Stereo Vision-Based Optimal Path Planning with Stochastic Maps for Mobile Robot Navigation

摘要

This paper addresses the problem of stereo vision-based environment mapping and optimal path planning for an autonomous mobile robot by using the methodology of getting 3D point cloud from disparity images, its transformation to 2D stochastic navigation map with occupancy grid cell values assigned from the set {obstacle, unoccupied, occupied}. We re-examined and extended this methodology with a combination of A-Star with binary heap algorithms for obstacle avoidance and indoor navigation of the Innopolis autonomous mobile robot with ZED stereo camera.