Disturbance observer based MEMS gyroscope prediction performance non-singular sliding-mode control method

Abstract

The invention relates to a disturbance observer based MEMS gyroscope prediction performance non-singular sliding-mode control method. The method comprises the following steps: in view of a gyrodynamics model of parameter perturbation and external time-varying disturbance, designing an adaptive update law of neutral network weight, and correcting the weight coefficient of the neutral network to realize effective estimation of unknown dynamics; introducing a performance function to limit the tracking error, and converting the limited tracking error into unlimited conversion error through errorconversion; designing a sliding-mode controller based on conversion error to realize prediction performance control on MEMS gyroscope; designing a non-singular terminal sliding-mode controller to realize feed-forward compensation of unknown dynamics while avoiding the problem of system singularity; designing a disturbance observer to perform estimation compensation on external disturbance. The method disclosed by the invention solves the problem that external disturbance, system singularity, overshooting and tracking error cannot be designed in advance, and enhances the control precision and further improves the performance of the MEMS gyroscope.

Description

technical field [0001] The invention relates to a modal control method of a MEMS gyroscope, in particular to a non-singular sliding mode control method based on an interference observer with preset performance of the MEMS gyroscope, and belongs to the field of intelligent instruments. Background technique [0002] As a sensor for measuring angular motion, MEMS gyroscopes are greatly affected by environmental factors and external disturbances, and conventional PID control methods are difficult to achieve high-precision control of gyroscopes. In order to further improve the measurement accuracy of MEMS gyroscopes, "Slidingmode control of MEMS gyroscopes using composite learning" (Rui Zhang, TianyiShao, Wanliang Zhao, Aijun Li, Bin Xu, "Neurocomputing", 2018) proposes a MEMS gyroscope sliding mode self- Adapt to the control method. On the one hand, a neural network that adaptively adjusts parameters is constructed to predict the nonlinearity of the system; on the other hand, s...

AI Technical Summary

Problems solved by technology

[0004] In order to overcome the shortcomings of the existing technology that does not consider external disturbances, singularity of sliding mode control, and system overshoot and tracking error cannot be pre-designed, the present invention proposes a non-singular sliding mode control based on the preset performance of MEMS gyroscopes based on disturbance observers method

Images