Risk-Averse Zero-Order Trajectory Optimization
2021
Conference Paper
al
We introduce a simple but effective method for managing risk in zero-order trajectory optimization that involves probabilistic safety constraints and balancing of optimism in the face of epistemic uncertainty and pessimism in the face of aleatoric uncertainty of an ensemble of stochastic neural networks. Various experiments indicate that the separation of uncertainties is essential to performing well with data-driven MPC approaches in uncertain and safety-critical control environments.
| Author(s): | Marin Vlastelica* and Sebastian Blaes* and Cristina Pinneri and Georg Martius |
| Book Title: | Conference on Robot Learning |
| Year: | 2021 |
| Department(s): | Autonomous Learning |
| Research Project(s): |
Model-based Reinforcement Learning and Planning
|
| Bibtex Type: | Conference Paper (inproceedings) |
| Paper Type: | Conference |
| Note: | *Equal Contribution |
| State: | Accepted |
| URL: | https://openreview.net/forum?id=WqUl7sNkDre |
| Links: |
OpenReview
|
| Attachments: | |
|
BibTex @inproceedings{VlastelicaBlaesEtal2021:riskaverse,
title = {Risk-Averse Zero-Order Trajectory Optimization},
author = {Vlastelica*, Marin and Blaes*, Sebastian and Pinneri, Cristina and Martius, Georg},
booktitle = {Conference on Robot Learning},
year = {2021},
note = {*Equal Contribution},
doi = {},
url = {https://openreview.net/forum?id=WqUl7sNkDre}
}
|
|
