Dynamic Control of Soft Robotic Arm: An Experimental Study

TL;DR

This paper presents an experimental study on a soft robotic arm using a new adaptive passivity control strategy, demonstrating improved trajectory tracking accuracy over traditional methods across various scenarios.

Contribution

The study introduces a novel adaptive passivity control method for soft robots, validated through extensive experiments showing enhanced tracking performance.

Findings

Adaptive passivity control outperforms PD control in accuracy

Effective across different speeds and disturbances

Identifies strengths and limitations of the proposed method

Abstract

In this paper, a reinforced soft robot prototype with a custom-designed actuator-space string encoder are created to investigate dynamic soft robotic trajectory tracking. The soft robot prototype embedded with the proposed adaptive passivity control and efficient dynamic model make the challenging trajectory tracking tasks possible. We focus on the exploration of tracking accuracy as well as the full potential of the proposed control strategy by performing experimental validations at different operation scenarios: various tracking speed and external disturbance. In all experimental scenarios, the proposed adaptive passivity control outperforms the conventional PD feedback linearization control. The experimental analysis details the advantage and shortcoming of the proposed approach, and points out the next steps for future soft robot dynamic control.