A new 3D map reconstruction based mobile robot navigation

In order to build a precise 3D map of outdoor environment, a novel map reconstruction algorithm is proposed based on a mobile robot. The robot used a four-layer laser radar (LD_ML) to obtain space information and used a DGPS/INS Difference GPS/Inertial Navigation System position system to obtain itself pose and position information. With the aim of building a complete 3D grid map, the range data of environment need be changed into a world coordinate system with the simultaneous position data of the robot. Since the outdoor environment is always complicated, the 3D grid map is built by updating z-coordinate value using obstacle detection results. And the obstacle detection is achieved by LD_ML according to the gradient and height information of detection regions. The results of 3D grip map reconstruction testified the 3D map reconstruction algorithm was reliable and stable.