Vision navigation of an autonomous vehicle by fuzzy reasoning

This paper presents a method for vision navigation of an autonomous vehicle based on fuzzy reasoning. Autonomous vehicles operating in real world require fast image processing and robustness with respect to noisy sensor readings and with respect to environment changes. The key subject of this paper is the integration of some special domain knowledge into a fuzzy rule base in order to enhance image segmentation performance. The proposed method is applied to navigate the THMR-III autonomous vehicle in out-door environments.