Aiming at the problems of the limitations of artificial landmark geometric relationships,poor robustness,and positioning failures in dynamic occlusion situations for the reflector localization algorithm based on multilateral strategy in indoor environments,the paper proposed a navigation and localization method based on improved correlative scan matching (CSM):by filtering the landmark reflection point clouds,the best match with the multi-resolution grid maps was searched hierarchically to improve the speed and accuracy of real-time robot positioning. Experimental results showed that the average positioning error of the improved CSM method under normal working conditions would be 2.571 mm,and it could also be maintained below 5.015 mm when some artificial landmarks are obstructed;furthermore,the proposed method could help complete synchronous positioning and map construction under complex and dynamic working conditions of obstacles.
indoor localizationartificial landmarksscan matchinglaser reflection navigationsimultaneous localization and mapping (SLAM)
NETL
NSTL
万方数据
