基于小视场固态激光雷达的融合定位研究OA

Solid-state LiDAR with narrow fields of view(FOV)faces challenges in GNSS-denied environments due to irregular scanning patterns and limited FOV,including poor feature extraction and low matching robustness.To address these limitations,this paper proposes a real-time robust localization algorithm that fuses LiDAR and an inertial measurement unit(IMU).The algorithm optimizes point cloud processing through scan-line reconstruction and adaptive spatial geometric feature extraction,employs an unscented Kalman filter(UKF)for efficient state estimation,and enhances matching accuracy using pre-built maps.Experimental results demonstrate that the algorithm achieves a maximum positioning error of 0.22 m(mean 0.14 m)in subway tunnel scenarios and further reduced errors to 0.09 m(mean 0.07 m)in open autonomous-rail rapid transit(ART)environments,with per-frame computation time below 50 ms,meeting real-time requirements.The research findings provide an effective solution for high-precision localization in intelligent and autonomous operation applications in the rail transit sector.