Vision-Based Docking for Biomimetic Wheeled Robots

Authors:	Manchester Ian, University of New South Wales, Australia Savkin Andrey, University of New South Wales, Australia
Topic:	7.5 Intelligent Autonomous Vehicles
Session:	Mobile Robots III
Keywords:	Robotics, Vision, Biocybernetics

Abstract

We present a new control law for the problem of docking a wheeled robot at a certain location with a desired heading. Recent research into insect navigation has inspired a solution which uses just one environment sensor: a video camera. The control law is of the ``behavioral'' or ``reactive'' type, in that no attempt is made to observe the relative pose of robot and target, all control actions are based on immediate visual information. Knowledge of the distance to the target is not required. Docking success under certain conditions is proved mathematically, and simulation studies show the control law to be robust to camera calibration errors.