Automatic vehicle light leveling system and method based on road surface semantics and slope joint estimation network
The automatic headlight leveling system, which uses a road surface semantic and slope joint estimation network, achieves forward-looking and precise adjustment of headlight beams. This solves the problems of lag and rigid control strategies in existing headlight leveling systems, and improves nighttime driving safety and system intelligence.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- 上海星宇智行技术有限公司
- Filing Date
- 2026-06-01
- Publication Date
- 2026-07-17
AI Technical Summary
Existing automatic headlight leveling systems cannot predict road surface undulations ahead, resulting in blind spots or glare. Furthermore, they cannot recognize changes in road surface material and vehicle posture, and their control strategies are simplistic and rigid, making it difficult to achieve refined and differentiated dimming, thus affecting nighttime driving safety.
An automatic headlight leveling system based on a road surface semantic and slope joint estimation network is adopted. The system synchronously collects multi-source heterogeneous data through the perception layer, the decision layer outputs road surface semantic segmentation results and slope estimation results through a multi-task road surface perception joint network, and the execution layer executes headlight control commands to achieve dynamic and precise adjustment of headlight pitch angle.
It enables forward-looking and precise adjustment of the headlight beam, eliminates blind spots and glare, improves nighttime driving safety and ride comfort, provides richer perception signals, and lays the foundation for headlight control and vehicle intelligence.
Smart Images
Figure CN122402371A_ABST