Adaptive-critic-design based tracking controller for boost complement glide vehicle with performance constraint and input saturation

Abstract

In order to achieve high precision and fast response flight control, an adaptive-critic-design (ACD) based tracking controller is proposed for boost complement glide vehicle (BCGV) with performance constraint and input saturation. The desired tracking performance is prescribed by error behavior boundary. Firstly, a tracking differentiator and a disturbance observer are proposed to allocate tracking command and resist disturbance, respectively. Then, a composite variable is constructed with the transformation mechanism and anti-saturation auxiliary variable, which quantifies the transformed tracking system with input saturation. Finally, a robust ACD with composite variable is proposed to enforce the performance constraint in a minimally invasive fashion, which optimizes the control actions with iterative learning online. Especially, the uncertainty is considered in the learning mechanism to improve the robustness. Besides, the proposed policy iteration is implemented by a single critic network to reduce the computing cost and complexity. Simulations show the effectiveness of the proposed controller and advantages compared with other methods.