English

Real-to-Sim: Predicting Residual Errors of Robotic Systems with Sparse Data using a Learning-based Unscented Kalman Filter

Robotics 2023-05-09 v3 Machine Learning Systems and Control Systems and Control

Abstract

Achieving highly accurate dynamic or simulator models that are close to the real robot can facilitate model-based controls (e.g., model predictive control or linear-quadradic regulators), model-based trajectory planning (e.g., trajectory optimization), and decrease the amount of learning time necessary for reinforcement learning methods. Thus, the objective of this work is to learn the residual errors between a dynamic and/or simulator model and the real robot. This is achieved using a neural network, where the parameters of a neural network are updated through an Unscented Kalman Filter (UKF) formulation. Using this method, we model these residual errors with only small amounts of data -- a necessity as we improve the simulator/dynamic model by learning directly from real-world operation. We demonstrate our method on robotic hardware (e.g., manipulator arm, and a wheeled robot), and show that with the learned residual errors, we can further close the reality gap between dynamic models, simulations, and actual hardware.

Keywords

Cite

@article{arxiv.2209.03210,
  title  = {Real-to-Sim: Predicting Residual Errors of Robotic Systems with Sparse Data using a Learning-based Unscented Kalman Filter},
  author = {Alexander Schperberg and Yusuke Tanaka and Feng Xu and Marcel Menner and Dennis Hong},
  journal= {arXiv preprint arXiv:2209.03210},
  year   = {2023}
}

Comments

Accepted to Ubiquitous Robotics 2023, Honolulu, Hawaii

R2 v1 2026-06-28T00:53:14.644Z