Output Second-order Sliding-mode Control for a Gecko Biomimetic Climbing Robot

Abstract

The aim of this study is to propose the output controller to solve the execution of the gait sequence for Gecko Bio-inspired Robotic Devices (GBRDs). The Twisting Controller (TC) serves as basis of the tracking trajectory designed algorithm. The TC uses both the tracking error and its estimated derivative, which is calculated by a set of distributed Super-Twisting Algorithms (STAs). Each STA is implemented as a robust and exact differentiator. The output-based controller structure corresponds to a sort of decentralized form for robotic devices. Consequently, each articulation is controlled by an independent TC. A set of proposed references trajectories reproduce the gait cycle of a Real Gecko-Lizard (RG-L). The reference trajectories are proposed as the superposition of sigmoid functions ful-filling the conditions of the Bezzier polynomials. Numerical simulations evaluate the GBRD movement enforced by the suggested controller in the horizontal as well as the vertical gaits. An own 3D printed GBRD is the experimental platform aimed to test the distributed controller. Two controlled vacuum pumps are used to adhere the GBRD to the wall surfaces. A set of experimental validations confirm the robustness and the reliability of the proposed controller when its performance is compared with classical output feedback controllers.