dc.contributor.author | Larsen, Thomas Nakken | |
dc.contributor.author | Hansen, Hannah | |
dc.contributor.author | Rasheed, Adil | |
dc.date.accessioned | 2023-11-29T07:53:08Z | |
dc.date.available | 2023-11-29T07:53:08Z | |
dc.date.created | 2023-10-25T13:50:10Z | |
dc.date.issued | 2023 | |
dc.identifier.isbn | 9781510850712 | |
dc.identifier.uri | https://hdl.handle.net/11250/3105133 | |
dc.description.abstract | In this work, we propose a novel policy network architecture for model-free Reinforcement Learning (RL)-based path-following and collision avoidance in marine surface vessels. By applying convolutional neural networks (CNNs) for mapping LiDAR-like distance measurements to Collision Risk Indices (CRIs), we evaluate the utility of risk-based pretraining of CNN feature extractors prior to RL. Where previous works required hand-crafted preprocessing of high-resolution distance measurements to train an autonomous RL agent successfully, the proposed approach achieves this goal in a data-driven fashion. Ultimately, we propose future directions to improve CNN-based perception models for collision avoidance in range sensing applications. | en_US |
dc.language.iso | eng | en_US |
dc.publisher | IFAC | en_US |
dc.relation.ispartof | 22nd IFAC World Congress | |
dc.rights | Attribution-NonCommercial-NoDerivatives 4.0 Internasjonal | * |
dc.rights.uri | http://creativecommons.org/licenses/by-nc-nd/4.0/deed.no | * |
dc.title | Risk-based Convolutional Perception Models for Collision Avoidance in Autonomous Marine Surface Vessels using Deep Reinforcement Learning | en_US |
dc.title.alternative | Risk-based Convolutional Perception Models for Collision Avoidance in Autonomous Marine Surface Vessels using Deep Reinforcement Learning | en_US |
dc.type | Chapter | en_US |
dc.description.version | publishedVersion | en_US |
dc.source.pagenumber | 10773-10778 | en_US |
dc.identifier.doi | 10.1016/j.ifacol.2023.10.870 | |
dc.identifier.cristin | 2188420 | |
dc.relation.project | Norges forskningsråd: 309230 | en_US |
cristin.ispublished | true | |
cristin.fulltext | postprint | |