Physics-informed disturbance estimation and nonlinear controller design for a multi-axis gimbal system

The increasing demand for target tracking, environmental surveys, surveillance and mapping requires multi-axis gimbal systems with high tracking and stabilization performance. In this study, a new torque estimation structure is proposed to compute the complex disturbances negatively affecting the performance of the system. Two different control strategies based on Active Disturbance Rejection Control (ADRC) and Estimated Torque Model are implemented on a two-axis gimbal system. In the first strategy, the purpose is to improve the robustness, environmental adaptability and tracking accuracy of the system. The tuning effort of the ADRC is reduced by integrating a Neural Network (NN) based disturbance compensator (NN assisted ADRC). In the second strategy, NN is replaced with an Estimated Torque Model (ETM assisted ADRC), whose inputs come from plant outputs. The simulation results show that, both NN and ETM assisted control structures decrease the tracking errors. However, the improvements achieved by the physics-informed neural network-based estimator are more significant.