Two-Dimensional Obstacle Avoidance Behavior Based on Three-Dimensional Environment Map for a Ground Vehicle

Hiroki Bingo, Hiroshi Ohtake
2018 Innovative Computing Information and Control Express Letters, Part B: Applications  
In order to drive robots automatically under various environments, the robots themselves have to recognize the surrounding environment, determine appropriate motions, and perform the determined motions. The robot has to have some functional capabilities such as creating the environment map and avoiding obstacles. In this paper, as the basic research for realizing our goal, we propose the obstacle avoidance method for a ground vehicle using three-dimensional Laplacian potential method based on
more » ... e information of self-localization and three-dimensional environment map which are obtained by RGB-D camera and simultaneous localization and mapping (SLAM) algorithm. However, the method needs high computational cost. GPGPU method is employed for solving this problem. The experimental results demonstrate the utility of this approach.
doi:10.24507/icicelb.09.06.485 fatcat:nycybhxr7zethfplwbe4acoxn4