Locally Weighted Regression Pseudo-Rehearsal for Adaptive Model Predictive Control
Abstract
We consider the problem of online adaptation of a neural network designed to represent system dynamics. The neural network model is intended to be used by an MPC control law for autonomous control. This problem is challenging because both input and target distributions are non-stationary, and naive approaches to online adaptation result in catastrophic forgetting. We present a novel online learning method, which combines the pseudo-rehearsal method with locally weighted projection regression. We demonstrate the effectiveness of the resulting Locally Weighted Projection Regression Pseudo-Rehearsal (LW-PR2) method on an autonomous vehicle in simulation and real world data collected with a 1/5 scale autonomous vehicle.
Cite
Text
Williams et al. "Locally Weighted Regression Pseudo-Rehearsal for Adaptive Model Predictive Control." Conference on Robot Learning, 2019.Markdown
[Williams et al. "Locally Weighted Regression Pseudo-Rehearsal for Adaptive Model Predictive Control." Conference on Robot Learning, 2019.](https://mlanthology.org/corl/2019/williams2019corl-locally/)BibTeX
@inproceedings{williams2019corl-locally,
title = {{Locally Weighted Regression Pseudo-Rehearsal for Adaptive Model Predictive Control}},
author = {Williams, Grady R. and Goldfain, Brian and Lee, Keuntaek and Gibson, Jason and Rehg, James M. and Theodorou, Evangelos A.},
booktitle = {Conference on Robot Learning},
year = {2019},
pages = {969-978},
volume = {100},
url = {https://mlanthology.org/corl/2019/williams2019corl-locally/}
}