车辆信息丢失的目标跟踪方法、系统、介质及设备
By combining lane information with surrounding target information to predict newly appearing targets, the problem of information loss caused by sensor occlusion in autonomous driving is solved, and more accurate target tracking and association are achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- 上海中海庭数智技术有限公司
- Filing Date
- 2022-12-26
- Publication Date
- 2026-07-17
AI Technical Summary
In existing technologies, sensors are prone to information loss or becoming new targets due to occlusion when tracking targets, and there is a lack of effective solutions, especially in autonomous driving, where the loss and re-association of vehicle information are inaccurate.
By combining actual road conditions with lane information and surrounding related target information, newly appearing targets can be predicted, thereby enhancing tracking accuracy. The location of lost targets can be predicted by using the vehicle's lane information and the correlation information of surrounding targets, and then associated with newly appearing targets.
It improves the predictive accuracy and reliability of target tracking, reduces information loss, and enhances target association capabilities in occluded and complex environments.
Smart Images
Figure CN116259033B_ABST