Development of a Modular and Submersible Soft Robotic Arm and Corresponding Learned Kinematics Models

TL;DR

This paper introduces a modular, 3D-printable, submersible soft robotic arm with hydraulic actuation, capable of multiple configurations, and presents initial learned kinematics models using deep neural networks for control.

Contribution

It develops a novel modular, waterproof soft robotic arm suitable for underwater use and provides the first learned kinematics models for such a system.

Findings

Successfully designed a modular, waterproof soft robotic arm

Demonstrated preliminary deep learning-based kinematics models

Enabled easy reconfiguration and actuator swapping

Abstract

Many soft-body organisms found in nature flourish underwater. Similarly, soft robots are potentially well-suited for underwater environments partly because the problematic effects of gravity, friction, and harmonic oscillations are less severe underwater. However, it remains a challenge to design, fabricate, waterproof, model, and control underwater soft robotic systems. Furthermore, submersible robots usually do not have configurable components because of the need for sealed electronics and mechanical elements. This work presents the development of a modular and submersible soft robotic arm driven by hydraulic actuators which consists of mostly 3D printable parts which can be assembled or modified in a relatively short amount of time. Its modular design enables multiple shape configurations and easy swapping of soft actuators. As a first step to exploring machine learning control algorithms on this system, we also present preliminary forward and inverse kinematics models developed using deep neural networks.