Global and local information fusion-based automatic parking optimal path planning method and system

Abstract

The application discloses a global and local information fusion automatic parking optimal path planning method and system, and relates to the technical field of automatic parking path planning. According to global information, a global optimal path of a vehicle is planned, and in the process that the vehicle travels according to the planned path, the surrounding environment is sensed in real time, obstacle prediction is carried out, the planned path is influenced, and obstacle avoidance is realized. When it is sensed that there is an obstacle in the parking space or beside the parking space entrance which influences normal parking, a parking path is re-planned or a new target parking space is selected, global and local combined optimal path planning is realized, the disadvantages of single planning in the global or local are avoided, and the influence of a future motion track of a dynamic obstacle on a travel path is considered.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic parking path planning, in particular to an automatic parking optimal path planning method and system based on global and local information fusion. BACKGROUND

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

[0003] In recent years, after people meet the basic travel demand through cars, the pursuit of car driving experience, comprehensive performance, intelligence, individualization and technology is also increasing. The intelligence and automation of cars not only have advantages such as safety, comfort, energy saving and high efficiency, but also can provide drivers with good driving experience and travel experience. The automatic parking technology in the parking scene has gradually become an important scene, helping drivers to quickly, safely and conveniently find a parking space and park in, and solving the problems of difficult parking and parking.

[0004] At present, the automatic parking system mainly uses parking lot information to select a parking space according to the distance and make a simple path planning guide or identify a parking space around the vehicle through a vehicle-mounted sensor and plan to park in the parking space. However, the automatic parking system does not combine global information acquisition, parking space selection and path planning with local obstacle avoidance path planning and optimal parking path planning, and does not predict the motion trajectory of dynamic obstacles, thereby affecting the local obstacle avoidance path planning and the path planning of the parking space.

[0005] In addition, for the selection of the target parking space, the existing automatic parking system cannot analyze the actual situation and select the optimal target parking space in real time. Therefore, there is an urgent need for a method and system that can realize global and local information fusion in real parking scenarios, select the optimal parking space and comprehensively plan the path, and realize highly intelligent, flexible and predictive automatic parking. SUMMARY

[0006] In view of the deficiencies in the prior art, the present application provides an automatic parking optimal path planning method and system based on global and local information fusion, which plans a global optimal path for a vehicle according to global information, and in the process of driving the vehicle according to the planned path, the surrounding environment is sensed in real time, the influence of obstacles on the planned path is planned, and obstacle avoidance is realized. When the target parking space is reached and it is sensed that there are obstacles in the parking space or beside the parking space entrance that affect normal parking, the parking path is re-planned or a new target parking space is selected, the optimal path planning based on global and local combination is realized, the disadvantages of separate planning in global or local are avoided, and the influence of the future motion trajectory of dynamic obstacles on the driving path is considered.

[0007] In order to achieve the above-mentioned purpose, the present application is realized by the following technical scheme:

[0008] The first aspect of the present disclosure provides an automatic parking optimal path planning method based on global and local information fusion, comprising the following steps:

[0009] Obtaining vehicle basic parameters and judging the scene where the vehicle is located;

[0010] Obtaining global information and selecting an optimal target parking space based on the global information;

[0011] Considering the kinematic constraints and spatial constraints of the vehicle, the optimal path planning is performed on the path of the vehicle to the target parking space and the path of parking in the parking space.

[0012] During the process of controlling the vehicle to drive to the optimal target parking space according to the planned global optimal path, the vehicle surrounding obstacles are sensed in real time, and the short-time motion trajectory of the dynamic obstacles is predicted.

[0013] According to the sensed obstacle information and the predicted short-time motion trajectory, local path planning is performed to achieve obstacle avoidance.

[0014] Further, the scene where the vehicle is located is judged by using GPS to obtain the vehicle position and map information to determine the scene where the vehicle is located.

[0015] Further, the global information includes the type, position and size information of each parking space, as well as the position information of the lane and the entrance and exit around the parking space, which are accurately obtained through the layout of the facility construction and the map information; the parking space type includes arrangement type and function type: the arrangement type includes vertical parking space, horizontal parking space and inclined parking space; the function type mainly includes ordinary parking space, electric vehicle dedicated parking space with charging facilities and special vehicle dedicated parking space.

[0016] Further, the selection of the optimal target parking space is specifically as follows: in the range of empty parking spaces, according to the size and function type information of each parking space, the parking space that meets the size requirement of the vehicle and has the most suitable function type is selected to form a first target parking space range; considering the distance between the parking space and the entrance or exit of the parking lot or the destination of the driver, the length of the parking path and the convenience of getting on and off the vehicle in the parking space, the optimal target parking space is selected in the first target parking space range.

[0017] Further, the kinematic constraints consider the maximum turning angle and the maximum turning rate of the front wheels of the vehicle; and the spatial constraints consider the non-collision between the four corners of the vehicle and the space obstacles.

[0018] Further, the obstacle information acquisition and short-time trajectory prediction specifically comprises the following steps: obtaining the type, position and size information of the surrounding obstacles through the vehicle-mounted camera and radar, and obtaining the motion trajectory of the dynamic obstacles; predicting the short-time future motion trajectory of the obstacles through the historical motion trajectory and historical motion speed of the dynamic obstacles.

[0019] Further, the specific strategy for realizing the obstacle avoidance is as follows: when the space constraint is satisfied, the vehicle turns to the side to bypass; when there is a slowly moving obstacle and the predicted trajectory conflicts with the turning, the vehicle slowly turns to bypass under the safe situation; when the turning space constraint is not satisfied, the vehicle slows down to follow or stops to wait.

[0020] The second aspect of the present disclosure provides an automatic parking optimal path planning system based on global and local information fusion, comprising:

[0021] An information acquisition module is configured to acquire basic parameters of the vehicle and determine the scene in which the vehicle is located;

[0022] A global information acquisition and planning module is configured to acquire global information, select an optimal target parking space based on the global information, and perform optimal path planning for the path of the vehicle to the target parking space and the path of the vehicle to the parking space, considering the kinematic constraint and the space constraint of the vehicle;

[0023] An obstacle perception module is configured to control the vehicle to perceive the obstacles around the vehicle in real time during the driving process of the vehicle to the optimal target parking space according to the planned global optimal path, and predict the short-time motion trajectory of the dynamic obstacle;

[0024] A local planning module is configured to perform local path planning to realize obstacle avoidance according to the perceived obstacle information and the predicted short-time motion trajectory.

[0025] The third aspect of the present disclosure provides a medium having a program stored thereon, and the program is executed by a processor to realize the steps in the global and local information fusion automatic parking optimal path planning method according to the first aspect of the present disclosure.

[0026] The fourth aspect of the present disclosure provides a device comprising a memory, a processor, and a program stored in the memory and executable on the processor, and the processor executes the program to realize the steps in the global and local information fusion automatic parking optimal path planning method according to the first aspect of the present disclosure.

[0027] The beneficial effects of the above embodiments of the present disclosure are as follows:

[0028] The application can fuse global and local information, select an optimal parking space according to multiple influence factors, plan a globally optimal path, then acquire obstacle information through a vehicle-mounted sensor, track and predict the future trajectory of a dynamic obstacle, plan a locally optimal obstacle avoidance path, realize the combination of global and local path planning, realize the selection of an optimal parking space and an optimal path under global and local conditions, avoid the disadvantages of separate planning under global or local conditions, and improve the safety, efficiency, applicability, flexibility, predictability and intelligence of the parking system.

[0029] For the selection of the current optimal parking space, the application selects the most suitable parking space type according to different parking environments and different vehicle types, and selects an optimal target parking space according to multiple influence factors such as the size of the parking space, the distance between the vehicle and the entrance of the parking lot or the destination of the driver, the length of the parking path and the convenience of getting out of the vehicle. Compared with the defect that the traditional automatic parking system cannot analyze the actual environment in real time, the optimal parking space selection method of the application is more flexible and practical. BRIEF DESCRIPTION OF DRAWINGS

[0030] The accompanying drawings, which form a part of this specification, are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification. The embodiments of these drawings are set to explain the application and do not constitute an improper limitation on the application.

[0031] Figure 1 The flow chart of the method for global and local information fusion of the automatic parking optimal path planning in embodiment one of the application. DETAILED DESCRIPTION

[0032] It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the present application. Unless otherwise indicated, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0033] It should be noted that the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, the singular form is intended to include the plural form unless the context clearly indicates otherwise, and it should also be understood that when the terms "comprise" and / or "include" are used in the specification, there is a presence of the features, steps, operations, devices, components and / or their combinations;

[0034] Embodiment one

[0035] Embodiment one of the present disclosure provides a global and local information fusion automatic parking optimal path planning method, as shown in Figure 1 The method comprises the following steps:

[0036] S1: Obtain basic vehicle parameters and determine the vehicle's current location.

[0037] Preferably, the basic vehicle parameters include vehicle dimensions and driver's fixed parking space location information.

[0038] S1.1: Vehicle Scene Determination: Use GPS to obtain vehicle location and map information to determine the vehicle's scene, and use features such as the number of lane lines in the scene to further confirm the vehicle's scene.

[0039] Preferably, parking scenarios can be categorized as: public parking lots; roadside parking spaces; and parking lots with fixed spaces, such as those in residential communities and companies.

[0040] In some embodiments, when the vehicle is in a public parking lot or roadside parking space, the subsequent steps are executed; when the vehicle is in a parking lot with fixed parking spaces, such as a residential area or company, the optimal parking route is directly planned to the fixed parking space, i.e., the process jumps directly to step S3.

[0041] S2: Obtain global information and select the optimal target parking space based on the global information.

[0042] S2.1: Global information acquisition. Information such as the type, size, and location of each parking space, as well as the location of the surrounding lanes and entrances / exits, is obtained through facility construction layout and map information.

[0043] Preferably, parking space types include arrangement type and function type: arrangement type mainly includes: perpendicular parking space, horizontal parking space, and x° angled parking space; function type mainly includes: ordinary parking space, electric vehicle-specific parking space with charging facilities, and special vehicle-specific parking space.

[0044] S2.1.1: When a vehicle is in a public parking lot, obtain the location information of the parking lot entrance and exit; when a vehicle is in a roadside parking space, obtain the driver's destination location information. Real-time parking space occupancy information is obtained through parking lot cameras or roadside cameras.

[0045] S2.1.2: Accurately acquire and determine whether there are parked vehicles or obstacles in parking spaces using cameras in parking lots and along roadsides to determine in real time whether a parking space is vacant. Specifically, use cameras in parking lots and along roadsides to accurately acquire information about each parking space. The detection range for parking lots is the entire parking area, and the detection range for roadsides is within a certain distance from the destination on the current road.

[0046] S2.3: Optimal target parking space selection. Consider the most suitable parking space type for the vehicle, and select the optimal target parking space from the empty parking spaces in the parking lot or roadside based on the parking space size, the distance between the parking space and the parking lot entrance / exit or the driver's destination, the length of the parking path, and the convenience of getting in and out of the parking space.

[0047] S2.3.1: Firstly, in the range of empty parking spaces, according to the size of each parking space and the type of parking space function information, a first target parking space range is selected which meets the size requirements of the vehicle and is the most suitable type of function. For example, the electric vehicle dedicated parking space with charging facilities is the most suitable type of function when the current SOC of the electric vehicle is less than 30%.

[0048] S2.3.2: If there is no first priority function type parking space in the existing empty parking spaces in the current parking lot or roadside, it is extended to the general parking space.

[0049] S2.3.3: Considering the distance between the parking space and the entrance of the parking lot or the destination of the driver, the length of the parking path and the convenience of getting on and off the vehicle in the parking space, the optimal target parking space is selected in the first target parking space range, which is specifically:

[0050] M=k1S+k2L+k3Q

[0051] In the formula, M is a quantitative evaluation index, the smaller the better, k1, k2, k3 are weight coefficients of the distance S between the parking space and the entrance of the parking lot or the destination of the driver, the length L of the parking path and the convenience Q of getting on and off the vehicle in the parking space, and k1+k2+k3=1.

[0052] Preferably, the distance between the vehicle and the entrance of the parking lot or the destination of the driver is the road distance between the parking space and the pedestrian entrance in the public parking lot environment, and the pedestrian road distance between the parking space and the destination in the roadside parking environment.

[0053] Preferably, the length of the parking path is the pre-planned path length, that is, the shortest path planned from the current position of the vehicle in the lane to the target parking space at the entrance of the parking lot or roadside.

[0054] Preferably, the convenience of getting on and off the vehicle in the parking space is whether there are parked vehicles or obstacles within a certain range on the left and right sides of the vehicle after parking.

[0055] In some embodiments, if there are parked vehicles or obstacles within a certain range on the left and right sides of the vehicle, the specific interval distance is calculated to quantify the convenience of getting on and off the vehicle, which is specifically:

[0056]

[0057] In the formula, Q is a quantitative index of the convenience of getting on and off the vehicle, k Q is a proportional coefficient, P is the closest distance between the left and right sides of the vehicle and the obstacle or parked vehicle, and P0 is the distance range threshold.

[0058] S3: Considering the kinematic constraints and spatial constraints of the vehicle, the optimal path planning is performed on the path of the vehicle to the target parking space and the path of parking in the parking space.

[0059] S3.1: On the basis of the target parking space pre-planned path, considering the vehicle kinematic constraints, spatial constraints, the overall path is optimally planned.

[0060] Preferably, the kinematic constraints consider the maximum turning angle and maximum turning rate of the front wheels of the vehicle; and the spatial constraints consider the collision-free between the four corners of the vehicle and the surrounding walls and other spatial obstacles.

[0061] S3.2: The shortest path of the path to the target parking space under the constraints and the comprehensive optimal path of the parking path are the most efficient paths.

[0062] S3.2.1: The shortest path is selected as the shortest path to the target parking space by comparing the lengths of each path.

[0063] S3.2.2: The specific selection process of the comprehensive optimal path of the parking path is as follows: the optimal one-time parking strategy is selected, i.e. the vehicle is parked in the appropriate position area without adjustment, if the one-time parking does not meet the constraint condition, the adjustment parking strategy is selected, i.e. the vehicle is parked in the parking area multiple times with adjustment, the path length and the number of adjustments are comprehensively optimized, and the specific process is as follows:

[0064] M L =k1L+k2N

[0065] In the formula, M L is a quantitative evaluation index, k1 and k2 are weight coefficients of the parking path length L and the number of adjustments N, and k1+k2=1.

[0066] S4: During the process of controlling the vehicle to drive to the optimal target parking space according to the planned global optimal path, the vehicle is real-time perceived around the obstacles, and the short-time motion trajectory of the dynamic obstacle is predicted.

[0067] S4.1: During the process of controlling the vehicle to drive to the optimal target parking space according to the planned global optimal path, the vehicle uses the camera and ultrasonic radar to real-time perceive the position, size, historical motion trajectory and other information of the surrounding obstacles.

[0068] S4.2: Through the information obtained by the vehicle-mounted camera and radar, the real-time motion trajectory of the dynamic obstacle is obtained, and through the historical motion trajectory and historical motion speed of the dynamic obstacle, the short-time future motion trajectory of the obstacle is predicted.

[0069] S5: According to the perceived obstacle information and the predicted short-time motion trajectory, the local path planning is carried out to realize obstacle avoidance.

[0070] S5.1: When the obstacle affects the parking space path, the degree of influence of the obstacle and the predicted trajectory is determined. When the obstacle or the predicted trajectory conflicts with the vehicle in the parking space, a target parking space is selected around the vehicle and a parking plan is made. When the obstacle or the predicted trajectory conflicts with the vehicle outside the parking space, a variety of strategies are used to re-plan the path for parking. If a new parking path cannot be planned, a target parking space is selected around the vehicle and a parking plan is made.

[0071] S5.2: When there is an obstacle in front of the vehicle path or a conflict with the predicted trajectory of the obstacle, the vehicle is planned to turn to the side or slow down.

[0072] Preferably, the specific trajectory planning strategy includes: vehicle turning to the side to bypass strategy, slow down and follow or stop and wait strategy, slow down and turn to bypass strategy.

[0073] S5.1.1: When there is a static obstacle in front of the trajectory or there is a slow-moving obstacle in front of the trajectory and its motion trajectory coincides with the vehicle planning path, the vehicle turns to the side to bypass strategy is adopted when the space constraint is met. When there is a slow-moving obstacle and the predicted trajectory conflicts with the turning, the vehicle slowly turns to bypass strategy under safe conditions is adopted. When the vehicle turning space constraint is not met, the slow down and follow or stop and wait strategy is adopted.

[0074] Embodiment two:

[0075] The embodiment two of the present disclosure provides an automatic parking optimal path planning system with global and local information fusion, which comprises:

[0076] An information acquisition module is configured to acquire vehicle basic parameters and judge the scene where the vehicle is located;

[0077] A global information acquisition and planning module is configured to acquire global information, select an optimal target parking space based on the global information, and consider the kinematic constraint and space constraint of the vehicle to plan an optimal path for the vehicle to reach the target parking space and park in the parking space.

[0078] An obstacle perception module is configured to control the vehicle to perceive the obstacles around the vehicle in real time during the driving process of the vehicle to the optimal target parking space according to the planned global optimal path, and predict the short-time motion trajectory of the dynamic obstacle.

[0079] A local planning module is configured to plan a local path to avoid obstacles according to the perceived obstacle information and the predicted short-time motion trajectory.

[0080] Embodiment three:

[0081] Embodiment three of the present disclosure provides a medium having a program stored thereon, which, when executed by a processor, implements the steps in the global and local information fusion automatic parking optimal path planning method according to embodiment one of the present disclosure, and the steps are as follows:

[0082] S1: acquiring vehicle basic parameters and judging the scene where the vehicle is located;

[0083] S2: acquiring global information and selecting an optimal target parking space based on the global information;

[0084] S3: considering the kinematic constraints and spatial constraints of the vehicle, performing optimal path planning on the path of the vehicle to the target parking space and the path of parking in the parking space;

[0085] S4: controlling the vehicle to perform real-time perception on the obstacles around the vehicle during the driving process of the vehicle to the optimal target parking space according to the planned global optimal path, and predicting the short-time motion trajectory of the dynamic obstacles;

[0086] S5: performing local path planning to achieve obstacle avoidance according to the perceived obstacle information and the predicted short-time motion trajectory.

[0087] The detailed steps are the same as those in the global and local information fusion automatic parking optimal path planning method provided in embodiment one, and will not be repeated here.

[0088] Embodiment four

[0089] Embodiment four of the present disclosure provides a device including a memory, a processor, and a program stored in the memory and executable on the processor, and the processor implements the steps in the global and local information fusion automatic parking optimal path planning method according to embodiment one of the present disclosure when executing the program, and the steps are as follows:

[0090] S1: acquiring vehicle basic parameters and judging the scene where the vehicle is located;

[0091] S2: acquiring global information and selecting an optimal target parking space based on the global information;

[0092] S3: considering the kinematic constraints and spatial constraints of the vehicle, performing optimal path planning on the path of the vehicle to the target parking space and the path of parking in the parking space;

[0093] S4: controlling the vehicle to perform real-time perception on the obstacles around the vehicle during the driving process of the vehicle to the optimal target parking space according to the planned global optimal path, and predicting the short-time motion trajectory of the dynamic obstacles;

[0094] S5: performing local path planning to achieve obstacle avoidance according to the perceived obstacle information and the predicted short-time motion trajectory.

[0095] The detailed steps are the same as the global and local information fusion automatic parking optimal path planning method provided in the first embodiment, and thus will not be described herein.

[0096] Those skilled in the art will appreciate that embodiments of the disclosure can be provided as methods, systems, or computer program products. Accordingly, the disclosure can take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Moreover, the disclosure can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, magnetic disk storage and optical storage) containing computer-usable program code.

[0097] The above only provides preferred embodiments of the present application and is not intended to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. An optimal path planning method for automatic parking that integrates global and local information, characterized in that, include: Obtain basic vehicle parameters and determine the vehicle's location within the given environment; When the vehicle is in a public parking lot or roadside parking space, continue with the subsequent steps; when the vehicle is in a parking lot with fixed parking spaces in a residential area or company, directly plan the optimal parking route to the fixed parking space. The process involves acquiring comprehensive information, including accurately determining the type, location, and size of each parking space through facility layout and map data, as well as the location of surrounding lanes and entrances / exits. It also utilizes parking lot and roadside cameras to accurately capture information about each parking space. Based on this comprehensive information, the optimal target parking space is selected. Specifically, within the range of available parking spaces, based on the size and function type of each space (including arrangement type, horizontal type, and angled type), and the main function types (including regular parking spaces, electric vehicle-only parking spaces with charging facilities, and special vehicle-only parking spaces), the first target parking space range is formed by selecting spaces that meet vehicle size requirements and have the most suitable function type. Finally, considering three influencing factors—distance from the parking space to the parking lot entrance / exit or the driver's destination, parking path length, and ease of getting in and out of the parking space—the optimal target parking space is selected from the first target parking space range. In the formula, M is a quantitative evaluation index, where a smaller value is better. k 1. k 2. k 3 represents the distance S between the parking space and the parking lot entrance / exit or the driver's destination, the length of the parking path L, and the ease of getting in and out of the parking space. Q The weighting coefficients, k 1+ k 2+ k 3=1; The convenience of getting on and off the vehicle in the parking space is determined by whether there are parked vehicles or obstacles within a certain range on both sides after the vehicle is parked in the space. If there are parked vehicles or obstacles within a certain range on both sides of the vehicle, the specific interval distance is calculated to quantify the convenience of getting on and off the vehicle. Specifically: In the formula, Q To quantify the ease of getting on and off the bus, k Q P is a proportionality coefficient, where P represents the closest distance between the vehicle and obstacles or parked vehicles on its left and right sides. P 0 represents the distance range threshold; Considering vehicle kinematic constraints and spatial constraints, optimal path planning is performed for the vehicle's path to the target parking space and the path to the parking space. During the process of controlling the vehicle to travel to the optimal target parking space according to the planned globally optimal path, real-time perception of obstacles around the vehicle is conducted, and short-term trajectories of dynamic obstacles are predicted. The specific steps for obstacle information acquisition and short-term trajectory prediction are as follows: The type, location, and size information of surrounding obstacles are acquired through onboard cameras and radar, and the movement trajectories of dynamic obstacles are obtained. Based on the historical movement trajectories and historical speeds of dynamic obstacles, the short-term future movement trajectories of obstacles are predicted. Based on the perceived obstacle information and the predicted short-term movement trajectories, local path planning is performed to achieve obstacle avoidance. When an obstacle affects the parking path, the degree of impact of the obstacle and the predicted trajectory is assessed. If the obstacle or predicted trajectory conflicts with the vehicle within the parking space, a new target parking space is selected around the vehicle and parking is planned. If the obstacle or predicted trajectory conflicts with the vehicle outside the parking space, multiple strategies are used to replan the parking path. If a new parking path cannot be planned, a new target parking space is selected around the vehicle and parking is planned. The specific strategies for obstacle avoidance are as follows: when the spatial constraints are met, a strategy of turning the vehicle to the side to bypass the obstacle is adopted; when there is a slowly moving obstacle and the predicted trajectory conflicts with the turning, a strategy of slowly turning to bypass the obstacle while maintaining a safe driving situation is adopted; when the vehicle's turning space constraints are not met, a strategy of slowing down and following or stopping and waiting is adopted.

2. The automatic parking optimal path planning method based on global and local information fusion as described in claim 1, characterized in that, The vehicle's location is determined by using GPS to obtain the vehicle's location and map information.

3. The automatic parking optimal path planning method based on global and local information fusion as described in claim 1, characterized in that, The kinematic constraints take into account the maximum turning angle and maximum turning rate of the vehicle's front wheels; the spatial constraints take into account that the vehicle's four corners do not collide with spatial obstacles.

4. An automatic parking optimal path planning system that integrates global and local information, characterized in that: include: The information acquisition module is configured to acquire basic vehicle parameters and determine the vehicle's current location. When the vehicle is in a public parking lot or roadside parking space, continue with the subsequent steps; when the vehicle is in a parking lot with fixed parking spaces in a residential area or company, directly plan the optimal parking route to the fixed parking space. The global information acquisition and planning module is configured to acquire global information, including accurately acquiring the type, location, and size information of each parking space through facility construction layout and map information, as well as the location of the surrounding lanes and entrances / exits of the parking space, accurately acquiring the information of each parking space using cameras in the parking lot and on the roadside, and selecting the optimal target parking space based on the global information; Considering vehicle kinematic constraints and spatial constraints, optimal path planning is performed for the vehicle's path to the target parking space and the path to the parking space. Specifically, the optimal target parking space is selected from the available parking spaces based on their size and function type. Parking space types include arrangement type and function type: arrangement type includes perpendicular parking spaces, horizontal parking spaces, and angled parking spaces; function types mainly include ordinary parking spaces, electric vehicle-only parking spaces with charging facilities, and special vehicle-only parking spaces. Parking spaces that meet the vehicle size requirements and have the most suitable function type are selected to form the first target parking space range. Taking into account three influencing factors—distance from the parking space to the parking lot entrance / exit or the driver's destination, parking path length, and ease of getting in and out of the parking space—the optimal target parking space is selected from the first target parking space range. In the formula, M is a quantitative evaluation index, where a smaller value is better. k 1. k 2. k 3 represents the distance S between the parking space and the parking lot entrance / exit or the driver's destination, the length of the parking path L, and the ease of getting in and out of the parking space. Q The weighting coefficients, k 1+ k 2+ k 3=1; The convenience of getting on and off the vehicle in the parking space is determined by whether there are parked vehicles or obstacles within a certain range on both sides after the vehicle is parked in the space. If there are parked vehicles or obstacles within a certain range on both sides of the vehicle, the specific interval distance is calculated to quantify the convenience of getting on and off the vehicle. Specifically: In the formula, Q To quantify the ease of getting on and off the bus, k Q P is a proportionality coefficient, where P represents the closest distance between the vehicle and obstacles or parked vehicles on its left and right sides. P 0 represents the distance range threshold; The obstacle perception module is configured to perceive obstacles around the vehicle in real time while controlling the vehicle to travel to the optimal target parking space according to the planned global optimal path, and predict the short-term movement trajectory of dynamic obstacles. The specific steps of obstacle information acquisition and short-term trajectory prediction are as follows: acquire the type, location and size information of surrounding obstacles through the vehicle camera and radar, acquire the movement trajectory of dynamic obstacles, and predict the short-term future movement trajectory of obstacles through the historical movement trajectory and historical movement speed of dynamic obstacles. The local planning module is configured to perform local path planning to achieve obstacle avoidance based on perceived obstacle information and predicted short-term motion trajectories. When an obstacle affects the parking path, the degree of impact of the obstacle and the predicted trajectory is determined. If the obstacle or predicted trajectory conflicts with the vehicle within the parking space, a new target parking space is selected around the vehicle and the parking is planned. If the obstacle or predicted trajectory conflicts with the vehicle outside the parking space, multiple strategies are used to replan the parking path. If a new parking path cannot be planned, a new target parking space is selected around the vehicle and the parking is planned. The specific strategies for obstacle avoidance are as follows: when the spatial constraints are met, the vehicle is turned to the side to detour; when there is a slow-moving obstacle and the predicted trajectory conflicts with the steering, the vehicle is turned slowly to detour while maintaining a safe situation; when the vehicle's steering spatial constraints are not met, the vehicle is slowed down to follow or stopped to wait.

5. A computer-readable storage medium, characterized in that, It stores multiple instructions, which are adapted to be loaded and executed by the processor of the terminal device, and are the automatic parking optimal path planning method based on the fusion of global and local information as described in any one of claims 1-3.

6. A terminal device, characterized in that, The system includes a processor and a computer-readable storage medium, wherein the processor implements various instructions; and the computer-readable storage medium stores multiple instructions adapted to be loaded by the processor and executed by the processor for the automatic parking optimal path planning method based on global and local information fusion as described in any one of claims 1-3.

Images