Study-based guaranteed preset performance cluster flight distribution attitude tracking control method

Abstract

The invention discloses a study-based guaranteed preset performance cluster flight distribution attitude tracking control method. For an attitude tracking distributed control problem in cluster flight, under the condition of not needing spacecraft inertia parameter information (i.e. inertia matrix parameter information), a feedback controller based on a spacecraft state is designed, thus the prior design of transient and steady information of a spacecraft attitude tracking error system is obtained. At the same time, based on an adaptive dynamic programming, with the help of input and output data of a spacecraft, the design of a robust compensation controller is realized, and so the robustness and adaptability of a control system of the spacecraft in a cluster are improved. According to the method, under the observation condition of no need of the online identification of system unknown dynamics inertia parameter and unknown external disturbance, the tracking control of a cluster flight attitude can be realized. Based on online model-free learning characteristics of an ADP, the robustness of a controlled system for unknown uncertainties is greatly improved.

Description

【Technical field】 [0001] The invention relates to a robust distributed control technology for spacecraft formation clusters, in particular to a learning-based distributed attitude tracking control method for cluster flight with preset performance. 【Background technique】 [0002] In recent decades, formation swarm technology consisting of multiple small, low-cost spacecraft has developed rapidly. Compared with a single large spacecraft, formation cluster flight has the advantages of high robustness, safety and reliability, and low cost. It is an effective and feasible method to carry out various space tasks, such as space science research, earth observation, space debris monitoring, etc. [0003] For attitude tracking control of formation flight, the existing control methods are distributed optimal control based on state Riccati equation, finite-time attitude cooperative control based on terminal sliding mode control, distributed attitude control based on backstepping method...

AI Technical Summary

Problems solved by technology

However, due to the fuel consumption of the spacecraft itself and the unknown uncertainty of the space environment, it is often difficult to accurately estimate / measure the inertial matrix parameters of the spacecraft

Although the unknown nonlinear system model of the spacecraft can be identified with the help of neural network or fuzzy logic system, the design complexity of the distributed attitude controller based on neural network is relatively high, which limits its practical application.

Images