Towards Continual Reinforcement Learning for Quadruped Robots
No Thumbnail Available
Date
2023
Journal Title
Journal ISSN
Volume Title
Publisher
The Eurographics Association
Abstract
Quadruped robots have emerged as an evolving technology that currently leverages simulators to develop a robust controller capable of functioning in the real-world without the need for further training. However, since it is impossible to predict all possible real-world situations, our research explores the possibility of enabling them to continue learning even after their deployment. To this end, we designed two continual learning scenarios, sequentially training the robot on different environments while simultaneously evaluating its performance across all of them. Our approach sheds light on the extent of both forward and backward skill transfer, as well as the degree to which the robot might forget previously acquired skills. By addressing these factors, we hope to enhance the adaptability and performance of quadruped robots in real-world scenarios.
Description
@inproceedings{10.2312:imet.20231258,
booktitle = {International Conference on Interactive Media, Smart Systems and Emerging Technologies (IMET)},
editor = {Pelechano, Nuria and Liarokapis, Fotis and Rohmer, Damien and Asadipour, Ali},
title = {{Towards Continual Reinforcement Learning for Quadruped Robots}},
author = {Minelli, Giovanni and Vassiliades, Vassilis},
year = {2023},
publisher = {The Eurographics Association},
ISBN = {978-3-03868-233-2},
DOI = {10.2312/imet.20231258}
}