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

Reinforcement learning (RL) is a powerful data-driven control method that hasbeen largely explored in autonomous driving tasks. However, conventional RLapproaches learn control policies through trial-and-error interactions with theenvironment and therefore may cause disastrous consequences such as collisionswhen testing in real-world traffic. Offline RL has recently emerged as apromising framework to learn effective policies from previously-collected,static datasets without the requirement of active interactions, making itespecially appealing for autonomous driving applications. Despite promising,existing offline RL algorithms such as Batch-Constrained deep Q-learning (BCQ)generally lead to rather conservative policies with limited explorationefficiency. To address such issues, this paper presents an enhanced BCQalgorithm by employing a learnable parameter noise scheme in the perturbationmodel to increase the diversity of observed actions. In addition, aLyapunov-based safety enhancement strategy is incorporated to constrain theexplorable state space within a safe region. Experimental results in highwayand parking traffic scenarios show that our approach outperforms theconventional RL method, as well as state-of-the-art offline RL algorithms.

Offline Reinforcement Learning for Autonomous Driving with Safety and Exploration Enhancement