摘要
针对变体无人机的控制问题,给出了Q学习控制方法。首先根据设计任务要求设计控制律控制变体无人机按给定航路完成飞行。同时根据飞行环境和飞行任务的变化,利用Q学习方法控制变体飞行器相应地改变外形(平直翼、小前掠翼、大前掠翼),使变体飞行器能始终保持最优飞行状态,以满足在变化很大的飞行环境里执行多种任务(如巡航、机动、盘旋、攻击等)的要求。仿真结果验证了该方法的正确性和有效性。
This paper develops a control methodology for morphing, which combines Q-Learning and PID Control. Sections 1 and 2 of the full paper explain our design mentioned in the title, which we believe is feasible or effective and whose core consists of： ＂The morphing control function, which uses Q-Learning, is integrated with the trajecto- ry tracking function, which uses PID Control. Optimality is addressed by cost functions representing optimal shapes corresponding to specified operating conditions, and an episodic ＇ reinforcement learning＇ simulation is developed to learn the optimal shape change policy. The methodology is demonstrated by a numerical example of a morphing air vehicle, which simultaneously tracks a specified trajectory and autonomously morphs over a set of shapes corre- sponding to flight conditions along the trajectory. ＂ Simulation results, presented in Figs. 4 and 5, and their analysis show preliminarily that this methodology is capable of learning the required shape and morphing into it and accu- rately tracking the reference trajectory, thus showing that our design is indeed feasible.
出处
《西北工业大学学报》
EI
CAS
CSCD
北大核心
2012年第3期340-344,共5页
Journal of Northwestern Polytechnical University
基金
航空科学基金(200907S3007)
国家国际科技合作专项基金(2011DFR81070)资助
关键词
变体无人机
Q学习
控制
control, design, efficiency,reinforcement learning, tracking（cles）
flowcharting, morphing UAV, Q-learning, simulationposition） , UAV （ unmanned aerial vehi-
