Abstract:Ensuring safety and adapting to the user's behavior are of paramount importance in physical human-robot interaction. Thus, incorporating elastic actuators in the robot's mechanical design has become popular, since it offers intrinsic compliance and additionally provide a coarse estimate for the interaction force by measuring the deformation of the elastic components. While observer-based methods have been shown to improve these estimates, they rely on accurate models of the system, which are challenging to obtain in complex operating environments. In this work, we overcome this issue by learning the unknown dynamics components using Gaussian process (GP) regression. By employing the learned model in a Bayesian filtering framework, we improve the estimation accuracy and additionally obtain an observer that explicitly considers local model uncertainty in the confidence measure of the state estimate. Furthermore, we derive guaranteed estimation error bounds, thus, facilitating the use in safety-critical applications. We demonstrate the effectiveness of the proposed approach experimentally in a human-exoskeleton interaction scenario.

What problem does this paper attempt to address?

This paper attempts to address the problem of ensuring safety and adapting to user behavior in physical human-robot interaction. Specifically, the authors focus on accurately estimating external forces (such as human active torque) when using robots with Series Elastic Actuators (SEAs), especially in complex operational environments where traditional model-based methods struggle to obtain accurate dynamic models. ### Problem Background 1. **Safety**: Ensuring safety is crucial in physical human-robot interaction, particularly in applications like rehabilitation robots. 2. **Adaptability**: Robots need to adapt to individual differences among users. 3. **Accuracy of Dynamic Models**: Traditional model-based methods struggle to obtain accurate dynamic models in complex environments, leading to estimation errors. ### Solution The authors propose a method that combines Gaussian Process (GP) regression and Bayesian filtering to learn unknown dynamic components and use them to improve state estimation. The specific steps are as follows: 1. **Data-Driven Model Learning**: Use Gaussian Process regression to learn unknown dynamic components, including human passive dynamics. 2. **Enhanced Kalman Filter**: Combine the learned model with an Extended Kalman Filter (EKF) to form an enhanced state observer (GP-AKF). 3. **Error Bound Guarantee**: Derive theoretical bounds for estimation errors using the ellipsoid set method to ensure reliability in safety-critical applications. ### Main Contributions 1. **Data-Driven Force Observer**: Propose a data-driven force observer that can accurately estimate external forces in complex environments. 2. **Uncertainty Quantification**: Explicitly consider local model uncertainty through Gaussian Process regression. 3. **Theoretical Guarantee**: Derive theoretical bounds for estimation errors, ensuring reliability in safety-critical applications. ### Experimental Validation The authors validate the effectiveness of the proposed method through experiments on a single-degree-of-freedom SEA elbow exoskeleton. The experimental results show that the method can accurately learn human passive dynamics and successfully estimate dynamic torque. ### Conclusion This paper proposes a method that combines Gaussian Process regression and Bayesian filtering to address the problem of accurately estimating external forces in physical human-robot interaction, especially in complex environments. Through a data-driven approach, the method can adapt to different users and provides theoretical error bounds to ensure reliability in safety-critical applications.

Data-driven Force Observer for Human-Robot Interaction with Series Elastic Actuators using Gaussian Processes

An adaptive compliance position control based on EKF for series elastic actuation

Decoupling Observer for Contact Force Estimation of Robot Manipulators Based on Enhanced Gaussian Process Model

Gaussian Process-Integrated State Space Model for Continuous Joint Angle Prediction from EMG and Interactive Force in a Human-Exoskeleton System

Safe Machine-Learning-supported Model Predictive Force and Motion Control in Robotics

Online Sparse Gaussian Process Based Human Motion Intent Learning for an Electrically Actuated Lower Extremity Exoskeleton.

Human Gait Trajectory Learning Using Online Gaussian Process for Assistive Lower Limb Exoskeleton

Real-time observer designs for elastic-joint industrial robots: Experimental comparison and new strategies

Motion Prediction with Gaussian Processes for Safe Human-Robot Interaction in Virtual Environments

Motion Prediction With Gaussian Processes for Safe Human–Robot Interaction in Virtual Environments

Contact Force Estimation of Robot Manipulators With Imperfect Dynamic Model: On Gaussian Process Adaptive Disturbance Kalman Filter

Output Constrained Control of Lower Limb Exoskeleton Based on Knee Motion Probabilistic Model with Finite-Time Extended State Observer

External Force Detection for Physical Human-Robot Interaction Using Dynamic Model Identification.

Data-Driven Kinematic Model for the End-Effector Pose Control of a Manipulator Robot

A Black-Box Physics-Informed Estimator Based on Gaussian Process Regression for Robot Inverse Dynamics Identification

Model-Based External Force/Moment Estimation for Humanoid Robots with no Torque Measurement

Learning User-Specific Control Policies for Lower-Limb Exoskeletons Using Gaussian Process Regression

Dependent-Gaussian-Process-Based Learning of Joint Torques Using Wearable Smart Shoes for Exoskeleton.

A Novel Control Law for Multi-Joint Human-Robot Interaction Tasks While Maintaining Postural Coordination

Force Observer for an Upper Limb Rehabilitation Robotic Device Using Iterative Learning Control

Whole-Body Joint Angle Estimation for Real-Time Humanoid Robot Imitation Based on Gaussian Process Dynamical Model and Particle Filter