Visual Servoing of a Wheeled Mobile Robot using Unconstrained Optimization with a Ceiling Mounted Camera

Abstract

We propose an image-based visual servoing algorithm of a wheeled mobile robot utilizing a ceiling mounted camera. The visual navigation of a mobile robot is defined as an unconstrained optimization problem to minimize the image difference between the goal position and the position of a mobile robot in the image plane. Full New ton's method is applied for unconstrained optimization. We also propose a method to find the Hessian term of the image error using a secant approximation method. The performance of visual servoing was evaluated using the simulation.