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

Deep reinforcement learning is actively used for training autonomous carpolicies in a simulated driving environment. Due to the large availability ofvarious reinforcement learning algorithms and the lack of their systematiccomparison across different driving scenarios, we are unsure of which ones aremore effective for training autonomous car software in single-agent as well asmulti-agent driving environments. A benchmarking framework for the comparisonof deep reinforcement learning in a vision-based autonomous driving will openup the possibilities for training better autonomous car driving policies. Toaddress these challenges, we provide an open and reusable benchmarkingframework for systematic evaluation and comparative analysis of deepreinforcement learning algorithms for autonomous driving in a single- andmulti-agent environment. Using the framework, we perform a comparative study ofdiscrete and continuous action space deep reinforcement learning algorithms. Wealso propose a comprehensive multi-objective reward function designed for theevaluation of deep reinforcement learning-based autonomous driving agents. Werun the experiments in a vision-only high-fidelity urban driving simulatedenvironments. The results indicate that only some of the deep reinforcementlearning algorithms perform consistently better across single and multi-agentscenarios when trained in various multi-agent-only environment settings. Forexample, A3C- and TD3-based autonomous cars perform comparatively better interms of more robust actions and minimal driving errors in both single andmulti-agent scenarios. We conclude that different deep reinforcement learningalgorithms exhibit different driving and testing performance in differentscenarios, which underlines the need for their systematic comparative analysis.The benchmarking framework proposed in this paper facilitates such acomparison.

Evaluating the Robustness of Deep Reinforcement Learning for Autonomous Policies in a Multi-agent Urban Driving Environment