Probabilistic Crash Prediction and Prevention of Vehicle Crash

Authors: Lavanya Annadi, Fahimeh Jafari

Abstract:

Transportation brings immense benefits to society, but it also has its costs. Costs include the cost of infrastructure, personnel, and equipment, but also the loss of life and property in traffic accidents on the road, delays in travel due to traffic congestion, and various indirect costs in terms of air transport. This research aims to predict the probabilistic crash prediction of vehicles using Machine Learning due to natural and structural reasons by excluding spontaneous reasons, like overspeeding, etc., in the United States. These factors range from meteorological elements such as weather conditions, precipitation, visibility, wind speed, wind direction, temperature, pressure, and humidity, to human-made structures, like road structure components such as Bumps, Roundabouts, No Exit, Turning Loops, Give Away, etc. The probabilities are categorized into ten distinct classes. All the predictions are based on multiclass classification techniques, which are supervised learning. This study considers all crashes in all states collected by the US government. The probability of the crash was determined by employing Multinomial Expected Value, and a classification label was assigned accordingly. We applied three classification models, including multiclass Logistic Regression, Random Forest and XGBoost. The numerical results show that XGBoost achieved a 75.2% accuracy rate which indicates the part that is being played by natural and structural reasons for the crash. The paper has provided in-depth insights through exploratory data analysis.

Keywords: Road safety, crash prediction, exploratory analysis, machine learning.

References:

[1] B. G. Lee, J. Park, C. C. Pu and W. Chung, "Smartwatch-Based Driver Vigilance Indicator with Kernel-Fuzzy-C-Means-Wavelet Method," in IEEE Sensors Journal, vol. 16, no. 1, pp. 242-253, Jan.1, 2016, doi: 10.1109/JSEN.2015.2475638.
[2] Ashtaiwi, "Intelligent Road Crashes Avoidance System," 2019 International Conference on Artificial Intelligence in Information and Communication (ICAIIC), 2019, pp. 280-286, doi: 10.1109/ICAIIC.2019.8668972.
[3] W. Tong, A. Hussain, W. X. Bo and S. Maharjan, "Artificial Intelligence for Vehicle-to-Everything: A Survey," in IEEE Access, vol. 7, pp. 10823-10843, 2019, doi: 10.1109/ACCESS.2019.2891073.
[4] E. Karlsson and N. Mohammadiha, "A Data-Driven Generative Model for GPS Sensors for Autonomous Driving," 2018 IEEE/ACM 1st International Workshop on Software Engineering for AI in Autonomous Systems (SEFAIAS), 2018, pp. 1-5.
[5] Y. Yoon, C. Kim, J. Lee and K. Yi, "Interaction-Aware Probabilistic Trajectory Prediction of Cut-In Vehicles Using Gaussian Process for Proactive Control of Autonomous Vehicles," in IEEE Access, vol. 9, pp. 63440-63455, 2021, doi: 10.1109/ACCESS.2021.3075677.
[6] X. Bai, L. Sang, J. Khan, H. Kang and Z. Liu, "Initial Thoughts and Application of AI in Geo-hazards Monitoring and Early Warning of Highroad in Beijing," 2019 5th International Conference on Transportation Information and Safety (ICTIS), 2019, pp. 1358-1363, doi: 10.1109/ICTIS.2019.8883743.
[7] Najjar, A., Kaneko, S., & Miyanaga, Y. (2017). Combining Satellite Imagery and Open Data to Map Road Safety. AAAI.
[8] Akin, Darcin & Akbas, Bulent. (2008). A Neural Network (NN) Model to Predict Intersection Crashes Based upon Crash Properties: Driver, Vehicle, and Roadway Surface Characteristics.
[9] Fancello, Gianfranco & Carta, Michele & Fadda, Paolo. (2015). A Decision Support System for Road Safety Analysis. Transportation Research Procedia. 5. 10.1016/j.trpro.2015.01.009.
[10] Silva, P.B., Andrade, M. and Ferreira, S., 2020. Machine learning applied to road safety modeling: a systematic literature review. Journal of Traffic and transportation engineering (English edition).
[11] Tonhauser, M. and Ristvej, J., 2021. Implementation of new technologies to improve safety of road transport. Transportation research procedia, 55, pp.1599-1604.
[12] DfT (2019), Road Safety Statement: Progress Report. A Lifetime of Road Safety, 2019 Road Safety Statement https://www.gov.uk/government/publications/road-safetystatement-2019-a-lifetime-of-road-safety.
[13] Haghani, M., Bliemer, M.C., Farooq, B., Kim, I., Li, Z., Oh, C., Shahhoseini, Z. and MacDougall, H., 2020. https://www.hindawi.com/journals/jat/si/892372/
[14] Li, H., Zhu, M., Graham, D.J. and Ren, G., 2021. Evaluating the speed camera sites selection criteria in the UK. Journal of safety research, 76, pp.90-100.
[15] Hughes, B.P., Anund, A. and Falkmer, T., 2015. System theory and safety models in Swedish, UK, Dutch and Australian road safety strategies. Accident Analysis & Prevention, 74, pp.271-278.
[16] Nogayeva, S., Gooch, J. and Frascione, N., 2020. The forensic investigation of vehicle-pedestrian collisions: a review. Science & Justice.
[17] Damani, J. and Vedagiri, P., 2021. Safety of motorised two wheelers in mixed traffic conditions: literature review of risk factors. Journal of Traffic and transportation engineering (English edition).
[18] Tony Yiu, “Understanding Random Forest”, Towards Data Science,2019
[19] Iyad Lahsen Cherif, Abdesselem Kortebi, “On using eXtreme Gradient Boosting (XGBoost) Machine Learning algorithm for Home Network Traffic Classification”, IEEE Xplore: 13 June 2019
[20] https://towardsdatascience.com/usa-accidents-data-analysis-d130843cde02