Finite-Time Distributed Observer-Based Robust Adaptive Dynamic Programming for Synchronous Control of Quadrotor Team

The synchronous control of a quadrotor team is a challenging issue due to disturbances and unknown reference information. This article investigates a robust adaptive dynamic programming method to achieve optimal synchronous control of a quadrotor team. A finite-time distributed observer is designed to estimate the reference position and reference velocity. The finite-time estimation of the reference state results in the decentralized optimal control design. An adaptive dynamic programming based algorithm is developed to solve a numerical solution to the Hamilton-Jacobi-Isaacs equation in a data driven manner. By solving the Hamilton-Jacobi-Isaacs equation, the analytical optimal control law is obtained. To ad dress the disturbances, an integral sliding mode control scheme and an H<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">∞</sub> technique are integrated to attenuate bounded matched disturbance and unmatched disturbance, respectively. Compared with current adaptive dynamic programming-based method, specialized activation functions are designed to approximate the optimal value function, control law, and disturbance law. Finally, numerical results verify the effectiveness of the proposed method.