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

Racing demands each vehicle to drive at its physical limits, when any safetyinfraction could lead to catastrophic failure. In this work, we study theproblem of safe reinforcement learning (RL) for autonomous racing, using thevehicle's ego-camera view and speed as input. Given the nature of the task,autonomous agents need to be able to 1) identify and avoid unsafe scenariosunder the complex vehicle dynamics, and 2) make sub-second decision in afast-changing environment. To satisfy these criteria, we propose to incorporateHamilton-Jacobi (HJ) reachability theory, a safety verification method forgeneral non-linear systems, into the constrained Markov decision process (CMDP)framework. HJ reachability not only provides a control-theoretic approach tolearn about safety, but also enables low-latency safety verification. Though HJreachability is traditionally not scalable to high-dimensional systems, wedemonstrate that with neural approximation, the HJ safety value can be learneddirectly on vision context -- the highest-dimensional problem studied via themethod, to-date. We evaluate our method on several benchmark tasks, includingSafety Gym and Learn-to-Race (L2R), a recently-released high-fidelityautonomous racing environment. Our approach has significantly fewer constraintviolations in comparison to other constrained RL baselines in Safety Gym, andachieves the new state-of-the-art results on the L2R benchmark task. We provideadditional visualization of agent behavior at the following anonymized paperwebsite: https://sites.google.com/view/safeautonomousracing/home

Safe Autonomous Racing via Approximate Reachability on Ego-vision