Mobile Robot Regulation With Image Based Visual Servoing

Abstract

The development of image processing algorithms grew exponentially over the past few decades with improvements in vision sensors and computational power. In this paper, a visual servo controller is designed and developed using the image-based method for a differential drive robot. The objective is to reach a desired pose relative to a target object placed in the world frame with four feature points. A full system model that includes the mobile base and camera is presented along with the design of a proportional controller. The system is implemented in the Husky A200 Robot by Clearpath Robotics. MATLAB Simulation and experimental results are analyzed and discussed with conclusion and future works recommendation in the end.