SVR based Obstacle Avoidance and Control of a Two Wheeled Mobile Robot

Abstract

Support vector regression (SVR) based obstacle avoidance and control of a two wheeled mobile robot is considered in this paper. Under the uncertain observation information, a method based on SVR technique is proposed for providing reliable control information in order to control the mobile robot by using a potential function, where the potential function is based on Lyapunov function and the flat plane is constructed to avoid the local minima for compensating the Lyapunov function. As a result, motion of the mobile robot can be controlled more accurately and smoothly. The proposed method is evaluated by demonstrating numerical simulation results.