Simulation and Optimization of a Reference Adaptive Control Model for a Gough-Stewart Platform

This paper presents the model reference adaptive control (MRAC), a combination of a Gough-Stewart platform with an adaptive control strategy for force control. For this purpose, firstly, the platform geometry is described followed by solving the inverse kinematic equations using the so-called vector method. Afterward, the dynamic robot model is built utilizing the MSC Adams software, with its linear form generated by the control plugin of the software, where the model takes the force as input and returns the displacement as output. As a next step, an adaptive control strategy is formulated to control the position of the robot operator and then simulated in MATLAB software. The control strategy has two components, namely the feedback and feedforward control strategies. The adaptive law is calculated based on the Lyapunov theory. Finally, the controller parameters are optimized through a genetic algorithm based on integral time squared error (ITSE) to minimize the system error over time. The results proved that the proposed system outputs are in agreement with the corresponding values for the model reference and that the controller performs well in presence of usual levels of noise and/or disturbance.