Vision and spatial modeling for mobile robots

Abstract

An analysis is conducted of a number of computer vision approaches that the author has been involved in constructing. These range from the ACRONYM model-based vision system through more recent mobile robot vision systems. In each case success has been achieved by treating vision as a source of constraint, rather than a deliverer of a world model. The way in which vision works is then entirely task dependent.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>