Hybrid Adaptive Disturbance Rejection Control for Inflatable Robotic Arms.

This paper aims to tackle the controller design issue of highly nonlinear and stochastic inflatable robotic arms (IRAs). A novel control scheme, i.e., hybrid adaptive disturbance rejection control (HADRC), is devised to handle the challenging tracking control of hard-to-model IRAs. The model-free adaptive control (MFAC) that linearizes the dynamics by leveraging solely the online input and output (I/O) data of plants is analytically enhanced for superlinear convergence. Both internal and external disturbances are rejected via the active disturbance rejection control (ADRC) that requires little prior model information. The fuzzy logic control (FLC) is subsequently implemented to correlate the two sub-controllers and contribute to attaining smooth motions. The superiority of the proposed scheme is demonstrated by the comparative simulations and experiments on a 2-degree-of-freedom (DOF) IRA.