Data-driven ship-borne rocket launcher optimal tracking control method

Abstract

The invention discloses a data-driven ship-borne rocket launcher optimal tracking control method, which combines a nonlinear internal model based on a neural network with adaptive dynamic programming to obtain an approximately optimal feedforward-feedback composite controller design method, and comprises the following two steps of: (1) utilizing output regulation and neural network theories to control the optimal tracking of a ship-borne rocket launcher; constructing a nonlinear neural network internal model, reconstructing a steady-state solution of an output regulation equation, and obtaining a feedforward controller; and (2) converting an optimal output adjustment problem into an optimal stabilization problem by using a steady-state solution of internal model reconstruction, then introducing an adaptive dynamic programming algorithm, and obtaining an approximate optimal feedback controller through algorithm iteration. According to the method, the approximate optimal controller can be obtained only by depending on the input-state data of the ship-borne rocket launcher launching device; a nonlinear internal model is introduced, uncertain parameters in the system can be processed, meanwhile, the influence of nonlinear disturbance such as external sea waves on the launching device can be resisted, the practicability is higher, and the application range is wider.

Description

technical field [0001] The invention relates to the technical field of shipboard weapon control, in particular to a data-driven optimal tracking control method for a shipboard rocket launcher launcher. Background technique [0002] With the rapid development of high technology and its wide application in the military, higher requirements are put forward for the control performance of the shipborne rocket launcher. However if figure 1 , due to the difficulty in obtaining an accurate system model of the launcher and the disturbance of the launcher by sea waves, the target striking performance of the system is reduced. In addition, ocean operations require more troublesome material supplies. Therefore, how to overcome the interference of the hull swing and realize the tracking control of the rocket launcher at the minimum cost without relying on the system model has become one of the problems that shipborne rocket launchers urgently need to overcome. . [0003] According to ...

AI Technical Summary

Problems solved by technology

[0004] The technical problem to be solved by the present invention is: to overcome the adverse effects of nonlinear disturbances such as unknown models and external waves on shipborne rocket launchers, and provide a data-driven The optimal tracking control method for rocket launcher launcher combines the internal model principle with adaptive dynamic programming, uses the internal model principle to reconstruct the state of the external system, and obtains the feedforward control items; uses the adaptive dynamic programming algorithm to design the optimal Feedback controller for target tracking at minimum cost

Images