Motion Planning Based on Learning From Demonstration for Multiple-Segment Flexible Soft Robots Actuated by Electroactive Polymers

TLDR

Multiple‑segment flexible and soft robotic arms composed of ionic polymer–metal composite (IPMC) actuators are compliant yet difficult to plan for due to redundant degrees of freedom, despite their promise for tasks such as navigating body cavities. We propose a learning‑from‑demonstration method that uses statistical machine‑learning algorithms to plan motion paths for IPMC‑based manipulators. The approach encodes demonstrated trajectories with Gaussian mixture model and Gaussian mixture regression, and derives forward and inverse kinematic models of the IPMC soft robotic arm for motion control. A six‑segment IPMC manipulator was built and successfully navigated a narrow keyhole, validating the learned paths.

Abstract

Multiple-segment flexible and soft robotic arms composed by ionic polymer--metal composite (IPMC) flexible actuators exhibit compliance but suffer from the difficulty of path planning due to their redundant degrees of freedom, although they are promising in complex tasks such as crossing body cavities to grasp objects. We propose a learning from demonstration method to plan the motion paths of IPMC-based manipulators, by statistics machine-learning algorithms. To encode demonstrated trajectories and estimate suitable paths for the manipulators to reproduce the task, models are built based on Gaussian mixture model and Gaussian mixture regression, respectively. The forward and inverse kinematic models of IPMC-based soft robotic arm are derived for the motion control. A flexible and soft robotic manipulator is implemented with six IPMC segments, and it verifies the learned paths by successfully completing a representative task of navigating through a narrow keyhole.