基于分布式并行伪谱-神经网络算法的双脉冲导弹多阶段协同轨迹优化  

Multi-stage Cooperative Trajectory Optimization of Dual-pulse Missile Based on Decentralized Parallel Pseudospectral-neural Network Algorithm

作　　者：刘超越 张成[1] LIU Chaoyue;ZHANG Cheng(Key Laboratory of Dynamics and Control of Flight Vehicle of Ministry of Education, Beijing Institute of Technology, Beijing 100081, China)

机构地区：[1]北京理工大学飞行器动力学与控制教育部重点实验室,北京100081

出　　处：《兵工学报》2020年第10期1988-2000,共13页Acta Armamentarii

基　　金：国家自然科学基金项目(11532002)。

摘　　要：为求解双脉冲导弹多阶段协同轨迹规划问题,并考虑将分离的发动机壳体投送至安全区域,提出一种基于高斯伪谱法和人工神经网络的分布式并行算法。针对双脉冲导弹两级脉冲的工作特点,将全弹道划分为发射段、爬升段、续航段和制导攻击段4个阶段;为预测分离发动机壳体的落点位置,建立射程预测函数,并利用人工神经网络对该函数进行离线拟合,以提高预测速度;在分布式并行算法中各导弹并行独立求解最优轨迹,并引入飞行时间下界约束保证导弹飞行时间一致性。通过两个仿真算例,将该分布式并行算法与集中式算法进行了比较,仿真结果表明,所提的分布式并行算法对于求解双脉冲导弹多阶段协同轨迹规划问题可得到更优的性能指标,以及更高的求解效率。A decentralized parallel algorithm based on Gauss pseudospectral method and artificial neural network is presented for the multi-stage cooperative trajectory optimization problem of dual-pulse missiles,in which considers dropping the detached engine shell to a safe area.According to the characteristic of two-stage pulse,the entire trajectory of dual-pulse missile is divided into four flight stages,such as launching,climb,endurance and attack.In order to predict the landing position of the detached engine shell,a range prediction function is established and is fitted off-line by artificial neural network to improve the prediction speed.In the decentralized parallel algorithm,each missile independently solves its optimal trajectory in parallel.The lower bound constraint of flight time is introduced to ensure the flight time consistency of missiles.The decentralized parallel algorithm is compared with the centralized algorithm by two simulation examples.The simulated results show that the decentralized parallel algorithm proposed in this paper can be used to obtain better performance index and higher solution efficiency for solving the multi-stage cooperative trajectory optimization problem of dual-pulse missiles.

关 键 词：双脉冲导弹 协同控制 多阶段轨迹优化 分布式并行算法 高斯伪谱法 神经网络 

分 类 号：TJ765.31[兵器科学与技术—武器系统与运用工程]