Modeling and control of McKibben artificial muscle enhanced with echo state networks

There has been a challenging work for using conventional techniques to model and control pneumatic artificial muscle (PM) due to poor knowledge and uncertainty of the process and/or complexity of the resulting mathematical model. Trying to deal with these problems, this study proposes a novel framework—Echo State Network (ESN) as a basis to implement the tasks in the PM's modeling and control. To describe the system dynamics and the external disturbance changes with time, the online ESN adaptation scheme is presented based on the recursive least squares (RLS) algorithm. Both simulation and experimental results show that the proposed procedure has better dynamic performance and strong robustness over the other typical/classical approaches.