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

Realistic traffic simulation is crucial for developing self-driving softwarein a safe and scalable manner prior to real-world deployment. Typically,imitation learning (IL) is used to learn human-like traffic agents directlyfrom real-world observations collected offline, but without explicitspecification of traffic rules, agents trained from IL alone frequently displayunrealistic infractions like collisions and driving off the road. This problemis exacerbated in out-of-distribution and long-tail scenarios. On the otherhand, reinforcement learning (RL) can train traffic agents to avoidinfractions, but using RL alone results in unhuman-like driving behaviors. Wepropose Reinforcing Traffic Rules (RTR), a holistic closed-loop learningobjective to match expert demonstrations under a traffic compliance constraint,which naturally gives rise to a joint IL + RL approach, obtaining the best ofboth worlds. Our method learns in closed-loop simulations of both nominalscenarios from real-world datasets as well as procedurally generated long-tailscenarios. Our experiments show that RTR learns more realistic andgeneralizable traffic simulation policies, achieving significantly bettertradeoffs between human-like driving and traffic compliance in both nominal andlong-tail scenarios. Moreover, when used as a data generation tool for trainingprediction models, our learned traffic policy leads to considerably improveddownstream prediction metrics compared to baseline traffic agents. For moreinformation, visit the project website: https://waabi.ai/rtr

Learning Realistic Traffic Agents in Closed-loop