English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Robot navigation is a task where reinforcement learning approaches are stillunable to compete with traditional path planning. State-of-the-art methodsdiffer in small ways, and do not all provide reproducible, openly availableimplementations. This makes comparing methods a challenge. Recent research hasshown that unsupervised learning methods can scale impressively, and beleveraged to solve difficult problems. In this work, we design ways in whichunsupervised learning can be used to assist reinforcement learning for robotnavigation. We train two end-to-end, and 18 unsupervised-learning-basedarchitectures, and compare them, along with existing approaches, in unseen testcases. We demonstrate our approach working on a real life robot. Our resultsshow that unsupervised learning methods are competitive with end-to-endmethods. We also highlight the importance of various components such as inputrepresentation, predictive unsupervised learning, and latent features. We makeall our models publicly available, as well as training and testingenvironments, and tools. This release also includes OpenAI-gym-compatibleenvironments designed to emulate the training conditions described by otherpapers, with as much fidelity as possible. Our hope is that this helps inbringing together the field of RL for robot navigation, and allows meaningfulcomparisons across state-of-the-art methods.

NavRep: Unsupervised Representations for Reinforcement Learning of Robot Navigation in Dynamic Human Environments