Learning Dynamic Tasks on a Large-scale Soft Robot in a Handful of Trials

TL;DR

This paper introduces a data-efficient Bayesian optimization method for controlling large-scale soft robots in dynamic tasks, bypassing complex modeling and enabling rapid learning from minimal trials.

Contribution

It presents a novel approach that directly optimizes control policies from pressure commands, avoiding detailed kinematic or dynamic models for large-scale soft robots.

Findings

Effective in simulated environments

Successful real-world experiments

Requires fewer trials than traditional methods

Abstract

Soft robots offer more flexibility, compliance, and adaptability than traditional rigid robots. They are also typically lighter and cheaper to manufacture. However, their use in real-world applications is limited due to modeling challenges and difficulties in integrating effective proprioceptive sensors. Large-scale soft robots ($\approx$ two meters in length) have greater modeling complexity due to increased inertia and related effects of gravity. Common efforts to ease these modeling difficulties such as assuming simple kinematic and dynamics models also limit the general capabilities of soft robots and are not applicable in tasks requiring fast, dynamic motion like throwing and hammering. To overcome these challenges, we propose a data-efficient Bayesian optimization-based approach for learning control policies for dynamic tasks on a large-scale soft robot. Our approach optimizes the task objective function directly from commanded pressures, without requiring approximate kinematics or dynamics as an intermediate step. We demonstrate the effectiveness of our approach through both simulated and real-world experiments.