CN-122015882-A - Navigation auxiliary method, device, system and medium based on scene mode

CN122015882ACN 122015882 ACN122015882 ACN 122015882ACN-122015882-A

Abstract

The invention relates to the technical field of vehicles, in particular to a scene mode-based pilot auxiliary method, a scene mode-based pilot auxiliary device, a scene mode-based pilot auxiliary system and a scene mode-based pilot auxiliary medium, wherein the scene mode-based pilot auxiliary method comprises the steps of obtaining driver input information, vehicle chassis domain information, cloud information and vehicle body domain information; the method comprises the steps of obtaining scene mode characteristic parameters by blurring input information of a driver, adjusting map grid weights based on the scene mode characteristic parameters to obtain map grids with adjusted weights, carrying out path search on the map grids with adjusted weights by adopting a heuristic search algorithm to generate a plurality of candidate paths, carrying out feasibility assessment on the candidate paths according to residual cruising parameters in vehicle chassis domain information, sorting the candidate paths which pass the feasibility assessment according to comprehensive scores, and outputting the sorted paths for the driver to select.

Inventors

SUN YINGJIE
LONG LI
WANG MINGYU
LI JIALING
Shan Baichuan
ZHOU ZHENYI
YU CHENGHAO

Assignees

中国第一汽车股份有限公司

Dates

Publication Date: 20260512
Application Date: 20251212

Claims (10)

1. A method of pilot assistance based on scene mode, the method comprising the steps of: Acquiring driver input information, vehicle chassis domain information, cloud information and vehicle body domain information; Blurring processing is carried out on the driver input information to obtain scene mode characteristic parameters; Adjusting the map grid weight based on the scene mode characteristic parameters to obtain a map grid with adjusted weight, wherein the map grid weight comprises scenic spot fitness weight and energy supplementing facility weight; Carrying out path search on the map grids with the adjusted weights by adopting a heuristic search algorithm to generate a plurality of candidate paths; carrying out feasibility evaluation on the candidate paths according to the remaining endurance parameters in the vehicle chassis domain information; and sorting the candidate paths passing through the feasibility evaluation according to the comprehensive scores, and outputting the sorted paths for the driver to select.
2. The method of claim 1, wherein the obtaining driver input information, vehicle chassis domain information, cloud information, and body domain information comprises: acquiring driver input information, wherein the driver input information comprises a travel destination, a travel starting point, a travel reason and scenic spot preference; Acquiring vehicle chassis domain information, wherein the vehicle chassis domain information comprises a vehicle motion information parameter, battery power/fuel information and driving cycle driving mileage; acquiring cloud information, wherein the cloud information comprises a navigation map, scenic spot information, along-road charging potential information/gas station information and weather information; and acquiring car body domain information, wherein the car body domain information comprises car window state information, child seat side member conditions, passenger position information and sunlight and rainfall sensor information.
3. The method according to claim 1, wherein said blurring the driver input information to obtain scene mode feature parameters includes: extracting travel reasons and scenic spot preference keywords in the driver input information; and matching the keywords with a preset scene mode feature library to generate scene mode feature parameters, wherein the scene mode feature parameters comprise leisure game feature parameters, commute feature parameters and shopping feature parameters.
4. The method of claim 1, wherein adjusting the map grid weight based on the scene mode feature parameter results in an adjusted map grid, comprising: When the scene mode characteristic parameter is a leisure game characteristic parameter, improving the scenic spot fitness weight of a scenic spot region matched with the scenic spot preference in the map grid; And according to the remaining endurance parameters in the vehicle chassis domain information, increasing the weight of energy supplementing facilities in the area where the energy supplementing facilities are located in the map grid to obtain the map grid with the adjusted weight, wherein the energy supplementing facilities comprise charging stations and gas stations.
5. The method of claim 1, wherein the performing a path search on the weighted map grid using a heuristic search algorithm to generate a plurality of candidate paths comprises: taking a travel starting point in the driver input information as a search starting point and taking the travel destination as a search destination to construct a map grid network; assigning the adjusted scenic spot fitness weight and energy supplementing facility weight to a cost function corresponding to the map grid; and carrying out path search on the map grid network through an A-algorithm, and generating at least one path with the minimum cost sum as a candidate path.
6. The method of claim 1, wherein the performing the feasibility assessment of the candidate path based on the remaining endurance parameters in the vehicle chassis domain information comprises: calculating the total driving mileage of each candidate path; Comparing the total driving mileage with the remaining endurance parameters, and judging that the candidate path passes the feasibility evaluation when the total driving mileage is smaller than or equal to the remaining endurance parameters; and when the total driving mileage is greater than the remaining endurance parameters, checking whether the candidate path contains energy supplementing facilities, if so, judging that the candidate path passes the feasibility evaluation, and if not, judging that the candidate path does not pass the feasibility evaluation.
7. The method of claim 1, wherein ranking candidate paths through feasibility assessment according to a composite score comprises: Calculating the comprehensive score of the candidate path through feasibility evaluation, wherein the calculation mode of the comprehensive score is that the comprehensive score = alpha x scenic spot fitness weight sum + beta x energy supplementing facility weight sum + gamma x (remaining endurance parameter-total driving range), wherein alpha, beta and gamma are preset weight coefficients; And sequencing the candidate paths according to the sequence of the comprehensive scores from high to low, and outputting a sequenced path list for the driver to select.
8. A scene mode based pilot assist device, the device comprising: The first module is used for acquiring driver input information, vehicle chassis domain information, cloud information and vehicle body domain information; The second module is used for carrying out fuzzification processing on the input information of the driver to obtain scene mode characteristic parameters; The third module is used for adjusting the map grid weight based on the scene mode characteristic parameters to obtain a map grid with adjusted weight, wherein the map grid weight comprises scenic spot fitness weight and energy supplementing facility weight; A fourth module, configured to perform path search on the map grid after the weight adjustment by using a heuristic search algorithm, so as to generate a plurality of candidate paths; a fifth module, configured to perform feasibility assessment on a candidate path according to the remaining endurance parameters in the vehicle chassis domain information; And a sixth module, configured to rank the candidate paths passing through the feasibility evaluation according to the composite score, and output the ranked paths for the driver to select.
9. A scene mode based pilot assistance system comprising a memory, a processor and a program stored on the memory and executable on the processor, which when executed by the processor implements the method of any one of claims 1 to 7.
10. A computer readable storage medium storing a computer program, characterized in that the computer program, when executed by a processor, implements the method of any one of claims 1 to 7.

Description

Navigation auxiliary method, device, system and medium based on scene mode Technical Field The invention relates to the technical field of vehicles, in particular to a scene mode-based pilot assistance method, device, system and medium. Background In the current ADAS equipped vehicles, the pilot auxiliary function has a larger and larger duty ratio, but still a lot of technical schemes rely on high-precision map information to plan paths in the range of preset electronic fences, compared with an SD map, the method lacks flexible pilot technology, and can recommend starting points and end points and use functions in the stage of preset routes of users. Disclosure of Invention In view of the above, an object of the embodiments of the present invention is to provide a navigation assistance method, apparatus, system and medium based on scene mode, which aims to solve the problems of high dependency on high-precision map, insufficient flexibility of path planning and difficulty in dynamically adjusting the actual state of the vehicle in combination with personalized requirements of the driver in the existing navigation assistance technology. By fusing multidimensional information, a dynamic path planning mechanism based on a scene mode is constructed, a path searching strategy can be intelligently adjusted according to key parameters such as travel purposes, preference of a driver and remaining endurance of a vehicle, a candidate path which is more fit with actual requirements is generated, and the practicability and personalized experience of pilot assistance are improved. In one aspect, an embodiment of the present invention provides a method for assisting navigation based on a scene mode, the method including the steps of: Acquiring driver input information, vehicle chassis domain information, cloud information and vehicle body domain information; Blurring processing is carried out on the driver input information to obtain scene mode characteristic parameters; Adjusting the map grid weight based on the scene mode characteristic parameters to obtain a map grid with adjusted weight, wherein the map grid weight comprises scenic spot fitness weight and energy supplementing facility weight; Carrying out path search on the map grids with the adjusted weights by adopting a heuristic search algorithm to generate a plurality of candidate paths; carrying out feasibility evaluation on the candidate paths according to the remaining endurance parameters in the vehicle chassis domain information; and sorting the candidate paths passing through the feasibility evaluation according to the comprehensive scores, and outputting the sorted paths for the driver to select. Optionally, the obtaining driver input information, vehicle chassis domain information, cloud information and vehicle body domain information includes: acquiring driver input information, wherein the driver input information comprises a travel destination, a travel starting point, a travel reason and scenic spot preference; Acquiring vehicle chassis domain information, wherein the vehicle chassis domain information comprises a vehicle motion information parameter, battery power/fuel information and driving cycle driving mileage; acquiring cloud information, wherein the cloud information comprises a navigation map, scenic spot information, along-road charging potential information/gas station information and weather information; and acquiring car body domain information, wherein the car body domain information comprises car window state information, child seat side member conditions, passenger position information and sunlight and rainfall sensor information. Optionally, the blurring processing is performed on the driver input information to obtain scene mode feature parameters, including: extracting travel reasons and scenic spot preference keywords in the driver input information; and matching the keywords with a preset scene mode feature library to generate scene mode feature parameters, wherein the scene mode feature parameters comprise leisure game feature parameters, commute feature parameters and shopping feature parameters. Optionally, the adjusting the weight of the map grid based on the scene mode feature parameter to obtain the map grid after the weight adjustment includes: When the scene mode characteristic parameter is a leisure game characteristic parameter, improving the scenic spot fitness weight of a scenic spot region matched with the scenic spot preference in the map grid; And according to the remaining endurance parameters in the vehicle chassis domain information, increasing the weight of energy supplementing facilities in the area where the energy supplementing facilities are located in the map grid to obtain the map grid with the adjusted weight, wherein the energy supplementing facilities comprise charging stations and gas stations. Optionally, the performing path search on the map grid after the weight adjustment by using a heuristic sear