Line of Sight Control Strategies for Inertial Stabilization Platforms: Comparative Study

In this paper, a double-axis gimbal system with two degrees of freedom, one over azimuth and the other over elevation, is considered. The system model is derived using Newton-Euler approach considering the base angular rotation and neglecting dynamic mass unbalance. Two controllers are designed to achieve line of sight (LOS) stabilization with minimum error in the presence of reference tracking and disturbance rejection. The first one is a Proportional Integral Derivative (PID) controller with low-pass filtering on the derivative term along with anti-reset windup on the integral term and is tuned based on minimization of the integral of time-weighted absolute error (ITAE) method. The second one is a linear Model Predictive Controller (MPC) which is designed using the gimbal system model and an optimization algorithm to minimize the cost function using the control signal. The performance of both controllers is compared in terms of reference tracking and disturbance rejection. Simulation results demonstrate that the designed controllers are capable of minimizing the error in the presence of reference tracking and disturbance rejection. However, the MPC outperformed the PID with less rise time, overshoot percentage, rms error and settling time.