Neural network dynamic surface control method for under-actuated rope-system compound system

Abstract

The invention discloses a neural network dynamic surface control method for an under-actuated rope-system compound system. The neural network dynamic surface control method comprises the following steps that the post-arresting spatial rope-system compound system is decomposed into an under-actuated subsystem Xi and a fully-actuated subsystem Xi, and a dynamical model of the under-actuated subsystem Xi and the fully-actuated subsystem Xi of a rope system is built; the under-actuated subsystem Xi and the fully-actuated subsystem Xi have coupling relationships that aiming at control over the disturbed under-actuated subsystem Xi, a virtual control law is designed through a neural network and PD control, and an expected augmentation tracking trajectory is constructed; andaiming at the disturbed fully-actuated subsystem Xi with constrained control input, the expected augmentation tracking trajectory serves as the tracking amount, and a neural network dynamic surfacecontrol law is designed. The neural network dynamic surface control method can be used for solving the stable control problem of a rope-system compound under the situations of external disturbance and saturated control input and the control problem of the under-actuated rope-system system.

Description

technical field [0001] The invention relates to a neural network dynamic surface control method for an underactuated tether complex system after target capture, belonging to the category of tether spacecraft stability control. Background technique [0002] The tethered spacecraft is a complex system formed by connecting the space platform with the spacecraft and the grabber (flying tongue, flying spear, flying net, etc.) through a light and soft tether. Tethered spacecraft can be used to complete orbital debris (such as space debris, rocket upper stages, etc.) clearance, in-orbit maintenance of failed spacecraft and other space tasks. The above-mentioned tasks need to be performed under the premise that the attitude of the complex system is stable. When the grabber captures the high-speed rotating unstable non-cooperative target and forms a complex, in order to achieve the attitude stability of the complex, the tension control device located on the space platform can be use...

AI Technical Summary

Problems solved by technology

Mainly focused on the following aspects: 1), Sun Guanghui from Harbin Institute of Technology, etc., and Zhu Zhenghong from York University in Canada designed a fractional order control law to realize the control of the tethered satellite "dumbbell body" underactuated model; 2), Nanjing Wen Hao of the University of Aeronautics and Astronautics studied the stability control of the tethered satellite during the recovery / release process of the electrodynamic tether, and designed the corresponding tether tension control force and current control law, but only considered the tether’s in-plane angle and out-of-plane angle. and the length of the tether, ignoring the attitude of the satellite; 3), the team of Professor Huang Panfeng of Northwestern Polytechnical University studied the stability control of the complex after the tether robot target capture, considering the three-dimensional attitude of the complex after capture , but it assumes that the post-capture complex system is an all-drive system

Images