Tilt Plane Control System Design

Abstract

In the paper a tilt plane with parallel kinematics control system development is considered. The performance of the system is estimated executing the task called "ball on plane" when an unstable object on the plane like a ball must be held in the required position on the plane. The kinematics is designed with servomotors with improved internal positional control loop. The control system design is performed with virtual model and genetic algorithm optimization. The input of control system is reference position on the plane. Neuroevolution approach is used in comparison with classical linear control with nonlinear plant. The tilt table system is going to be used in education to improve student’s intuition in control systems and mechanics.