基于MPC的无人机航迹跟踪控制器设计

针对固定翼无人机航迹跟踪问题,采用基于状态扩展的双反馈模型预测控制理论对控制器进行设计。首先推导基于侧向偏差的无人机侧向航迹跟踪模型,采用动态逆方法对模型进行线性化处理,在此基础上设计基于状态扩展的双反馈模型预测控制器,并采用量子粒子群优化(quantum particle swarm optimization, QPSO)算法对控制器参数进行优化,考虑无人机飞行过程中受到的未知干扰,引入扩张状态观测器(extended states observer, ESO)对干扰进行观测,进一步提高系统的鲁棒性,并结合实际工程应用对系统进行数学仿真。仿真结果表明,基于状态扩展双反馈模型预测控制的无人机侧向航迹跟踪控制器,能够在系统存在模型不确定性与受到动态干扰时对期望航迹进行准确、稳定的跟踪。

Design of trajectory tracking controller for UAV based on MPC

Aiming at the problem of trajectory tracking of fixed-wing unmanned aerial vehicle, the controller is designed by using the dual-feedback model predictive control theory based on state expansion. Firstly, the unmanned aerial vehicle side trajectory tracking model based on the lateral deviation is derived, and the dynamic inverse method is used to linearize the model. Based on this, a dual feedback model predictive controller based on state expansion is designed, and the parameters of the controller are optimized using quantum particle swarm optimization (QPSO). Then, considering the unknown interference encountered during the flight, the extended states observer (ESO) is introduced to observe the interference, which further improves the robustness of the system. Finally, the system is mathematically simulated in combination with actual engineering applications. Simulation results show that the side trajectory tracking controller of unmanned aerial vehicle based on the state expansion dual feedback model predictive control can accurately and stably track the expected trajectory when the system has model uncertainty and is subject to dynamic interference.