Regresar

Performance analysis of a continuous vision-based control system for the navigation of a mobile robot

Abstract:

in the last year, we have proposed a new continuous visual navigation approach defining the navigation as a global visual control task. It implies continuity problems produced by the visibility changes of the image features during the navigation. To avoid the discontinuities in the control law, a smooth task function based on weighted features was proposed and a continuous control law was obtained and tested. in this paper, a performance analysis of this visual navigation approach is presented. First of all, the influence of the parameters selection in the weight functions is analyzed. The visual navigation approach presented in this paper is used to control the robot to repeat the walk learned before. The departure position of the robot must be near the initial position memorized before. The word near is in some way quantified in this paper. an exhaustive number of experiments using virtual reality worlds to simulate a typical indoor environment have been carried out to justify the affirmations made. Copyright © 2005 IFAC.