CN-121989920-A - Parking method, device, medium, product, chip system and vehicle

Abstract

The disclosure provides a parking method, a device, a medium, a product, a chip system and a vehicle, and relates to the technical field of intelligent driving. The parking method comprises the steps of obtaining the current position of a vehicle when the vehicle arrives in a preset range outside a target parking space, wherein the target parking space is the parking space where the vehicle is to be parked, planning a parking path based on the current position of the vehicle and loading the parking space data of the target parking space which is subjected to implicit reconstruction, and controlling the vehicle to park in the target parking space according to the parking path. Based on the method, the problem that effective automatic parking assistance cannot be provided is solved, a parking path is planned based on the parking space data of the target parking space with implicit reconstruction completed, and the vehicle is controlled to park according to the path, so that the vehicle can park into the parking space smoothly, the application range of the auxiliary parking technology is improved, and the user experience is improved.

Inventors

• YE HANGJUN
• XIANG HENGYONG
• LI BICHAO

Assignees

• 小米汽车科技有限公司

Dates

Publication Date

20260508

Application Date

20260403

Claims (20)

1. 1. A method of parking, comprising: when a vehicle arrives in a preset range outside a target parking space, acquiring the current position of the vehicle, wherein the target parking space is a parking space into which the vehicle is to be parked; Planning a parking path based on the current position of the vehicle and loading the parking space data of the target parking space which is subjected to implicit reconstruction; and controlling the vehicle to park in the target parking space according to the parking path.
2. 2. The method as recited in claim 1, further comprising: when a vehicle first parks into the target parking space, acquiring environmental characteristic data of the target parking space and space position information of the vehicle; And carrying out implicit reconstruction on the target parking space according to the environmental characteristic data and the spatial position information, and storing an implicit reconstruction result, wherein the implicit reconstruction result comprises parking space data for planning to park in the target parking space.
3. 3. The method of claim 2, wherein the environmental characteristic data comprises picture data and the spatial location information comprises a global pose of the vehicle; the obtaining the environmental characteristic data of the target parking space and the spatial position information of the vehicle comprises the following steps: The method comprises the steps of collecting picture data in a preset distance range of a target parking space through a vehicle-mounted sensor, and synchronously collecting the global position and the attitude of the vehicle when the picture data are collected, wherein the global position and the attitude of the vehicle comprise three-dimensional position coordinates and attitude angle information of the vehicle, and the attitude angle information comprises a course angle, a pitch angle and a roll angle.
4. 4. The method according to claim 2, wherein the environmental feature data includes image data acquired by a vehicle, and the implicitly reconstructing the target parking space and storing an implicit reconstruction result according to the environmental feature data and the spatial position information includes: Preprocessing the picture data to obtain the environmental characteristics of the target parking space; Correlating the extracted environmental features with the global pose of the vehicle, and constructing a three-dimensional implicit representation model of the target parking space so as to implicitly reconstruct the target parking space; And storing an implicit reconstruction result of the target parking space.
5. 5. The method of claim 1, wherein the controlling the vehicle to park in the target parking space according to the parking path comprises: Determining the position of a parking start point corresponding to the parking path; acquiring a position deviation between the current position of the vehicle and the position of the parking start point; and adjusting the current position of the vehicle according to the position deviation so that the vehicle starts from the position of the parking start point and parks into the target parking space according to the parking path.
6. 6. The method as recited in claim 1, further comprising: acquiring current environmental characteristic data of the vehicle; The planning a parking path based on the current position of the vehicle and loading the parking space data of the target parking space with the implicit reconstruction completed comprises the following steps: Planning a parking path based on the current position of the vehicle, the current environmental characteristic data of the vehicle and the parking space data of the target parking space with the implicit reconstruction completed.
7. 7. The method of claim 6, wherein the current environmental characteristic data comprises current environmental picture data collected by a vehicle; The method further comprises the steps of: Extracting two-dimensional feature points in the current environment picture data and three-dimensional feature points of each parking space data which have been subjected to implicit reconstruction; Obtaining the matching degree between the two-dimensional feature points and each three-dimensional feature point through a preset algorithm based on the point constraint relation between the two-dimensional feature points and the three-dimensional feature points; and responding to the matching degree between the two-dimensional characteristic points and the target three-dimensional characteristic points being larger than a preset matching threshold, and taking the parking space data corresponding to the target three-dimensional characteristic points as the parking space data of the target parking space.
8. 8. The method as recited in claim 1, further comprising: acquiring parking space data in the preset range in an implicit reconstruction result based on the current position of the vehicle; and loading the data of each parking place sequentially according to the sequence from near to far of the distance between the parking place coordinates and the current position of the vehicle.
9. 9. The method of claim 1, further comprising, prior to planning a parking path based on the current location of the vehicle and loading the target parking space data for the target parking space for which implicit reconstruction has been completed: Acquiring current external parameters of a vehicle-mounted camera, wherein the current external parameters comprise relative position parameters and angle parameters between the camera and a vehicle; and obtaining the target position of the target parking space under the own coordinate system of the current vehicle through coordinate conversion according to the position of the vehicle-mounted camera in the parking space data of the target parking space and the current external parameters of the vehicle-mounted camera.
10. 10. The method of claim 1, wherein after obtaining the current location of the vehicle, the method further comprises: collecting the topographic features and the mark object types of the areas corresponding to the current position of the vehicle based on the current position of the vehicle; According to the topographic features and the mark object types, matching the topographic features and the mark object types corresponding to the parking space data; and loading the parking space data sequentially from high to low according to the matching degree.
11. 11. The method as recited in claim 1, further comprising: collecting environment feedback data in the parking process, wherein the environment feedback data comprises distance data of a vehicle and surrounding obstacles and steering angle data of wheels; And adjusting the parking path based on the environment feedback data.
12. 12. The method of any one of claims 1-11, wherein the target spot is an irregular spot.
13. 13. A parking apparatus, characterized by comprising: the first acquisition module is used for acquiring the current position of the vehicle when the vehicle reaches a preset range outside a target parking space, wherein the target parking space is a parking space into which the vehicle is to be parked; the path planning module is used for planning a parking path based on the current position of the vehicle and loading the parking space data of the target parking space which is subjected to implicit reconstruction; and the parking control module is used for controlling the vehicle to park in the target parking space according to the parking path.
14. 14. The apparatus as recited in claim 13, further comprising: The second acquisition module is used for acquiring environmental characteristic data of the target parking space and space position information of the vehicle when the vehicle is first parked in the target parking space; And the implicit reconstruction module is used for performing implicit reconstruction on the target parking space according to the environmental characteristic data and the space position information and storing an implicit reconstruction result, wherein the implicit reconstruction result comprises parking space data for planning to park in the target parking space.
15. 15. The apparatus of claim 13, wherein the current environmental characteristic data comprises current picture data collected by a vehicle, and wherein the path planning module is further configured to: Extracting two-dimensional feature points in the current environment picture data and three-dimensional feature points of each parking space data which have been subjected to implicit reconstruction; Obtaining the matching degree between the two-dimensional feature points and each three-dimensional feature point through a preset algorithm based on the point constraint relation between the two-dimensional feature points and the three-dimensional feature points; and responding to the matching degree between the two-dimensional characteristic points and the target three-dimensional characteristic points being larger than a preset matching threshold, and taking the parking space data corresponding to the target three-dimensional characteristic points as the parking space data of the target parking space.
16. 16. The apparatus of claim 13, further comprising a coordinate scaling module; The first acquisition module is further used for acquiring current external parameters of the vehicle-mounted camera, wherein the current external parameters comprise relative position parameters and angle parameters between the camera and the vehicle; the coordinate conversion module is used for obtaining the target position of the target parking space under the own coordinate system of the current vehicle through coordinate conversion according to the position of the vehicle-mounted camera in the target parking space data and the current external parameters of the vehicle-mounted camera.
17. 17. A non-transitory computer readable storage medium storing a computer program or instructions which, when executed by a processor of an electronic device, enable the electronic device to perform the method steps of any one of claims 1-12.
18. 18. A computer program product comprising a computer program which, when executed by a processor, implements the method steps of any of claims 1-12.
19. 19. A chip system, characterized in that it comprises a processing unit and an interface circuit, the processing unit obtaining program instructions via the interface circuit, the program instructions being executed by the processing unit, the processing unit being adapted to perform the steps of the method according to any of claims 1-12.
20. 20. A vehicle, characterized by comprising: A memory for storing processor-executable instructions; Wherein the processor is configured to implement the method steps of any one of claims 1-12.

Description

Parking method, device, medium, product, chip system and vehicle Technical Field The disclosure relates to the technical field of intelligent driving, in particular to a parking method, a device, a medium, a product, a chip system and a vehicle. Background With the continuous development and popularization of parking assist technology, users can more efficiently and conveniently complete parking operation in the daily driving process. However, the existing parking assist technology is still not ideal in identifying parking spaces and parking effects in practical application, and particularly, accurate detection and effective assistance are difficult to achieve when the existing parking assist technology faces irregular parking spaces. Disclosure of Invention In order to overcome the problems in the related art, the application discloses a parking method, a device, a medium, a product, a chip system and a vehicle. According to a first aspect of an embodiment of the present disclosure, there is provided a parking method including: when a vehicle arrives in a preset range outside a target parking space, acquiring the current position of the vehicle, wherein the target parking space is a parking space into which the vehicle is to be parked; Planning a parking path based on the current position of the vehicle and loading the parking space data of the target parking space which is subjected to implicit reconstruction; and controlling the vehicle to park in the target parking space according to the parking path. According to the scheme, the problems that the existing parking auxiliary technology is difficult to accurately detect and effective automatic parking auxiliary cannot be provided can be effectively solved, the parking path fitting the actual parking space shape is accurately planned through implicitly reconstructing the parking space data, vehicles can smoothly park into the parking space, the application range of the automatic parking auxiliary technology is widened, and the driving experience of users is improved. Optionally, the method further comprises the steps of acquiring environmental characteristic data of the target parking space and space position information of the vehicle when the vehicle is first parked in the target parking space; And carrying out implicit reconstruction on the target parking space according to the environmental characteristic data and the spatial position information, and storing an implicit reconstruction result, wherein the implicit reconstruction result comprises parking space data for planning to park in the target parking space. According to the scheme disclosed by the invention, when the vehicle arrives at the target parking space again, the parking path can be planned by directly utilizing the parking space data in the stored implicit reconstruction result, complex environment sensing and data processing are not required to be repeated, the accuracy and efficiency of parking assistance of the parking space are improved, the system resources and the calculation time are effectively saved, and a smoother, more convenient and reliable parking experience is brought to a user. Optionally, the environmental feature data includes picture data, and the spatial position information includes a global pose of the vehicle; the obtaining the environmental characteristic data of the target parking space and the spatial position information of the vehicle comprises the following steps: The method comprises the steps of collecting picture data in a preset distance range of a target parking space through a vehicle-mounted sensor, and synchronously collecting the global position and the attitude of the vehicle when the picture data are collected, wherein the global position and the attitude of the vehicle comprise three-dimensional position coordinates and attitude angle information of the vehicle, and the attitude angle information comprises a course angle, a pitch angle and a roll angle. According to the scheme disclosed by the invention, the relation between the environmental characteristics of the target parking space and the vehicle space position can be accurately acquired, multidimensional and high-precision basic data are provided for the subsequent implicit reconstruction of the parking space, and the accuracy and reliability of the implicit reconstruction of the parking space are ensured. Optionally, the environmental feature data includes image data collected by a vehicle, and the implicitly reconstructing the target parking space and storing an implicit reconstruction result according to the environmental feature data and the spatial position information includes: Preprocessing the picture data to obtain the environmental characteristics of the target parking space; Correlating the extracted environmental features with the global pose of the vehicle, and constructing a three-dimensional implicit representation model of the target parking space so as to implicitly reconstr