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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Autonomous Driving requires high levels of coordination and collaborationbetween agents. Achieving effective coordination in multi-agent systems is adifficult task that remains largely unresolved. Multi-Agent ReinforcementLearning has arisen as a powerful method to accomplish this task because itconsiders the interaction between agents and also allows for decentralizedtraining -- which makes it highly scalable. However, transferring policies fromsimulation to the real world is a big challenge, even for single-agentapplications. Multi-agent systems add additional complexities to theSim-to-Real gap due to agent collaboration and environment synchronization. Inthis paper, we propose a method to transfer multi-agent autonomous drivingpolicies to the real world. For this, we create a multi-agent environment thatimitates the dynamics of the Duckietown multi-robot testbed, and trainmulti-agent policies using the MAPPO algorithm with different levels of domainrandomization. We then transfer the trained policies to the Duckietown testbedand compare the use of the MAPPO algorithm against a traditional rule-basedmethod. We show that the rewards of the transferred policies with MAPPO anddomain randomization are, on average, 1.85 times superior to the rule-basedmethod. Moreover, we show that different levels of parameter randomization havea substantial impact on the Sim-to-Real gap.

Transferring Multi-Agent Reinforcement Learning Policies for Autonomous Driving using Sim-to-Real