Planning method and device of parking path, storage medium and electronic device

Abstract

The invention provides a parking path planning method and device, a storage medium and an electronic device, and the method comprises the steps: determining the parking direction of a vehicle, and obtaining the environment perception information of the vehicle before parking; a path end point channel of the vehicle is calculated according to the parking-out direction and the environment perception information, and the path end point channel corresponds to the distribution range of path end points of a parking-out path of the vehicle; and planning a parking path meeting a collision constraint condition based on the path end point channel. According to the embodiment of the invention, the technical problem of low flexibility of planning the parking path of the vehicle in related technologies is solved, the flexibility of planning the parking path is improved, and the planning success rate of the parking path is also improved.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology, and in particular to a method and device for planning a parking path, a storage medium, and an electronic device. Background Art

[0002] In the related art, the rationality and success rate of the path planning of the automatic parking assist (APA) system are key factors affecting the product experience. Various path planning methods have been applied to automatic parking products. Typical parking path planning algorithms include two-segment, three-segment, and five-segment path planning based on arc and straight line combinations, geometric path planning based on roundabouts, path planning based on state lattices for spatial sampling, and path planning based on search. At present, in addition to traditional geometric planning, the search-based hybrid a star algorithm is widely used in automatic parking systems because of its great environmental adaptability and high success rate. However, most of the research and application of the planning algorithm for the parking system is limited to solving the automatic parking scene. The automatic parking process is simply processed as the reverse process of automatic parking, and the path planning method applied is almost the same as that of automatic parking, which reduces the success rate of the parking path planning and the flexibility of the path.

[0003] The entire automatic parking path planning process includes the following steps: the driver patrols the garage (step 1), the system models the environment based on sensor information while the vehicle is moving (step 2), the driver selects the target parking space after the vehicle stops (step 3), calculates the target point based on the target parking space (step 4), and finally plans the path (5). The parking exit path planning process is similar to the parking entry process, but the patrol process is missing. At the same time, step 3 corresponding to parking entry is for the driver to select the parking exit direction. In the planning step, a path from the starting point to the end point is sought. At present, the mainstream planning method for parking entry and parking exit is based on hybrid a star, which finds a path from a fixed starting point to a fixed end point. The general process of Hybrid a star is: expand multiple nodes from the starting point, determine whether the expanded nodes collide, and calculate the cost function for the non-colliding nodes; sort the expanded nodes according to the cost function in the new cycle, and select the node with the lowest cost as the current node; at the current node, determine whether it has reached the end point (attempt 1), or determine whether the parsable planner based on the improved planning speed can successfully plan directly from the current node to the end point (attempt 2). If attempt 1 is successful, output the search path; if attempt 2 is successful, output the concatenated path of the search path and the parsable planning path. If either attempt 1 or attempt 2 is satisfied, the planning can be terminated, indicating that the planning is successful. The commonly used parsable planners in the industry are basically RS planners. RS planners use a combination of different numbers of arcs and straight lines to connect a path from a fixed starting point to a fixed end point.

[0004] However, in fact, the automatic parking process is very different from the automatic parking process in many aspects. First, the perception range of the parking process is different. When the automatic parking is patrolling the warehouse, the driver will keep driving the vehicle forward along the warehouse. The sensors (lidar, radar, ultrasonic radar, camera) installed around the vehicle will scan a large area outside and inside the warehouse. The environmental modeling constructs a large number of obstacles around the target parking space. Based on this, when planning the path, the system will make more accurate processing of whether there will be a collision inside and outside the warehouse. However, for the parking process, the starting position of the vehicle is inside the warehouse. It has not undergone the dynamic environment modeling process. It is limited by the perception range of the sensor and the occupancy of the vehicle next to it. It cannot form a large perception range for the surrounding area of ​​the warehouse, especially the outside of the warehouse. In many cases, the vehicle’s perception of obstacles outside the warehouse is unknown. Therefore, it is unknown whether the calculated parking target point outside the warehouse will collide with the environment, and it is also unknown whether the path planned based on the current perception situation will collide with the environment after leaving the warehouse. Therefore, when the vehicle's automatic parking is activated, it is difficult to select a suitable target point, so it is impossible to ensure that the path planned at the beginning is appropriate. Secondly, the goals to be achieved for parking out and parking in are different. The goal of automatic parking in is for the automatic parking system to park the car accurately in the selected storage space. There are relatively high requirements for whether the parking space is on the line and the distance from the obstacles next to the storage space. In other words, once the storage space is selected, the final parking posture has been determined. However, parking out should not be like this. The purpose of parking out is to park the vehicle outside the storage space in the direction selected by the driver, and at the same time adjust the posture to be flush with the storage space opening as much as possible, and to be at a certain distance from the storage space opening. That is, after the direction of parking out is selected, there are only relatively vague requirements for the target posture, which is not as strict as parking in.

[0005] In related technologies, automatic parking planning solutions all require a definite target point (target posture), without considering the greater uncertainty of the external environment of the storage location and the ambiguity of the target posture, and are highly dependent on the preset parking end point (parking target posture) information. In order to achieve the purpose of non-collision of the path due to the uncertain environment and the definite target posture, the target point needs to be adjusted multiple times after the vehicle leaves the warehouse. The adjustment of the target point will cause the re-planned path to constantly adjust its shape and mode, resulting in poor consistency of the path, affecting the experience. In addition, the definite target posture will greatly sacrifice the flexibility and success rate of path planning.

[0006] With respect to the above-mentioned problems existing in the related technologies, no efficient and accurate solutions have been found yet. Summary of the invention

[0007] The present invention provides a method and device for planning a parking path, a storage medium, and an electronic device to solve technical problems in related technologies.

[0008] According to one embodiment of the present invention, a method for planning a parking path is provided, comprising: determining a parking direction of a vehicle, and acquiring environmental perception information of the vehicle before parking; calculating a path end channel of the vehicle according to the parking direction and the environmental perception information, wherein the path end channel corresponds to a distribution range of path end points of the parking path of the vehicle; and planning a parking path that satisfies collision constraints based on the path end channel.

[0009] Optionally, calculating the path endpoint channel of the vehicle according to the parking direction and the environmental perception information includes: calculating a reference target posture of the vehicle according to the parking direction and the environmental perception information, wherein the reference target posture is a reference posture of the vehicle after parking is completed; moving a posture frame where the reference target posture is located by a preset distance forward, backward, left and right, respectively, to obtain a rectangular envelope frame; and determining the rectangular envelope frame as the path endpoint channel.

[0010] Optionally, calculating the reference target posture of the vehicle according to the parking direction and the environmental perception information includes: determining an opening edge of the storage location of the vehicle according to the environmental perception information; determining a vector of the opening edge along the parking direction as a heading vector, and extending a preset distance in the horizontal and vertical directions with the opening edge as the starting position to obtain a parking position; and using the heading vector and the parking position to calculate the reference target posture of the vehicle.

[0011] Optionally, planning a parking path that satisfies collision constraints based on the path endpoint channel includes: initializing an open list and obtaining a source positioning point of the vehicle; configuring the source positioning point as an initial search point and adding it to the open list; iteratively executing the following steps until a parking path that satisfies collision constraints is successfully planned or the search termination condition is reached: selecting a current node with the lowest cost in the open list and moving the current node to a closed list; and determining whether the current node has successfully planned the parking path.

[0012] Optionally, determining whether the current node has successfully planned the parking path includes: determining whether the current node satisfies the endpoint channel constraint; if the current node satisfies the endpoint channel constraint, generating the parking path based on the current node; if the current node does not satisfy the endpoint channel constraint, determining whether a resolvable path can be planned using the current node, wherein the constraint condition of the resolvable path is that the path endpoint is within the path endpoint channel and the heading is parallel to the length direction of the path endpoint channel; if a resolvable path can be planned using the current node, generating the parking path based on the current node.

[0013] Optionally, determining whether the current node satisfies the endpoint channel constraint includes: determining whether the coordinates of the current node are within the path endpoint channel, and determining whether an angle between the heading of the current node and the length direction of the path endpoint channel is less than a set threshold; if the coordinates of the current node are within the path endpoint channel, and an angle between the heading of the current node and the length direction of the path endpoint channel is less than a set threshold, it is determined that the current node satisfies the endpoint channel constraint.

[0014] Optionally, determining whether a resolvable path can be planned using the current node includes: determining a node position and a first heading of the current node, determining a minimum turning radius of the vehicle, and determining a second heading of the end point of the path; inputting the node position, the first heading, the minimum turning radius, and the second heading into a first geometric planner to solve a first arc in a forward direction and a second arc in a backward direction; determining whether a collision constraint condition is satisfied in the first arc or the second arc; if the collision constraint condition is satisfied in the first arc or the second arc, determining that a resolvable path can be planned using the current node.

[0015] Optionally, determining whether a resolvable path can be planned using the current node includes: determining the node position and a first heading of the current node, determining the minimum turning radius of the vehicle, and determining the second heading of the end point of the path; inputting the node position, the first heading, the minimum turning radius, and the second heading into a second geometric planner to solve for a first spiral in a forward direction and a second spiral in a backward direction; determining whether a collision constraint condition is satisfied in the first spiral or the second spiral; if the collision constraint condition is satisfied in the first spiral or the second spiral, determining that a resolvable path can be planned using the current node.

[0016] Optionally, selecting the current node with the smallest cost in the open list includes: for each extended node in the open list, calculating the distance difference and heading difference between the extended node and the reference target posture; calculating the sum of the costs of the distance difference and the heading difference; and selecting the node in the open list with the smallest sum of costs as the current node.

[0017] Optionally, after determining whether a resolvable path can be planned using the current node, the method further includes: if a resolvable path cannot be planned using the current node, performing node expansion based on the current node to obtain multiple child nodes with the current node as the parent node and satisfying collision constraints; selecting a target child node with the lowest cost and not in the closed list from the multiple child nodes; adding the target child node to the open list; and updating the current node to the target child node.

[0018] According to another embodiment of the present invention, a parking path planning device is provided, comprising: a processing module, used to determine a parking direction of a vehicle, and to obtain environmental perception information of the vehicle before parking; a calculation module, used to calculate a path end channel of the vehicle according to the parking direction and the environmental perception information, wherein the path end channel corresponds to a distribution range of path end points of the parking path of the vehicle; and a planning module, used to plan a parking path that satisfies collision constraints based on the path end channel.

[0019] Optionally, the calculation module includes: a calculation unit, used to calculate a reference target posture of the vehicle according to the parking direction and the environmental perception information, wherein the reference target posture is a reference posture of the vehicle after parking is completed; a moving unit, used to move a posture frame where the reference target posture is located by a preset distance in the front, back, left and right directions respectively to obtain a rectangular envelope frame; and a determination unit, used to determine the rectangular envelope frame as the path endpoint channel.

[0020] Optionally, the calculation unit includes: a determination subunit, used to determine the opening edge of the storage location of the vehicle according to the environmental perception information; a processing subunit, used to determine the vector of the opening edge along the parking direction as a heading vector, and extend a preset distance in the horizontal and vertical directions with the opening edge as the starting position to obtain a parking position; a calculation subunit, used to use the heading vector and the parking position to calculate the reference target posture of the vehicle.

[0021] Optionally, the planning module includes: an initialization unit, used to initialize an open list and obtain a source positioning point of the vehicle; a configuration unit, used to configure the source positioning point as an initial search point and add it to the open list; an iteration unit, used to iteratively execute the following steps until a parking path that satisfies collision constraints is successfully planned or the search termination condition is reached: selecting a current node with the lowest cost in the open list and moving the current node out to a closed list; and determining whether the current node has successfully planned the parking path.

[0022] Optionally, the iteration unit includes: a first judgment subunit, used to judge whether the current node satisfies the endpoint channel constraint; a second judgment subunit, used to generate the parking path based on the current node if the current node satisfies the endpoint channel constraint; if the current node does not satisfy the endpoint channel constraint, judge whether a resolvable path can be planned using the current node, wherein the constraint condition of the resolvable path is that the path endpoint is within the path endpoint channel and the heading is parallel to the length direction of the path endpoint channel; a generation subunit, used to generate the parking path based on the current node if a resolvable path can be planned using the current node.

[0023] Optionally, the first judgment subunit is also used to: determine whether the coordinates of the current node are within the path endpoint channel, and determine whether the angle between the heading of the current node and the length direction of the path endpoint channel is less than a set threshold; if the coordinates of the current node are within the path endpoint channel, and the angle between the heading of the current node and the length direction of the path endpoint channel is less than a set threshold, it is determined that the current node satisfies the endpoint channel constraint.

[0024] Optionally, the second judgment subunit is also used to: determine the node position and first heading of the current node, determine the minimum turning radius of the vehicle, and determine the second heading of the path end point; input the node position, the first heading, the minimum turning radius, and the second heading into the first geometric planner to solve the first arc in the forward direction and the second arc in the backward direction; determine whether the collision constraint condition is met in the first arc or the second arc; if the collision constraint condition is met in the first arc or the second arc, determine that a resolvable path can be planned using the current node.

[0025] Optionally, the second judgment subunit is also used for: a first determination subunit, used to determine the node position and first heading of the current node, determine the minimum turning radius of the vehicle, and determine the second heading of the path end point; input the node position, the first heading, the minimum turning radius, and the second heading into a second geometric planner to solve for a first spiral in the forward direction and a second spiral in the backward direction; determine whether the first spiral or the second spiral satisfies a collision constraint; if the first spiral or the second spiral satisfies a collision constraint, determine that a resolvable path can be planned using the current node.

[0026] Optionally, the iteration unit includes: a first calculation subunit, used to calculate, for each expansion node in the open list, the distance difference and the heading difference between the expansion node and the reference target posture; a second calculation subunit, used to calculate the sum of the costs of the distance difference and the heading difference; and a selection subunit, used to select the node with the smallest sum of costs in the open list as the current node.

[0027] Optionally, the iteration unit also includes: an expansion sub-unit, which is used to expand the node based on the current node after the second judgment sub-unit determines whether a resolvable path can be planned using the current node. If a resolvable path cannot be planned using the current node, a plurality of child nodes with the current node as the parent node and satisfying collision constraints are obtained; a selection sub-unit, which is used to select a target child node with the smallest cost and which is not in the closed list from the plurality of child nodes; an addition sub-unit, which is used to add the target child node to the open list; and an update sub-unit, which is used to update the current node to the target child node.

[0028] According to another aspect of an embodiment of the present application, a storage medium is further provided, which includes a stored program, and the above steps are executed when the program is run.

[0029] According to another aspect of an embodiment of the present application, there is also provided an electronic device, including a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; wherein: the memory is used to store computer programs; and the processor is used to execute the steps in the above method by running the program stored in the memory.

[0030] The embodiment of the present application also provides a computer program product including instructions, which, when executed on a computer, enables the computer to execute the steps in the above method.

[0031] Beneficial effects of the present invention:

[0032] 1. By utilizing the large-scale constraints of the path endpoint channel, the more complex considerations for dynamic adjustment of the parking endpoint in the design phase are eliminated, which improves the flexibility of path planning and reduces the difficulty of engineering;

[0033] 2. Improved the solution space of planning, thereby improving the planning success rate of the parking path;

[0034] 3. The end point of the path is not required to coincide with the set reference target position, which is more in line with the fuzzy characteristics of the human driver's parking process of "just driving the car out of the parking space flat", and is more humanized;

[0035] 4. The improved geometry planner reduces the number of path shifts, makes the path simpler, and improves the experience;

[0036] 5. Thanks to the ambiguity of the parking target, the sensitivity to environmental changes is reduced, and minor environmental changes can avoid major adjustments in path shape and gear shifting times. The replanned path after the vehicle leaves the parking lot is more consistent with the original path. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0038] Figure 1 is a hardware structure block diagram of a car according to an embodiment of the present invention;

[0039] Figure 2 is a flow chart of a method for planning a parking path according to an embodiment of the present invention;

[0040] Figure 3 is a system architecture diagram of an embodiment of the present invention;

[0041] Figure 4 is a schematic diagram of a vehicle parking out according to an embodiment of the present invention;

[0042] Figure 5 is a schematic diagram of parking path planning in an embodiment of the present invention;

[0043] Figure 6 This is a flow chart of the expansion of the outbound node in an embodiment of the present invention;

[0044] Figure 7 is a flow chart of automatic parking path planning in an embodiment of the present invention;

[0045] Figure 8 It is a structural block diagram of a device for planning a parking path according to an embodiment of the present invention. DETAILED DESCRIPTION

[0046] In order to enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only embodiments of a part of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work should fall within the scope of protection of the present application. It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0047] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application 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 interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, product or device comprising 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.

[0048] Example 1

[0049] The method embodiment provided in the first embodiment of the present application can be executed in a car, a server, a processor, a controller or a similar processing device. Taking running on a car as an example, Figure 1 1 is a hardware structure diagram of a car according to an embodiment of the present invention. Figure 1 As shown, a car may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data. Optionally, the above-mentioned automobile may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is for illustration only and does not limit the structure of the above-mentioned automobile. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations shown.

[0050] The memory 104 can be used to store automobile programs, for example, software programs and modules of application software, such as an automobile program corresponding to a method for planning a parking path for an automobile in an embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the automobile program stored in the memory 104, that is, to implement the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the automobile via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0051] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the car. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0052] In this embodiment, a method for planning a parking path is provided. Figure 2 is a flow chart of a method for planning a parking path according to an embodiment of the present invention. Figure 2 As shown, the process includes the following steps:

[0053] Step S202, determining the parking direction of the vehicle, and obtaining the environment perception information of the vehicle before parking;

[0054] The environmental perception information of this embodiment is information perceived by sensors such as cameras and radars of the vehicle, such as information about the parking space where the vehicle is located.

[0055] Step S204, calculating a path end channel of the vehicle according to the parking direction and the environment perception information, wherein the path end channel corresponds to a distribution range of a path end of the parking path of the vehicle;

[0056] The path end point of this embodiment is not a fixed point, but can be flexibly selected within the path end point channel. As long as the end point of the currently searched path falls within the path end point channel and meets the preset constraints, the end point of the currently searched path can be used as the final path end point of the parking path.

[0057] Step S206, planning a parking path that meets the collision constraint condition based on the path endpoint channel.

[0058] Through the above steps, the parking direction of the vehicle is determined, and the environmental perception information of the vehicle before parking is obtained; the path end channel of the vehicle is calculated according to the parking direction and the environmental perception information, wherein the path end channel corresponds to the distribution range of the path end of the parking path of the vehicle; based on the path end channel, a parking path that meets the collision constraint condition is planned, and the large-range constraint of the path end channel is utilized to avoid the limitation of a fixed end point, thereby solving the technical problem of low flexibility in planning the parking path of the vehicle in the related technology, improving the flexibility of parking path planning, and also improving the success rate of parking path planning.

[0059] Figure 3The system architecture diagram of an embodiment of the present invention includes a perception system (1), a positioning system (2), a vehicle computer (3), and a path planning (4). The perception system (1) uses ultrasonic, millimeter wave, camera, laser radar and other sensors to obtain obstacle information and environmental information such as the parking space where the vehicle is located; the positioning system (2) is used to provide information such as the vehicle position and posture in a local coordinate system or a global coordinate system, and to perform conversion between local and global coordinate systems; the vehicle computer (3) is used to obtain instruction information such as the parking direction selected by the driver. The path planning (4) plans a collision-free and smooth parking path based on the environmental information output by (1) and the parking direction instruction of (3).

[0060] In one implementation of the present embodiment, calculating the path endpoint channel of the vehicle according to the parking direction and the environmental perception information includes: calculating a reference target posture of the vehicle according to the parking direction and the environmental perception information, wherein the reference target posture is a reference posture of the vehicle after parking is completed; moving a posture frame where the reference target posture is located by a preset distance forward, backward, left and right respectively to obtain a rectangular envelope frame; and determining the rectangular envelope frame as the path endpoint channel.

[0061] The reference target posture of this embodiment is the best posture / ideal posture for the vehicle to be parked.

[0062] According to the reference target posture, a rectangle parallel to the reference target posture is set as the path end channel. Specifically, the reference target posture is moved along the preset distance forward, backward, left and right to form a rectangular envelope. The moving distance takes into account the perception range and the distribution range of the parking posture of the human driver during the parking process, and can be directly set as an empirical value.

[0063] In one example, calculating the reference target posture of the vehicle according to the parking direction and the environmental perception information includes: determining the opening edge of the parking location of the vehicle according to the environmental perception information; determining the vector of the opening edge along the parking direction as the heading vector, and extending a preset distance in the horizontal and vertical directions with the opening edge as the starting position to obtain the parking position; and using the heading vector and the parking position to calculate the reference target posture of the vehicle.

[0064] Figure 4 Schematic diagram of vehicle parking in an embodiment of the present invention, including: other vehicles 1, the vehicle 2, the parking space 3, other parking spaces 4, obstacles 5, walls 6, target parking direction 7, reference target posture 8, and path end channel 9. The reference target posture is calculated based on the parking space information and the parking target direction. The reference target posture has the following characteristics: it does not need to consider whether it will collide with the environment (feature one); it will not be used as the end point of the path (feature two); it will be used as an anchor point in the search process (feature three).

[0065] By utilizing feature 1, the target point calculation does not need to introduce too complex constraints, and the calculation rules can be greatly simplified. It only needs to consider the ideal situation, where the end point of the vehicle's path is located relative to the storage location. Specific implementation method: In terms of direction, consider that the heading vector of the target posture 8 is consistent with the vector of the opening edge of the storage location 3 along the parking direction 7. For example, if the parking direction is to the right, the direction of the target posture 8 is parallel to the opening edge of the storage location 3, and the front of the vehicle is to the right. In terms of coordinates, considering the horizontal and vertical distances of the human driver relative to the opening edge of the storage location when driving out of the storage location in most cases, these two distances can be directly set as an empirical value.

[0066] By using feature 2, the calculated reference target pose is not used as the pose of the last point of the path planning result. This makes the target pose considered only as a reference for the path rather than an absolute constraint on the path. This feature greatly improves the diversity and success rate of the path.

[0067] Using feature 3, the calculated reference target pose, under the premise of satisfying feature 2, is only used as a reference value of an heuristic cost in the path search process (such as hybrid a star). The cost specifically calculates the distance difference and heading difference between the extended node of the search process and the considered target pose. The smaller the two differences, the smaller the cost, and the closer the node pose is to the considered target pose. This makes the path guide the search to get as close to the target pose as possible, but does not take the considered target pose as the final pose. This ensures that it is as close to the preset considered target pose as possible.

[0068] In this embodiment, planning a parking path that meets the collision constraint condition based on the path endpoint channel includes: initializing an open list and obtaining the source positioning point of the vehicle; configuring the source positioning point as an initial search point and adding it to the open list; iteratively executing the following steps until a parking path that meets the collision constraint condition is successfully planned or the search termination condition is reached: selecting a current node with the lowest cost in the open list and moving the current node out to the closed list; and determining whether the current node successfully plans the parking path.

[0069] According to the positioning information and environmental information of the vehicle and the calculated target posture information, starting from the starting point in the parking space, the nodes are expanded outside the parking space. In the process of expanding the nodes, if the collision detection of the nodes passes, when calculating the cost of the current node, in addition to considering the heuristic cost, the distance cost of the obstacle, the path gear switching cost, the path smoothing cost, etc. can also be considered.

[0070] In one implementation of the present embodiment, determining whether the current node has successfully planned the parking path includes: determining whether the current node satisfies the endpoint channel constraint; if the current node satisfies the endpoint channel constraint, generating the parking path based on the current node; if the current node does not satisfy the endpoint channel constraint, determining whether a resolvable path can be planned using the current node, wherein the constraint condition of the resolvable path is that the path endpoint is within the path endpoint channel and the heading is parallel to the length direction of the path endpoint channel; if a resolvable path can be planned using the current node, generating the parking path based on the current node.

[0071] In one example, determining whether the current node satisfies the endpoint channel constraint includes: determining whether the coordinates of the current node are within the path endpoint channel, and determining whether the angle between the heading of the current node and the length direction of the path endpoint channel is less than a set threshold; if the coordinates of the current node are within the path endpoint channel, and the angle between the heading of the current node and the length direction of the path endpoint channel is less than a set threshold, it is determined that the current node satisfies the endpoint channel constraint.

[0072] During the path search process, as long as the current node is inside the path end channel and the heading difference between the heading and the target posture is less than a certain threshold, the planning can be terminated and the complete path can be obtained by backtracking the current node. This greatly improves the success rate and speed of planning and makes the path more flexible. As long as the cost function is set reasonably, the end point setting of a large area reduces the difference between the re-planned path and the original path after the path is disturbed by obstacles after leaving the warehouse, that is, the path consistency will be higher.

[0073] When planning a resolvable path, according to the set path end channel, the segment arc geometric planner 1 and the spiral geometric planner 2 can be used for planning. These two geometric planners set the end point constraint as the path end channel constraint, that is, the coordinates of the end point are within the rectangular frame of the end point channel, and the heading is parallel to the length direction of the rectangular frame.

[0074] In one embodiment, determining whether a resolvable path can be planned using the current node includes: determining a node position and a first heading of the current node, determining a minimum turning radius of the vehicle, and determining a second heading of the end point of the path; inputting the node position, the first heading, the minimum turning radius, and the second heading into a first geometric planner to solve a first arc in a forward direction and a second arc in a backward direction; determining whether a collision constraint condition is satisfied in the first arc or the second arc; if the collision constraint condition is satisfied in the first arc or the second arc, determining that a resolvable path can be planned using the current node.

[0075] In this embodiment, a collision-free arc path is planned from the current node so that the end point of the path is within the path end channel, satisfying the constraints of the resolvable path. In the solution process: given the position and heading of the current node coordinates, p0 (x0, y0, headng0), given the minimum turning radius r of the vehicle, given the heading 1 of the end point, solve the arc that meets the above conditions. You can try to solve two arcs, corresponding to the forward and backward directions of the vehicle respectively. For any arc that will not collide, if its end point is within the rectangular box of the end channel, then the path planning is successful.

[0076] In another embodiment, determining whether a resolvable path can be planned using the current node includes: determining a node position and a first heading of the current node, determining a minimum turning radius of the vehicle, and determining a second heading of the end point of the path; inputting the node position, the first heading, the minimum turning radius, and the second heading into a second geometric planner to solve for a first spiral in a forward direction and a second spiral in a backward direction; determining whether a collision constraint condition is satisfied in the first spiral or the second spiral; if the collision constraint condition is satisfied in the first spiral or the second spiral, determining that a resolvable path can be planned using the current node.

[0077] During the solving process of the second geometric planner and the first geometric planner, the first geometric planner obtains the arc based on the fixed minimum turning radius, and the second geometric planner obtains the clothoid based on the variable turning radius (the turning radius starts from the minimum turning radius and gradually increases).

[0078] In this embodiment, a clothoid is used as a geometric shape. The properties of the clothoid itself make it possible to calculate the parameters related to the clothoid based on a fixed starting point p0 (x0, y0, headng0) and a given vehicle's minimum turning radius r, and a given end point's heading 1. Then, numerical calculations are performed based on the clothoid parameters and the information of the starting and end points to obtain the discretized points on the clothoid. Similarly, you can try to solve two clothoids, corresponding to the forward and backward directions of the vehicle, respectively. For any clothoid that will not collide, if its end point is within the rectangular box of the path end channel, then the path planning is successful.

[0079] If either the first geometric planner or the second geometric planner succeeds, a resolvable path can be directly returned and joined with the search path obtained by backtracking from the current node to obtain a complete path. Figure 5It is a schematic diagram of the parking path planning in an embodiment of the present invention, the currently searched expansion node 1 (current node), the searched backtracking path 2, the resolvable path 3 / 3', the path end channel 4, the reference target posture 5, the resolvable planner on the left is the first geometric planner, and the resolvable planner on the right is the second geometric planner. After expanding a certain number of nodes, both successfully plan a complete path that is collision-free based on the current perception and satisfies the end channel constraint. The end points of both paths do not coincide with the target posture, achieving flexibility and rapidity in planning. For scenarios where gear shifting is required for parking in narrow spaces, it is not necessary to rely on the RS planner to achieve gear shifting. The necessary gear shifting will be achieved during the search phase. There is an obstacle on one side of the external channel of the scene storage location, and a search backtracking path with necessary gear shifting is successfully planned at the storage location entrance.

[0080] Optionally, selecting the current node with the smallest cost in the open list includes: for each extended node in the open list, calculating the distance difference and heading difference between the extended node and the reference target posture; calculating the sum of the costs of the distance difference and the heading difference; and selecting the node in the open list with the smallest sum of costs as the current node.

[0081] In addition to the heuristic cost, the distance cost from the vehicle to the obstacle, the path gear switching cost, and the path smoothing cost can also be considered at the same time, and the total cost is weighted and calculated.

[0082] In the iterative process of this embodiment, after determining whether a resolvable path can be planned using the current node, it also includes: if a resolvable path cannot be planned using the current node, performing node expansion based on the current node to obtain multiple child nodes with the current node as the parent node and satisfying collision constraints; selecting a target child node with the lowest cost and not in the closed list from the multiple child nodes; adding the target child node to the open list; and updating the current node to the target child node.

[0083] Figure 6 It is a flowchart of node expansion in an embodiment of the present invention, including: traversing all nodes expanded based on the current node; judging whether the new child node is available. This step judges whether the path to the expanded child node collides, and if so, judges whether the new child node is in the close list. If so, it indicates that the child node has been visited and abandoned, otherwise, calculate the cost, heuristic cost and total cost of the new child node, and judge whether the new child node is in the open list, otherwise put it in the open list. Put the new child node into the open list, and reorder the open list according to the cost, judge whether the child node cost is smaller than the previous cost, select the target child node with the smallest cost, reconnect the child node, and update the child node to the current node.

[0084] The solution of this embodiment can be applied to the path planning of automatic parking. The present invention improves the existing hybridastar parking path planning, so that it can be applied to the path planning of automatic parking. This embodiment calculates a reference target posture outside the storage location based on the environmental perception information obtained by the sensors installed in the vehicle inside the storage location and the parking direction selected by the driver, generates a rectangular path end channel based on the target posture, and uses a search and combined analyzable geometric planner that satisfies the path end channel constraints to plan an initial path that is as simple as possible that satisfies the collision constraints; after the vehicle drives out of the storage location, the original path may collide with the update of the perception information. When a possible collision is detected, the vehicle stops and plans a new path.

[0085] Figure 7 is a flow chart of automatic parking path planning in an embodiment of the present invention, including:

[0086] Step 1: Get the parking direction. When the driver clicks to enter the automatic parking system, if the system detects that the vehicle is in the parking space, it will automatically jump to the automatic parking function interface and wait for the user to select the parking direction. When the parking direction is selected, it goes to step 2.

[0087] Step 2: Calculate the parking target posture. Refer to the characteristics of human drivers and calculate the best recommended reference target posture. Go to step 3.

[0088] Step 3: Calculate the path end channel. As mentioned above, based on the target pose, calculate a rectangular area where the end point of the constraint path is located as the end channel. Go to step 4.

[0089] Step 4: Initialize the open list. Put the starting point of the path planning, that is, the current positioning point, as the first search node into the open list. Go to step 5.

[0090] Step 5: Start the loop and enter condition 1.

[0091] Condition 1: Determine whether the segment has reached the termination condition. The termination condition includes whether the search has timed out and whether the open list is empty (i.e. whether the current space can no longer expand nodes). If not, go to step 6; otherwise, go to step 9 to plan the failure state.

[0092] Step 6: Select the node with the lowest cost from the open list as the new current node and put it into the visited node close node. Go to step 7.

[0093] Condition 2: Determine whether the segment meets the terminal channel constraint. Determine whether the current node of the segment meets the terminal channel constraint. The specific constraint is that the current node coordinates are within the rectangular frame of the terminal channel, and the angle between the heading of the current node and the length direction of the rectangular frame is less than a set threshold. If yes, enter step 8 to plan successfully; otherwise, enter condition 3.

[0094] Enter condition three: determine whether a resolvable path can be planned. As mentioned above, use a specific geometric planner to try to plan a one-way path from the current node to the end point. If the planning is successful, go to step 8; otherwise, go to step 7.

[0095] Step 7: Node expansion. Take the current node as the parent node and expand multiple child nodes without collision. Go to step 5.

[0096] Step 8: Planning is successful. If the path planning is successful, a complete path is generated. Go to step 10.

[0097] Step 9: Planning failed. Report planning failure information to the system. Go to step 10.

[0098] Step 10: End the loop.

[0099] The solution of this embodiment uses the large-scale constraint of the end point, which eliminates the more complicated considerations for the dynamic adjustment of the parking end point in the design stage, improves the flexibility of path planning and reduces the engineering difficulty; the solution space of planning is improved, thereby improving the planning success rate; there is no restriction that the path end point must coincide with the set target, which is more in line with the fuzzy characteristics of the human driver's parking process of "just driving the car out of the parking space flat", and is more humanized; compared with the multiple gear shifts that may exist in the RS planner, the use of the improved unidirectional resolvable planner reduces the number of path shifts, makes the path simpler, and improves the experience; the RS path is very sensitive to environmental changes, and even slight changes may bring about large adjustments to the path shape and the number of gear shifts. Thanks to the fuzziness of the parking target, the re-planned path after leaving the warehouse is more consistent with the original path.

[0100] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0101] Example 2

[0102] In this embodiment, a device for planning a parking path is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware of a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0103] Figure 8 is a structural block diagram of a parking path planning device according to an embodiment of the present invention, such as Figure 8 As shown, the device comprises:

[0104] The processing module 80 is used to determine the parking direction of the vehicle and obtain the environmental perception information of the vehicle before parking;

[0105] A calculation module 82, configured to calculate a path end channel of the vehicle according to the parking direction and the environment perception information, wherein the path end channel corresponds to a distribution range of a path end of the parking path of the vehicle;

[0106] The planning module 84 is used to plan a parking path that meets the collision constraint condition based on the path end channel.

[0107] Optionally, the calculation module includes: a calculation unit, used to calculate a reference target posture of the vehicle according to the parking direction and the environmental perception information, wherein the reference target posture is a reference posture of the vehicle after parking is completed; a moving unit, used to move a posture frame where the reference target posture is located by a preset distance in the front, back, left and right directions respectively to obtain a rectangular envelope frame; and a determination unit, used to determine the rectangular envelope frame as the path endpoint channel.

[0108] Optionally, the calculation unit includes: a determination subunit, used to determine the opening edge of the storage location of the vehicle according to the environmental perception information; a processing subunit, used to determine the vector of the opening edge along the parking direction as a heading vector, and extend a preset distance in the horizontal and vertical directions with the opening edge as the starting position to obtain a parking position; a calculation subunit, used to use the heading vector and the parking position to calculate the reference target posture of the vehicle.

[0109] Optionally, the planning module includes: an initialization unit, used to initialize an open list and obtain a source positioning point of the vehicle; a configuration unit, used to configure the source positioning point as an initial search point and add it to the open list; an iteration unit, used to iteratively execute the following steps until a parking path that satisfies collision constraints is successfully planned or the search termination condition is reached: selecting a current node with the lowest cost in the open list and moving the current node out to a closed list; and determining whether the current node has successfully planned the parking path.

[0110] Optionally, the iteration unit includes: a first judgment subunit, used to judge whether the current node satisfies the endpoint channel constraint; a second judgment subunit, used to generate the parking path based on the current node if the current node satisfies the endpoint channel constraint; if the current node does not satisfy the endpoint channel constraint, judge whether a resolvable path can be planned using the current node, wherein the constraint condition of the resolvable path is that the path endpoint is within the path endpoint channel and the heading is parallel to the length direction of the path endpoint channel; a generation subunit, used to generate the parking path based on the current node if a resolvable path can be planned using the current node.

[0111] Optionally, the first judgment subunit is also used to: determine whether the coordinates of the current node are within the path endpoint channel, and determine whether the angle between the heading of the current node and the length direction of the path endpoint channel is less than a set threshold; if the coordinates of the current node are within the path endpoint channel, and the angle between the heading of the current node and the length direction of the path endpoint channel is less than a set threshold, it is determined that the current node satisfies the endpoint channel constraint.

[0112] Optionally, the second judgment subunit is also used to: determine the node position and first heading of the current node, determine the minimum turning radius of the vehicle, and determine the second heading of the path end point; input the node position, the first heading, the minimum turning radius, and the second heading into the first geometric planner to solve the first arc in the forward direction and the second arc in the backward direction; determine whether the collision constraint condition is met in the first arc or the second arc; if the collision constraint condition is met in the first arc or the second arc, determine that a resolvable path can be planned using the current node.

[0113] Optionally, the second judgment subunit is also used for: a first determination subunit, used to determine the node position and first heading of the current node, determine the minimum turning radius of the vehicle, and determine the second heading of the path end point; input the node position, the first heading, the minimum turning radius, and the second heading into a second geometric planner to solve for a first spiral in the forward direction and a second spiral in the backward direction; determine whether the first spiral or the second spiral satisfies a collision constraint; if the first spiral or the second spiral satisfies a collision constraint, determine that a resolvable path can be planned using the current node.

[0114] Optionally, the iteration unit includes: a first calculation subunit, used to calculate, for each expansion node in the open list, the distance difference and the heading difference between the expansion node and the reference target posture; a second calculation subunit, used to calculate the sum of the costs of the distance difference and the heading difference; and a selection subunit, used to select the node with the smallest sum of costs in the open list as the current node.

[0115] Optionally, the iteration unit also includes: an expansion sub-unit, which is used to expand the node based on the current node after the second judgment sub-unit determines whether a resolvable path can be planned using the current node. If a resolvable path cannot be planned using the current node, a plurality of child nodes with the current node as the parent node and satisfying collision constraints are obtained; a selection sub-unit, which is used to select a target child node with the smallest cost and which is not in the closed list from the plurality of child nodes; an addition sub-unit, which is used to add the target child node to the open list; and an update sub-unit, which is used to update the current node to the target child node.

[0116] It should be noted that the above modules can be implemented by software or hardware. For the latter, it can be implemented in the following ways, but not limited to: the above modules are all located in the same processor; or the above modules are located in different processors in any combination.

[0117] Example 3

[0118] An embodiment of the present invention further provides a storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when running.

[0119] Optionally, in this embodiment, the storage medium may be configured to store a computer program for performing the following steps:

[0120] S1, determining a parking direction of the vehicle, and obtaining environmental perception information of the vehicle before parking;

[0121] S2, calculating a path end channel of the vehicle according to the parking direction and the environment perception information, wherein the path end channel corresponds to a distribution range of a path end of the parking path of the vehicle;

[0122] S3, planning a parking path that meets collision constraints based on the path endpoint channel.

[0123] Optionally, in this embodiment, the above-mentioned storage medium may include but is not limited to: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store computer programs.

[0124] An embodiment of the present invention further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0125] Optionally, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0126] Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:

[0127] S1, determining a parking direction of the vehicle, and obtaining environmental perception information of the vehicle before parking;

[0128] S2, calculating a path end channel of the vehicle according to the parking direction and the environment perception information, wherein the path end channel corresponds to a distribution range of a path end of the parking path of the vehicle;

[0129] S3, planning a parking path that meets collision constraints based on the path endpoint channel.

[0130] Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments and optional implementation modes, and this embodiment will not be described in detail here.

[0131] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0132] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a general hardware platform, and of course, by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the relevant technology can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0133] It should be understood that the terms used herein are only for the purpose of describing specific example embodiments and are not intended to be limiting. Unless the context clearly indicates otherwise, the singular forms "one", "an" and "said" as used herein may also be meant to include plural forms. The terms "include", "comprise", "contain", and "have" are inclusive, and therefore specify the existence of stated features, steps, operations, elements and / or parts, but do not exclude the existence or addition of one or more other features, steps, operations, elements, parts, and / or combinations thereof. The method steps, processes, and operations described herein are not interpreted as necessarily requiring them to be performed in the specific order described or illustrated, unless the execution order is clearly indicated. It should also be understood that additional or alternative steps may be used.

[0134] The foregoing is merely a specific embodiment of the present invention, which enables those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A method for planning a parking path, characterized in that: include: Determining a parking direction of the vehicle, and obtaining environmental perception information of the vehicle before parking; Calculating a path end channel of the vehicle according to the parking direction and the environment perception information, wherein the path end channel corresponds to a distribution range of a path end of the parking path of the vehicle; A parking path that satisfies collision constraints is planned based on the path endpoint channel.

2. The method according to claim 1, characterized in that: Calculating the path end channel of the vehicle according to the parking direction and the environmental perception information includes: Calculating a reference target posture of the vehicle according to the parking direction and the environmental perception information, wherein the reference target posture is a reference posture of the vehicle after the parking is completed; The pose frame where the reference target pose is located is moved forward, backward, left, and right by a preset distance to obtain a rectangular envelope frame; The rectangular envelope is determined as the path end channel.

3. The method according to claim 2, characterized in that Calculating the reference target pose of the vehicle according to the parking direction and the environmental perception information includes: Determine an opening edge of the storage location of the vehicle according to the environmental perception information; Determine the vector of the opening edge along the parking direction as the heading vector, and extend the preset distance in the horizontal and vertical directions with the opening edge as the starting position to obtain the parking position; The heading vector and the unparking position are used to calculate a reference target pose of the vehicle.

4. The method according to claim 1, characterized in that: Planning a parking path that satisfies the collision constraint condition based on the path endpoint channel includes: Initialize an open list and obtain the source positioning point of the vehicle; configuring the source positioning point as an initial search point and adding it to the open list; The following steps are iteratively performed until a parking path that satisfies the collision constraint condition is successfully planned or the search termination condition is reached: a current node with the lowest cost is selected from the open list, and the current node is moved out to the closed list; and it is determined whether the parking path is successfully planned for the current node.

5. The method according to claim 4, characterized in that Determining whether the current node successfully plans the exit path includes: Determine whether the current node satisfies the end channel constraint; If the current node satisfies the terminal channel constraint, the outbound path is generated based on the current node; if the current node does not satisfy the terminal channel constraint, it is determined whether a resolvable path can be planned using the current node, wherein the constraint condition of the resolvable path is that the path terminal is within the path terminal channel and the heading is parallel to the length direction of the path terminal channel; If a resolvable path can be planned using the current node, the ported path is generated based on the current node.

6. The method according to claim 5, characterized in that Determining whether the current node satisfies the end channel constraint includes: Determine whether the coordinates of the current node are within the path end channel, and determine whether the angle between the heading of the current node and the length direction of the path end channel is less than a set threshold; If the coordinates of the current node are within the path endpoint channel, and the angle between the heading of the current node and the length direction of the path endpoint channel is less than a set threshold, it is determined that the current node satisfies the endpoint channel constraint.

7. The method according to claim 5, characterized in that Determining whether a resolvable path can be planned using the current node includes: Determining a node position and a first heading of the current node, determining a minimum turning radius of the vehicle, and determining a second heading of a path end point; Inputting the node position, the first heading, the minimum turning radius, and the second heading into a first geometric planner to solve and obtain a first arc in a forward direction and a second arc in a backward direction; Determining whether the first arc or the second arc satisfies a collision constraint condition; If the collision constraint condition is satisfied in the first arc or the second arc, it is determined that a resolvable path can be planned using the current node.

8. The method according to claim 5, characterized in that Determining whether a resolvable path can be planned using the current node includes: Determining a node position and a first heading of the current node, determining a minimum turning radius of the vehicle, and determining a second heading of a path end point; Inputting the node position, the first heading, the minimum turning radius, and the second heading into a second geometric planner to solve for obtaining a first clothoid in a forward direction and a second clothoid in a backward direction; Determining whether a collision constraint condition is satisfied in the first clothoid or the second clothoid; If the collision constraint condition is satisfied in the first clothoid or the second clothoid, it is determined that a resolvable path can be planned using the current node.

9. The method according to claim 4, characterized in that Selecting the current node with the minimum cost in the open list includes: For each extended node in the open list, calculating the distance difference and heading difference between the extended node and the reference target posture; Calculating the cost sum of the distance difference and the heading difference; The node with the smallest cost sum in the open list is selected as the current node.

10. The method according to claim 5, characterized in that After determining whether a resolvable path can be planned using the current node, the method further includes: If a resolvable path cannot be planned using the current node, node expansion is performed based on the current node to obtain multiple child nodes that have the current node as a parent node and satisfy collision constraints; Selecting a target child node with the smallest cost and not in the closed list from among the multiple child nodes; Adding the target child node to the open list; Update the current node to the target child node.

11. A device for planning a parking path, characterized in that: include: A processing module, used to determine the parking direction of the vehicle and obtain the environmental perception information of the vehicle before parking; a calculation module, configured to calculate a path end channel of the vehicle according to the parking direction and the environment perception information, wherein the path end channel corresponds to a distribution range of a path end of the parking path of the vehicle; A planning module is used to plan a parking path that meets the collision constraint condition based on the path end channel.

12. The device according to claim 11, characterized in that The calculation module comprises: a calculation unit, configured to calculate a reference target posture of the vehicle according to the parking direction and the environmental perception information, wherein the reference target posture is a reference posture of the vehicle after the parking is completed; A moving unit, used for moving the posture frame where the reference target posture is located by a preset distance in the front, back, left, and right directions respectively to obtain a rectangular envelope frame; A determination unit is used to determine the rectangular envelope as a path end channel.

13. A storage medium, characterized in that: The storage medium stores a computer program, wherein the computer program is configured to execute the method according to any one of claims 1 to 10 when executed.

14. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to run the computer program to perform the method according to any one of claims 1 to 10.

Images