Parameter design and optimization of a flight attitude simulator system based on PILCO framework

被引：0
作者：
Yang Y.-F. [1 ]
Deng K. [2 ]
Zuo Y.-Q. [1 ]
Ban X.-J. [1 ]
Huang X.-L. [1 ]
机构：
[1] Center for Control Theory and Guidance Technology, Harbin Institute of Technology, Harbin
[2] Sichuan Academy of Aerospace Technology, Chengdu
来源：
Guangxue Jingmi Gongcheng/Optics and Precision Engineering | 2019年 / 27卷 / 11期
关键词：
Aircraft control; Parameter optimization; Probabilistic Inference for Learning Control(PILCO); Reinforcement learning;
D O I：
10.3788/OPE.20192711.2365
中图分类号：
学科分类号：
摘要：
Proportional-integral-derivative (PID) controllers are widely used in flight control systems. However, it is often very cumbersome to adjust the parameters of a PID controller. In this study, we use Probabilistic Inference for Learning Control (PILCO) to optimize the parameters of a PID controller. As the first step, we develop a probabilistic dynamics model of the flight control system using input and output data. Next, the existing PID controller is evaluated using the policy evaluation method. Finally, the evaluated PID controller is optimized by policy update. The sampling frequency of the system is 100 Hz and the data acquisition time per round is 8 s. The optimized PID controller can achieve stable control post 10 rounds of offline training. Through PILCO optimization, the flight attitude simulator performed robustly in a fixed-point experiment, indicating that PILCO has tremendous potential in solving nonlinear control and parameter optimization problems. © 2019, Science Press. All right reserved.
引用
收藏
页码：2365 / 2373
页数：8
相关论文
共 14 条
  • [1] Gao J.Z., Jia H.G., Design of longitudinal control law for small fixed-wing UAV during auto landing, Opt. Precision Eng., 24, 7, pp. 1799-1806, (2016)
  • [2] Li D., Chen X.J., Xu Z.J., Gain adaptive sliding mode controller used for flight attitude control of MAV, Opt. Precision Eng., 21, 5, pp. 1183-1191, (2013)
  • [3] Yin H., Yang Y.F., Zhao Y., Self-tuning controller design for a 2-DOF flight attitude simulator, Electric Machines and Control, 22, 4, pp. 105-112, (2018)
  • [4] Busoniua L., Bruinb T.D., Tolicc D., Et al., Reinforcement learning for control: performance, stability, and deep approximators, Annual Reviews in Control, 5, pp. 1-18, (2018)
  • [5] Recht B., A tour of reinforcement learning: the view from continuous control
  • [6] Dong L., Guang-Hong Y., Model-free adaptive control design for nonlinear discrete-time processes with reinforcement learning techniques, International Journal of Systems Science, 49, 11, pp. 2298-2308, (2018)
  • [7] Levine S., Reinforcement learning and control as probabilistic inference: tutorial and review
  • [8] Zhang T., Research on Path Planning Method of Quadrotor UAV Based on Reinforcement Learning, (2018)
  • [9] Kaelbling L.P., Littman M.L., Moore A.W., Reinforcement learning: a survey, Artificial Intelligence Research, 4, 1, pp. 237-285, (1996)
  • [10] Chua K., Calandra R., McAllister R., Et al., Deep reinforcement learning in a handful of trials using probabilistic dynamics models
12