Enhancing fuzzy robot navigation systems by mimicking human visual perception of natural terrain traversability

Abstract
©2001 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/ republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works.Presented at the Joint 9th IFSA World Congress and 20th NAFIPS International Conference, Vancouver, Canada, July 2001.DOI: 10.1109/NAFIPS.2001.944218This paper presents a technique for learning to assess terrain traversability for outdoor mobile robot navigation using human-embedded logic and real-time perception of terrain features extracted from image data. The methodology utilizes a fuzzy logic framework and vision algorithms for analysis of the terrain. The terrain assessment and learning methodology is tested and validated with a set of realworld image data acquired by an onboard vision system