analiza aiqbench
End2Race presents a significant advancement in multi-vehicle autonomous racing by replacing traditional rule-based logic with an end-to-end learning framework capable of sub-millisecond inference. This approach effectively addresses the critical gap in handling complex interactions between multiple agents, demonstrating superior speed and safety on standard research platforms like F1Tenth. The work highlights how deep reinforcement learning is evolving from single-agent tasks to coordinated multi-agent systems essential for future autonomous vehicle deployment.
End2Race: An End-to-End Learning Framework for Multi-Vehicle Autonomous Racing
arXiv:2509.16894v2 Announce Type: replace Abstract: Autonomous racing serves as a compelling testbed for advancing autonomous vehicle systems. The 1/10-scale F1Tenth platform is widely adopted for education and research, with annual competitions held worldwide. Yet, leading solutions in these tournaments remain dominated by rule-based approaches. While learning-based methods have been proposed, they focus primarily on single-vehicle settings, leaving challenging multi-vehicle interactions largely unexplored. To address this gap, we introduce End2Race, an end-to-end learning framework designed for head-to-head autonomous racing. Using this framework, we develop a computationally efficient policy network that achieves an inference latency below 1 ms on an F1Tenth onboard computer. Extensive evaluations show that the policy generalizes to both novel tracks and unseen opponent behaviors, demonstrating high racing speeds, robust safety, and adaptive overtaking maneuvers. These results substantially outperform both prior learning-based methods and rule-based baselines. Codebase is available at https://github.com/michigan-traffic-lab/End2Race.
Abstract:Autonomous racing serves as a compelling testbed for advancing autonomous vehicle systems. The 1/10-scale F1Tenth platform is widely adopted for education and research, with annual competitions held worldwide. Yet, leading solutions in these tournaments remain dominated by rule-based approaches. While learning-based methods have been proposed, they focus primarily on single-vehicle settings, leaving challenging multi-vehicle interactions largely unexplored. To address this gap, we introduce End2Race, an end-to-end learning framework designed for head-to-head autonomous racing. Using this framework, we develop a computationally efficient policy network that achieves an inference latency below 1 ms on an F1Tenth onboard computer. Extensive evaluations show that the policy generalizes to both novel tracks and unseen opponent behaviors, demonstrating high racing speeds, robust safety, and adaptive overtaking maneuvers. These results substantially outperform both prior learning-based methods and rule-based baselines. Codebase is available at this https URL.
Robotics (cs.RO)
arXiv:2509.16894 [cs.RO]
(or arXiv:2509.16894v2 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2509.16894 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 230 cuvinte
Subiecte conexe
NVIDIA GTC Taipei at COMPUTEX: Live Updates on What’s Next in AI
NVIDIA · 21 mai 2026
SGTP: Sampling-based Game-Theoretic Planning for Real-Time Multi-Vehicle Autonomous Racing
arXiv · 29 iul. 2026
A Multi-Modal Perception Pipeline for Object Detection and Tracking in Autonomous Racing
arXiv · 9 sept. 2026
High-Fidelity Multi-Body Simulator for Autonomous Racing
arXiv · 14 sept. 2026