Vehicle Environment Mapping With Classified Occupancy Grids
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle environment mapping technologies face challenges in accurately positioning and classifying objects due to high uncertainty in passive positioning systems like cameras and inconsistencies in integrating semantic information with high-accuracy active positioning systems like radar, lidar, and ultrasound.
Innovation Solution
Combining semantic information from camera systems with high-accuracy active positioning systems using sensor fusion models, specifically through Dempster-Shafer evidence theory, to create classified occupancy grids that enhance object classification and positioning accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If semantic information from camera systems is used for object classification, then object classification capability is improved, but positioning accuracy deteriorates due to high uncertainty in passive positioning systems
Solution Approach 1:
The patent combines semantic information from camera systems with occupancy grid data from active positioning systems (radar, lidar, ultrasound) to create classified occupancy grids. This merging allows the system to leverage the object classification capabilities of passive systems while maintaining the high positioning accuracy of active systems, resolving the contradiction between classification capability and positioning precision.
2Measurement precision
If occupancy grids from active positioning systems are used, then positioning accuracy is improved, but object classification capability deteriorates due to lack of semantic information
Solution Approach 1:
The system merges occupancy grid data containing high-precision positioning information with semantic information containing object classification data. The classified occupancy grid integrates both types of information, allowing the system to simultaneously achieve high positioning accuracy and object classification capability that neither system could provide alone.
3Measurement precision
If semantic information from multiple sources is integrated, then object classification accuracy is improved, but system complexity increases
Solution Approach 1:
The patent uses Dempster-Shafer evidence theory as an intermediary framework to integrate semantic information from multiple sources (cameras, mapping applications, etc.). This mathematical framework provides a systematic method for combining evidence from different sources while handling uncertainty, improving object classification accuracy without requiring complex custom integration logic.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A computer-implemented method and device for mapping a vehicle environment of a vehicle are disclosed. The method comprising determining an occupancy grid representing the vehicle environment, the occupancy grid comprising occupancy probability information of a first set of object detections, wherein the occupancy probability information is determined from first positioning information obtained from a first sensor system, the first positioning information indicating one or more positions of the first set of object detections with respect to the vehicle; obtaining semantic information and second positioning information associated with the semantic information from one or more semantic information sources, the semantic information comprising object classification information of a second set of object detections, the second positioning system indicating one or more positions of the second set of object detections with respect to the vehicle; and combining the object classification information of the second set of object detections with the occupancy probability information of the occupancy grid to generate a classified occupancy grid.