A parking APA planning control method and device

Abstract

The present invention provides a parking APA planning control method and device, which solves the problems of poor parking path planning accuracy, weak adaptability and high computing power requirements in the prior art. By obtaining the coordinate values ​​of the parking space type, parking space corner points, obstacle corner points and limit block corner points in the world coordinate system, and constructing the parking space coordinate system, the parking starting point posture, parking pre-end point posture and parking end point posture are formed, and then a parking path is generated through a variety of path planning algorithms. While completing high-precision parking path planning, low computing power parking planning and strong adaptability of parking path planning are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of parking planning, and in particular to a parking APA planning control method and device. Background Art

[0002] With the rapid development of high-tech technologies such as automobile intelligence, electrification, networking, and sharing, the automobile market is booming, and the number of domestic passenger cars is continuing to increase. However, the number of parking lots in China is currently limited, especially in first-tier cities. The shortage of parking spaces cannot meet the current parking needs, resulting in more and more narrow parking spaces in China. Especially for many novice drivers, drivers with less skilled driving skills, etc., "difficult parking, difficult parking" has become a headache for them. In addition, due to the small parking spaces and narrow spacing between parking spaces, the difficulty of parking in and out of the vehicle and the inability to open the door after parking are becoming increasingly serious. Therefore, there is a need for a stable, reliable, safe and high-precision automatic parking method and system.

[0003] Existing automatic parking systems include technologies such as automatic parking assist (APA), remote parking, and automated valet parking (AVP). Among them, APA (Automatic Parking Assist) is an advanced automotive technology designed to help drivers complete parking operations more easily and safely. This system usually uses sensors on the vehicle (such as ultrasonic sensors or cameras) to detect parking spaces and automatically controls the steering wheel to guide the vehicle into the parking space.

[0004] However, in the field of automatic parking technology in the current automotive industry, there are the following problems: (1) Strict control accuracy requirements: Even if an ideal parking path is available, how to make the vehicle drive along the expected trajectory is still a difficult problem. Due to the influence of vehicle dynamics and the existence of external interference factors, it is often difficult to achieve the theoretically perfect effect in the actual implementation process, especially in narrow spaces or complex road conditions, which places more stringent requirements on the control system; (2) Insufficient adaptability: During the automatic parking process, when the parking scene changes, it is impossible to adapt to the changed parking scene; (3) High computing power requirements: In the process of parking path planning, in order to achieve a better path planning, the path planning process is completed through a path planning algorithm with high computing power requirements, which results in a certain degree of computing power waste. Summary of the invention

[0005] The present invention provides a parking APA planning control method and device to solve the problems of poor parking path planning accuracy, weak adaptability and high computing power requirements in the prior art.

[0006] In a first aspect, the present invention provides a parking APA planning control method, which specifically includes the following steps:

[0007] Step S1, obtaining external environment information required for parking, and generating a parking starting point pose, a parking pre-end pose and a parking end pose in a world coordinate system according to the external environment information;

[0008] Step S2, performing parking path planning processing according to the parking starting point pose, the parking pre-end pose and the parking end pose in the world coordinate system to form a parking path; wherein the parking path includes N+1 track point information;

[0009] Step S3: transmitting the parking path to the regulation and control interaction processor for processing, and completing the control of the parking path through the regulation and control interaction processor.

[0010] Wherein, N is a positive integer.

[0011] Preferably, in step S1, external environment information required for parking is obtained, and a parking start pose, a parking pre-end pose, and a parking end pose in a world coordinate system are generated according to the external environment information, which specifically includes the following steps:

[0012] Step S101, obtaining the coordinate values ​​of the parking space type, parking space corner points, and obstacle corner points in the world coordinate system. If a limit block is provided in the parking space (in this application, the "limit block" is also called a parking block or a limiter, which is a physical obstacle installed behind the parking space and is used to limit the distance the vehicle moves backward to ensure that the vehicle does not overdrive into the parking space, thereby avoiding collision with objects behind such as walls, other vehicles or structures), it is also necessary to obtain the coordinate values ​​of the limit block corner points in the world coordinate system and construct a parking space coordinate system;

[0013] Step S102: forming a parking end point pose in the world coordinate system (in this application, the “pose” refers to the position and posture of the vehicle in the coordinate system) according to the parking space type, the parking space corner point coordinate values ​​in the parking space coordinate system and the world coordinate system, and the obstacle corner point coordinate values. If a limit block is provided in the parking space, forming a parking end point pose in the world coordinate system according to the parking space type, the parking space corner point coordinate values ​​in the parking space coordinate system and the world coordinate system, the obstacle corner point coordinate values ​​and the limit block corner point coordinate values;

[0014] Step S103, processing obstacles inside and outside a region of interest (ROI, where, in the present application, the “region of interest” refers to a portion of the image or sensor data that requires special attention, i.e., a region that is critical to the automatic parking process, and contains features that are helpful in identifying parking spaces, obstacles, or other key information) and performing real-time monitoring of dynamic obstacles to form a pre-end pose of parking in the world coordinate system; wherein the pre-end pose is within the parking space;

[0015] Step S104: obtain a positioning source used for parking (in this application, the "positioning source" refers to a method used to obtain positioning, such as SLAM positioning and ODOM positioning. The errors between the positioning results generated by different positioning methods and the actual positioning values ​​are different), and determine the parking starting point pose in the world coordinate system.

[0016] Preferably, in step S102, generating a parking end pose in the world coordinate system according to the parking space type, the parking space coordinate system and the parking space corner coordinate values ​​in the world coordinate system, the obstacle corner coordinate values ​​and the limit block corner coordinate values ​​specifically includes the following steps:

[0017] Step S102a, the parking space corner point coordinate values, obstacle corner point coordinate values ​​and limit block corner point coordinate values ​​in the world coordinate system are subjected to coordinate system conversion processing to form parking space corner point coordinate values, obstacle corner point coordinate values ​​and limit block corner point coordinate values ​​in the parking space coordinate system;

[0018] Step S102b, removing obstacle points inside the parking space according to the parking space corner point coordinate values ​​and obstacle corner point coordinate values ​​in the parking space coordinate system;

[0019] Step S102c, forming a parking pre-end position and a parking end position in the parking space coordinate system according to the parking space corner coordinate values, obstacle corner coordinate values, limit block corner coordinate values ​​and the parking space type in the parking space coordinate system;

[0020] Step S102d: performing coordinate system conversion processing on the parking pre-end pose and the parking end pose in the parking space coordinate system to form the parking pre-end pose and the parking end pose in the world coordinate system.

[0021] Preferably, in step S102c, the parking space types include vertical parking spaces, horizontal parking spaces and inclined parking spaces.

[0022] Among them, the oblique training positions include the left-side herringbone oblique training position, the left-side reverse herringbone oblique training position, the right-side herringbone oblique training position and the right-side reverse herringbone oblique training position.

[0023] Preferably, in step S2, the information of each trajectory point includes the position, direction (in the present application, the direction includes a first direction and a second direction, the first direction indicates the direction in which the vehicle is parked into the parking space, and the second direction indicates the direction in which the vehicle is parked out of the parking space), speed, heading, curvature and acceleration of the trajectory point.

[0024] Preferably, in step S2, a parking path planning process is performed according to the parking starting point pose, the parking pre-end pose and the parking end pose in the world coordinate system to form a parking path, which specifically includes the following steps:

[0025] Step S201, planning and processing a parking path between the parking starting point pose and the parking pre-end point pose by using the Hybird A Star algorithm to form a first parking trajectory;

[0026] Step S202, planning a parking path between the parking pre-end pose and the parking end pose through a straight line to form a second parking trajectory;

[0027] Step S203: forming a parking path according to the first parking trajectory and the second parking trajectory.

[0028] Preferably, step S2 further includes:

[0029] Step S204: Optimizing the parking path: optimizing the parking path according to a preset cost function of the parking path to form an optimized parking path.

[0030] Preferably, in step S204, the preset cost function is specifically expressed as follows:

[0031] Cost total =W l *Cost l +W ref *Cost ref +W smooth *Cost smooth +W slack*Cost slack ;

[0032] in, Cost total represents the preset cost function (i.e., the total cost function when optimizing the parking path), W l represents the weight of the optimized parking path length cost, W ref The weight representing the offset cost of the optimized parking path relative to the original path, W smooth represents the weight of the smoothness cost of the optimized parking path, W slack represents the weight of the slack variable, Cost l represents the optimized parking path length cost, Cost ref It represents the offset cost of the optimized parking path relative to the original parking path. Cost smooth represents the smoothness cost of the optimized parking path, Cost slack represents the slack variable cost.

[0033] Assume that the positions of N+1 original trajectory points (in this application, the original trajectory points refer to the trajectory points before trajectory optimization, and the optimized trajectory points refer to the trajectory points after trajectory optimization) in the world coordinate system are respectively , the positions of the N+1 original trajectory points in the world coordinate system can be expressed as , the optimization variable is , then the optimized parking path length cost Cost l , specifically expressed as follows:

[0034] ;

[0035] in, i is a positive integer; x i Represents the first i The horizontal coordinate of the trajectory point, y i Represents the first i The vertical coordinate of each trajectory point;

[0036] The cost of the optimized parking path offset from the original parking path Cost ref , specifically expressed as follows:

[0037] ;

[0038] in, x i-ref Represents the original i The horizontal coordinate of the trajectory point, y i-ref Represents the original i The vertical coordinates of the trajectory points; the smoothness cost of the parking path after optimization Cost smooth , specifically expressed as follows:

[0039] ;

[0040] During the optimization of the preset cost function, the constraint conditions include boundary constraints and curvature constraints;

[0041] The boundary constraint is specifically expressed as follows:

[0042] ;

[0043] in, lower bound-i Indicates the world coordinate system i The maximum lower boundary change value of the trajectory point, upper bound-i Indicates the world coordinate system i The maximum upper boundary change value of the trajectory points;

[0044] The curvature constraint is specifically expressed as follows:

[0045] ;

[0046] in, represents the average value of the sum of the Euclidean distances between two points of the original trajectory in the world coordinate system, k max represents the maximum permissible curvature;

[0047] In order to avoid the optimization problem being unsolvable, the curvature constraint is corrected. The corrected curvature constraint is specifically expressed as follows:

[0048] ;

[0049] in, Indicates i The slack variables of the trajectory points are ;

[0050] Slack variable cost Cost slack , specifically expressed as follows:

[0051] .

[0052] Preferably, in step S3, the regulation and control interaction processor includes a parking planning information processing module and a parking control module, the parking control module is connected to the parking planning information processing module, the parking planning information processing module is used to transmit parking trajectory information (in the present application, the parking trajectory information includes parking trajectory and trajectory point information); the parking control module is used to issue vehicle control instructions and transmit feedback status to the parking planning information processing module.

[0053] In the present application, the states of the parking planning information processing module include No Op and Execute; wherein No Op indicates a no-operation state, i.e., a state where no action is required, in which the parking planning information processing module does not perform the action of transmitting the parking trajectory information; Execute represents an execution state, in which the parking planning information processing module is used to track and control multiple (in the present application, the "multiple" means at least two) trajectory points, and perform the action of transmitting the parking trajectory information to the parking control module.

[0054] In the present application, the states of the parking control module include Pending, Active and Success; wherein Pending indicates a suspended state, in which the parking control module waits to receive the parking trajectory information transmitted by the parking planning information processing module, and does not execute any control action of issuing control instructions; Active indicates an active state, in which the parking control module generates a control instruction according to the received parking trajectory information and executes the action of issuing the control instruction; Success indicates a completed state, in which the parking control module determines that all actions of the current trajectory tracking control have been completed according to the remaining distance information of the trajectory point, and does not execute the control action of issuing the control instruction.

[0055] Preferably, in step S3, the parking path is transmitted to a regulation and control interaction processor for processing, and the parking path is controlled by the regulation and control interaction processor, which specifically includes the following steps:

[0056] Step S301: when a parking path is not formed, the parking planning information processing module is set to No Op and the parking control module is set to Pending. When a parking path is formed, the parking planning information processing module stores the start pose, the pre end pose, the end pose and the parking path and transmits them to the parking control module, and the parking control module is set to Execute.

[0057] Step S302: dividing the parking path into a plurality of parking trajectory segments by the parking planning information processing module;

[0058] Step S303: the parking planning information processing module obtains the current parking trajectory segment, and transmits the current parking trajectory segment to the parking control module;

[0059] Step S304: After the parking control module receives the current parking trajectory segment, the parking control module status is Active, the parking control module issues a control instruction and tracks the parking process according to the current parking trajectory segment. After the current parking trajectory segment is tracked, the parking control module status is Success. After the status Success is transmitted to the parking planning information processing module, the parking control module status is Pending.

[0060] Step S305: After receiving the status Success, the parking planning information processing module switches the current parking trajectory segment and sets the next parking trajectory segment as the current parking trajectory segment;

[0061] Step S306: Repeat steps S303 to S305 until the parking control module completes tracking of all parking trajectory segments.

[0062] Preferably, in step S302, the parking path is divided into a plurality of parking trajectory segments according to the position, speed and direction of the trajectory points in the parking path.

[0063] Preferably, a parking APA planning control method further includes:

[0064] Step S4: When the parking planning scenario changes, step S1 to step S2 are re-executed to form a new parking path.

[0065] Preferably, when the parking planning scenario changes, the new parking path is transmitted to the regulation and control interaction processor for processing, and the parking path is controlled by the regulation and control interaction processor, which specifically includes the following steps:

[0066] Step S307: when the parking planning scenario changes, the parking planning information processing module execution action is canceled, the parking control module status is Pending, and the parking planning information processing module status is Execute;

[0067] Step S308 , completing the control of the new parking path through steps S301 - S306 .

[0068] Preferably, when parking is continued after a pause in the parking process, the control of the parking path is completed by the regulation and control interaction processor, which specifically includes the following steps:

[0069] Step S309: when the parking process is paused and then continued, the state of the parking planning information processing module is Execute, and the state of the parking control module is Active;

[0070] Step S3010, continue to complete the control of the parking path through steps S303 to S306.

[0071] In a second aspect, the present invention further provides a parking APA planning control device, which specifically includes the following modules:

[0072] A parking posture generation module is used to obtain the external environment information required for parking, and generate a parking start pose, a parking pre-end pose and a parking end pose according to the external environment information;

[0073] A parking path generation module, configured to perform parking path planning processing according to the parking starting point pose, the parking pre-end pose and the parking end pose in the world coordinate system to form a parking path; wherein the parking path includes N+1 track point information;

[0074] The parking regulation and control interaction module is used to transmit the parking path to the regulation and control interaction processor for processing, and complete the control of the parking path through the regulation and control interaction processor.

[0075] Wherein, N is a positive integer.

[0076] Preferably, the parking posture generation module specifically includes the following submodules:

[0077] The first submodule of the parking posture generation is used to obtain the coordinate values ​​of the corner points of the parking space, the corner points of the obstacles and the corner points of the limit blocks in the world coordinate system. If there are limit blocks in the parking space, it is also necessary to obtain the coordinate values ​​of the corner points of the limit blocks in the world coordinate system and construct the parking space coordinate system;

[0078] The second parking pose generation submodule is used to form a parking end pose in the world coordinate system according to the parking space type, the parking space corner coordinate values ​​in the parking space coordinate system and the world coordinate system, and the obstacle corner coordinate values. If a limit block is provided in the parking space, the parking end pose in the world coordinate system needs to be formed according to the parking space type, the parking space corner coordinate values ​​in the parking space coordinate system and the world coordinate system, the obstacle corner coordinate values ​​and the limit block corner coordinate values;

[0079] The third submodule of parking pose generation is used to process obstacles inside and outside the roi and monitor dynamic obstacles in real time to form a parking pre-end pose in the world coordinate system; wherein the pre-end pose is in the parking space;

[0080] The fourth submodule of parking posture generation is used to obtain the positioning source used for parking and determine the parking starting point posture in the world coordinate system.

[0081] Preferably, the second submodule for generating parking posture specifically includes the following submodules:

[0082] The first submodule is used for converting the parking space corner point coordinate values, obstacle corner point coordinate values ​​and limit block corner point coordinate values ​​in the world coordinate system into the parking space corner point coordinate values, obstacle corner point coordinate values ​​and limit block corner point coordinate values ​​in the parking space coordinate system;

[0083] The second submodule is used to remove the obstacle points inside the parking space according to the parking space corner point coordinate values ​​and the obstacle corner point coordinate values ​​in the parking space coordinate system;

[0084] The third submodule is used to form a parking pre-end position and a parking end position in the parking space coordinate system according to the parking space corner point coordinate value, the obstacle corner point coordinate value, the limit block corner point coordinate value and the parking space type in the parking space coordinate system;

[0085] The fourth submodule is used to perform coordinate system conversion processing on the parking pre-end pose and the parking end pose in the parking space coordinate system to form the parking pre-end pose and the parking end pose in the world coordinate system.

[0086] Preferably, in the third grandchild module, the parking space types include vertical parking spaces, horizontal parking spaces and inclined parking spaces.

[0087] Among them, the oblique training positions include the left-side herringbone oblique training position, the left-side reverse herringbone oblique training position, the right-side herringbone oblique training position and the right-side reverse herringbone oblique training position.

[0088] Preferably, in the parking path generation module, each trajectory point information includes the position, direction, speed, heading, curvature and acceleration of the trajectory point.

[0089] Preferably, the parking path generation module specifically includes the following submodules:

[0090] The first parking path planning submodule is used to plan and process the parking path between the parking starting point pose and the parking pre-end pose by using the Hybird A Star algorithm to form a first parking trajectory;

[0091] The second parking path planning submodule is used to plan and process the parking path between the parking pre-end pose and the parking end pose through a straight line to form a second parking trajectory;

[0092] The third submodule of parking path planning is used to form a parking path according to the first parking trajectory and the second parking trajectory.

[0093] Preferably, the parking path planning module further includes:

[0094] The fourth submodule of parking path planning is used to optimize the parking path, and optimize the parking path according to a preset cost function of the parking path to form an optimized parking path.

[0095] Preferably, in the fourth submodule of parking path planning, the preset cost function is specifically expressed as follows:

[0096] Cost total =W l *Cost l +W ref *Cost ref +W smooth *Cost smooth +W slack *Cost slack ;

[0097] in, Cost total represents the preset cost function (i.e., the total cost function when optimizing the parking path), W l represents the weight of the optimized parking path length cost, Wref The weight representing the offset cost of the optimized parking path relative to the original path, W smooth represents the weight of the smoothness cost of the optimized parking path, W slack represents the weight of the slack variable, Cost l represents the optimized parking path length cost, Cost ref It represents the offset cost of the optimized parking path relative to the original parking path. Cost smooth represents the smoothness cost of the optimized parking path, Cost slack represents the slack variable cost.

[0098] Assume that the positions of N+1 original trajectory points in the world coordinate system are ,…, , the positions of the N+1 original trajectory points in the world coordinate system can be expressed as , , the optimization variable is , then the optimized parking path length cost Cost l , specifically expressed as follows:

[0099] ;

[0100] in, i is a positive integer; x i Represents the first i The horizontal coordinate of the trajectory point, y i Represents the first i The vertical coordinate of each trajectory point;

[0101] The cost of the optimized parking path offset from the original parking path Cost ref , specifically expressed as follows:

[0102] ;

[0103] in, x i-ref Represents the original i The horizontal coordinate of the trajectory point, y i-ref Represents the original i The vertical coordinates of the trajectory points; the smoothness cost of the parking path after optimization Cost smooth, specifically expressed as follows:

[0104] ;

[0105] During the optimization of the preset cost function, the constraint conditions include boundary constraints and curvature constraints;

[0106] The boundary constraint is specifically expressed as follows:

[0107] ;

[0108] in, lower bound-i Indicates the world coordinate system i The maximum lower boundary change value of the trajectory point, upper bound-i Indicates the world coordinate system i The maximum upper boundary change value of the trajectory points;

[0109] The curvature constraint is specifically expressed as follows:

[0110] ;

[0111] in, represents the average value of the sum of the Euclidean distances between two points of the original trajectory in the world coordinate system, k max represents the maximum permissible curvature;

[0112] In order to avoid the optimization problem being unsolvable, the curvature constraint is corrected. The corrected curvature constraint is specifically expressed as follows:

[0113] ;

[0114] in, Indicates i The slack variables of the trajectory points are ;

[0115] Slack variable cost Cost slack , specifically expressed as follows:

[0116] .

[0117] Preferably, in the parking regulation and control interaction module, the regulation and control interaction processor includes a parking planning information processing module and a parking control module, the parking control module is connected to the parking planning information processing module, the parking planning information processing module is used to transmit parking trajectory information; the parking control module is used to issue vehicle control instructions and transmit feedback status to the parking planning information processing module.

[0118] Preferably, the parking control interaction module specifically includes the following submodules:

[0119] The first parking planning and control interaction submodule is used for, when a parking path is not formed, setting the parking planning information processing module status to No Op and the parking control module status to Pending; when a parking path is formed, the parking planning information processing module stores the start pose, the pre end pose, the end pose and the parking path and transmits them to the parking control module, and the parking control module status is Execute;

[0120] A second parking control interaction submodule, configured to divide the parking path into a plurality of parking trajectory segments through the parking planning information processing module;

[0121] A third parking planning and control interaction submodule, which is used for the parking planning information processing module to obtain the current parking trajectory segment and transmit the current parking trajectory segment to the parking control module;

[0122] The fourth parking control interaction submodule is used for when the parking control module receives the current parking trajectory segment, the parking control module status is Active, the parking control module issues a control instruction and tracks the parking process according to the current parking trajectory segment, after the current parking trajectory segment is tracked, the parking control module status is Success, after the status Success is transmitted to the parking planning information processing module, the parking control module status is Pending;

[0123] The fifth parking control interaction submodule is used for switching the current parking trajectory segment and setting the next parking trajectory segment as the current parking trajectory segment after the parking planning information processing module receives the status Success;

[0124] The parking control interaction sixth submodule is used to repeatedly execute the parking control interaction third submodule, the parking control interaction fourth submodule and the parking control interaction fifth submodule until the parking control module completes tracking of all parking trajectory segments.

[0125] Preferably, in the second parking control interaction submodule, the parking path is divided into a plurality of parking trajectory segments according to the position, speed and direction of the trajectory points in the parking path.

[0126] Preferably, a parking APA planning control device further includes:

[0127] The parking path re-planning module is used to re-execute the parking posture generation module and the parking path generation module to form a new parking path when the parking planning scenario changes.

[0128] Preferably, the parking control interaction module further includes:

[0129] The re-planning parking regulation and control interaction module is used to transmit the new parking path to the regulation and control interaction processor for processing when the parking planning scenario changes, and complete the control of the parking path through the regulation and control interaction processor, which specifically includes the following sub-modules:

[0130] The first submodule of the re-planning parking control interaction is used to cancel the execution action of the parking planning information processing module when the parking planning scenario changes, the state of the parking control module is Pending, and the state of the parking planning information processing module is Execute;

[0131] The second parking control interaction submodule is replanned to complete the control of the new parking path through the first parking control interaction submodule, the second parking control interaction submodule, the third parking control interaction submodule, the fourth parking control interaction submodule, the fifth parking control interaction submodule and the sixth parking control interaction submodule.

[0132] Preferably, the parking control interaction module further includes:

[0133] The pause and restart parking control interaction module is used to complete the control of the parking path through the control interaction processor when the parking process is paused and then continued. It specifically includes the following submodules:

[0134] The first submodule of the parking control interaction of pausing and restarting is used for when the parking process is paused and then continued, the state of the parking planning information processing module is Execute, and the state of the parking control module is Active;

[0135] The parking regulation and control interaction first submodule is paused and restarted to continue to complete the control of the parking path through the parking regulation and control interaction third submodule, the parking regulation and control interaction fourth submodule, the parking regulation and control interaction fifth submodule and the parking regulation and control interaction sixth submodule.

[0136] In a third aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a parking APA planning control method as described in any one of the first aspects of the present application.

[0137] In a fourth aspect, the present invention further provides an electronic device, comprising: a memory storing a computer program; and a processor communicatively connected to the memory, for executing a parking APA planning control method described in any one of the first aspects of the present application when the computer program is called.

[0138] Compared with the prior art, the present invention has the following obvious outstanding substantive features and significant advantages:

[0139] The present invention provides a parking APA planning control method and device, which solves the problems of poor parking path planning accuracy, weak adaptability and high computing power requirements in the prior art. By obtaining the coordinate values ​​of the parking space type, parking space corner points, obstacle corner points and limit block corner points in the world coordinate system, and constructing the parking space coordinate system, the parking starting point posture, parking pre-end point posture and parking end point posture are formed, and then a parking path is generated through a variety of path planning algorithms. While completing high-precision parking path planning, low computing power parking planning and strong adaptability of parking path planning are achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0140] The accompanying drawings, which constitute part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention. In the accompanying drawings:

[0141] Figure 1 It is a flow chart of a parking APA planning control method according to a preferred embodiment of the present invention.

[0142] Figure 2 It is a schematic diagram of the interaction logic of a normal parking scene control interaction processor in a parking APA planning control method in a preferred embodiment of the present invention.

[0143] Figure 3 It is a schematic diagram of interaction logic of a re-planning scene control interaction processor in a parking APA planning control method in a preferred embodiment of the present invention.

[0144] Figure 4 It is a schematic diagram of the interaction logic of the parking pause and continue scenario control interaction processor in a parking APA planning control method in a preferred embodiment of the present invention.

[0145] Figure 5 It is a schematic diagram of the structure of a parking APA planning control device according to a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0146] The present invention provides a parking APA planning control method and device. In order to make the purpose, technical solution and effect of the present invention clearer and more specific, the present invention is further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described here are only used to explain the present invention and are not used to limit the present invention.

[0147] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0148] Example 1:

[0149] like Figures 1 - 4 As shown, a parking APA planning control method described in this embodiment specifically includes the following steps:

[0150] Step S1, obtaining external environment information required for parking, and generating a parking start pose, a parking pre-end pose, and a parking end pose in a world coordinate system according to the external environment information.

[0151] Optionally, in step S1, external environment information required for parking is obtained, and a parking start pose, a parking pre-end pose, and a parking end pose in a world coordinate system are generated according to the external environment information, specifically comprising the following steps:

[0152] Step S101, obtaining the coordinate values ​​of the parking space type, parking space corner points, and obstacle corner points in the world coordinate system. If a limit block is provided in the parking space, it is also necessary to obtain the coordinate values ​​of the limit block corner points in the world coordinate system, and construct a parking space coordinate system.

[0153] Step S102: forming an end position of parking in the world coordinate system according to the parking space type, the parking space coordinate system and the parking space corner point coordinate values ​​and the obstacle corner point coordinate values ​​in the world coordinate system; if a limit block is provided in the parking space, forming an end position of parking in the world coordinate system according to the parking space type, the parking space coordinate system and the parking space corner point coordinate values, the obstacle corner point coordinate values ​​and the limit block corner point coordinate values.

[0154] Optionally, in step S102, generating a parking end pose in the world coordinate system according to the parking space type, the parking space coordinate system and the parking space corner coordinate values ​​in the world coordinate system, the obstacle corner coordinate values ​​and the limit block corner coordinate values, specifically includes the following steps:

[0155] Step S102a, the parking space corner point coordinate values, obstacle corner point coordinate values ​​and limit block corner point coordinate values ​​in the world coordinate system are subjected to coordinate system conversion processing to form the parking space corner point coordinate values, obstacle corner point coordinate values ​​and limit block corner point coordinate values ​​in the parking space coordinate system.

[0156] Step S102b: remove obstacles inside the parking space according to the parking space corner point coordinate values ​​and the obstacle corner point coordinate values ​​in the parking space coordinate system.

[0157] Step S102c, forming a parking pre-end position and a parking end position in the parking space coordinate system according to the parking space corner coordinate values, obstacle corner coordinate values, limit block corner coordinate values ​​and the parking space type in the parking space coordinate system; wherein the parking space types include vertical parking spaces, horizontal parking spaces and oblique parking spaces; the oblique parking spaces include left-side herringbone oblique parking spaces, left-side reverse herringbone oblique parking spaces, right-side herringbone oblique parking spaces and right-side reverse herringbone oblique parking spaces.

[0158] Step S102d: performing coordinate system conversion processing on the parking pre-end pose and the parking end pose in the parking space coordinate system to form the parking pre-end pose and the parking end pose in the world coordinate system.

[0159] Step S103: Process obstacles inside and outside the ROI and monitor dynamic obstacles in real time to form a pre-end pose of parking in the world coordinate system; wherein the pre-end pose is in the parking space.

[0160] Step S104: Acquire the positioning source used for parking, and determine the parking starting point pose startpose in the world coordinate system.

[0161] Step S2: performing parking path planning processing according to the parking start pose, the parking pre-end pose and the parking end pose in the world coordinate system to form a parking path; wherein the parking path includes N+1 (where N is a positive integer) track point information; wherein each track point information includes the position, direction, speed, heading, curvature and acceleration of the track point.

[0162] Wherein, step S2 specifically includes the following steps:

[0163] Step S201: planning a parking path between the parking starting point pose and the parking pre-end point pose by using the Hybird A Star algorithm to form a first parking trajectory.

[0164] Step S202: planning a parking path between the parking pre-end pose and the parking end pose through a straight line to form a second parking trajectory.

[0165] In this embodiment, in step S201 and step S202, different path planning algorithms are used for planning. Optionally, the generation of the first parking trajectory is planned by a Hybrid A Star algorithm.

[0166] Step S203: forming a parking path according to the first parking trajectory and the second parking trajectory.

[0167] The parking path planning process is implemented through steps S201-S203. Compared with directly connecting the parking starting point to the parking end point through the Hybrid A Star algorithm, the search space can be greatly reduced. At the same time, it will also consume a lot of computing power. For the low computing power platform carried by the vehicle, the above-mentioned segmented parking path planning can better achieve low computing power and high efficiency parking effect. The parking path between the parking pre-end pose and the parking end pose is planned and processed by a straight line, which has the following advantages: 1) The safety level inside the parking space is high and the trajectory diversity is low, and there is no need to use the Hybrid A Star algorithm for parking path planning; 2) It has strong plasticity. By setting the pre-end pose, the search space of the Hybrid A Star algorithm can be better adjusted.

[0168] Optionally, step S2 further includes step S204, optimizing the parking path, optimizing the parking path according to a preset cost function of the parking path to form an optimized parking path.

[0169] The preset cost function is specifically expressed as follows:

[0170] Cost total =W l *Cost l +W ref *Cost ref+W smooth *Cost smooth +W slack *Cost slack ;

[0171] in, Cost total represents the preset cost function (i.e., the total cost function when optimizing the parking path), W l represents the weight of the optimized parking path length cost, W ref The weight representing the offset cost of the optimized parking path relative to the original path, W smooth represents the weight of the smoothness cost of the optimized parking path, W slack represents the weight of the slack variable, Cost l represents the optimized parking path length cost, Cost ref It represents the offset cost of the optimized parking path relative to the original parking path. Cost smooth represents the smoothness cost of the optimized parking path, Cost slack represents the slack variable cost.

[0172] Assume that the positions of N+1 original trajectory points in the world coordinate system are ,…, , the positions of the N+1 original trajectory points in the world coordinate system can be expressed as , , the optimization variable is , then the optimized parking path length cost Cost l , specifically expressed as follows:

[0173] ;

[0174] in, i is a positive integer; x i Represents the first i The horizontal coordinate of the trajectory point, y i Represents the first i The vertical coordinate of each trajectory point;

[0175] The cost of the optimized parking path offset from the original parking path Cost ref, specifically expressed as follows:

[0176] ;

[0177] in, x i-ref Represents the original i The horizontal coordinate of the trajectory point, y i-ref Represents the original i The vertical coordinates of the trajectory points; the smoothness cost of the parking path after optimization Cost smooth , specifically expressed as follows:

[0178] ;

[0179] During the optimization of the preset cost function, the constraint conditions include boundary constraints and curvature constraints;

[0180] The boundary constraint is specifically expressed as follows:

[0181] ;

[0182] in, lower bound-i Indicates the world coordinate system i The maximum lower boundary change value of the trajectory point, upper bound-i Indicates the world coordinate system i The maximum upper boundary change value of the trajectory points;

[0183] The curvature constraint is specifically expressed as follows:

[0184] ;

[0185] in, represents the average value of the sum of the Euclidean distances between two points of the original trajectory in the world coordinate system, k max represents the maximum permissible curvature;

[0186] In order to avoid the optimization problem being unsolvable, the curvature constraint is corrected. The corrected curvature constraint is specifically expressed as follows:

[0187] ;

[0188] in, Indicates i The slack variables of the trajectory points are ;

[0189] Slack variable costCost slack , specifically expressed as follows:

[0190] .

[0191] Step S3: transmitting the parking path to the regulation and control interaction processor for processing, and completing the control of the parking path through the regulation and control interaction processor.

[0192] Among them, the regulation and control interaction processor includes a parking planning information processing module and a parking control module. The parking control module is connected to the parking planning information processing module, and the parking planning information processing module is used to transmit parking trajectory information; the parking control module is used to issue vehicle control instructions and transmit feedback status to the parking planning information processing module.

[0193] Among them, Figure 3 As shown, step S3 specifically includes the following steps:

[0194] Step S301: When a parking path is not formed, the parking planning information processing module status is No Op and the parking control module status is Pending. When a parking path is formed, the parking planning information processing module stores the start pose, the pre end pose, the end pose and the parking path and transmits them to the parking control module, and the parking control module status is Execute.

[0195] Step S302: The parking path is divided into a plurality of parking trajectory segments by the parking planning information processing module; wherein the parking path is divided into a plurality of parking trajectory segments according to the position, speed and direction of the trajectory points in the parking path. Due to the problem of insufficient computing power that some processors may have, the communication load cannot completely send down (transmit to) all the trajectory points in the entire parking path to the control module, so the regulation and control interaction processor will divide the parking path and send down the divided multiple parking trajectory segments to the control module, and the control module will track the trajectory segments. At the same time, the regulation and control interaction processor is also responsible for the interaction between planning and control after special problems such as re-planning and unexpected pause occur.

[0196] Step S303: the parking planning information processing module obtains the current parking trajectory segment, and transmits the current parking trajectory segment to the parking control module.

[0197] Step S304: After the parking control module receives the current parking trajectory segment, the parking control module status is Active, the parking control module issues a control instruction and tracks the parking process according to the current parking trajectory segment. After completing the tracking of the current parking trajectory segment, the parking control module status is Success. After transmitting the status Success to the parking planning information processing module, the parking control module status is Pending.

[0198] Step S305: After receiving the status Success, the parking planning information processing module switches the current parking trajectory segment and sets the next parking trajectory as the current parking trajectory segment.

[0199] Step S306: Repeat steps S303 to S305 until the parking control module completes tracking of all parking trajectory segments.

[0200] Optionally, a parking APA planning control method further includes step S4, when the parking planning scenario changes, re-executing steps S1 to S2 to form a new parking path.

[0201] like Figure 3 As shown, when the parking planning scenario changes, the new parking path is transmitted to the regulation and control interaction processor for processing, and the parking path is controlled by the regulation and control interaction processor, which specifically includes the following steps:

[0202] Step S307: When the parking planning scenario changes, the parking planning information processing module execution action is canceled, the parking control module status is Pending, and the parking planning information processing module status is Execute.

[0203] Step S308 , completing the control of the new parking path through steps S301 - S306 .

[0204] like Figure 4 As shown, when a parking process is paused and then the parking process is continued, the parking path is controlled by the regulation and control interaction processor, which specifically includes the following steps:

[0205] Step S309: When the parking process is paused and then continued, the state of the parking planning information processing module is Execute, and the state of the parking control module is Active.

[0206] Step S3010, continue to complete the control of the parking path through steps S303 to S306.

[0207] Example 2:

[0208] like Figure 5 As shown, the parking APA planning control device described in this embodiment specifically includes the following modules:

[0209] The parking posture generation module is used to obtain the external environment information required for parking, and generate the parking starting posture start pose, the parking pre-end posture pre end pose and the parking end posture end pose according to the external environment information.

[0210] Among them, the parking posture generation module specifically includes the following sub-modules:

[0211] The first submodule of parking posture generation is used to obtain the coordinate values ​​of the corner points of the parking space, the corner points of the obstacles and the corner points of the limit blocks in the world coordinate system. If there are limit blocks in the parking space, it is also necessary to obtain the coordinate values ​​of the corner points of the limit blocks in the world coordinate system and construct the parking space coordinate system.

[0212] The second parking posture generation submodule is used to form a parking end pose in the world coordinate system according to the parking space type, the parking space coordinate system and the parking space corner point coordinate values ​​and the obstacle corner point coordinate values ​​in the world coordinate system. If a limit block is provided in the parking space, the parking end pose in the world coordinate system needs to be formed according to the parking space type, the parking space coordinate system and the parking space corner point coordinate values, the obstacle corner point coordinate values ​​and the limit block corner point coordinate values ​​in the world coordinate system.

[0213] Among them, the second submodule of parking posture generation includes the following submodules:

[0214] The first submodule is used for converting the parking space corner point coordinate values, obstacle corner point coordinate values ​​and limit block corner point coordinate values ​​in the world coordinate system into the parking space corner point coordinate values, obstacle corner point coordinate values ​​and limit block corner point coordinate values ​​in the parking space coordinate system;

[0215] The second submodule is used to remove obstacles inside the parking space according to the parking space corner point coordinate values ​​and the obstacle corner point coordinate values ​​in the parking space coordinate system;

[0216] The third submodule is used to form a parking pre-end position and a parking end position in the parking space coordinate system according to the parking space corner point coordinate value, the obstacle corner point coordinate value, the limit block corner point coordinate value and the parking space type in the parking space coordinate system;

[0217] The fourth submodule is used to perform coordinate system conversion processing on the parking pre-end pose and the parking end pose in the parking space coordinate system to form the parking pre-end pose and the parking end pose in the world coordinate system.

[0218] The third submodule of parking posture generation is used to process obstacles inside and outside the roi and monitor dynamic obstacles in real time to form a parking pre-end pose in the world coordinate system; wherein the pre-end pose is in the parking space.

[0219] The fourth submodule of parking posture generation is used to obtain the positioning source used for parking and determine the parking starting point posture in the world coordinate system.

[0220] The parking path generation module is used to perform parking path planning processing according to the parking start pose, the parking pre-end pose and the parking end pose to form a parking path; wherein the parking path includes N+1 (where N is a positive integer) track point information, and each track point information includes the position, direction, speed, heading, curvature and acceleration of the track point.

[0221] The parking path generation module specifically includes the following submodules:

[0222] The first parking path planning submodule is used to plan and process the parking path between the parking starting point pose and the parking pre-end pose by using the Hybird A Star algorithm to form a first parking trajectory.

[0223] The second parking path planning submodule is used to plan and process the parking path between the parking pre-end pose and the parking end pose through a straight line to form a second parking trajectory.

[0224] The third submodule of parking path planning is used to form a parking path according to the first parking trajectory and the second parking trajectory.

[0225] Optionally, the parking path planning module further includes a parking path planning fourth submodule, which is used to optimize the parking path, and optimize the parking path according to a preset cost function of the parking path to form an optimized parking path.

[0226] The preset cost function is specifically expressed as follows:

[0227] Cost total =W l *Cost l +Wref *Cost ref +W smooth *Cost smooth +W slack *Cost slack ;

[0228] in, Cost total represents the preset cost function (i.e., the total cost function when optimizing the parking path), W l represents the weight of the optimized parking path length cost, W ref The weight representing the offset cost of the optimized parking path relative to the original path, W smooth represents the weight of the smoothness cost of the optimized parking path, W slack represents the weight of the slack variable, Cost l represents the optimized parking path length cost, Cost ref It represents the offset cost of the optimized parking path relative to the original parking path. Cost smooth represents the smoothness cost of the optimized parking path, Cost slack represents the slack variable cost.

[0229] Assume that the positions of N+1 original trajectory points in the world coordinate system are ,…, , the positions of the N+1 original trajectory points in the world coordinate system can be expressed as , , the optimization variable is , then the optimized parking path length cost Cost l , specifically expressed as follows:

[0230] ;

[0231] in, i is a positive integer; x i Represents the first i The horizontal coordinate of the trajectory point, y i Represents the first i The vertical coordinate of each trajectory point;

[0232] The cost of the optimized parking path offset from the original parking path Cost ref , specifically expressed as follows:

[0233] ;

[0234] in, x i-ref Represents the original i The horizontal coordinate of the trajectory point, y i-ref Represents the original i The vertical coordinates of the trajectory points; the smoothness cost of the parking path after optimization Cost smooth , specifically expressed as follows:

[0235] ;

[0236] During the optimization of the preset cost function, the constraint conditions include boundary constraints and curvature constraints;

[0237] The boundary constraint is specifically expressed as follows:

[0238] ;

[0239] in, lower bound-i Indicates the world coordinate system i The maximum lower boundary change value of the trajectory point, upper bound-i Indicates the world coordinate system i The maximum upper boundary change value of the trajectory points;

[0240] The curvature constraint is specifically expressed as follows:

[0241] ;

[0242] in, represents the average value of the sum of the Euclidean distances between two points of the original trajectory in the world coordinate system, k max represents the maximum permissible curvature;

[0243] In order to avoid the optimization problem being unsolvable, the curvature constraint is corrected. The corrected curvature constraint is specifically expressed as follows:

[0244] ;

[0245] in, Indicates i The slack variables of the trajectory points are ;

[0246] Slack variable cost Cost slack , specifically expressed as follows:

[0247] .

[0248] The parking regulation and control interaction module is used to transmit the parking path to the regulation and control interaction processor for processing, and complete the control of the parking path through the regulation and control interaction processor.

[0249] The regulation and control interaction processor includes a parking planning information processing module and a parking control module. The parking control module is connected to the parking planning information processing module. The parking planning information processing module is used to transmit parking trajectory information. The parking control module is used to issue vehicle control instructions and transmit feedback status to the parking planning information processing module, and includes the following submodules:

[0250] The first parking planning and control interaction submodule is used to, when a parking path is not formed, set the status of the parking planning information processing module to No Op and the status of the parking control module to Pending; when a parking path is formed, the parking planning information processing module stores the start pose, the pre end pose, the end pose and the parking path and transmits them to the parking control module, and the status of the parking control module is Execute.

[0251] The second parking control interaction submodule is used to divide the parking path into a plurality of parking trajectory segments through the parking planning information processing module; wherein the parking path is divided into a plurality of parking trajectory segments according to the position, speed and direction of the trajectory points in the parking path.

[0252] The third parking planning and control interaction submodule is used for the parking planning information processing module to obtain the current parking trajectory segment and transmit the current parking trajectory segment to the parking control module.

[0253] The fourth sub-module of parking control interaction is used for when the parking control module receives the current parking trajectory segment, the parking control module status is Active, the parking control module issues a control instruction and tracks the parking process according to the current parking trajectory segment, after completing the tracking of the current parking trajectory segment, the parking control module status is Success, after transmitting the status Success to the parking planning information processing module, the parking control module status is Pending.

[0254] The fifth parking control interaction submodule is used for switching the current parking trajectory segment and setting the next parking trajectory segment as the current parking trajectory segment after the parking planning information processing module receives the status Success.

[0255] The parking control interaction sixth submodule is used to repeatedly execute the parking control interaction second submodule, the parking control interaction third submodule and the parking control interaction fourth submodule until the parking control module completes tracking of all parking trajectory segments.

[0256] Optionally, a parking APA planning control device further includes a parking path re-planning module, which is used to re-execute the parking posture generation module and the parking path generation module to form a new parking path when the parking planning scenario changes.

[0257] Optionally, the parking regulation and control interaction module further includes a re-planning parking regulation and control interaction module, which is used to transmit the new parking path to the regulation and control interaction processor for processing when the parking planning scenario changes, and complete the control of the parking path through the regulation and control interaction processor, which specifically includes the following sub-modules:

[0258] The first submodule of the re-planning parking control interaction is used to cancel the execution action of the parking planning information processing module when the parking planning scenario changes, the status of the parking control module is Pending, and the status of the parking planning information processing module is Execute.

[0259] The second parking control interaction submodule is replanned to complete the control of the new parking path through the first parking control interaction submodule, the second parking control interaction submodule, the third parking control interaction submodule, the fourth parking control interaction submodule, the fifth parking control interaction submodule and the sixth parking control interaction submodule.

[0260] Optionally, the parking regulation and control interaction module further includes a pause and restart parking regulation and control interaction module, which is used to complete the control of the parking path through the regulation and control interaction processor when parking is resumed after a pause during the parking process, and specifically includes the following submodules:

[0261] The first submodule of the parking control interaction of pausing and restarting is used for when the parking process is paused and then continued, the state of the parking planning information processing module is Execute, and the state of the parking control module is Active.

[0262] The parking regulation and control interaction first submodule is paused and restarted to continue to complete the control of the parking path through the parking regulation and control interaction third submodule, the parking regulation and control interaction fourth submodule, the parking regulation and control interaction fifth submodule and the parking regulation and control interaction sixth submodule.

[0263] The specific embodiments of the present invention are described in detail above, but they are only examples, and the present invention is not limited to the specific embodiments described above. For those skilled in the art, any equivalent modifications and substitutions made to the present invention are also within the scope of the present invention. Therefore, the equalization changes and modifications made without departing from the spirit and scope of the present invention should be included in the scope of the present invention.

Claims

1. A parking APA planning and control method, characterized in that: The specific steps include: Step S1, obtaining external environment information required for parking, and generating a parking start pose, a parking pre-end pose, and a parking end pose in a world coordinate system according to the external environment information; Step S2, performing parking path planning processing according to the parking starting point pose, the parking pre-end pose and the parking end pose in the world coordinate system to form a parking path; wherein the parking path includes N+1 track point information; Step S3, transmitting the parking path to the regulation and control interaction processor for processing, and completing the control of the parking path through the regulation and control interaction processor; Wherein, N is a positive integer; In step S3, the regulation and control interaction processor includes a parking planning information processing module and a parking control module, the parking control module is connected to the parking planning information processing module, the parking planning information processing module is used to transmit parking trajectory information; the parking control module is used to issue a vehicle control instruction and transmit a feedback status to the parking planning information processing module; the regulation and control interaction processor includes a parking planning information processing module and a parking control module; the status of the parking planning information processing module includes No Op and Execute; the status of the parking control module includes Pending, Active and Success; Wherein, step S3 comprises: Step S301: when a parking path is not formed, the parking planning information processing module is set to No Op and the parking control module is set to Pending. When a parking path is formed, the parking planning information processing module stores the start pose, the pre end pose, the end pose and the parking path and transmits them to the parking control module, and the parking control module is set to Execute. In step S1, the external environment information required for parking is obtained, and the parking start pose, parking pre-end pose and parking end pose in the world coordinate system are generated according to the external environment information, which specifically includes the following steps: Step S101, obtaining the coordinate values ​​of the parking space type, parking space corner points, and obstacle corner points in the world coordinate system. If a limit block is provided in the parking space, the coordinate values ​​of the limit block corner points in the world coordinate system are also obtained, and a parking space coordinate system is constructed; Step S102: forming a parking end pose in the world coordinate system according to the parking space type, the parking space corner coordinate values ​​in the parking space coordinate system and the world coordinate system, and the obstacle corner coordinate values; if a limit block is provided in the parking space, forming a parking end pose in the world coordinate system according to the parking space type, the parking space corner coordinate values ​​in the parking space coordinate system and the world coordinate system, the obstacle corner coordinate values, and the limit block corner coordinate values; Step S103, processing obstacles inside and outside the region of interest and monitoring dynamic obstacles in real time to form a pre-end pose of parking in the world coordinate system; wherein the pre-end pose is within the parking space; Step S104: Acquire a positioning source used for parking, and determine a parking starting point pose in a world coordinate system.

2. A parking APA planning control method according to claim 1, characterized in that: In step S102, according to the parking space type, the parking space corner coordinate values ​​in the parking space coordinate system and the world coordinate system, the obstacle corner coordinate values ​​and the limit block corner coordinate values, the parking end pose in the world coordinate system is generated, which specifically includes the following steps: Step S102a, the parking space corner point coordinate values, obstacle corner point coordinate values ​​and limit block corner point coordinate values ​​in the world coordinate system are subjected to coordinate system conversion processing to form parking space corner point coordinate values, obstacle corner point coordinate values ​​and limit block corner point coordinate values ​​in the parking space coordinate system; Step S102b, removing obstacle points inside the parking space according to the parking space corner point coordinate values ​​and obstacle corner point coordinate values ​​in the parking space coordinate system; Step S102c, forming a parking pre-end position and a parking end position in the parking space coordinate system according to the parking space corner coordinate values, obstacle corner coordinate values, limit block corner coordinate values ​​and the parking space type in the parking space coordinate system; Step S102d: performing coordinate system conversion processing on the parking pre-end pose and the parking end pose in the parking space coordinate system to form the parking pre-end pose and the parking end pose in the world coordinate system.

3. A parking APA planning control method according to claim 2, characterized in that: In step S102c, the parking space types include vertical parking spaces, horizontal parking spaces and diagonal parking spaces; wherein the diagonal parking spaces include left herringbone diagonal parking spaces, left reverse herringbone diagonal parking spaces, right herringbone diagonal parking spaces and right reverse herringbone diagonal parking spaces.

4. The parking APA planning control method according to claim 1, characterized in that: In step S2, the information of each trajectory point includes the position, direction, speed, heading, curvature and acceleration of the trajectory point.

5. The parking APA planning control method according to claim 1, characterized in that: In step S2, a parking path planning process is performed according to the parking starting point pose, the parking pre-end pose and the parking end pose in the world coordinate system to form a parking path, which specifically includes the following steps: Step S201, planning and processing a parking path between the parking starting point pose and the parking pre-end point pose by using the Hybird A Star algorithm to form a first parking trajectory; Step S202, planning and processing a parking path between the parking pre-end pose and the parking end pose through a straight line to form a second parking trajectory; Step S203: forming a parking path according to the first parking trajectory and the second parking trajectory.

6. A parking APA planning control method according to claim 5, characterized in that: Step S2 further includes step S204, optimizing the parking path, optimizing the parking path according to a preset cost function of the parking path, and forming an optimized parking path.

7. The parking APA planning control method according to claim 6, characterized in that: In step S204, the preset cost function is specifically represented as follows: Cost total =W l *Cost l +W ref *Cost ref +W smooth *Cost smooth +W slack *Cost slack ； in, Cost total represents the preset cost function, W l represents the weight of the optimized parking path length cost, W ref The weight representing the offset cost of the optimized parking path relative to the original path, W smooth represents the weight of the smoothness cost of the optimized parking path, W slack represents the weight of the slack variable, Cost l represents the optimized parking path length cost, Cost ref It represents the offset cost of the optimized parking path relative to the original parking path. Cost smooth represents the smoothness cost of the optimized parking path, Cost slack represents the slack variable cost.

8. The parking APA planning control method according to claim 7, characterized in that: The positions of the N+1 original trajectory points are , then the optimized parking path length cost Cost l , specifically expressed as follows: ； in, i is a positive integer; x i Represents the first i The horizontal coordinate of the trajectory point, y i Represents the first i The vertical coordinate of each trajectory point; The cost of the optimized parking path offset from the original parking path Cost ref , specifically expressed as follows: ； in, x i-ref Represents the original i The horizontal coordinate of the trajectory point, y i-ref Represents the original i The vertical coordinates of the trajectory points; the smoothness cost of the parking path after optimization Cost smooth , specifically expressed as follows: ； During the optimization of the preset cost function, the constraint conditions include boundary constraints and curvature constraints; The boundary constraint is specifically expressed as follows: ; in, lower bound-i Indicates the world coordinate system i The maximum lower boundary change value of the trajectory point, upper bound-i Indicates the world coordinate system i The maximum upper boundary change value of the trajectory points; The curvature constraint is specifically expressed as follows: ; in, represents the average value of the sum of the Euclidean distances between two points of the original trajectory in the world coordinate system, k max represents the maximum permissible curvature; In order to avoid the optimization problem being unsolvable, the curvature constraint is corrected. The corrected curvature constraint is specifically expressed as follows: ； in, Indicates i The slack variables of the trajectory points are ; Slack variable cost Cost slack , specifically expressed as follows: 。 9. The parking APA planning control method according to claim 1, characterized in that: Step S3 also includes: Step S302: dividing the parking path into a plurality of parking trajectory segments by the parking planning information processing module; Step S303: the parking planning information processing module obtains the current parking trajectory segment, and transmits the current parking trajectory segment to the parking control module; Step S304: After the parking control module receives the current parking trajectory segment, the parking control module status is Active, the parking control module issues a control instruction and tracks the parking process according to the current parking trajectory segment. After the current parking trajectory segment is tracked, the parking control module status is Success. After the status Success is transmitted to the parking planning information processing module, the parking control module status is Pending. Step S305: After receiving the status Success, the parking planning information processing module switches the current parking trajectory segment and sets the next parking trajectory segment as the current parking trajectory segment; Step S306: Repeat steps S303 to S305 until the parking control module completes tracking of all parking trajectory segments.

10. The parking APA planning control method according to claim 9, characterized in that: In step S302, the parking path is divided into a plurality of parking trajectory segments according to the position, speed and direction of the trajectory points in the parking path.

11. The parking APA planning control method according to claim 1, characterized in that: The method further includes step S4, when the parking planning scenario changes, re-executing steps S1 to S2 to form a new parking path.

12. The parking APA planning control method according to claim 11, characterized in that: When the parking planning scenario changes, the new parking path is transmitted to the regulation and control interaction processor for processing, and the parking path is controlled by the regulation and control interaction processor, which specifically includes the following steps: Step S307: when the parking planning scenario changes, the parking planning information processing module execution action is canceled, the parking control module status is Pending, and the parking planning information processing module status is Execute; Step S308 , completing the control of the new parking path through steps S301 - S306 .

13. The parking APA planning control method according to claim 11, characterized in that: When the parking process is paused and then continued, the parking path is controlled by the regulation and control interaction processor, which specifically includes the following steps: Step S309: when the parking process is paused and then continued, the state of the parking planning information processing module is Execute, and the state of the parking control module is Active; Step S3010, continue to complete the control of the parking path through steps S303 to S306.

14. A parking APA planning control device, characterized in that: The modules include: A parking posture generation module is used to obtain external environment information required for parking, and generate a parking start pose, a parking pre-end pose, and a parking end pose according to the external environment information; A parking path generation module, configured to perform parking path planning processing according to the parking starting point pose, the parking pre-end pose and the parking end pose in the world coordinate system to form a parking path; wherein the parking path includes N+1 track point information; A parking regulation and control interaction module, used for transmitting the parking path to the regulation and control interaction processor for processing, and completing the control of the parking path through the regulation and control interaction processor; Wherein, N is a positive integer; Among them, the parking control interaction module includes: The first parking planning and control interaction submodule is used for, when a parking path is not formed, setting the parking planning information processing module status to No Op and the parking control module status to Pending; when a parking path is formed, the parking planning information processing module stores the start pose, the pre end pose, the end pose and the parking path and transmits them to the parking control module, and the parking control module status is Execute; Among them, the parking posture generation module specifically includes the following sub-modules: The first submodule of the parking posture generation is used to obtain the coordinate values ​​of the corner points of the parking space, the corner points of the obstacles and the corner points of the limit blocks in the world coordinate system. If there are limit blocks in the parking space, it is also necessary to obtain the coordinate values ​​of the corner points of the limit blocks in the world coordinate system and construct the parking space coordinate system; The second parking pose generation submodule is used to form a parking end pose in the world coordinate system according to the parking space type, the parking space corner coordinate values ​​in the parking space coordinate system and the world coordinate system, and the obstacle corner coordinate values. If a limit block is provided in the parking space, the parking end pose in the world coordinate system needs to be formed according to the parking space type, the parking space corner coordinate values ​​in the parking space coordinate system and the world coordinate system, the obstacle corner coordinate values ​​and the limit block corner coordinate values; The third submodule of parking posture generation is used to process obstacles inside and outside the ROI and monitor dynamic obstacles in real time to form a parking pre-end pose in the world coordinate system; wherein the pre-end pose is in the parking space; The fourth submodule of parking posture generation is used to obtain the positioning source used for parking and determine the parking starting point posture in the world coordinate system.

15. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, a parking APA planning control method as described in any one of claims 1 to 13 is implemented.

16. An electronic device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, a parking APA planning control method as described in any one of claims 1 to 13 is implemented.

Images