# A Sensorimotor Reinforcement Learning Framework for Physical Human-Robot Interaction

arXiv: https://arxiv.org/abs/1607.07939 · HTML: https://mlsanity.com/paper/1607.07939/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 27 Jul 2016.

## Built on by (most cited first, 5 of 5)

- [Human-robot collaboration and machine learning: a systematic review of recent research](https://mlsanity.com/paper/2110.07448/) (2021)
- [Back to the Manifold: Recovering from Out-of-Distribution States](https://mlsanity.com/paper/2207.08673/) (2022)
- [It Takes Two: Learning to Plan for Human-Robot Cooperative Carrying](https://mlsanity.com/paper/2209.12890/) (2022)
- [An adaptive admittance controller for collaborative drilling with a robot based on subtask classification via deep learning](https://mlsanity.com/paper/2205.14457/) (2022)
- [Human-Machine Co-Adaptation for Robot-Assisted Rehabilitation via Dual-Agent Multiple Model Reinforcement Learning (DAMMRL)](https://mlsanity.com/paper/2407.21734/) (2024)

## Abstract (arXiv)

Modeling of physical human-robot collaborations is generally a challenging problem due to the unpredictive nature of human behavior. To address this issue, we present a data-efficient reinforcement learning framework which enables a robot to learn how to collaborate with a human partner. The robot learns the task from its own sensorimotor experiences in an unsupervised manner. The uncertainty of the human actions is modeled using Gaussian processes (GP) to implement action-value functions. Optimal action selection given the uncertain GP model is ensured by Bayesian optimization. We apply the framework to a scenario in which a human and a PR2 robot jointly control the ball position on a plank based on vision and force/torque data. Our experimental results show the suitability of the proposed method in terms of fast and data-efficient model learning, optimal action selection under uncertainties and equal role sharing between the partners.