Neural Control Design of Robust Resonance Entrainment and Experimental Validation with Elastic Flapper

In engineering designs of mechanical systems, efficiency can often be increased by exploiting resonance. Motivated by biological control mechanisms of neuronal circuits called central pattern generator (CPG), recent studies have shown that a controller with a CPG structure can be designed to detect and achieve resonance modes of oscillation via sensory feedback. The objective of this study is to extract the essential dynamics of the CPG control and propose a nonlinear damping compensator for robust resonance entrainment. Existence and properties of stable oscillations are examined by the harmonic balance method, based on which a graphical design method is also proposed to construct the feedback control system. To experimentally validate the design method, a flapper mechanism for underwater vehicle applications is designed, built, and modeled, with open-loop frequency response tests characterizing its performance. Closed-loop oscillations for desired entrainment are then generated and analyzed. Experimental results in air and water both reveal that the resonance entrainment can be maintained robustly against parameter perturbations. Furthermore, we perform mode switching tests and mass-perturbation tests of the flapper prototype to demonstrate the utility and robustness of the proposed controller