Downloads

Dai, W. . (2021). Safety Evaluation of Traffic System with Historical Data Based on Markov Process and Deep-Reinforcement Learning. Journal of Computational Methods in Engineering Applications, 1(1), 1–14. https://ojs.sgsci.org/journals/jcmea/article/view/141
Volume 1, Issue 1, 2021

Safety Evaluation of Traffic System with Historical Data Based on Markov Process and Deep-Reinforcement Learning

This study introduces a comprehensive framework for discerning and enhancing traffic system safety through a multifaceted approach that integrates Markov processes and deep reinforcement learning (DRL). The foundation of this framework lies in the establishment of key safety evaluation metrics, predicated upon a thorough survey of historical data and pertinent regulations. These metrics, once identified, are systematically scored within discrete nodes, ensuring a nuanced and precise assessment. Subsequently, areas displaying lower scores are pinpointed, prompting the implementation of tailored improvement strategies aimed at mitigating identified shortcomings. This process culminates in the rigorous testing of the efficacy of these interventions, effectuated through a comparative analysis of the established scores pre- and post-improvement. By leveraging the synergy between Markov processes and DRL, this approach not only gauges the system's current safety status but also enables the formulation and execution of targeted enhancements. This framework forms an integral step toward achieving a traffic system that is both responsive and resilient, ultimately fostering a safer and more efficient urban mobility landscape.

traffic systems safety deep learning reinforcement learning markov process

References

  1. Zhu S. Safety Evaluation on Slow Traffic Facilities Around an Intersection. Journal of Huaiyin Institute of Technology 2014; 22: 30–03.
  2. Austin PC. Are (the Log-Odds of) Hospital Mortality Rates Normally Distributed? Implications for Studying Variations in Outcomes of Medical Care. Journal of Evaluation in Clinical Practice 2009; 15: 514–523.
  3. Park B-J, Lord D, Lee C. Finite Mixture Modeling for Vehicle Crash Data With Application to Hotspot Identification. Accident Analysis and Prevention 2014; 71: 319–326.
  4. Teschke K, Dennis J, Reynolds COO, Winters M, Harris MA. Bicycling Crashes on Streetcar (Tram) Or Train Tracks: Mixed Methods to Identify Prevention Measures. Journal of BMC Public Health 2016; 16: 617.
  5. Sun C, Li W, Shao Y. Fault Tolerant Design and Management of Urban Road Traffic Safety. Journal of Papers on Urban Traffic Planning in China 2017; 15: 199–342.
  6. Da Costa S, Qu X, Parajuli PM. A Crash Severity-Based Black Spot Identification Model. Journal of Transportation Safety & Security 2015; 7(3): 268–277.
  7. Elvik R. New Approach to Accident Analysis for Hazardous Road Locations. Transportation Research Record 2006; 1953: 50–55.
  8. Cheng W, Washington S. New Criteria for Evaluating Methods of Identifying Hot Spots. Transportation Research Record 2008; 2083: 76–85.
  9. Hoff PD. A First Course in Bayesian Statistical Methods; Springer: Berlin, Germany, 2009.
  10. Kim D, Washington S, Oh J. Modeling Crash Outcomes: New Insights into the Effects of Covariates on Crashes at Rural Intersections. Journal of Transportation Engineering 2006; 132(4): 282–292.
  11. Lyon C, Gotts B, Wong W, Persaud B. Comparison of Alternative Methods for Identifying Sites with High Proportion of Specific Accident Types. Transportation Research Record. 2007; 2019: 212–218.
  12. Mannering FL, Bhat CR. Analytic Methods in Accident Research: Methodological Frontier and Future Directions. Analytic Methods in Accident Research 2014; 1: 1–22.
  13. Mengersen KL, Robert CP. Testing for Mixtures: a Bayesian Entropic Approach. In Bayesian Statistics 5; Oxford University Press: New York, NY, USA, 1996.
  14. Bandyopadhyaya R, Mitra S. Fuzzy Cluster--Based Method of Hotspot Detection with Limited Information. Journal of Transportation Safety & Security 2015; 7(4): 307–323.
  15. Milton JC, Shankar VN, Mannering FL. Highway Accident Severities and the Mixed Logit Model: An Exploratory Empirical Analysis. Accident Analysis and Prevention 2008; 40: 260–266.
  16. Miranda-Moreno LF, Fu L, Ukkusuri SV, Lord D. How to Incorporate Accident Severity and Vehicle Occupancy into the Hot Spot Identification Process?. Transportation Research Record 2009; 2102: 53–60.
  17. Luo Z, Xu H, Chen F. Audio Sentiment Analysis by Heterogeneous Signal Features Learned from Utterance-Based Parallel Neural Network. Proceedings of the 2nd Workshop on Affective Content Analysis (AffCon 2019) co-located with Thirty-Third AAAI Conference on Artificial Intelligence (AAAI 2019), Honolulu, HI, USA, January 27 2019.
  18. Chen F, Luo Z, Xu Y, Ke D. Complementary Fusion of Multi-Features and Multi-Modalities in Sentiment Analysis. AAAI-2020 Workshop on Affective Content Analysis, New York, NY, USA, 7 February 2020.
  19. Elvik R. Comparative Analysis of Techniques for Identifying Locations of Hazardous Roads. Transportation Research Record 2008; 2083: 72–75.
  20. AASHTO. Highway Safety Manual, 1st ed.; American Association of State Highway and Transportation Officials (AASHTO): Washington, DC, USA, 2010.
  21. Luo Z, Zeng X, Bao Z, Xu M. Deep Learning-Based Strategy for Macromolecules Classification with Imbalanced Data from Cellular Electron Cryotomography. 2019 International Joint Conference on Neural Networks (IJCNN), Budapest, Hungary, 14–19 July 2019.
  22. Luo Z. Knowledge-guided Aspect-based Summarization. 2023 International Conference on Communications, Computing and Artificial Intelligence (CCCAI), Shanghai, China, 23–25 June 2023.