Efficient Jacobian-Based Inverse Kinematics with Sim-to-Real Transfer of Soft Robots by Learning

TL;DR

This paper introduces a learning-based method using neural networks to efficiently solve inverse kinematics for soft robots, incorporating a sim-to-real transfer to enhance practical applicability and accuracy.

Contribution

The work develops a neural network approach to learn forward kinematics and Jacobians for soft robots, enabling efficient IK solutions with a novel sim-to-real transfer strategy.

Findings

Effective IK solution for soft robots demonstrated

Successful sim-to-real transfer with limited real data

Path following and positioning validated on pneumatic soft robots

Abstract

This paper presents an efficient learning-based method to solve the inverse kinematic (IK) problem on soft robots with highly non-linear deformation. The major challenge of efficiently computing IK for such robots is due to the lack of analytical formulation for either forward or inverse kinematics. To address this challenge, we employ neural networks to learn both the mapping function of forward kinematics and also the Jacobian of this function. As a result, Jacobian-based iteration can be applied to solve the IK problem. A sim-to-real training transfer strategy is conducted to make this approach more practical. We first generate a large number of samples in a simulation environment for learning both the kinematic and the Jacobian networks of a soft robot design. Thereafter, a sim-to-real layer of differentiable neurons is employed to map the results of simulation to the physical hardware, where this sim-to-real layer can be learned from a very limited number of training samples generated on the hardware. The effectiveness of our approach has been verified on pneumatic-driven soft robots for path following and interactive positioning.