the two superpositions are combined as the random sampling range; Random sampling is performed within the trajectory range, and the distance constraint to the obstacle and the shortest overall path are used as constraints. Path points are selected starting from the intermediate vehicle posture to obtain a reverse path planning sequence, which is then flipped to obtain a forward path planning sequence. The path point selection process is as follows: starting from the position corresponding to the intermediate vehicle posture and ending at the position corresponding to the initial vehicle posture, a search random tree is constructed based on the sampling points, and a search box is set. Each time, the nodes within a search box are searched, and the node that minimizes the cost function is selected as the next path point. Path points The cost function formula is: in, is the distance from the target pose to the path point distance, Is a waypoint The inverse of the distance to the nearest obstacle, The vehicle arrives at the waypoint The deviation of the front wheel angle change between the time when the path point is reached and the time when the path point is reached is: The vehicle arrives at the waypoint The heading angle deviation from the target posture at time , Is a waypoint The distance deviation from the target pose, 、 、 、 are all weight coefficients.

2. The reverse path planning method for automatic parking according to claim 1, characterized in that: The vehicle performance includes an optional range of the front wheel steering angle and the driving step length. Sampling is performed within the optional range of the front wheel steering angle and the driving step length of the current vehicle to obtain multiple combinations of front wheel steering angle and driving step length. Based on each group of front wheel steering angle and driving step length, the vehicle running trajectory is generated in different driving directions to obtain the trajectory range in different driving directions.

3. The reverse path planning method for automatic parking according to claim 1, characterized in that: Based on the size of the vehicle and the maximum trajectory range of the vehicle when moving forward, the initial position of the adjustment direction that can be used to drive out of the parking space without crossing the line is determined, and this is used as the setting position of the intermediate vehicle posture.

4. The reverse path planning method for automatic parking according to claim 1, wherein: If the trajectory range that meets the above conditions cannot be found, the intermediate vehicle posture is reset; if the trajectory range that meets the above conditions is still not found after multiple resets, it is considered necessary to switch direction.

5. An automatic parking reverse path planning device, characterized in that: include: The vehicle trajectory range calculation module is configured to: determine the maximum trajectory range of the current vehicle when traveling in each direction based on a given starting position according to vehicle performance; The target vehicle posture determination module is configured to: obtain an obstacle grid map and determine the target parking space position and the target vehicle posture in the obstacle grid map; The sampling range determination module is configured to: take the set position when the front of the vehicle drives out of the parking space as the intermediate vehicle posture, take the intermediate vehicle posture as the starting posture, superimpose the maximum trajectory range in the direction of the target vehicle posture on the obstacle grid map, and analyze to obtain the trajectory range that can avoid obstacles; after obtaining the trajectory range that can avoid obstacles, it is also judged whether the trajectory range can cover the vehicle in the initial vehicle posture. If not, on the basis of the last superimposed trajectory range, the maximum trajectory range of the vehicle is superimposed again on the obstacle grid map, and it is ensured that: the trajectory range obtained after the second superposition and obstacle avoidance analysis can cover the vehicle in the initial vehicle posture, can avoid obstacles, intersects with the trajectory range obtained by the last superposition, and the intersection is larger than the set size of the vehicle; if a trajectory range that meets the above conditions is found, the trajectory ranges obtained by the two superpositions are merged as the random sampling range; The path point reverse optimization module is configured as follows: random sampling is performed within the trajectory range, and the distance constraints to obstacles and the shortest overall path are used as constraints. The path point is selected from the intermediate vehicle posture to obtain a reverse path planning sequence, and the forward path planning sequence is obtained by flipping it. The path point selection process is as follows: starting from the position corresponding to the intermediate vehicle posture and ending at the position corresponding to the initial vehicle posture, a search random tree is constructed based on the sampling points, and a search box is set. Each time, the nodes in the search box are searched, and the node that minimizes the cost function is selected as the next path point. The cost function formula is: in, is the distance from the target pose to the path point distance, Is a waypoint The inverse of the distance to the nearest obstacle, The vehicle arrives at the waypoint The deviation of the front wheel angle change between the time when the path point is reached and the time when the path point is reached is: The vehicle arrives at the waypoint The heading angle deviation from the target posture at time , Is a waypoint The distance deviation from the target pose, 、 、 、 are all weight coefficients.

6. An electronic device comprising a processor and a memory, wherein the memory stores computer instructions, characterized in that: When the computer instructions are executed by the processor, the electronic device is caused to perform the method according to any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

8. A computer program product, comprising a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images