A Distributed Reinforcement Learning Guidance Method under Impact Angle Constraints

被引：0
|
作者：
Li B. [1 ,2 ,3 ]
An X. [1 ,2 ,3 ]
Yang X. [1 ,2 ,3 ]
Wu Y. [1 ,2 ,3 ]
Li G. [4 ]
机构：
[1] State Key Laboratory of Virtual Reality Technology and System, Beihang University, Beijing
[2] School of Automation Science and Electrical Engineering, Beihang University, Beijing
[3] Science and Technology on Aircraft Control Laboratory, Beijing
[4] School of Astronautics, Northwestern Polytechnical University, Xi'an
来源：
Yuhang Xuebao/Journal of Astronautics | 2022年 / 43卷 / 08期
关键词：
Gradient algorithm; Impact angle; Missile guidance; Reinforcement learning;
D O I：
10.3873/j.issn.1000-1328.2022.08.008
中图分类号：
学科分类号：
摘要：
In order to improve the target hitting effect of missile with the impact angle fixed, a distributed reinforcement learning guidance strategy based on deep deterministic policy gradient algorithm is proposed. To minimize the impact angle error, a new reward function is designed to make the line of sight angle converge to the expected value while meeting the field of view angle constraint. In addition, in order to enhance the generalization ability of the reinforcement learning model, a distributed exploration strategy is proposed to improve the efficiency of environment exploration during model training. The simulation results verify that the proposed distributed reinforcement learning guidance method can achieve accurate attack on the target under the constraint of fixed impact angle. Compared with the traditional guidance law, the impact angle error of the proposed guidance law is smaller and the convergence rate is faster. © 2022 China Spaceflight Society. All rights reserved.
引用
收藏
页码：1061 / 1069
页数：8
相关论文
共 23 条
  • [1] HU Q, HAN T, XIN M., Three-dimensional guidance for various target motions with terminal angle constraints using twisting control!, J IEEE Transactions on Industrial Electronics, 67, 2, pp. 1242-1253, (2019)
  • [2] SEO M, LEE C, TAHK M., New design methodology for impact angle control guidance for various missile and target motions J |, IEEE Transactions on Control Systems Technology, 26, 6, pp. 2190-2197, (2017)
  • [3] EAN S P, QI Q, LU K, Et al., Autonomous collision avoidance technique of cruise missiles based on modified artificial potential method ∗ J ], Transaction of Beijing Institute of Technology, 38, 8, pp. 828-834, (2018)
  • [4] WANG Liguo, MA Guoxin, JIAO Yongkang, Cooperative interception guidance law design for multi-aircrafts formation with angular constraints [J], Journal of Naval Aeronautical and Astro- nautical University, 33, 3, pp. 289-296, (2018)
  • [5] LIU L, WANG D, PENG Z, Et al., Predictor-based los guidance law for path following of underactuated marine surface vehicles with sideslip) compensation∗ J], Ocean Engineering, 124, 15, pp. 340-348, (2016)
  • [6] GUO H, FU W, FU B, Et al., Smart homing guidance strategy with control saturation against a cooperative target-defender team [J], Journal of Systems Engineering and Electronics, 30, 2, pp. 366-383, (2019)
  • [7] ZENG Yaohua, GU Yongyan, LI Juan, Terminal guidance law design with view angle and impact angle constraints associated with attack angle[J], Aerospace Control, 36, 1, pp. 14-18, (2018)
  • [8] LI G, WU Y., Nonsingular adaptive-gain super-twisting guidance with an impact angle constraint ∗ J ], Proceedings of the Institution of Mechanical Engineers Part G-Journal of Aerospace Engin eering, 233, 5, pp. 1705-1714, (2019)
  • [9] ZHANG W, FU S, LI W, Et al., An impact angle constraint integral sliding mode guidance law for maneuvering targets inter ception ∗ J ], Journal of Systems Engineering and Electronics, 31, 1, pp. 168-184, (2020)
  • [10] LI Bo, YUE Kaiqiang, PAN Zhigang, Et al., Multi UAV cooperative autonomous navigation based on multi agent deep deterministic policy gradient∗ J], Journal of Astronautics, 42, 6, pp. 757-765, (2021)
123