Preprint · 2026 · Open access

Safe Reinforcement Learning of Autonomous Highway Driving: A Unified Framework for Safety and Efficiency

Chufei Yan, Zhihao Cui, Yiyan Lv, T X Chen, Ning Bian, Yulei Wang

arXiv (Cornell University) · Published 12 June 2026 · DOI 10.48550/arxiv.2606.14609

Open PDFView at the publisherCite this paper

In brief

“Safe Reinforcement Learning of Autonomous Highway Driving: A Unified Framework for Safety and Efficiency” is a 2026 preprint by Chufei Yan and 5 co-authors, published in arXiv (Cornell University). OpenAlex counts 0 citations as of 22 November 2026. It is open access: the full text is free to read at arXiv (Cornell University).

Citations
0
References
39
Authors
6

Abstract

Deep reinforcement learning (DRL) offers a compelling route to decision-making for advanced autonomous vehicles (AVs), yet its trial-and-error nature makes it difficult to guarantee safety during training and to achieve both safety and efficiency at deployment. We propose a unified safe reinforcement learning (SRL) framework that integrates safe distance (SD), reward machines (RM), and mixture-of-experts (MoE), termed MoE-RM-SRL. For deployment, SD and RM jointly shape a rule-aware reward that encodes highway traffic regulations and stage-wise objectives, enabling safe and reliable behavior without sacrificing efficiency. For training, we introduce a sparsely gated MoE layer comprising up to 11 deep Q-networks (DQNs); an SD-based gating rule activates a minimal set of experts for lane-keeping and lane-changing, mitigating the instability, discontinuities, and impulsive transients commonly induced by switching between heterogeneous controllers (e.g., MPC/rule-based modules and learned policies). We implement the proposed architecture in CARLA and integrate it with a 6-DoF driver-in-the-loop virtual-reality (DiL-VR) platform. Experiments in stochastic two-lane traffic show that MoE-RM-SRL substantially improves safety and efficiency over state-of-the-art baselines, and the framework naturally extends to multi-lane driving as well as on-ramp merging and exiting scenarios.

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Yan, C., Cui, Z., Lv, Y., Chen, T. X., Bian, N., & Wang, Y. (2026). Safe Reinforcement Learning of Autonomous Highway Driving: A Unified Framework for Safety and Efficiency. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2606.14609

MLA 9

Yan, Chufei, et al. “Safe Reinforcement Learning of Autonomous Highway Driving: A Unified Framework for Safety and Efficiency.” arXiv (Cornell University), 2026. https://doi.org/10.48550/arxiv.2606.14609.

Chicago

Yan, Chufei, Zhihao Cui, Yiyan Lv, T X Chen, Ning Bian, and Yulei Wang. 2026. “Safe Reinforcement Learning of Autonomous Highway Driving: A Unified Framework for Safety and Efficiency.” arXiv (Cornell University). https://doi.org/10.48550/arxiv.2606.14609.

Harvard

Yan, C. et al. (2026) ‘Safe Reinforcement Learning of Autonomous Highway Driving: A Unified Framework for Safety and Efficiency’, arXiv (Cornell University). doi:10.48550/arxiv.2606.14609.

IEEE

C. Yan, Z. Cui, Y. Lv, T. X. Chen, N. Bian, and Y. Wang, “Safe Reinforcement Learning of Autonomous Highway Driving: A Unified Framework for Safety and Efficiency,” arXiv (Cornell University), 2026, doi: 10.48550/arxiv.2606.14609.

BibTeX

@misc{yan2026safe,
  title     = {{Safe Reinforcement Learning of Autonomous Highway Driving: A Unified Framework for Safety and Efficiency}},
  author    = {Yan, Chufei and Cui, Zhihao and Lv, Yiyan and Chen, T X and Bian, Ning and Wang, Yulei},
  howpublished = {arXiv (Cornell University)},
  year      = {2026},
  doi       = {10.48550/arxiv.2606.14609},
  url       = {https://doi.org/10.48550/arxiv.2606.14609}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Topic and keywords

Physical Sciences › Engineering › Automotive Engineering › Autonomous Vehicle Technology and Safety

  • safe reinforcement learning
  • safety distance
  • reward machines
  • mixture of experts
  • deep Q-network
  • lane keeping
  • lane changing
  • driver-in-the-loop

Work with this paper

Most-cited papers in Engineering

All highly cited Engineering papers

Questions about this paper

How do I cite this paper in APA style?

Yan, C., Cui, Z., Lv, Y., Chen, T. X., Bian, N., & Wang, Y. (2026). Safe Reinforcement Learning of Autonomous Highway Driving: A Unified Framework for Safety and Efficiency. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2606.14609

How many times has this paper been cited?

0 times, according to OpenAlex on 22 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

Yes. It is open access at arXiv (Cornell University): sign in and open the PDF from this page.

Who wrote this paper?

Chufei Yan, Zhihao Cui, Yiyan Lv, T X Chen, Ning Bian, Yulei Wang.

What is the DOI of this paper?

10.48550/arxiv.2606.14609, permanently at https://doi.org/10.48550/arxiv.2606.14609. Sign in and open it from this page.

Bibliographic data from OpenAlex (CC0), last checked 22 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.