PIMBS: Efficient Body Schema Learning for Musculoskeletal Humanoids with Physics-Informed Neural Networks

TL;DR

This paper introduces PIMBS, a physics-informed neural network approach that efficiently learns the body schema of musculoskeletal humanoids, achieving high accuracy with limited data by incorporating physical laws.

Contribution

The study presents a novel PINNs-based method for musculoskeletal humanoid body schema learning that reduces data requirements and improves accuracy over traditional data-only approaches.

Findings

Effective learning with limited data demonstrated in simulation.

High-accuracy body schema learned in real humanoid experiments.

Physics laws improve learning efficiency and robustness.

Abstract

Musculoskeletal humanoids are robots that closely mimic the human musculoskeletal system, offering various advantages such as variable stiffness control, redundancy, and flexibility. However, their body structure is complex, and muscle paths often significantly deviate from geometric models. To address this, numerous studies have been conducted to learn body schema, particularly the relationships among joint angles, muscle tension, and muscle length. These studies typically rely solely on data collected from the actual robot, but this data collection process is labor-intensive, and learning becomes difficult when the amount of data is limited. Therefore, in this study, we propose a method that applies the concept of Physics-Informed Neural Networks (PINNs) to the learning of body schema in musculoskeletal humanoids, enabling high-accuracy learning even with a small amount of data. By utilizing not only data obtained from the actual robot but also the physical laws governing the relationship between torque and muscle tension under the assumption of correct joint structure, more efficient learning becomes possible. We apply the proposed method to both simulation and an actual musculoskeletal humanoid and discuss its effectiveness and characteristics.