Development of Control System for Inverted Wheeled Robot on a Ship: Predicting Ship´s Rolling Motion

This paper proposes a predictive control system for balancing an inverted wheeled robot on a ship using both the autoregressive model and a Kalman filter. In general, the control system of inverted wheeled robots is designed on the assumption that the ground does not move. However, a ship has six degrees of freedom. Therefore, a method of predicting the ship´s motion was required in order to control the balancing inverted wheeled robot on the ship. This study was focused on the ship´s rolling motion. On the ship, the rolling motion was predicted by the autoregressive model and a Kalman filter. Then, the inverted wheeled robot was controlled by adding accelerations to the inverted wheeled robot for canceling out the inertia forces of the predicted rolling motion. The effectiveness of the proposed control system was indicated with experimental results.