Magneto-mechanical actuation model for fin-based locomotion

TL;DR

This paper presents a simplified magneto-mechanical model for fin-based robotic locomotion, enabling direct fin bending with potential for system optimization and minimal actuation control.

Contribution

It introduces a coupled elastic-fin and magnetic actuator model, analyzing limit cycles for fin bending and thrust generation, aiding initial parameter estimation.

Findings

System's frequency and amplitude depend on fin stiffness and magnetic forcing

Limit cycle analysis reveals conditions for periodic fin bending

Model aids in preliminary design and control of magneto-mechanical actuators

Abstract

In this paper, we report the results from the analysis of a numerical model used for the design of a magnetic linear actuator with applications to fin-based locomotion. Most of the current robotic fish generate bending motion using rotary motors which implies at least one mechanical conversion of the motion. We seek a solution that directly bends the fin and, at the same time, is able to exploit the magneto-mechanical properties of the fin material. This strong fin-actuator coupling blends the actuator and the body of the robot, allowing cross optimization of the system's elements. We study a simplified model of an elastic element, a spring-mass system representing a flexible fin, subjected to nonlinear forcing, emulating magnetic interaction. The dynamics of the system is studied under unforced and periodic forcing conditions. The analysis is focused on the limit cycles present in the system, which allows the periodic bending of the fin and the generation of thrust. The frequency, maximum amplitude and center of the periodic orbits (offset of the bending) depend directly on the stiffness of the fin and the intensity of the forcing; we use this dependency to sketch a simple parameter controller. Although the model is strongly simplified, it provides means to estimate first values of the parameters for this kind of actuator and it is useful to evaluate the feasibility of minimal actuation control of such systems.