Active Uncertainty Learning for Human-Robot Interaction: An Implicit Dual Control Approach

We provide an MATLAB implementation of implicit dual control-based active uncertainty learning for autonomous driving applications, which can be found here.

Click to watch our spotlight video:

• MPT3 (Toolbox for MPC and parametric optimization.)
• SNOPT (Nonlinear programming solver. Academic licenses are available.)

• Robotics Toolbox for MATLAB (Tools for plotting the vehicles. You need to first install it in MATLAB and then replace plot_vehicle.m in the root directory with ours.)

In this repository, we provide an example of our method applied for human-robot interactive driving scenarios.

1. Clone the repo

   git clone https://github.com/SafeRoboticsLab/Dual_Control_HRI.git

2. In MATLAB, run the main.m script to reproduce our results.

Distributed under the BSD 3-Clause License. See LICENSE for more information.

Haimin Hu - @HaiminHu - [email protected]

Project Link: https://github.com/SafeRoboticsLab/Dual_Control_HRI

Homepage Link: https://haiminhu.org/dual_control_hri

Available on arXiv: https://arxiv.org/abs/2202.07720

@inproceedings{hu2023active,
  title={Active Uncertainty Reduction for Human-Robot Interaction: An Implicit Dual Control Approach},
  author={Hu, Haimin and Fisac, Jaime F},
  booktitle={Algorithmic Foundations of Robotics XV},
  pages={385--401},
  year={2023},
  publisher={Springer International Publishing}
}

Our prior work:

@article{hu2022sharp,
  author={Hu, Haimin and Nakamura, Kensuke and Fisac, Jaime F.},
  journal={IEEE Robotics and Automation Letters}, 
  title={SHARP: Shielding-Aware Robust Planning for Safe and Efficient Human-Robot Interaction}, 
  year={2022},
  volume={7},
  number={2},
  pages={5591-5598},
  doi={10.1109/LRA.2022.3155229}
}