Autonomous Land Vehicle Guidance in Complex Room Environments By Computer Vision Techniques

Abstract

An integrated approach to autonomous land vehicle (ALV) guidance in complex room environments by judging the conditions of routes to generate appropriate navigation paths using computer vision techniques is proposed. The judgement of the existence of a route or a collision is based on the information of route surfaces which is learned in advance. Techniques for model learning are proposed to extract the route features, record the information about turnings, and learn the park location environments. And techniques for path generation are proposed to generate pates among objects in rooms automatically and to guide the ALV to the central path of a route. Furthermore, three strategies for the ALV to park in desired locations are also proposed. Lots of successful experiments with a real ALV confirm the feasibility of the proposed approach.