Method and device for planning parking path, vehicle and storage medium

Abstract

The invention provides a parking path planning method and device, a vehicle and a storage medium. In the method, after a plurality of parking paths are determined based on an initial pose of a vehicle, a target pose of a target parking space and a preset parking operation of the vehicle in a parking process, the quality of each parking path is evaluated by the method. The invention discloses a quality evaluation method. The angle deviation (capable of being mapped as the parking precision of the path end point) between the vehicle head orientation of the last section of parking operation corresponding to the parking path and the orientation of the target parking space, the number of short paths in the parking path (capable of being mapped as the operation efficiency corresponding to the parking path) and the proportion of path nodes meeting preset conditions in the parking path are introduced. The safety corresponding to the parking path can be mapped, the three planning indexes can well and comprehensively quantify multiple aspects of the parking path, the limitation of single path planning can be well overcome, and the adaptability of the parking path is improved.

Description

Technical Field

[0001] The present application relates to the field of vehicle technology, and more particularly, to a method, device, vehicle, and storage medium for planning a parking path in the field of vehicle technology. Background Art

[0002] With the development of autonomous driving technology, automatic parking systems have become an essential component of intelligent vehicles. By integrating perception, decision-making, and control subsystems, these systems enable fully automated parking for intelligent vehicles without the driver having to directly control the steering wheel, accelerator, or brake pedals.

[0003] In related technologies, automated parking systems use a single-path algorithm to plan parking paths. However, due to local optima, environmental complexity, or algorithmic flaws, single-path algorithms can only find a single feasible path under specific assumptions. Therefore, the adaptability of the planned path needs to be improved.

[0004] Therefore, a solution that can overcome the limitations of single path planning is urgently needed. Summary of the Invention

[0005] The present application provides a method, device, vehicle, and storage medium for planning a parking path. The method can effectively overcome the limitations of single path planning and improve the adaptability of the parking path.

[0006] In a first aspect, a method for planning a parking path is provided. The method includes: determining, in response to a parking instruction from a vehicle, multiple parking paths based on an initial position of the vehicle at a current position, a target position of a target parking space, and a preset parking operation of the vehicle during parking; determining a quality score for each parking path on a first planning objective, the first planning objective comprising an angular deviation between a vehicle head orientation in a last parking operation corresponding to the parking path and an orientation of the target parking space, the number of short paths in the parking path, and a proportion of path nodes in the parking path that meet a preset condition, wherein a short path is a path corresponding to a parking operation having a length less than a preset length, the preset condition being used to indicate that a curvature of a path node corresponding to the position is less than a preset curvature; and controlling parking of the vehicle based on the parking path with the highest quality score among the multiple parking paths.

[0007] In the above technical solution, after determining multiple parking paths based on the vehicle's initial position, the target position of the target parking space, and the vehicle's pre-set parking maneuvers during the parking process, the method then evaluates the quality of each parking path. This quality evaluation incorporates the angular deviation between the vehicle's heading during the final parking maneuver and the heading of the target parking space (which can be mapped to parking accuracy at the path endpoint), the number of short paths within the parking path (which can be mapped to operational efficiency), and the proportion of path nodes within the parking path that meet pre-set requirements (which can be mapped to safety). These three planning metrics effectively and comprehensively quantify multiple aspects of the parking path, effectively overcoming the limitations of single-path planning and improving the adaptability of parking paths. Furthermore, while enhancing the comprehensiveness of parking path scoring, controlling vehicle parking based on the parking path with the highest quality score can significantly improve parking efficiency and quality.

[0008] In combination with the first aspect, in certain possible implementations, multiple parking paths are determined based on the initial posture of the vehicle at the current position, the target posture of the target parking space, and the preset parking operation of the vehicle during the parking process, including: starting from the initial posture, using the preset parking operation, determining multiple first predicted postures; taking any first predicted posture as the second predicted posture, and discretizing the second predicted posture based on a preset grid resolution to obtain grid coordinates corresponding to the second predicted posture; determining whether the second predicted posture should be discarded based on whether the grid coordinates corresponding to the second predicted posture are the same as the grid coordinates corresponding to the first posture, so as to correct the multiple first predicted postures, and the path node corresponding to the first posture is the parent node of the path node corresponding to the second predicted posture; determining the multiple parking paths based on the initial posture, the target posture, and the corrected multiple first predicted postures.

[0009] In the above technical solution, multiple first predicted poses are generated based on discrete actions of a preset parking operation, ensuring that the parking path complies with the vehicle's kinematic constraints and avoiding the generation of invalid paths that exceed the vehicle's physical limits. Secondly, a grid resolution is introduced to discretize the first predicted poses, mapping the continuous pose states to grid coordinates. By comparing the grid coordinates corresponding to any first predicted pose (second predicted pose) with the grid coordinates corresponding to the first pose of its parent node, it is possible to determine whether the second predicted pose and the first pose are mapped to the same grid. If they are identical, the second predicted pose is discarded, which effectively eliminates redundant states, compresses the pose space, and improves the search efficiency of path nodes. Furthermore, based on the corrected multiple first predicted poses, the initial pose, and the target pose, multiple parking paths are constructed, which retains physically feasible candidate paths with discrete and independent states. This provides a rich and efficient path input source for parking path quality assessment, assisting the automatic parking system in determining parking paths that excel in all planning objectives.

[0010] In combination with the first aspect and the above-mentioned implementation methods, in some possible implementation methods, the preset grid resolution includes position resolution and orientation resolution. Based on the preset grid resolution, the second predicted posture is discretized to obtain the grid coordinates corresponding to the second predicted posture, including: determining the ratio of the first horizontal axis coordinate in the second predicted posture to the first length corresponding to the position resolution as the target first horizontal axis coordinate in the grid coordinate; determining the ratio of the first vertical axis coordinate in the second predicted posture to the first length as the target first vertical axis coordinate in the grid coordinate; and determining the ratio of the first orientation in the second predicted posture to the first orientation corresponding to the orientation resolution as the target first orientation in the grid coordinate.

[0011] In this technical solution, the grid coordinates corresponding to the second predicted pose are determined by mapping the rigid ratio between the predicted pose and the resolution. This eliminates floating-point arithmetic errors and ensures the accuracy of the mesh coordinates. The dimensional decoupling process described above allows the horizontal and vertical coordinates to be discretized independently, avoiding mesh distortion caused by joint quantization.

[0012] In combination with the first aspect and the above-mentioned implementation manner, in some possible implementation manners, based on the initial posture, the target posture and the revised multiple first predicted postures, the multiple parking paths are determined, including: determining the second posture corresponding to the target posture from the revised multiple first predicted postures, and connecting the path node corresponding to the target posture with the path node corresponding to the second posture, and the parent node of the path node corresponding to the target posture is the path node corresponding to the second posture; deleting the second posture from the revised multiple first predicted postures to obtain at least one third predicted posture, and determining the third posture corresponding to each third predicted posture, and connecting the path node corresponding to each third predicted posture with the path node corresponding to the third posture. The path nodes are connected, and the parent node of the path node corresponding to the third predicted posture is the path node corresponding to the third posture, so as to determine the fourth predicted postures corresponding to the initial posture, and the path nodes corresponding to the fourth predicted postures should be connected with the path node corresponding to the initial posture, and the parent node of the path nodes corresponding to the fourth predicted postures is the path node corresponding to the initial posture; based on the connection relationship between the multiple path nodes formed by the initial posture, the target posture and the corrected multiple first predicted postures, starting from the path node corresponding to the target posture, passing through the path nodes corresponding to the corrected multiple first predicted postures, and tracing back to the path node corresponding to the initial posture in sequence, the multiple parking paths are obtained.

[0013] In the above technical solution, the second posture corresponding to the target posture is identified and a parent-child relationship is established, that is, the path node corresponding to the target posture is connected to the path node corresponding to the second posture. This can ensure that each parking path ends accurately at the target parking space and avoid end point deviation. Furthermore, at least one third predicted posture remaining after the second posture is deleted and processed, and the parent node is recursively connected to the path node corresponding to the initial posture. This can cover multiple first predicted postures that have been corrected after discretization processing and quickly generate multiple parking paths. In addition, the complete connection relationship from the target posture back to the initial posture can directly output multiple physically feasible parking paths, providing a low-redundancy and high-diversity path input source for the quality assessment of the parking path, significantly improving the generation efficiency and integrity of multiple parking paths.

[0014] In combination with the first aspect and the above-mentioned implementation methods, in certain possible implementation methods, determining the quality score of each parking path on the first planning target includes: outputting a weight configuration reminder, the weight configuration reminder being used to remind the target object in the vehicle to configure the weights corresponding to each first planning target; upon receiving the weights corresponding to each first planning target, performing a weighted fusion on the first score, the second score, and the third score based on the first configuration weight, the second configuration weight, and the third configuration weight to obtain the quality score, the first score being obtained based on the angle deviation, the second score being obtained based on the number of short paths, and the third score being obtained based on the proportion of path nodes that meet preset conditions; upon not receiving the weights corresponding to each first planning target, outputting a target configuration reminder, the target configuration reminder being used to remind the target object to use driving style as a planning target; upon receiving a confirmation operation for the target configuration reminder, determining, based on the driving style, a fourth configuration weight, a fifth configuration weight, and a sixth configuration weight for weighted fusion of the first score, the second score, and the third score to determine the quality score.

[0015] In the above technical solution, the weight configuration reminder allows the target object to adjust the angular deviation between the direction of the vehicle head in the last parking operation and the direction of the target parking space, the number of short paths, and the proportion of path nodes that meet the preset conditions in the evaluation of path quality as needed. This enables the quality score (scoring result) to be personalized to match the parking preferences of the driving target object. When not configured, it intelligently guides the choice of driving style, converts the abstract driving style into specific weight parameters, and solves the pain point of insufficient universality caused by the default weight. In addition, during manual configuration, multiple scores are weighted and fused based on the configured weights; when no configuration is done, dynamic weights are derived through driving style. This two-stage scoring mechanism can cover a variety of weight determination scenarios and significantly improve the accuracy of parking path screening.

[0016] In combination with the first aspect and the above-mentioned implementation manner, in some possible implementation manners, based on the driving style, determining the fourth configuration weight, the fifth configuration weight and the sixth configuration weight for weighted fusion of the first score, the second score and the third score, including: when the driving style is an aggressive style, determining the fourth configuration weight as the first preset weight, determining the fifth configuration weight as the second preset weight, and determining the sixth configuration weight as the third preset weight, the first preset weight is less than the second preset weight, and the third preset weight is less than the second preset weight; when the driving style is a conservative style, determining the fourth configuration weight as the fourth preset weight, determining the fifth configuration weight as the fifth preset weight, and determining the sixth configuration weight as the sixth preset weight, the fifth preset weight is less than the fourth preset weight, and the fifth preset weight is less than the sixth preset weight; when the driving style is a balanced style, determining the fourth configuration weight, the fifth configuration weight and the sixth configuration weight as the target preset weights.

[0017] In the above technical solution, under the aggressive style, the configuration weight corresponding to the angle deviation is reduced (the first preset weight) and the configuration weight corresponding to the above proportion is reduced (the third preset weight), and the weight corresponding to the number of short paths is increased (the second preset weight). The setting of the above configuration weights is more suitable for the target object of pursuing parking efficiency. Under the robust style, the curvature safety of the path nodes is strengthened (the sixth preset weight is larger) and the importance of angle deviation is emphasized, which can ensure the smoothness of the vehicle when turning. Under the balanced style, the above three planning goals are balanced with the same weight. This solution can significantly optimize the scenario adaptability of the quality score of the parking path through a directional mapping mechanism between driving style and configuration weights.

[0018] In combination with the first aspect and the above-mentioned implementations, in certain possible implementations, determining a quality score for each parking path on a first planning objective includes: determining a target path cost for each parking path, where the parking path is determined based on the initial position, the target position, and at least one first predicted position related to the preset parking operation, the target path cost being the sum of a first path cost for the vehicle traveling from the initial position to any first predicted position and a second path cost for traveling from any first predicted position to the target position, the first path cost being the sum of a cumulative path cost for the vehicle traveling from the initial position to a fifth predicted position and an operation cost for traveling from the fifth predicted position to any first predicted position, the path node corresponding to the fifth predicted position being the parent node of the path node corresponding to any first predicted position; determining the target path cost, total path length, and number of path nodes for each parking path as a second planning objective when planning the parking path; and determining a quality score for each parking path on the first planning objective and the second planning objective.

[0019] In the above technical solution, the total path cost (target path cost) of each parking path is determined. The target path cost starts from the initial posture, goes to the first predicted posture corresponding to the intermediate path node, and finally to the target posture, integrating all historical and future costs. This can circumvent the limitations of only evaluating the end state (such as angular deviation), accurately quantify the cost of the path segment, and improve the accuracy of determining the quality score. Furthermore, incorporating the total path length and the number of paths into the second planning goal can increase the comprehensiveness of the planning goals and make the quality score more accurate. This solution can significantly improve the comprehensiveness and practicality of parking path scoring through the fusion mechanism of hierarchical path cost calculation and multiple planning goals.

[0020] In a second aspect, a device for planning a parking path is provided. The device includes: a determination module for: determining, in response to a parking instruction from a vehicle, a plurality of parking paths based on an initial position of the vehicle at a current position, a target position of a target parking space, and a preset parking operation of the vehicle during parking; determining a quality score of each parking path on a first planning objective, wherein the first planning objective includes an angular deviation between a vehicle head orientation in a last parking operation corresponding to the parking path and an orientation of the target parking space, the number of short paths in the parking path, and a proportion of path nodes in the parking path that meet a preset condition, wherein the short path is a path corresponding to a parking operation having a length less than a preset length, wherein the preset condition is used to indicate that a curvature of a path node corresponding to the position is less than a preset curvature; and a control module for controlling the parking of the vehicle based on the parking path with the highest quality score among the plurality of parking paths.

[0021] In combination with the second aspect, in some possible implementations, the determination module is specifically used to: starting from the initial posture, use the preset parking operation to determine multiple first predicted postures; take any first predicted posture as the second predicted posture, and discretize the second predicted posture based on a preset grid resolution to obtain grid coordinates corresponding to the second predicted posture; determine whether the second predicted posture should be discarded based on whether the grid coordinates corresponding to the second predicted posture are the same as the grid coordinates corresponding to the first posture, so as to correct the multiple first predicted postures, and the path node corresponding to the first posture is the parent node of the path node corresponding to the second predicted posture; determine the multiple parking paths based on the initial posture, the target posture and the corrected multiple first predicted postures.

[0022] In combination with the second aspect and the above-mentioned implementation methods, in some possible implementation methods, the preset grid resolution includes position resolution and orientation resolution, and the device also includes: a processing module, used to: determine the ratio between the first horizontal axis coordinate in the second predicted posture and the first length corresponding to the position resolution as the target first horizontal axis coordinate in the grid coordinate; determine the ratio between the first vertical axis coordinate in the second predicted posture and the first length as the target first vertical axis coordinate in the grid coordinate; determine the ratio between the first orientation in the second predicted posture and the first orientation corresponding to the orientation resolution as the target first orientation in the grid coordinate.

[0023] In combination with the second aspect and the above-mentioned implementation method, in some possible implementation methods, the determination module is further specifically used to: determine the second posture corresponding to the target posture from the corrected multiple first predicted postures, and connect the path node corresponding to the target posture with the path node corresponding to the second posture, and the parent node of the path node corresponding to the target posture is the path node corresponding to the second posture; delete the second posture from the corrected multiple first predicted postures to obtain at least one third predicted posture, and determine the third posture corresponding to each third predicted posture, and connect the path node corresponding to each third predicted posture with the path node corresponding to the third posture, and the third predicted posture The parent node of the path node corresponding to the measured posture is the path node corresponding to the third posture, so as to determine the fourth predicted postures corresponding to the initial posture. The path nodes corresponding to the fourth predicted postures should be connected with the path node corresponding to the initial posture, and the parent node of the path node corresponding to the fourth predicted posture is the path node corresponding to the initial posture; based on the connection relationship between the multiple path nodes formed by the initial posture, the target posture and the corrected multiple first predicted postures, starting from the path node corresponding to the target posture, passing through the path nodes corresponding to the corrected multiple first predicted postures, and tracing back to the path node corresponding to the initial posture in sequence, multiple parking paths are obtained.

[0024] In combination with the second aspect and the above-mentioned implementation manner, in some possible implementation manners, the device also includes: an output module for outputting a weight configuration reminder, the weight configuration reminder being used to remind the target object in the vehicle to configure the weights corresponding to each first planning target; a fusion module for, upon receiving the weights corresponding to each first planning target, performing weighted fusion on the first score, the second score, and the third score based on the first configuration weight, the second configuration weight, and the third configuration weight to obtain the quality score, the first score being obtained based on the angle deviation, the second score being obtained based on the number of short paths, and the third score being obtained based on the proportion of path nodes that meet preset conditions; the output module is also used to output a target configuration reminder when the weights corresponding to each first planning target are not received, the target configuration reminder being used to remind the target object to use the driving style as a planning target; the determination module is specifically further used to determine, based on the driving style, a fourth configuration weight, a fifth configuration weight, and a sixth configuration weight for weighted fusion of the first score, the second score, and the third score to determine the quality score when receiving a confirmation operation for the target configuration reminder.

[0025] In combination with the second aspect and the above-mentioned implementation methods, in some possible implementation methods, the determination module is further specifically used to: when the driving style is an aggressive style, determine the fourth configuration weight as the first preset weight, determine the fifth configuration weight as the second preset weight, and determine the sixth configuration weight as the third preset weight, the first preset weight is less than the second preset weight, and the third preset weight is less than the second preset weight; when the driving style is a conservative style, determine the fourth configuration weight as the fourth preset weight, determine the fifth configuration weight as the fifth preset weight, and determine the sixth configuration weight as the sixth preset weight, the fifth preset weight is less than the fourth preset weight, and the fifth preset weight is less than the sixth preset weight; when the driving style is a balanced style, determine the fourth configuration weight, the fifth configuration weight and the sixth configuration weight as the target preset weights.

[0026] In combination with the second aspect and the above-mentioned implementation manner, in some possible implementation manners, the determination module is further specifically used to: determine a target path cost for each parking path, where the parking path is determined based on the initial posture, the target posture, and at least one first predicted posture related to the preset parking operation, the target path cost being the sum of a first path cost for the vehicle traveling from the initial posture to any first predicted posture and a second path cost for traveling from any first predicted posture to the target posture, the first path cost being the sum of a cumulative path cost for the vehicle traveling from the initial posture to a fifth predicted posture and an operation cost for traveling from the fifth predicted posture to any first predicted posture, the path node corresponding to the fifth predicted posture being the parent node of the path node corresponding to any first predicted posture; determining the target path cost, total path length, and number of path nodes of each parking path as a second planning target when planning the parking path; and determining a quality score for each parking path based on the first planning target and the second planning target.

[0027] In a third aspect, a vehicle is provided, comprising a memory and a processor. The memory is configured to store executable program code, and the processor is configured to retrieve and execute the executable program code from the memory, so that the vehicle executes the method of the first aspect or any possible implementation of the first aspect.

[0028] In a fourth aspect, a computer-readable storage medium is provided, which stores an executable program code. When the executable program code runs on a computer, the computer executes the method in the above-mentioned first aspect or any possible implementation of the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 is a schematic diagram of a parking scenario provided in an embodiment of the present application;

[0030] Figure 2 is a schematic flow chart of a method for planning a parking path provided in an embodiment of the present application;

[0031] Figure 3 is a schematic diagram of determining an initial posture and a target posture provided by an embodiment of the present application;

[0032] Figure 4 is a schematic diagram of determining multiple parking paths provided by an embodiment of the present application;

[0033] Figure 5 1 is a schematic structural diagram of a device for planning a parking path provided in an embodiment of the present application;

[0034] Figure 6 It is a structural schematic diagram of a vehicle provided in an embodiment of the present application. DETAILED DESCRIPTION

[0035] The following will clearly and thoroughly describe the technical solutions in this application in conjunction with the accompanying drawings. In the description of the embodiments of this application, unless otherwise specified, " / " means or, for example, A / B can mean A or B: "and / or" in the text is only a description of the association relationship of associated objects, indicating that there can be three relationships, for example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the description of the embodiments of this application, "multiple" means two or more than two.

[0036] In the following, the terms "first" and "second" are used for descriptive purposes only and should not be understood to imply or suggest relative importance or implicitly indicate the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features.

[0037] Figure 1 This is a schematic diagram of a parking scenario provided by an embodiment of the present application. Figure 1 Describe the problems in parking scenarios and related technologies.

[0038] For example, Figure 1 The parking lot shown in FIG. 1 includes multiple parking spaces. Vehicle A drives to the parking lot shown in FIG. Figure 1 After locating the vehicle A, the vehicle user determines that there is an available parking space B to the right of vehicle A. At this point, the user triggers vehicle A's automatic parking function, and the automatic parking system begins operating. Through its integrated perception, decision-making, and control subsystems, it plans a parking path for vehicle A from its current position into parking space B. Based on the parking path, the system then automatically controls vehicle A into parking space B. During this process, the user does not need to directly control the steering wheel, accelerator, or brake pedals, achieving fully automated parking for vehicle A.

[0039] However, in the above process, the automatic parking system in the related art uses a single path algorithm to plan the parking path, that is, it can only find one feasible parking path under specific assumptions, for example, Figure 1 Parking path a is shown. This is an exemplary parking path that takes into account the impact of obstacles. However, this parking path does not balance other path planning objectives, such as the number of path turns. It often focuses on only one aspect or makes simple compromises. Therefore, the adaptability of the planned path needs to be improved.

[0040] In order to solve the above problems, this application proposes a method for planning a parking path, which can be found in the following Figure 2 The method steps in .

[0041] Figure 2This is a schematic flowchart of a method for planning a parking path provided in an embodiment of the present application.

[0042] It should be understood that the method for planning a parking path provided in the embodiment of the present application can be applied to Figure 1 The vehicle shown (eg, vehicle A) may also be applied to the automatic parking system in vehicle A.

[0043] For example, Figure 2 As shown, the method 200 includes the following steps 201 to 203.

[0044] Step 201 : In response to a parking instruction from a vehicle, a plurality of parking paths are determined based on an initial posture of the vehicle at a current position, a target posture of a target parking space, and a preset parking operation of the vehicle during parking.

[0045] It should be understood that the initial posture in the above step 201 can be represented by the position coordinates and the direction of the vehicle's head at the current position, that is, it can be represented by (x0, y0, theta0), wherein the position coordinates (x0, y0) can be the coordinates of a first feature point on the vehicle. Optionally, the first feature point is the center of the vehicle (the geometric center of the vehicle body), and the position coordinates (x0, y0) are the coordinates in a two-dimensional coordinate system with the target point as the origin. Optionally, the target point is the center point of the parking lot entrance. In addition, the orientation is specifically represented by degrees, and a direction of 0° can indicate the east direction, and a direction of 90° can indicate the north direction.

[0046] It should also be understood that the target posture in the above step 201 can be represented by the position coordinates and orientation of the target parking space, that is, by (x g ,y g ,theta g ) characterization, where the position coordinate (x g ,y g ) can be the coordinates of a second feature point on the target parking space. Optionally, the second feature point is the center of the target parking space, and the position coordinates (x g ,y g ) is the coordinate in the two-dimensional coordinate system with the target point as the origin. The orientation of the target parking space refers to the direction the vehicle head should face when the vehicle is parked in the target parking space. The orientation of the target parking space is parallel to the long side of the target parking space. In this application, there is only one orientation of the parking space. In addition, x0, y0, x g and y g The unit of is m. And the target parking space is an empty parking space.

[0047] It should be noted that the preset parking operation of the vehicle during the parking process in the above step 201 refers to a pre-set basic movement instruction, which is used to guide the vehicle to complete the parking process from the starting position (current position) to the target parking space. The preset parking operation is used to control the driving direction of the vehicle and the driving step length in the driving direction. Optionally, the preset parking operation includes forward straight ahead (steering angle is 0°, forward 1m), forward left turn (steering angle is 30°, forward 1m), forward right turn (steering angle is -30°, forward 1m), backward straight ahead (steering angle is 0°, backward 1m), backward left turn (steering angle is 30°, backward 1m) and backward right turn (steering angle is -30°, backward 1m). Among them, forward left turn (steering angle of 30°, forward 1m) means that when the current gear is in forward gear and the steering wheel is turned 30° to the left, the operation of driving 1m, and rearward right turn (steering angle of -30°, backward 1m) means that when the current gear is in reverse gear and the steering wheel is turned 30° to the right, the operation of driving 1m.

[0048] It should also be noted that each of the multiple parking paths in step 201 is a path connecting a starting path node, at least one first predicted node, and a target path node. The starting path node corresponds to the starting posture, and the starting path node can be represented as S0 = Node(x0, y0, theta0). The target path node corresponds to the target posture, and the target path node can be represented as S g =Node(x g ,y g ,theta g ), each first prediction node in the at least one first prediction node corresponds to a first prediction posture, and the at least one first prediction posture is determined based on the initial posture and the preset parking operation, such as the first prediction posture is represented by (x 11 ,y 11 ,theta 11 ), the corresponding first prediction node is represented as S 11 =Node(x 11 ,y 11 ,theta 11 ).

[0049] Optionally, the parking command is triggered when an automatic parking button in the vehicle is pressed.

[0050] Figure 3 This is a schematic diagram of determining the initial posture and target posture provided by the embodiment of the present application. Figure 3 The process of determining the initial and target poses is described by taking the target point as the center point of the parking lot entrance, the first feature point as the center of the vehicle, the second feature point as the center of the target parking space, and the direction 0° indicating the east direction as an example.

[0051] For example, Figure 3 As shown in the figure, the position indicated by b is the entrance of the parking lot, which is also the exit of the parking lot. C is the center point of the parking lot entrance. A two-dimensional coordinate system is constructed with c as the origin. At this time, a vehicle C travels to position D. The position coordinates of the center d of vehicle C are (x d ,y d ), the front of vehicle C is facing due east. At this time, the initial position of vehicle C after triggering the parking command can be determined as (x d ,y d ,0°). At this time, there is an empty parking space E in front of the right side of vehicle C. The position coordinates of the center e of parking space E are (x e ,y e ), when vehicle C is parked in parking space E, the direction the vehicle head should face is due north. At this time, the target position of parking space E can be determined as (x e ,y e ,90°).

[0052] It should be noted here that if Figure 3 As shown, when vehicle C is parked in parking space E, the front of the vehicle can face due north or due south. However, if the front of the vehicle faces due north, it is more convenient for vehicle C to leave parking space E. Therefore, the front of the vehicle should face due north.

[0053] Furthermore, if the vehicle's front is facing due north, a more convenient parking scenario for vehicle C when leaving parking space E is when there is an obstacle to the south of parking space E, such as a wall or guardrail. Convenience here means that vehicle C moves forward rather than backward when leaving parking space E. When moving forward, the target's line of sight is wider, allowing them to directly see the road conditions and pedestrians ahead, making forward movement simple and smooth.

[0054] Furthermore, if the vehicle's front is facing due north or due south, vehicle C can easily exit parking space E. In this case, a prompt is displayed on the vehicle's screen, prompting the target user to select either a north or south orientation for vehicle C when parking in parking space E. If the prompt indicates that the target user has selected a heading, the heading selected by the target user is determined as the direction the vehicle's front should face, either due north or due south. If the prompt indicates that the target user has not selected a heading, the automatic parking system selects one of these directions as the direction the vehicle's front should face. Furthermore, if the vehicle's front is facing due north or due south, the parking scenario in which vehicle C can easily exit parking space E is when there are no obstacles to the south or north of parking space E.

[0055] In one possible implementation, step 201 determines multiple parking paths based on the initial posture of the vehicle at the current position, the target posture of the target parking space, and the preset parking operation of the vehicle during parking, including: starting from the initial posture, using the preset parking operation to determine multiple first predicted postures; using any first predicted posture as a second predicted posture, and discretizing the second predicted posture based on a preset grid resolution to obtain grid coordinates corresponding to the second predicted posture; determining whether the second predicted posture should be discarded based on whether the grid coordinates corresponding to the second predicted posture are the same as the grid coordinates corresponding to the first posture, so as to correct the multiple first predicted postures, and the path node corresponding to the first posture is the parent node of the path node corresponding to the second predicted posture; and determining the multiple parking paths based on the initial posture, the target posture, and the corrected multiple first predicted postures.

[0056] It should be understood that in the above scheme, during the parking process, the vehicle's posture is continuous, and continuous postures have infinite possibilities. Computer equipment cannot handle infinite states. Therefore, this method uses a discretization strategy to divide the posture space into grid units to manage the state of the path node corresponding to the first predicted posture. Among them, the discretization strategy divides the posture space into grid units of fixed size by presetting the grid resolution. Optionally, the preset grid resolution includes a position resolution d1 m and an orientation resolution θ°, that is, if the two first predicted postures are in the same d1 m×d1 m grid and the orientation deviation is less than θ°, the two first predicted postures can be mapped to the same grid, and then the two first predicted postures are considered to correspond to the same path node. Among them, when the two first predicted postures are mapped to the same grid, the grid coordinates corresponding to the two first predicted postures are the same.

[0057] In the above technical solution, multiple first predicted poses are generated based on the discrete actions of the preset parking operation, ensuring that the parking path conforms to the vehicle's kinematic constraints and avoiding invalid paths. Secondly, grid resolution is introduced to discretize the first predicted poses, mapping the continuous pose states into grid coordinates. By comparing the grid coordinates corresponding to any first predicted pose (second predicted pose) with the grid coordinates corresponding to the first pose corresponding to its parent node, it is possible to determine whether the second predicted pose and the first pose are mapped to the same grid. If they are the same, the second predicted pose is discarded, which effectively eliminates redundant states, compresses the pose space, and improves the search efficiency of path nodes. Furthermore, based on the corrected multiple first predicted poses, the initial pose, and the target pose, multiple parking paths are constructed, which can retain physically feasible candidate paths with discrete and independent states. This can provide a rich and efficient path input source (multiple candidate paths) for parking path quality assessment, assisting the automatic parking system in determining a parking path that is relatively excellent in all planning objectives.

[0058] In some embodiments, starting from the initial posture, using the preset parking operation, multiple first predicted postures are determined, including: based on the initial posture, the driving step length of the preset parking operation and the corresponding steering angle, determining at least one first candidate posture, the at least one first candidate posture being a predicted posture corresponding to a child node of the path node corresponding to the initial posture; using each first candidate posture in the at least one first candidate posture as a target candidate posture, and based on the target candidate posture, the driving step length of the preset parking operation and the corresponding steering angle, determining at least one second candidate posture, the at least one second candidate posture being a child node corresponding to the path node corresponding to the target candidate posture. each second candidate posture of the at least one second candidate posture is used as a reference candidate posture, and based on the reference candidate posture, the driving step length of the preset parking operation and the corresponding steering angle, at least one third candidate posture is determined, and the at least one third candidate posture is the predicted posture corresponding to the child node of the path node corresponding to the reference candidate posture; and so on, until the predicted posture corresponding to the child node of the path node corresponding to the determined at least one fourth candidate posture is the target posture, the multiple first predicted postures include the at least one first candidate posture, the at least one second candidate posture, the at least one third candidate posture and the at least one fourth candidate posture.

[0059] It should be understood that the above scheme only exemplifies that the target posture is reached after four rounds of prediction of the first predicted posture. Generally, it takes more than four rounds to reach the target posture, depending on the actual situation.

[0060] It should also be understood that the predicted posture corresponding to the child node of the path node corresponding to at least one of the fourth candidate postures mentioned above is the target posture, which means that the fourth candidate posture meets the termination conditions, specifically, the position error between the fourth candidate posture and the target posture is less than 0.5m, and the orientation error between the fourth candidate posture and the target posture is less than 10°.

[0061] In one possible implementation, the preset grid resolution includes a position resolution and an orientation resolution. Based on the preset grid resolution, the second predicted posture is discretized to obtain the grid coordinates corresponding to the second predicted posture, including: determining the ratio of the first horizontal axis coordinate in the second predicted posture to the first length corresponding to the position resolution as the target first horizontal axis coordinate in the grid coordinate; determining the ratio of the first vertical axis coordinate in the second predicted posture to the first length as the target first vertical axis coordinate in the grid coordinate; and determining the ratio of the first orientation in the second predicted posture to the first orientation corresponding to the orientation resolution as the target first orientation in the grid coordinate.

[0062] It should be understood that the position resolution in the above solution refers to the spatial size of the grid cell, specifically the size in the x-direction and the size in the y-direction, and the orientation resolution refers to the accuracy of the grid cell division in the orientation dimension. Optionally, the position resolution is 0.5m and the orientation resolution is 15°.

[0063] In the above technical solution, the grid coordinates corresponding to the second predicted posture are determined by mapping the rigid ratio between the predicted posture and the resolution, which can eliminate the error during floating-point operations and ensure the accuracy of the grid coordinates. The dimensional decoupling process mentioned above can make the horizontal axis coordinates and the horizontal axis coordinates independently discretized, avoiding the grid distortion caused by joint quantization. In addition, the specific parameters in the preset grid resolution are strongly related to the physical characteristics of the vehicle. This can make the spatial size of the grid unit accurately match the parking requirements, distinguish the differences between key postures, and greatly compress the search space of the path nodes to generate efficient multiple parking paths.

[0064] In some embodiments, after determining multiple first predicted postures, the method 200 also includes: obtaining obstacle coordinates around the vehicle and the target parking space; and determining whether the second predicted posture should be discarded based on whether the position coordinates in the second predicted posture fall within the obstacle area formed by the obstacle coordinates, so as to correct the multiple first predicted postures.

[0065] Optionally, the obstacle coordinates include first coordinates of static obstacles and second coordinates of dynamic obstacles, the first coordinates are obtained through a high-precision map around the target parking space, and the second coordinates are obtained through a target sensor in the vehicle.

[0066] Optionally, the target sensor includes a lidar and a vehicle-mounted camera.

[0067] Optionally, static obstacles include fixed vehicles and guardrails, and dynamic obstacles include pedestrians and animals.

[0068] In some embodiments, whether the second predicted pose should be discarded is determined based on whether the position coordinates in the second predicted pose fall within the obstacle area formed by the obstacle coordinates, including: when the position coordinates in the second predicted pose fall within the obstacle area formed by the obstacle coordinates, determining that the second predicted pose should be discarded; when the position coordinates in the second predicted pose do not fall within the obstacle area formed by the obstacle coordinates, determining that the second predicted pose should not be discarded.

[0069] In some embodiments, whether the second predicted posture should be discarded is determined based on whether the grid coordinates corresponding to the second predicted posture are the same as the grid coordinates corresponding to the first posture, including: when the grid coordinates corresponding to the second predicted posture are the same as the grid coordinates corresponding to the first posture, determining that the second predicted posture should be discarded; when the grid coordinates corresponding to the second predicted posture are different from the grid coordinates corresponding to the first posture, determining that the second predicted posture should not be discarded.

[0070] It should be understood that after determining that the second predicted pose should be discarded, the second predicted pose is not included in the plurality of first predicted poses.

[0071] In one possible implementation, the multiple parking paths are determined based on the initial posture, the target posture, and the revised multiple first predicted postures, including: determining a second posture corresponding to the target posture from the revised multiple first predicted postures, and connecting the path node corresponding to the target posture with the path node corresponding to the second posture, wherein the parent node of the path node corresponding to the target posture is the path node corresponding to the second posture; deleting the second posture from the revised multiple first predicted postures to obtain at least one third predicted posture, and determining a third posture corresponding to each third predicted posture, and connecting the path node corresponding to each third predicted posture with the path node corresponding to the third posture. Then, the parent node of the path node corresponding to the third predicted posture is the path node corresponding to the third posture, so as to determine the fourth predicted postures corresponding to the initial posture. The path nodes corresponding to the fourth predicted postures should be connected with the path node corresponding to the initial posture, and the parent node of the path nodes corresponding to the fourth predicted postures is the path node corresponding to the initial posture; based on the connection relationship between the multiple path nodes formed by the initial posture, the target posture and the corrected multiple first predicted postures, starting from the path node corresponding to the target posture, passing through the path nodes corresponding to the corrected multiple first predicted postures, and tracing back to the path node corresponding to the initial posture in sequence, the multiple parking paths are obtained.

[0072] It should be understood that the above scheme describes starting from the target posture, exploring the parent node of the path node corresponding to the target posture to determine the corrected first predicted posture (i.e., the second posture) corresponding to the parent node; deleting the second posture from the corrected multiple first predicted postures to obtain at least one third predicted posture, and for each third predicted posture in at least one third predicted posture, exploring the parent node of the path node corresponding to the third predicted posture to determine the corrected first predicted posture (i.e., the third posture) corresponding to the parent node, and so on, after multiple rounds of exploration, until the posture corresponding to the parent node is determined to be the initial posture. In addition, the path nodes corresponding to the postures with a parent-child relationship are connected to determine multiple parking paths by backtracking.

[0073] In the above technical solution, the second posture corresponding to the target posture is identified and a parent-child relationship is established, that is, the path node corresponding to the target posture is connected to the path node corresponding to the second posture. This can ensure that each parking path ends accurately at the target parking space and avoid end point deviation. Furthermore, at least one third predicted posture remaining after the second posture is deleted and processed, and the parent node is recursively connected to the path node corresponding to the initial posture. This can cover multiple first predicted postures that have been corrected after discretization processing and quickly generate multiple parking paths. In addition, the complete connection relationship from the target posture back to the initial posture can directly output multiple physically feasible parking paths, providing a low-redundancy and high-diversity path input source for the quality assessment of the parking path, significantly improving the generation efficiency and integrity of multiple parking paths.

[0074] Step 202 determines the quality score of each parking path based on a first planning objective. The first planning objective includes the angular deviation between the vehicle's head orientation during the last parking maneuver corresponding to the parking path and the orientation of the target parking space, the number of short paths in the parking path, and the percentage of path nodes in the parking path that meet a preset condition. A short path is a path whose length corresponding to the parking maneuver is less than a preset length. The preset condition indicates that the curvature of the path node corresponding to the posture is less than a preset curvature.

[0075] It should be understood that the vehicle's heading for the final parking maneuver corresponding to the parking path in step 202 refers to the vehicle's heading at the start of the final parking maneuver (typically the final reverse parking maneuver) within the parking path. Step 202 measures the difficulty and accuracy of parking the vehicle into the target parking space using angular deviation. A larger angular deviation indicates that the vehicle will require a greater or more urgent steering adjustment during the final parking maneuver to properly position the vehicle into the target space. A high-quality parking path should minimize this angular deviation, ensuring that the vehicle is in a near-ideal parking position at the start of the final parking maneuver. Angular deviation serves as a measure of the parking path's "readiness."

[0076] For example, the target parking space is a standard vertical space with a 90° orientation. During the planned parking path, the vehicle's frontal orientation during the final parking maneuver (reversing) is not completely aligned with the parking space, but deviates 10° to the right. The vehicle's body forms a 10° angle with the parking space line, and the vehicle's frontal orientation is now 80°. Therefore, the angular deviation is 10° (90° - 80°).

[0077] It should also be understood that the number of short paths in the parking path in step 202 above refers to the number of segments within a single, complete parking path where the length of a continuous forward or reverse operation segment (i.e., a path segment corresponding to a single operation) is less than a preset length. The preset length can be determined based on the vehicle body length or the minimum safe movement distance of the vehicle. Optionally, the preset length is 0.35 vehicle body lengths or 1 meter. The number of short paths is used to measure the redundancy of parking path operations. A greater number of short paths indicates inefficient parking path planning or unnecessary fine-tuning. Automatic parking systems require frequent gear shifts while only moving a short distance. At the same time, a high-quality parking path should minimize the number of short paths (even to zero). Frequent short forward and reverse operations can cause discomfort (a sense of frustration) to the target user and appear to be unskilled or unsmooth parking operations.

[0078] Among them, the above-mentioned one operation refers to the continuous movement process of the vehicle in the same gear (such as forward gear or reverse gear) without shifting gears, and its core lies in the continuity of the gear.

[0079] It should also be understood that the percentage of path nodes in the parking path that meet the preset conditions in step 202 refers to the percentage of path nodes with curvature less than the preset curvature among the N path nodes in a single, complete parking path. The curvature of a path node is the reciprocal of the radius of curvature R at the path node, which is the radius of the best-fit circle tangent to the parking path at that path node. The smaller the radius of curvature, the greater the curvature at that path node, the more severe the curvature of the parking path at that path node, and the sharper the turn, i.e., the sharp turn. The curvature at a path node is used to measure the ride smoothness of the parking path. A high-quality parking path should minimize this percentage to avoid large-angle steering during sharp turns. Furthermore, the larger the percentage corresponding to the parking path, the better the comfort for the target vehicle. Therefore, this percentage can be used to measure the comfort of the parking path.

[0080] In some embodiments, a method for determining the number of short paths in a parking path includes: dividing the parking path into continuous forward path segments and reverse path segments; using each path segment in the forward path segment and the reverse path segment as a target path segment, and accumulating the Euclidean distances between all adjacent path nodes included in the target path segment to obtain the path length of the target path segment; if the path length is less than a preset length, determining that the target path segment is a short path; and counting all short paths in the parking path to obtain the number of short paths.

[0081] In some embodiments, a method for determining the proportion of path nodes in a parking path that meet a preset condition includes: determining a first total number of path nodes in the parking path; determining a first number of path nodes in the parking path having a curvature less than a preset curvature; and determining a ratio between the first number and the first total number as the proportion of path nodes in the parking path that meet the preset condition.

[0082] Optionally, the first planning objective also includes the maximum curvature of path nodes in the parking path and the monotonicity of the orientation during single-handle operation.

[0083] It should be understood that, generally, for a single, complete parking path, the maximum curvature of the path nodes should be as small as possible. The monotonicity of the vehicle's heading during a single-handle operation in the above scheme refers to whether the vehicle's heading changes consistently and non-repeatedly over the path segment corresponding to a continuous forward or reverse operation (i.e., a single operation). Monotonic increase refers to a continuous increase in the vehicle's heading over the path segment (e.g., a continuous left turn); monotonic decrease refers to a continuous decrease in the vehicle's heading over the path segment (e.g., a continuous right turn); and non-monotonic refers to a situation where the vehicle's heading first increases and then decreases over the path segment (e.g., first turning left, then right). Monotonic heading changes indicate smoother steering wheel rotation, eliminating the need for significant steering adjustments during forward / reverse movements and reducing the number of steering wheel operations. Non-monotonic heading changes are more like hesitation or repeated corrections, which can cause the target to feel abrupt, disjointed, and uncomfortable. When evaluating parking paths, parking paths with a single heading change trend during single-handle operation should be selected whenever possible.

[0084] It should be noted that, for any path segment corresponding to an operation, the orientations of all path nodes in the path segment are extracted in chronological order, and an orientation sequence is generated; the orientation change between adjacent path nodes is determined to obtain multiple orientation changes; when multiple orientation changes are all positive or negative, it is determined that the orientation change trend of all path nodes in the path segment is single; when there are positive and negative orientations in multiple orientation changes, it is determined that the orientation change trend of all path nodes in the path segment is not single.

[0085] In one possible implementation, determining the quality score of each parking path for the first planning objective in step 202 includes: outputting a weight configuration reminder, the weight configuration reminder being used to remind the target object in the vehicle to configure weights corresponding to each first planning objective; upon receiving the weights corresponding to each first planning objective, performing a weighted fusion of the first score, the second score, and the third score based on the first configuration weight, the second configuration weight, and the third configuration weight to obtain the quality score, wherein the first score is obtained based on the angle deviation, the second score is obtained based on the number of short paths, and the third score is obtained based on the proportion of path nodes that meet preset conditions; upon not receiving the weights corresponding to each first planning objective, outputting a target configuration reminder, the target configuration reminder being used to remind the target object to use driving style as a planning objective; and upon receiving a confirmation operation for the target configuration reminder, determining, based on the driving style, a fourth configuration weight, a fifth configuration weight, and a sixth configuration weight for weighted fusion of the first score, the second score, and the third score to determine the quality score.

[0086] It should be understood that the sum of the first configuration weight, the second configuration weight, and the third configuration weight in the above solution is a preset value, which is 1. Optionally, the first configuration weight is 0.35, the second configuration weight is 0.35, and the third configuration weight is 0.3.

[0087] It should also be understood that the driving style in the above solution refers to the target object's expectations of the vehicle's operation and behavior during parking. Optionally, the driving style includes an aggressive style, a steady style, and a balanced style.

[0088] Among them, when the driving style is aggressive, the target subject often allows the vehicle to make sharp turns, the above-mentioned angle deviation is large, and the path length of the path segment corresponding to the one-shot operation is not very acceptable. Therefore, the fourth configuration weight corresponding to the first score can be configured to be smaller, indicating that the target subject does not attach much importance to the degree of alignment between the vehicle head orientation and the orientation of the target parking space in the last parking segment corresponding to a single parking path (each parking path); the fifth configuration weight corresponding to the second score can be configured to be larger, that is, the target subject prefers long path operations and attaches more importance to the proportion of non-short paths in the total number of single parking paths; the sixth configuration weight corresponding to the third score can be configured to be smaller, indicating that the target subject does not care much about the proportion of path nodes in a single parking path that meet the preset conditions and can tolerate larger curvatures and more aggressive steering angles at path nodes.

[0089] When the driving style is conservative, the target object tends to be less tolerant of sharp turns, large angle deviations, and shorter path segments corresponding to single-use maneuvers. Therefore, the fourth configuration weight corresponding to the first score can be assigned a larger weight, the fifth configuration weight corresponding to the second score can be assigned a smaller weight, and the sixth configuration weight corresponding to the third score can be assigned a larger weight. When the driving style is balanced, the target object tends to maintain a neutral attitude towards various planning objectives. Therefore, the fourth configuration weight corresponding to the first score, the fifth configuration weight corresponding to the second score, and the sixth configuration weight corresponding to the third score can be assigned the same weight. Furthermore, after determining the fourth, fifth, and sixth configuration weights, the first, second, and third scores can be weighted and fused based on the fourth, fifth, and sixth configuration weights to obtain a single parking path quality score.

[0090] In addition, in the above solution, appropriately configuring the fifth configuration weight to a smaller / larger value, and appropriately configuring the sixth configuration weight to a smaller / larger value, are both relative to each configuration weight.

[0091] In the above technical solution, the weight configuration reminder allows the target driver to adjust the angular deviation between the vehicle's heading during the last parking maneuver and the target parking space, the number of short paths, and the proportion of path nodes that meet preset conditions as needed when assessing path quality. This allows the quality score (scoring result) to be personalized to match the driver's parking preferences. When not configured, it intelligently guides driving style selection (for example, an aggressive style can automatically reduce the penalty weight for short paths), converting abstract driving styles into specific weight parameters, resolving the pain point of insufficient universality caused by default weights. In addition, during manual configuration, multiple scores are weighted and fused based on the configured weights; when unmanned configuration is performed, dynamic weights are derived based on driving style. This two-stage scoring mechanism can cover a variety of weight determination scenarios and significantly improve the accuracy of parking path screening. Therefore, through dynamic weight configuration and user interaction mechanism, the above solution can significantly optimize the adaptability and accuracy of parking path quality assessment.

[0092] In addition, while outputting the weight configuration reminder, each first planning target and its corresponding specific meaning can also be output. For example, the first reference planning target (the angular deviation between the vehicle's front orientation in the last parking operation corresponding to the parking path and the orientation of the target parking space) is used to measure the difficulty and accuracy of parking the vehicle in the target parking space; the second reference planning target (the number of short paths in the parking path) is used to measure the redundancy of the parking path operation; and the third reference planning target (the proportion of path nodes in the parking path that meet preset conditions) is used to measure the driving smoothness and comfort of the parking path to assist in configuring the weights of the target objects.

[0093] In some embodiments, after outputting the target configuration reminder, the method 200 further includes: in the absence of receiving a confirmation operation for the target configuration reminder, determining the sum of the first score, the second score, and the third score as the total score, and determining the product of the total score and the target preset weight as the quality score.

[0094] It should be understood that the "target preset weight" in the above scheme is the ratio between the preset value and the preset number. The preset number is 3, and the target preset weight is 1 / 3.

[0095] In some embodiments, based on the first configuration weight, the second configuration weight and the third configuration weight, the first score, the second score and the third score are weightedly fused to obtain the quality score, including: weighting the first score based on the first configuration weight to obtain a first quality score; weighting the second score based on the second configuration weight to obtain a second quality score; weighting the third score based on the third configuration weight to obtain a third quality score; and determining the sum of the first quality score, the second quality score and the third quality score as the quality score.

[0096] In some embodiments, the method for determining the first score includes: determining the angular difference between the maximum angular deviation and the angular deviation, where the maximum angular deviation is the maximum angular deviation between the orientation of the vehicle head in the last parking operation corresponding to the parking path and the orientation of the target parking space; and determining the first score as the ratio of the angular difference to the maximum angular deviation.

[0097] It should be understood that the maximum angular deviation mentioned above is 180°, i.e., the maximum theoretical angular deviation. The angular deviation mentioned above can be regarded as an actual angular deviation. The first score is used to reflect the degree of alignment (the degree of alignment) between the vehicle's head direction and the target parking space's direction in the last parking maneuver corresponding to a single parking path. The maximum value of the first score is 1 point, in which case the angular deviation is 0°; the minimum value of the first score is 0 point, in which case the angular deviation is 180°. In addition, the greater the angular deviation between the vehicle's head direction and the target parking space's direction in the last parking maneuver corresponding to a single parking path, the lower the first score of the single parking path.

[0098] In some embodiments, a method for determining the second score includes: dividing a single parking path into continuous forward path segments and reverse path segments; determining a second total number of forward path segments and reverse path segments; determining a quantity deviation between the second total number and the number of short paths in the single parking path; and determining the second score as a ratio of the quantity deviation to the second total number.

[0099] It should be understood that the aforementioned number deviation refers to the number of non-short paths (not short paths) within a single parking path. The second score reflects the proportion of non-short paths within a single parking path to the second total number. The maximum value of the second score is 1, which indicates that there are no short paths within the single parking path. The fewer non-short paths within a single parking path, the lower the second score.

[0100] In some embodiments, the method for determining the third score includes: determining the third score as the proportion of path nodes in a single parking path that meet the preset condition.

[0101] It should be understood that the greater the above proportion in a single parking path, the higher the third score of the single parking path.

[0102] In some embodiments, the method for determining the driving style includes: when the target object is a driver, analyzing the target object's driving style based on the target object's driving operation data during historical driving; or, when the target object is not a driver, analyzing the target object's behavioral changes or external attire to determine the target object's desired driving style.

[0103] In some embodiments, the driving operation data includes the total number of vehicle accelerations and the first number of sudden accelerations within a historical preset time period, as well as the peak value of the steering angular velocity. Based on the driving operation data of the target object during the historical driving process, the driving style of the target object is analyzed, including: determining the ratio between the first number and the total number as the sudden acceleration frequency; when the sudden acceleration frequency is greater than the first preset frequency, determining the driving style as an aggressive style; when the sudden acceleration frequency is less than or equal to the first preset frequency, determining whether the peak value of the steering angular velocity is within a first range less than or equal to the preset peak value, the first range being used to indicate that the target object steers the wheel slowly when turning; when the peak value of the steering angular velocity is within the first range, determining the driving style as a steady style; when the peak value of the steering angular velocity is within a second range greater than the preset peak value, determining the driving style as a balanced style, the second range being used to indicate that the target object steers the wheel at a medium speed when turning.

[0104] It should be understood that the "slowly" mentioned above means that the steering angular velocity of the steering wheel is less than the first steering angular velocity, such as 80° / s; the "medium speed" mentioned above means that the steering angular velocity of the steering wheel is greater than the first steering angular velocity, such as 80° / s, and can also be between 80° / s and 120° / s.

[0105] In some embodiments, the target object's behavioral changes or external attire are analyzed to determine the target object's desired driving style, including: determining whether the target object has nervous behavior when the vehicle is in a rapid acceleration condition; determining the driving style as an aggressive style if the target object does not have nervous behavior; determining the duration of the target object's nervous behavior when the target object has nervous behavior; determining the driving style as a conservative style if the duration is greater than or equal to a preset duration, the preset duration being the product of the total duration corresponding to the rapid acceleration condition and a preset coefficient; and determining the driving style as a conservative style if the duration is greater than or equal to a preset duration. When the duration is less than the preset duration, the driving style is determined to be a balanced style, or; a color analysis is performed on the external clothing, and when the color of the external clothing is bright, the dressing style of the external clothing is determined to be; when the dressing style is bold and avant-garde, the driving style is determined to be an aggressive style; when the color of the external clothing is bright and the dressing style is not bold and avant-garde, the driving style is determined to be a steady style; when the color of the external clothing is not bright and the dressing style is not bold and avant-garde, the driving style is determined to be a balanced style.

[0106] Optionally, the preset coefficient is 0.5.

[0107] In one possible implementation, based on the driving style, a fourth configuration weight, a fifth configuration weight, and a sixth configuration weight for weighted fusion of the first score, the second score, and the third score are determined, including: when the driving style is an aggressive style, the fourth configuration weight is determined as the first preset weight, the fifth configuration weight is determined as the second preset weight, and the sixth configuration weight is determined as the third preset weight, the first preset weight is less than the second preset weight, and the third preset weight is less than the second preset weight; when the driving style is a conservative style, the fourth configuration weight is determined as the fourth preset weight, the fifth configuration weight is determined as the fifth preset weight, and the sixth configuration weight is determined as the sixth preset weight, the fifth preset weight is less than the fourth preset weight, and the fifth preset weight is less than the sixth preset weight; when the driving style is a balanced style, the fourth configuration weight, the fifth configuration weight, and the sixth configuration weight are determined as the target preset weights.

[0108] In the above technical solution, under the aggressive style, the configuration weight corresponding to the angle deviation is reduced (the first preset weight) and the configuration weight corresponding to the above proportion is reduced (the third preset weight), and the weight corresponding to the number of short paths is increased (the second preset weight). The setting of the above configuration weights is more suitable for the target object of pursuing parking efficiency. Under the robust style, the curvature safety of the path nodes is strengthened (the sixth preset weight is larger) and the importance of angle deviation is emphasized, which can ensure the smoothness of the vehicle when turning. Under the balanced style, the above three planning goals are balanced with the same weight. This solution can significantly optimize the scenario adaptability of the quality score of the parking path through a directional mapping mechanism between driving style and configuration weights.

[0109] Optionally, the first preset weight is 0.3, the second preset weight is 0.5, and the third preset weight is 0.2.

[0110] Optionally, the fourth preset weight is 0.4, the fifth preset weight is 0.2, and the sixth preset weight is 0.4.

[0111] In one possible implementation, determining the quality score of each parking path based on the first planning objective in step 202 includes: determining a target path cost for each parking path, where the parking path is determined based on the initial position, the target position, and at least one first predicted position related to the preset parking operation, the target path cost being the sum of a first path cost for the vehicle traveling from the initial position to any first predicted position and a second path cost for traveling from any first predicted position to the target position, the first path cost being the sum of a cumulative path cost for the vehicle traveling from the initial position to a fifth predicted position and an operation cost for traveling from the fifth predicted position to any first predicted position, the path node corresponding to the fifth predicted position being the parent node of the path node corresponding to any first predicted position; determining the target path cost, total path length, and number of path steps for each parking path as a second planning objective when planning the parking path; and determining a quality score for each parking path based on the first planning objective and the second planning objective.

[0112] It should be understood that the target path cost f in the above scheme includes three costs. The first cost is the cumulative path cost g.acc of the vehicle traveling from the initial position to the fifth predicted position. The second cost is the operating cost g.oper of the vehicle traveling from the fifth predicted position to a certain first predicted position. The third cost is the second path cost h of the vehicle traveling from a certain first predicted position to the target position. The sum of the cumulative path cost and the operating cost is the first path cost g. In addition, the cumulative path cost of the vehicle traveling from the initial position to the fifth predicted position can be further obtained by summing the cumulative path cost of the vehicle traveling from the initial position to the sixth predicted position and the operating cost from the sixth predicted position to the fifth predicted position. Among them, the path node corresponding to the sixth predicted position is the parent node of the path node corresponding to the fifth predicted position.

[0113] It should also be understood that the number of paths in the above scheme refers to the total number of times a single parking path needs to switch between forward gear and reverse gear during the parking process. Similarly, when adding the second planning goal, it is necessary to determine the configuration weights corresponding to the target path cost, the total path length, and the number of paths, such as the configuration weight corresponding to the target path cost is the seventh configuration weight, the configuration weight corresponding to the total path length is the eighth configuration weight, and the configuration weight corresponding to the number of paths is the ninth configuration weight. When determining the quality score of each parking path on the first planning goal and the second planning goal, the sum of the first configuration weight, the second configuration weight, the third configuration weight, the seventh configuration weight, the eighth configuration weight, and the ninth configuration weight is a preset value, or the sum of the fourth configuration weight, the fifth configuration weight, the sixth configuration weight, the seventh configuration weight, the eighth configuration weight, and the ninth configuration weight is a preset value.

[0114] In the above technical solution, the cumulative path cost of each parking path from the vehicle's initial position to the fifth predicted position, the operating cost of the vehicle from the fifth predicted position to a certain first predicted position, and the second path cost of the vehicle from a certain first predicted position to the target position are determined. This determines the total path cost (target path cost) for each parking path. The target path cost starts from the initial position, passes through the first predicted positions corresponding to the intermediate path nodes, and finally reaches the target position, integrating all historical and future costs. This avoids the limitations of evaluating only the endpoint state (such as angular deviation), accurately quantifies the cost of each path segment, and improves the accuracy of the quality score. Furthermore, incorporating the total path length and the number of path segments into the second planning objective increases the comprehensiveness of the planning objectives and makes the quality score more accurate. By integrating hierarchical path cost calculation with multiple planning objectives, this solution can significantly improve the comprehensiveness and practicality of parking path scoring.

[0115] Optionally, the operation cost includes a forward cost of 1.0 / m, a backward cost of 1.5 / m (reversing penalty), and a turning cost of 0.2 / 30° (turning change penalty).

[0116] Figure 4 This is a schematic diagram of determining multiple parking paths provided by an embodiment of the present application. Figure 4 Taking an automatic parking scenario as an example, the process of determining multiple parking paths in this application is described.

[0117] For example, Figure 4 The scene setting shown in (a) is:

[0118] The center point of the parking lot entrance is taken as the origin of the two-dimensional coordinate system;

[0119] The initial position of the vehicle is (1, 4, 0°), which means the horizontal coordinate of the center of the vehicle is 1, the vertical coordinate of the center of the vehicle is 4, and the orientation of the vehicle (i.e. the direction of the front) theta is due east, with an orientation of 0°.

[0120] The target pose of the parking space is (4, 2, 90°), which means the horizontal coordinate of the center of the parking space is 4, the vertical coordinate of the center of the parking space is 2, and the orientation of the parking space is due north, that is, the direction the front of the vehicle should face when parked in the parking space, and the orientation of the parking space is 90°;

[0121] Vehicle parameters: The vehicle's wheelbase L is 2.5m, the vehicle length is 4.8m, and the vehicle width is 1.8m;

[0122] Operation set (including the following multiple parking operations): step length d is 1m;

[0123] Go straight forward (steering angle is 0°), turn left forward (steering angle is +30°), turn right forward (steering angle is -30°); go straight backward (steering angle is 0°), turn left backward (steering angle is +30°), turn right backward (steering angle is -30°);

[0124] Operation cost: forward cost is 1.0 / m, backward cost is 1.5 / m (reversing penalty), and turning cost is 0.2 / 30° (turning change penalty);

[0125] Termination conditions: Position error with target pose < 0.5m, orientation error < 10°;

[0126] Grid resolution: position resolution is 0.5m, orientation resolution is 15°;

[0127] The following is a process for determining multiple path nodes in the parking path based on the above multiple known conditions:

[0128] 1. Initialization process: Construct the starting path node based on the initial pose

[0129] Starting path node: S0 = Node(1,4,0°);

[0130] The path cost of the vehicle from the initial position to the current position (from the starting path node to the current path node (i.e., the starting path node)); g = 0; specifically, the sum of the cumulative cost g.acc from the starting path node to the parent node of the current path node and the operation cost from the parent node to the current path node. Where the cumulative cost is 0, the operation cost is 0;

[0131] The second path cost of the vehicle traveling from the initial position to the target position (current path node to target path node) is: h = ReedsShepp (1, 4, 0°; 4, 2, 90°) = 5.2;

[0132] Cost f: f = g + h = 5.2;

[0133] The starting path node has no parent node;

[0134] Initialize the data structure: Open list: add the starting path node and the corresponding total cost; Close list: initially empty.

[0135] The grid coordinates of the discretized starting path node are: (2,8,0);

[0136] 2. First cycle: Expand sub-path nodes based on the starting path node

[0137] 1) Take any path node in the previous round (i.e., the starting path node) from the open list, i.e., S0 = Node(1,4,0°); based on the target path node S corresponding to the starting path node and the target posture g =Node(4,2,90°) determines whether the termination condition is met: If 90°-0° is greater than 10°, the termination condition is not met and the path node expansion continues;

[0138] 2) Taking the starting path node S0 as the benchmark, expand the sub-path nodes based on the operation set:

[0139] Go straight ahead: new_x = 1 + 1 * cos (0 °) = 2; new_y = 4 + 1 * sin (0 °) = 4; new_theta = 0 °; S 11 =Node(2,4,0°); Operation cost: g.oper=1*1=1; Total cost: f 11 =g.acc+g.oper+h=0+1+4.1=5.1; where h=ReedsShepp(1,4,0°;2,4,0°), the path node S11 The discretized grid coordinates are: (4,8,0);

[0140] Turn left forward (steering angle is +30°): turning radius: R = L / tan(30°) = 4.33m; new heading change: Δθ = d / R rad = 13.2°; new_theta = 0° + Δθ = 13.2°; 13.2° / 2 = 6.6°; new_x = 1 + 1 * cos(6.6°) = 1.99; new_y = 4 + 1 * sin(6.6°) = 4.12; S 12 =Node(1.99,4.12,13.2°); Operation cost: g.oper=1*1+0.2=1.2; Total cost: f 12 =g.acc+g.oper+h=0+1.2+4.3=5.5; h=ReedsShepp(1,4,0°;1.99,4.12,13.2°), the path node S 12 The discretized grid coordinates are: (3,8,0);

[0141] Turn right forward (steering angle is -30°): new_theta = 0° - Δθ = -13.2°; -13.2° / 2 = -6.6°; new_x = 1 - 1*cos(-6.6°) = 0.01; new_y = 4 - 1*sin(-6.6°) = 3.88; S 13 =Node(0.01,3.88,-13.2°); Operation cost: g.oper=1*1.5+0.2=1.7; Total cost: f 13 =g.acc+g.oper+h=0+1.7+5.8=7.5; h=ReedsShepp(1,4,0°;0.01,3.88,-13.2°), and the path node S 13 The discretized grid coordinates are: (0,7,-1);

[0142] Go straight back: new_x = 1-1*cos(0°) = 0; new_y = 4-1*sin(0°) = 4; new_theta = 0°; S 14 =Node(0,4,0°); Operation cost: g.oper=1*1.5=1.5; Total cost: f 14 =g.acc+g.oper+h=0+1.5+5.2=6.7; h=ReedsShepp(1,4,0°;0,4,0°), the path node S 14 The discretized grid coordinates are: (0,8,0);

[0143] Turn left (steering angle is +30°): new_theta = (d*tan(30°)) / L*(-1) = -13.24°; -13.24° / 2 = -6.62°; new_x = 1+1*cos(-6.62°)*(-1) = 0.007; new_y = 4+1*sin(-6.62°)*(-1) = 4.115; S 15 =Node(0.007,4.115,-13.24°); Operation cost: g.oper=1*5+0.2=1.7; Total cost: f 15 =g.acc+g.oper+h=0+1.7+5.5=7.2; where h=ReedsShepp(1,4,0°;0.007,4.15,-13.24°), the path node S 15 The discretized grid coordinates are: (0,8,-1);

[0144] Turn right (steering angle is -30°): new_theta = (d*tan(-30°)) / L*(-1) = 13.24°; 13.24° / 2 = 6.62°; new_x = 1+1*cos(6.62°)*(-1) = 0.007; new_y = 4+1*sin(6.62°)*(-1) = 3.885; S 16 =Node(0.007,3.885,13.24°); Operation cost: g.oper=1*1.5+0.2=1.7; Total cost: f 16 =g.acc+g.oper+h=0+1.7+5.8=7.5; where h=ReedsShepp(1,4,0°;0.007,3.885,13.24°), the path node S 16 The discretized grid coordinates are: (0,7,0);

[0145] Path node S 11 The grid coordinates of (4,8,0); path node S 12 The grid coordinates of the path node S are (3,8,0); 13 The grid coordinates of the path node S are (0,7,-1); 14 The grid coordinates of the path node S are (0,8,0); 15 The grid coordinates of (0,8,-1), the path node S 16 The grid coordinates are (0,7,0); these six path nodes S 11 、S 12 、S 13 、S 14 、S 15 and S 16are all different from the grid coordinates (2,8,0) of the starting path node S0, that is, the path node S 11 , path node S 12 , path node S 13 , path node S 14 , path node S 15 , path node S 16 and their corresponding total costs f are added; close the list: add the starting path node S0.

[0146] 3. Second round of loop: continue to expand subpath nodes

[0147] 1) Take any path node from the previous round from the open list (you can first take the path node with the smallest f), that is, S 11 =Node(2,4,0°); based on path node S 11 The target path node S corresponding to the target pose g =Node(4,2,90°) determines whether the termination condition is met: If 90°-0° is greater than 10°, the termination condition is not met and the path node expansion continues;

[0148] 2) Take path node S 11 As a benchmark, the sub-path nodes are expanded based on the operation set. The specific implementation process can be found in the above steps 2 and 2). After determining the path node S 11 After the sub-path node is found, similarly, any path node in the previous round can be taken from the open list (the path node with the smallest f can be taken first), that is, S 12 =Node(1.99,4.12,13.2°); based on path node S 12 and the target path node S g =Node(4,2,90°) determines whether the termination condition is met: 90°-13.2° is greater than 10°, which means the termination condition is not met, and the path node expansion continues; and so on, until the path node S 11 , path node S 12 , path node S 13 , path node S 14 , path node S 15 , path node S 16 After the sub-path nodes of each path node in are determined, their sub-path nodes can be determined according to the above steps until the determined path node is consistent with the target path node S g The termination condition is met.

[0149] In addition, when considering the collision of obstacles, the above-mentioned multiple path nodes can discard the path nodes that overlap with the positions of obstacles.

[0150] Assume that after multiple path node expansions, the obtained path nodes include: path node S0, and the sub-path nodes of path node S0 include path node S 11 , path node S 12 and path node S 13 , path node S 11 The subpath nodes include the path node S 21 , path node S 12 The subpath nodes include the path node S 31 and path node S 32 , path node S 13 The subpath nodes include the path node S 41 , path node S 21 The subpath nodes include the path node S 51 , path node S 31 The subpath nodes include the path node S 61 , path node S 41 The subpath nodes include the path node S 71 , path node S 51 The subpath nodes include the path node S 81 , path node S 61 The subpath nodes include the path node S 91 and path node S 92 , path node S 71 The subpath nodes include the path node S 101 , path node S 81 Subpath node, path node S 91 and path node S 92 The subpath nodes and path nodes S 101 The sub-path nodes are all target path nodes S g , and connect the path nodes that have a parent-child relationship among the above multiple path nodes.

[0151] Starting from the path node corresponding to the target posture (target path node), through the expanded multiple path nodes, backtracking to the starting path node corresponding to the initial posture in sequence, we can obtain Figure 4 The multiple parking paths shown in (b) include: S0->S 11 ->S 21 ->S 51 ->S 81 ->S g ;S0->S 12 ->S 31 ->S61 ->S 91 ->S g ;S0->S 12 ->S 31 ->S 61 ->S 92 ->S g ;S0->S 13 ->S 41 ->S 71 ->S 101 ->S g , that is, 4 parking paths are obtained.

[0152] It should be understood that the above Figure 4 In the embodiment shown, the automatic parking system can be considered to search for possible path nodes (or multiple first predicted poses corresponding to the path nodes) through the Hybrid A* algorithm, specifically using the vehicle motion model (the above-mentioned preset parking operation) and the Reeds Shepp heuristic algorithm (the ReedsShepp() function mentioned above).

[0153] The Hybrid A* algorithm is an improvement on the traditional A* algorithm, a grid-based heuristic search algorithm. It evaluates the cost of each grid node and prioritizes nodes with the lowest overall cost, effectively planning the shortest driving path in discrete space. It is widely used in robot path planning scenarios using static maps.

[0154] The Hybrid A* algorithm builds upon the traditional A* algorithm by incorporating vehicle kinematic constraints (such as forward straight ahead, forward right turns, and backward left turns). This ensures that the generated driving path complies with the vehicle's dynamic characteristics. In path planning, the Hybrid A* algorithm is widely used in complex environments (such as narrow garages) in fields such as autonomous driving and robotics, generating smooth, executable, and obstacle-avoiding driving paths.

[0155] Step 203 : Controlling parking of the vehicle based on the parking path with the highest quality score among the multiple parking paths.

[0156] It should be understood that in step 203, the parking path with the highest quality score needs to be converted into a vehicle control action, and the vehicle is then parked into the target parking space based on the vehicle control action. Vehicle control actions include steering actions such as the steering angle of the steering wheel and the state of the driving speed.

[0157] In some embodiments, a method for determining the parking path with the highest quality score among the multiple parking paths includes: sorting the parking paths corresponding to the quality scores among the multiple parking paths in descending order of quality scores to obtain a plurality of sorted parking paths; and determining the first parking path among the sorted plurality of parking paths as the parking path with the highest quality score.

[0158] In some embodiments, step 203 includes: smoothing the parking path with the highest quality score to obtain a first target path; converting the first target path into multiple candidate vehicle control actions; adjusting the multiple candidate vehicle control actions based on preset dynamic control constraints to obtain multiple target vehicle control actions; and controlling the parking of the vehicle based on the multiple target vehicle control actions.

[0159] It should be understood that the aforementioned smoothing of the parking path with the highest quality score refers to processing the parking path itself to optimize its geometry and output a smoother first target path. Adjusting multiple candidate vehicle control actions based on preset dynamic control constraints to obtain multiple target vehicle control actions refers to the process of resolving path tracking execution issues to ensure that the vehicle smoothly tracks the first target path.

[0160] Furthermore, the above scheme can smooth the parking path with the highest quality score using a preset gradient descent method. Multiple candidate vehicle control actions can be adjusted based on preset dynamic control constraints using a model predictive control algorithm.

[0161] Optionally, the dynamic control constraint includes limiting the steering angular velocity of the steering wheel to ≤70° / s.

[0162] In some embodiments, the first target path is converted into multiple candidate vehicle control actions, including: based on adjacent path nodes, decomposing the first target path into multiple continuous path segments, each path segment is defined by two adjacent path nodes; for each path segment, extracting the vehicle state corresponding to the adjacent first node and the vehicle state corresponding to the second node, the vehicle state including candidate position coordinates and candidate orientation; based on the first candidate position coordinates and the second candidate position coordinates, and a preset time step, determining the candidate speed for traveling from the first candidate position coordinates to the second candidate position coordinates; based on the first candidate orientation and the second candidate orientation, determining the steering wheel angle when traveling from the first candidate position coordinates to the second candidate position coordinates; determining the candidate speed and steering wheel angle corresponding to each path segment as multiple candidate vehicle control actions.

[0163] Figure 5 Schematic diagram of the structure of a device for planning a parking path provided in an embodiment of the present application.

[0164] For example, Figure 5 As shown, the apparatus 500 includes:

[0165] The determination module 501 is configured to:

[0166] In response to a parking instruction from the vehicle, determining a plurality of parking paths based on an initial posture of the vehicle at a current position, a target posture of a target parking space, and a preset parking operation of the vehicle during parking;

[0167] Determining a quality score for each parking path based on a first planning objective, the first planning objective comprising the angular deviation between the vehicle's heading during the last parking maneuver corresponding to the parking path and the orientation of the target parking space, the number of short paths in the parking path, and the percentage of path nodes in the parking path that meet a preset condition. A short path is a path whose length corresponding to the parking maneuver is less than a preset length. The preset condition indicates that the curvature of the path node corresponding to the posture is less than a preset curvature.

[0168] The control module 502 is configured to control parking of the vehicle based on the parking path with the highest quality score among the multiple parking paths.

[0169] Optionally, the determination module 501 is specifically used to: starting from the initial posture, use the preset parking operation to determine multiple first predicted postures; take any first predicted posture as the second predicted posture, and discretize the second predicted posture based on a preset grid resolution to obtain grid coordinates corresponding to the second predicted posture; determine whether the second predicted posture should be discarded based on whether the grid coordinates corresponding to the second predicted posture are the same as the grid coordinates corresponding to the first posture, so as to correct the multiple first predicted postures, and the path node corresponding to the first posture is the parent node of the path node corresponding to the second predicted posture; determine the multiple parking paths based on the initial posture, the target posture and the corrected multiple first predicted postures.

[0170] Optionally, the preset grid resolution includes position resolution and orientation resolution, and the device 500 also includes: a processing module, used to: determine the ratio between the first horizontal axis coordinate in the second predicted posture and the first length corresponding to the position resolution as the target first horizontal axis coordinate in the grid coordinate; determine the ratio between the first vertical axis coordinate in the second predicted posture and the first length as the target first vertical axis coordinate in the grid coordinate; determine the ratio between the first orientation in the second predicted posture and the first orientation corresponding to the orientation resolution as the target first orientation in the grid coordinate.

[0171] Optionally, the determination module 501 is further specifically used to: determine the second posture corresponding to the target posture from the corrected multiple first predicted postures, and connect the path node corresponding to the target posture with the path node corresponding to the second posture, and the parent node of the path node corresponding to the target posture is the path node corresponding to the second posture; delete the second posture from the corrected multiple first predicted postures to obtain at least one third predicted posture, and determine the third posture corresponding to each third predicted posture, and connect the path node corresponding to each third predicted posture with the path node corresponding to the third posture, and the path node corresponding to the third predicted posture The parent node is the path node corresponding to the third posture, so as to determine the fourth predicted postures corresponding to the initial posture. The path nodes corresponding to the fourth predicted postures should be connected with the path node corresponding to the initial posture, and the parent node of the path nodes corresponding to the fourth predicted postures is the path node corresponding to the initial posture; based on the connection relationship between the multiple path nodes formed by the initial posture, the target posture and the corrected multiple first predicted postures, starting from the path node corresponding to the target posture, passing through the path nodes corresponding to the corrected multiple first predicted postures, and tracing back to the path node corresponding to the initial posture in sequence, a plurality of parking paths are obtained.

[0172] Optionally, the device 500 also includes: an output module for outputting a weight configuration reminder, which is used to remind the target object in the vehicle to configure the weights corresponding to each first planning target; a fusion module for, upon receiving the weights corresponding to each first planning target, performing weighted fusion on the first score, the second score, and the third score based on the first configuration weight, the second configuration weight, and the third configuration weight to obtain the quality score, wherein the first score is obtained based on the angle deviation, the second score is obtained based on the number of short paths, and the third score is obtained based on the proportion of path nodes that meet preset conditions; the output module is also used to output a target configuration reminder when the weights corresponding to each first planning target are not received, and the target configuration reminder is used to remind the target object to use the driving style as a planning target; the determination module 501 is specifically further used to determine, based on the driving style, a fourth configuration weight, a fifth configuration weight, and a sixth configuration weight for weighted fusion of the first score, the second score, and the third score to determine the quality score when a confirmation operation for the target configuration reminder is received.

[0173] Optionally, the determination module 501 is further specifically used to: when the driving style is an aggressive style, determine the fourth configuration weight as the first preset weight, determine the fifth configuration weight as the second preset weight, and determine the sixth configuration weight as the third preset weight, the first preset weight is less than the second preset weight, and the third preset weight is less than the second preset weight; when the driving style is a conservative style, determine the fourth configuration weight as the fourth preset weight, determine the fifth configuration weight as the fifth preset weight, and determine the sixth configuration weight as the sixth preset weight, the fifth preset weight is less than the fourth preset weight, and the fifth preset weight is less than the sixth preset weight; when the driving style is a balanced style, determine the fourth configuration weight, the fifth configuration weight and the sixth configuration weight as the target preset weights.

[0174] Optionally, the determination module 501 is further configured to: determine a target path cost for each parking path, where the parking path is determined based on the initial posture, the target posture, and at least one first predicted posture related to the preset parking operation, the target path cost being the sum of a first path cost for the vehicle traveling from the initial posture to any first predicted posture and a second path cost for traveling from any first predicted posture to the target posture, the first path cost being the sum of a cumulative path cost for the vehicle traveling from the initial posture to a fifth predicted posture and an operation cost for traveling from the fifth predicted posture to any first predicted posture, the path node corresponding to the fifth predicted posture being the parent node of the path node corresponding to any first predicted posture; determine the target path cost, total path length, and number of path nodes of each parking path as a second planning target when planning the parking path; and determine a quality score for each parking path based on the first planning target and the second planning target.

[0175] Figure 6 It is a structural schematic diagram of a vehicle provided in an embodiment of the present application.

[0176] For example, Figure 6 As shown, the vehicle 600 includes: a memory 601 and a processor 602, wherein the memory 601 stores an executable program code 603, and the processor 602 is used to call and execute the executable program code 603 to perform a method for planning a parking path.

[0177] In addition, an embodiment of the present application also protects a device, which may include a memory and a processor, wherein the memory stores executable program code, and the processor is used to call and execute the executable program code to perform a method for planning a parking path provided in an embodiment of the present application.

[0178] In this embodiment, the device can be divided into functional modules based on the above-described method examples. For example, each functional module can be mapped to a specific functional module, or two or more functions can be integrated into a single processing module. The integrated module can be implemented in hardware. It should be noted that the module division in this embodiment is illustrative and represents only a logical functional division. In actual implementation, other division methods may be used.

[0179] In the case of dividing the functional modules into corresponding functional modules, the device may further include a determination module, a control module, a processing module, an output module, and a fusion module. It should be noted that all relevant contents involved in the above method embodiments can be referred to the functional description of the corresponding functional modules and will not be repeated here.

[0180] It should be understood that the device provided in this embodiment is used to execute the above-mentioned method for planning a parking path, and thus can achieve the same effect as the above-mentioned implementation method.

[0181] In the case of an integrated unit, the device may include a processing module and a storage module. When the device is used in a vehicle, the processing module may be used to control and manage the vehicle's movements. The storage module may be used to support the vehicle's execution of relevant executable program code, etc.

[0182] The processing module may be a processor or controller that implements or executes the various exemplary logic blocks, modules, and circuits described in conjunction with the present disclosure. The processor may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a digital signal processing (DSP) and a microprocessor, and the storage module may be a memory.

[0183] In addition, the device provided in the embodiments of the present application can specifically be a chip, component or module, and the chip may include a connected processor and memory; wherein the memory is used to store instructions, and when the processor calls and executes the instructions, the chip can execute a method for planning a parking path provided in the above embodiment.

[0184] This embodiment also provides a computer-readable storage medium, which stores executable program code. When the executable program code is run on a computer, the computer executes the above-mentioned related method steps to implement a method for planning a parking path provided in the above embodiment.

[0185] This embodiment further provides a computer program product. When the computer program product is run on a computer, the computer is caused to execute the above-mentioned related steps to implement the method for planning a parking path provided in the above embodiment.

[0186] Among them, the device, computer-readable storage medium, computer program product or chip provided in this embodiment are all used to execute the corresponding methods provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding methods provided above, and will not be repeated here.

[0187] Through the description of the above implementation methods, technical personnel in the relevant field can understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0188] In the embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of modules or units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0189] The above content is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A method for planning a parking path, characterized in that: The method comprises: In response to a parking instruction from a vehicle, determining a plurality of parking paths based on an initial posture of the vehicle at a current position, a target posture of a target parking space, and a preset parking operation of the vehicle during parking; Determining a quality score for each parking path based on a first planning objective, the first planning objective including the angular deviation between the vehicle's heading during the last parking maneuver corresponding to the parking path and the orientation of the target parking space, the number of short paths in the parking path, and the percentage of path nodes in the parking path that meet a preset condition. A short path is a path whose length corresponding to the parking maneuver is less than a preset length. The preset condition indicates that the curvature of the path node corresponding to the posture is less than a preset curvature. Parking of the vehicle is controlled based on a parking path having a highest quality score among the plurality of parking paths.

2. The method according to claim 1, characterized in that The determining of multiple parking paths based on the initial posture of the vehicle at the current position, the target posture of the target parking space, and the preset parking operation of the vehicle during parking includes: Starting from the initial posture, using the preset parking maneuver, determining a plurality of first predicted postures; Taking any first predicted pose as a second predicted pose, and discretizing the second predicted pose based on a preset grid resolution to obtain grid coordinates corresponding to the second predicted pose; determining whether the second predicted pose should be discarded to correct the plurality of first predicted poses based on whether the grid coordinates corresponding to the second predicted pose are the same as the grid coordinates corresponding to the first pose, wherein the path node corresponding to the first predicted pose is a parent node of the path node corresponding to the second predicted pose; The plurality of parking paths are determined based on the initial posture, the target posture, and the corrected plurality of first predicted postures.

3. The method according to claim 2, characterized in that The preset grid resolution includes a position resolution and an orientation resolution. The discretization processing of the second predicted posture based on the preset grid resolution to obtain the grid coordinates corresponding to the second predicted posture includes: determining a ratio between the first transverse axis coordinate in the second predicted pose and the first length corresponding to the position resolution as the target first transverse axis coordinate in the grid coordinate; determining a ratio between the first longitudinal axis coordinate in the second predicted pose and the first length as the target first longitudinal axis coordinate in the grid coordinate; A ratio between the first orientation in the second predicted pose and the first orientation corresponding to the orientation resolution is determined as the target first orientation in the grid coordinates.

4. The method according to claim 2, characterized in that The determining of the plurality of parking paths based on the initial posture, the target posture, and the corrected plurality of first predicted postures includes: Determining a second posture corresponding to the target posture from the corrected multiple first predicted postures, and connecting the path node corresponding to the target posture with the path node corresponding to the second posture, where the parent node of the path node corresponding to the target posture is the path node corresponding to the second posture; Delete the second posture from the corrected multiple first predicted postures to obtain at least one third predicted posture, determine the third posture corresponding to each third predicted posture, and connect the path node corresponding to each third predicted posture with the path node corresponding to the third posture, and the parent node of the path node corresponding to the third predicted posture is the path node corresponding to the third posture, so as to determine each fourth predicted posture corresponding to the initial posture, and the path node corresponding to each fourth predicted posture should be connected with the path node corresponding to the initial posture, and the parent node of the path node corresponding to each fourth predicted posture is the path node corresponding to the initial posture; Based on the connection relationship between multiple path nodes formed by the initial posture, the target posture and the corrected multiple first predicted postures, starting from the path node corresponding to the target posture, passing through the path nodes corresponding to the corrected multiple first predicted postures, and tracing back to the path node corresponding to the initial posture in sequence, the multiple parking paths are obtained.

5. The method according to claim 1, wherein Determining the quality score of each parking path on the first planning objective includes: Outputting a weight configuration reminder, wherein the weight configuration reminder is used to remind the target object in the vehicle to configure the weight corresponding to each first planning target; Upon receiving the weights corresponding to the respective first planning objectives, performing weighted fusion on the first score, the second score, and the third score based on the first configuration weight, the second configuration weight, and the third configuration weight to obtain the quality score, where the first score is obtained based on the angle deviation, the second score is obtained based on the number of short paths, and the third score is obtained based on the proportion of path nodes that meet a preset condition; outputting a target configuration reminder when the weights corresponding to the respective first planning targets are not received, wherein the target configuration reminder is used to remind the target subject to use the driving style as a planning target; Upon receiving a confirmation operation for the target configuration reminder, based on the driving style, a fourth configuration weight, a fifth configuration weight, and a sixth configuration weight are determined for weighted fusion of the first score, the second score, and the third score to determine the quality score.

6. The method according to claim 5, characterized in that The determining, based on the driving style, a fourth configuration weight, a fifth configuration weight, and a sixth configuration weight for weighted fusion of the first score, the second score, and the third score includes: When the driving style is an aggressive style, the fourth configuration weight is determined to be a first preset weight, the fifth configuration weight is determined to be a second preset weight, and the sixth configuration weight is determined to be a third preset weight, the second preset weight is greater than the first preset weight, and the second preset weight is greater than the third preset weight; When the driving style is a conservative style, the fourth configuration weight is determined as a fourth preset weight, the fifth configuration weight is determined as a fifth preset weight, and the sixth configuration weight is determined as a sixth preset weight, the fifth preset weight is smaller than the fourth preset weight, and the fifth preset weight is smaller than the sixth preset weight; In a case where the driving style is a balanced style, the fourth configuration weight, the fifth configuration weight, and the sixth configuration weight are determined as target preset weights.

7. The method according to claim 1, characterized in that Determining the quality score of each parking path on the first planning objective includes: determining a target path cost for each parking path, the parking path being determined based on the initial posture, the target posture, and at least one first predicted posture associated with the preset parking operation, the target path cost being the sum of a first path cost of the vehicle traveling from the initial posture to any first predicted posture and a second path cost of traveling from any first predicted posture to the target posture, the first path cost being the sum of a cumulative path cost of the vehicle traveling from the initial posture to a fifth predicted posture and an operation cost of traveling from the fifth predicted posture to any first predicted posture, the path node corresponding to the fifth predicted posture being a parent node of the path node corresponding to any first predicted posture; Determining the target path cost, total path length, and number of paths of each parking path as a second planning target when planning the parking path; A quality score of each of the parking paths based on the first planning objective and the second planning objective is determined.

8. A device for planning a parking route, characterized in that: The device comprises: Identify modules for: In response to a parking instruction from a vehicle, determining a plurality of parking paths based on an initial posture of the vehicle at a current position, a target posture of a target parking space, and a preset parking operation of the vehicle during parking; Determining a quality score for each parking path based on a first planning objective, the first planning objective including the angular deviation between the vehicle's heading during the last parking maneuver corresponding to the parking path and the orientation of the target parking space, the number of short paths in the parking path, and the percentage of path nodes in the parking path that meet a preset condition. A short path is a path whose length corresponding to the parking maneuver is less than a preset length. The preset condition indicates that the curvature of the path node corresponding to the posture is less than a preset curvature. A control module is configured to control parking of the vehicle based on a parking path with a highest quality score among the plurality of parking paths.

9. A vehicle, characterized in that: The vehicle comprises: a memory for storing executable program code; A processor is configured to call and run the executable program code from the memory, so that the vehicle executes the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores an executable program code, and when the executable program code is executed, the method according to any one of claims 1 to 7 is implemented.

Images