%0 Journal Article %T 基于强化学习的小型无人直升机有限时间收敛控制设计 %T Finite time control based on reinforcement learning for a small-size unmanned helicopter %A 鲜斌,林嘉裕 %A XIAN,Bin %A LIN,Jia yu %J 控制与决策 %J Control and Decision %@ 1001-0920 %V 35 %N 11 %D 2020 %P 2646-2652 %K 无人直升机;强化学习;鲁棒控制;未知外部扰动;有限时间收敛;实验验证 %K helicopter;reinforcement learning;robust control;unknown external disturbances;finite time convergence;experimental verification %X 针对小型无人直升机精确动力学模型难以获取以及姿态控制易受未知外界风扰影响的问题,设计一种基于强化学习(reinforcement learning,RL)与super twisting相结合的非线性控制算法.利用直升机在线飞行数据,训练执行者-评价者(actor-critic,AC)网络以逼近系统建模不确定部分.为了抑制未知外界风扰,提高系统鲁棒性,同时补偿AC网络逼近误差,设计基于super twisting的鲁棒控制算法.进而,利用Lyapunov稳定性分析方法证明无人直升机姿态误差能在有限时间内收敛到零.最后对所提出的算法进行实验验证,实验结果表明,所提出算法具有良好的控制效果,对系统不确定性和外界扰动具有良好的鲁棒性. %X This paper presents a nonlinear control law based on the combination of reinforcement learning(RL) and super twisting methodology for the attitude control of a small-size unmanned helicopter, which is subjected to modeling uncertainties and unknown external disturbances. The proposed control law only uses input and output data of the helicopter to train the actor-critic(AC) neural networks to compensate for modeling uncertainties. Then a nonlinear robust controller based on super twisting methodology is developed to compensate for the unknown external disturbances. The Lyapunov based stability analysis is used to prove that the attitude error of a unmanned helicopter can converge to zero in finite time. Finally, the proposed control law is verified on a self-built hardware in the loop testbed. The experimental results show that the proposed control law can achieve good control performance together with good robustness for modeling uncertainties and wind disturbances. %R 10.13195/j.kzyjc.2019.0328 %U http://kzyjc.alljournals.cn/kzyjc/home %1 JIS Version 3.0.0