Miniature magnetic helical microrobot. Credit: SIAT
The usage of magnetic microrobots with miniature size and the quality of swimming successful liquids with debased Reynolds numbers is promising successful targeted therapy, since these robots tin determination successful constrictive environments flexibly.
However, owed to the impacts of the nonlinearity and diverseness of the desired analyzable trajectories, it is simply a situation to warrant the microrobot tracking accuracy without predominant controller adjustment.
Recently, a probe squad led by Xu Sheng and Xu Tiantian from the Shenzhen Institute of Advanced Technology of the Chinese Academy of Sciences has developed a caller method based connected wide learning strategy (BLS), which tin recognize close and flexible trajectory tracking of microrobot.
The survey was published successful IEEE Transactions connected Cybernetics connected Nov. 1.
Compared with the accepted heavy learning method, BLS—which features a flexible and elemental structure—could execute satisfactory accuracy.
The probe squad developed a learning-based microrobot servo power algorithm utilizing BLS, and combined the Lyapunov mentation with the analyzable learning method to deduce the controller parameters' constraints.
They besides developed a BLS-based controller grooming algorithm that uses aggregate desired tracking trajectories arsenic demonstrations, and obtained the controller parameters by the grooming algorithm.
According to the simulation and experimental results, this BLS-based method had a accelerated grooming speed, taking lone astir 6 seconds.
The trained BLS-based controller tin way the trajectories with antithetic shapes and velocities with amended accuracy, and it needs nary parameter re-adjustment owed to its beardown quality of generalization. Furthermore, since the BLS method is applied, the fig of nodes tin beryllium flexibly adjusted erstwhile caller demonstrations are required.
More information: Sheng Xu et al, A Learning-Based Stable Servo Control Strategy Using Broad Learning System Applied for Microrobotic Control, IEEE Transactions connected Cybernetics (2021). DOI: 10.1109/TCYB.2021.3121080
Citation: New microrobotic trajectory tracking method utilizing wide learning strategy (2021, November 19) retrieved 20 November 2021 from https://techxplore.com/news/2021-11-microrobotic-trajectory-tracking-method-broad.html
This papers is taxable to copyright. Apart from immoderate just dealing for the intent of backstage survey oregon research, no portion whitethorn beryllium reproduced without the written permission. The contented is provided for accusation purposes only.
English (US) ·