Traversability Assessment of Terrain for Autonomous Robot Navigation

Abstract

Autonomous robot navigation with the capability to perceive the surrounding environment of the robot enhances the efficiency and safety of the robot. A technique for terrain classification for traversability assessment of mobile robot navigating in natural terrain by extracting the textural features from visual sensing of terrain data using co-occurrence matrix is presented in this paper. The algorithm uses crisp rule based (CRB) classifier to detect the navigable terrain thereby planning the optimum path for the robot.