Visual guidance control based on velocity mapping

Abstract

A control scheme is proposed for an autonomous vehicle to track a moving object using a CCD camera. In this scheme, visual velocity mapping is exploited for guidance control of the vehicle. The advantage of this approach is that the nonlinear calibration as well as coordinate calculation is much simplified for vision-based navigation. Therefore, fast response to the immediate environment can be obtained. To reduce the computation burden of the on-board computer, a frame-rate image board was developed to provide edge information for guidance control of the autonomous vehicle. Experimental results show that the developed system can track a moving person in the laboratory environment. Copyright (C) 1998 IFAC.