车辆驾驶数据的处理方法、装置、设备及存储介质
By extracting and optimizing information from vehicle driving datasets and images, a candidate camera pose information set is obtained, which solves the attitude drift problem of autonomous vehicles when GPS and radar data fail, and realizes real-time positioning of autonomous vehicles.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- GUANGZHOU WERIDE TECH LTD CO
- Filing Date
- 2023-03-31
- Publication Date
- 2026-07-17
AI Technical Summary
In tunnel traffic jams, when GPS and radar data are temporarily unavailable, the autonomous vehicle's attitude will drift, causing the positioning system to be unable to suppress the attitude drift.
By acquiring vehicle driving datasets and images to be processed, vehicle driving motion information is extracted, images are optimized to obtain a candidate camera pose information set, and the candidate camera pose information set is used to optimize the vehicle driving dataset to obtain target pose information and suppress the attitude drift of autonomous vehicles.
Without reducing the fusion output frequency of the positioning system, the attitude drift of the autonomous vehicle is effectively suppressed, thus meeting the real-time positioning requirements of the autonomous vehicle.
Smart Images
Figure CN116363173B_ABST