Neural Network Compensation Method for Path Tracking Control of a Spherical Mobile Robot

This paper presents a neural network compensation strategy for the path tracking control of a spherical mobile robot BHQ-2 including a pendulum with two degrees of freedom. Based on our previous work, we propose a simplified method to decompose the dynamics model of BHQ-2 to be two sub-dynamics models. Applying the fuzzy guidance control method and a neural network compensation strategy, a path tracking controller for robot BHQ-2 is designed.