LS-SVM based motion control of mobile robot with Lyapunov function compensation

In this paper, least squares support vector machine (LS-SVM) based motion control and obstacle avoidance of a mobile robot are proposed under measured data with uncertainty. The proposed scheme can control the mobile robot by using a potential field function, where the potential field function is based on Lyapunov function candidate. The Lyapunov function candidate includes a flat plane, the flat plane is constructed to avoid the local minima. LS-SVM method is used for estimating the controller parameters by using the measured data with uncertainty. As a result, the motion of the mobile robot can be controlled smoothly. The proposed scheme is evaluated by demonstrating numerical simulation results.