关键词: 时间协同, 最优误差动力学, 速度时变, 迁移学习, 指定时间收敛, 一致性动力学

Abstract:

This article addresses the problem of time-coordinated strike against weakly maneuvering targets. The implicit cooperative guidance problem based on the constant velocity motion model and the explicit cooperative guidance problem based on the velocity time-varying model are studied to enhance the breakthrough and destruction capabilities of anti-tank missiles against high-value targets. Firstly， based on the assumption of missile motion at a constant velocity， the idea of predictive correction and the optimal error dynamics theory are employed. Time-constrained bias terms are added to the proportional navigation guidance law， and a time-constrained guidance law is derived based on optimal error dynamics. An implicit time-coordinated guidance law scenarios depending weakly on communication is then proposed. To address the issue that inaccurate time prediction affects the performance of implicit cooperation caused by variable speed missile strikes， a neural network-based explicit time-coordinated guidance method is further proposed. This method adopts the predictive correction guidance framework， and designs the predictor by using transfer learning integrated neural networks according to the characteristics of proportional navigation guidance law to accurately predict the remaining flight time. Consistency dynamics with specified time convergence is proposed to coordinate the lengths of missile trajectories in the distributed communication architecture， ensuring that multiple missiles simultaneously hit the target. Simulation results show that the proposed implicit time-coordinated guidance method can ensure that multiple missiles simultaneously hit the target accurately， and the neural network-based explicit time-coordinated guidance method proposed can ensure coordinated strikes against multiple missiles under variable speed conditions.

Key words: time coordination, optimal error dynamics, velocity time-varying, transfer learning, specified time convergence, consistency dynamics