Deep Learning with Predictive Control for Human Motion Tracking

Don Joven Agravante; Giovanni De Magistris; Asim Munawar; Phongtharin Vinayavekhin; Ryuki Tachibana

Deep Learning with Predictive Control for Human Motion Tracking

Robotics 2018-08-08 v1

Authors: Don Joven Agravante , Giovanni De Magistris , Asim Munawar , Phongtharin Vinayavekhin , Ryuki Tachibana

Abstract

We propose to combine model predictive control with deep learning for the task of accurate human motion tracking with a robot. We design the MPC to allow switching between the learned and a conservative prediction. We also explored online learning with a DyBM model. We applied this method to human handwriting motion tracking with a UR-5 robot. The results show that the framework significantly improves tracking performance.

Keywords

human motion prediction model predictive control robot control

Cite

@article{arxiv.1808.02200,
  title  = {Deep Learning with Predictive Control for Human Motion Tracking},
  author = {Don Joven Agravante and Giovanni De Magistris and Asim Munawar and Phongtharin Vinayavekhin and Ryuki Tachibana},
  journal= {arXiv preprint arXiv:1808.02200},
  year   = {2018}
}

Comments

To appear in 36th Annual Conference of the Robotics Society of Japan (RSJ 2018)

Related papers

View all related →