Could traffic factors enhance autonomous vehicle safety?

To answer this question, researchers at Imperial College London, developed a new method to combine infrastructure-based traffic data with vehicle-based data. They demonstrate that adding traffic covariates increases accuracy and the use of the No-U-Turn Sampler (NUTS) reduces the computational running time.

They published their study on vehicle-based conflict prediction in dynamic traffic environments on 18 September 2025, in Communications in Transportation Research.

This study introduces a novel, context-aware conflict prediction algorithm using a hierarchical Bayesian threshold excess model.

The research addresses the longstanding tradeoff between simplifying assumptions in physics-based models and the interpretability challenges of learning algorithms. One of the key contributions of this study is the use of a Bayesian model, which incorporates uncertainty and offers interpretability—a stark contrast to the "black box" nature of many machine learning approaches. This model allows for a more transparent and understandable assessment of collision risks, making it easier for practitioners to trust and implement the findings.

Moreover, the inclusion of traffic variables in vehicle-based applications is a rare and significant contribution of this paper. By considering these factors, the research produces more robust safety assessments, accounting for the dynamic and often unpredictable nature of real-world traffic conditions. The results are promising, demonstrating that the inclusion of traffic covariates enhances the model's goodness-of-fit by 4.80% in terms of Deviance Information Criterion. Additionally, the model shows improved generalizability with a decrease of 1.6% in mean absolute error. However, the research also highlights a nuanced finding: partially pooled models, while enhancing goodness-of-fit, can sometimes reduce generalization capabilities.

One of the key innovations of this study is the implementation of the Hamiltonian Monte Carlo No-U-Turn Solver (NUTS). The No-U-Turn Sampler compiled in JAX has shown sufficient performance for both online training and inference, making this methodology a feasible solution for real-time deployment in vehicle-based applications. This contrasts with traditional Markov Chain Monte Carlo sampling methods which are often deemed too slow for real-time applications.

The implications of this study extend beyond academic circles. For policymakers, vehicle manufacturers, and technology developers, these findings provide a robust framework for enhancing vehicle safety systems. As we move towards an era of smart and automated transportation, such advancements are pivotal in ensuring that our roads become safer for everyone.

About Communications in Transportation Research

Communications in Transportation Research was launched in 2021, with academic support provided by Tsinghua University and China Intelligent Transportation Systems Association. The Editors-in-Chief are Professor Xiaobo Qu, a member of the Academia Europaea from Tsinghua University and Professor Shuai’an Wang from Hong Kong Polytechnic University. The journal mainly publishes high-quality, original research and review articles that are of significant importance to emerging transportation systems, aiming to serve as an international platform for showcasing and exchanging innovative achievements in transportation and related fields, fostering academic exchange and development between China and the global community.

It has been indexed in SCIE, SSCI, Ei Compendex, Scopus, CSTPCD, CSCD, OAJ, DOAJ, TRID and other databases. It was selected as Q1 Top Journal in the Engineering and Technology category of the Chinese Academy of Sciences (CAS) Journal Ranking List. In 2022, it was selected as a High-Starting-Point new journal project of the “China Science and Technology Journal Excellence Action Plan”. In 2024, it was selected as the Support the Development Project of “High-Level International Scientific and Technological Journals”. The same year, it was also chosen as an English Journal Tier Project of the “China Science and Technology Journal Excellence Action Plan Phase Ⅱ”. In 2024, it received the first impact factor (2023 IF) of 12.5, ranking Top1 (1/58, Q1) among all journals in “TRANSPORTATION” category. In 2025, its 2024 IF was announced as 14.5, maintaining the Top 1 position (1/61, Q1) in the same category. Tsinghua University Press will cover the open access fee for all published papers in 2025.