Toward domain-independent navigation: dynamic vision and control

Abstract

Outlines a set of problems associated with constructing a robust, domain-independent vision-based navigation system suitable for both structured and unstructured environments. The system utilizes visual tracking to monitor a set of automatically selected image features (markers), and employs vision-based control to guide the motion of the robot from the image trajectory of a set of markers. An environment is represented as a graph which may be constructed either under human control (e.g. by giving the system a tour) or autonomously as the system explores. In the paper, we review the system architecture and present two image-based mobile robot controllers for following visually-defined trajectories.