Physics-Informed Neural Networks with Unscented Kalman Filter for Sensorless Joint Torque Estimation in Humanoid Robots

2025-07-14

AI Summary

Estimating joint torques in torque-sensor-less humanoid robots—particularly those employing high-ratio harmonic drive motor systems—remains challenging due to complex, nonlinear friction dynamics and the absence of direct torque measurements. Method: This paper proposes a real-time whole-body torque estimation and control framework integrating Physics-Informed Neural Networks (PINNs) with Unscented Kalman Filtering (UKF). PINNs jointly learn friction dynamics from data and physical constraints, and their outputs serve as key virtual measurements for UKF, enabling high-fidelity, model-augmented state estimation without hardware torque sensors. Contribution/Results: Embedded within a real-time control architecture and validated on the ergoCub platform, the framework reduces torque tracking error by 32% and energy consumption by 18% compared to conventional Recursive Newton–Euler Algorithm (RNEA)-based methods, while significantly improving robustness against external disturbances. Moreover, it demonstrates cross-platform transferability, having been successfully deployed across multiple identical robot platforms.

Technology Category

• Intelligent Robots: State Estimation
• Computer Vision: Low Level & Physics-based Vision
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

This paper presents a novel framework for whole-body torque control of humanoid robots without joint torque sensors, designed for systems with electric motors and high-ratio harmonic drives. The approach integrates Physics-Informed Neural Networks (PINNs) for friction modeling and Unscented Kalman Filtering (UKF) for joint torque estimation, within a real-time torque control architecture. PINNs estimate nonlinear static and dynamic friction from joint and motor velocity readings, capturing effects like motor actuation without joint movement. The UKF utilizes PINN-based friction estimates as direct measurement inputs, improving torque estimation robustness. Experimental validation on the ergoCub humanoid robot demonstrates improved torque tracking accuracy, enhanced energy efficiency, and superior disturbance rejection compared to the state-of-the-art Recursive Newton-Euler Algorithm (RNEA), using a dynamic balancing experiment. The framework's scalability is shown by consistent performance across robots with similar hardware but different friction characteristics, without re-identification. Furthermore, a comparative analysis with position control highlights the advantages of the proposed torque control approach. The results establish the method as a scalable and practical solution for sensorless torque control in humanoid robots, ensuring torque tracking, adaptability, and stability in dynamic environments.

Problem

• Estimating joint torque without sensors in humanoid robots
• Modeling friction using Physics-Informed Neural Networks
• Improving torque estimation robustness with Unscented Kalman Filter

Innovation

• Physics-Informed Neural Networks for friction modeling
• Unscented Kalman Filter for joint torque estimation
• Real-time torque control without joint sensors