检测车辆的运动方向的方法、装置、电子设备和介质
By using visual sensing methods and neural network models to process ground patterns, the inaccuracy of traditional sensing methods in detecting vehicle movement direction in nighttime environments with ambiguous wheel directions is solved, enabling more accurate vehicle direction prediction and planning.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- ZF COMMERCIAL VEHICLE SYSTEMS (QINGDAO) CO LTD
- Filing Date
- 2022-12-29
- Publication Date
- 2026-07-17
AI Technical Summary
Existing sensing methods and devices cannot accurately detect the direction of vehicle movement in nighttime environments where wheel direction is unclear, resulting in low confidence and poor intuitiveness.
Using a visual sensing method, regions of interest (ROIs) are identified from ground maps based on vehicle velocity vectors. A neural network model is then used to process the ground maps, calculate the representation vector of the ROI, and associate it with the velocity vector. The trained neural network model is stored to calculate the direction of the velocity vector.
It improves the accuracy of vehicle direction detection, enabling early detection of changes in vehicle steering angle, reducing uncertainty, and improving the accuracy of experience-based direction guessing.
Smart Images
Figure CN116311109B_ABST