Multi-Joint Physics-Informed Deep Learning Framework for Time-Efficient Inverse Dynamics

TL;DR

This paper introduces a physics-informed deep learning framework with a novel cross-attention module for efficient, label-free estimation of muscle activations and forces in multi-joint systems, improving inter-joint coordination modeling.

Contribution

The paper presents a new physics-informed deep learning model with a Multi-Joint Cross-Attention module that captures inter-joint coordination without needing labeled data.

Findings

Achieves performance comparable to supervised methods without ground-truth labels.

The MJCA module significantly improves inter-joint coordination modeling.

Enables time-efficient inference for muscle activation and force estimation.

Abstract

Time-efficient estimation of muscle activations and forces across multi-joint systems is critical for clinical assessment and assistive device control. However, conventional approaches are computationally expensive and lack a high-quality labeled dataset for multi-joint applications. To address these challenges, we propose a physics-informed deep learning framework that estimates muscle activations and forces directly from kinematics. The framework employs a novel Multi-Joint Cross-Attention (MJCA) module with Bidirectional Gated Recurrent Unit (BiGRU) layers to capture inter-joint coordination, enabling each joint to adaptively integrate motion information from others. By embedding multi-joint dynamics, inter-joint coupling, and external force interactions into the loss function, our Physics-Informed MJCA-BiGRU (PI-MJCA-BiGRU) delivers physiologically consistent predictions without labeled data while enabling time-efficient inference. Experimental validation on two datasets demonstrates that PI-MJCA-BiGRU achieves performance comparable to conventional supervised methods without requiring ground-truth labels, while the MJCA module significantly enhances inter-joint coordination modeling compared to other baseline architectures.