Physics-Informed Learning for Human Whole-Body Kinematics Prediction via Sparse IMUs

2025-09-29

AI Summary

This work addresses the challenge of high-fidelity, physically plausible full-body human motion prediction using only five IMUs. Methodologically, it introduces a physics-guided deep learning framework that encodes forward and differential kinematics as differentiable loss terms; incorporates joint-state caching and iterative optimization during inference to jointly enforce kinematic constraints and dynamic consistency; and integrates spatiotemporal IMU features with motion priors, regularized by inverse kinematics. Experiments demonstrate that the approach significantly outperforms state-of-the-art methods in cross-subject generalization, achieving superior prediction accuracy, smooth motion transitions, and strong physical plausibility. Moreover, it satisfies the stringent real-time performance and safety requirements essential for human–robot collaboration.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Computer Vision: Motion & Tracking
• Intelligent Robots: Motion and Path Planning

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Accurate and physically feasible human motion prediction is crucial for safe and seamless human-robot collaboration. While recent advancements in human motion capture enable real-time pose estimation, the practical value of many existing approaches is limited by the lack of fu- ture predictions and consideration of physical constraints. Conventional motion prediction schemes rely heavily on past poses, which are not always available in real-world scenarios. To address these limitations, we present a physics-informed learning framework that integrates domain knowledge into both training and inference to predict human motion using inertial measurements from only 5 IMUs. We propose a network that accounts for the spatial characteristics of human movements. During training, we incorporate forward and differential kinematics functions as additional loss components to regularize the learned joint predictions. At the inference stage, we refine the prediction from the previous iteration to update a joint state buffer, which is used as extra inputs to the network. Experimental results demonstrate that our approach achieves high accuracy, smooth transitions between motions, and generalizes well to unseen subjects

Problem

• Predicting human motion using only five sparse IMUs
• Ensuring physical feasibility in whole-body kinematics prediction
• Overcoming limitations of past pose dependency in motion prediction

Innovation

• Physics-informed learning integrates domain knowledge
• Sparse IMUs enable whole-body kinematics prediction
• Forward kinematics regularizes joint predictions via loss