Makine Öğrenmesi Tekniklerinin Sürüş Stili Sınıflandırmasında Kullanımı

Abstract

Sürücü davranışlarının trafik güvenliğine önemli derecede etkisi vardır. Bu nedenle, sürücülerin davranışsal örüntüleri ve bu örüntüleri etkileyen etmenler tanımlanmalıdır. Sürücüler, araçlarını daha verimli ve kurallara uygun kullanmaya yönlendirilmelidir. Bu bağlamda, sürücünün aracını nasıl kullandığı gözlemlenerek, sürücülerin sürüş risk derecelerine uygun olarak sigorta ya da kasko ücretleri belirlenebilir. Bu çalışmada, Araç İçi Teşhis (On Board Diagnostics-II, OBD-II) ve Küresel Konumlandırma Sistemi (Global Positioning System, GPS) cihazlarından alınan işlenmiş ve etiketlenmiş telemetri verileri ile Makine Öğrenmesi (Machine Learning, ML) algoritmaları kullanılarak sürücülerin risk gruplarının sınıflandırılması gerçekleştirilmiştir. OBD-II ile araçtan elde edilen verilerin işlenerek sürücülerin risk derecesinin belirlenip bu risk derecesine göre sigorta şirketlerinin kişiye özel sigorta ücreti belirlemesinde yardımcı rol oynanması planlanmaktadır. Çalışmada Destek Vektör Makinesi (Support Vector Machine, SVM, CatBoost, k-NN (k-nearest neighbors, K-En yakın komşuluk) ve Hafif Gradyan Artırma Makinesi (Light Gradient Boosting Machine, LGBM)) olmak üzere dört model kullanılmıştır. Test sonuçları incelendiğinde en iyi tanımaya SVM modeli ile erişilmiştir.

Keywords

• Sınıflandırma
• Tanıma
• Makine öğrenmesi
• Sürücü davranışları

Driving Style Classification Using Machine Learning Techniques

Abstract

Driver behavior has a significant impact on traffic safety. Therefore, drivers' behavioral patterns and the factors affecting these patterns should be identified. Drivers should be directed to use their vehicles more efficiently and in accordance with the rules. In this context, by observing how the driver uses his vehicle, insurance or car insurance fees can be determined in accordance with the drivers' driving risk levels. In this study, the risk groups of drivers are classified using Machine Learning (ML) algorithms with processed and labeled telemetry data obtained from On Board Diagnostics-II (OBD-II) and Global Positioning System (GPS) devices. It is planned to determine the drivers' risk level by processing the data obtained from the vehicle with OBD-II and to play an auxiliary role in determining the personal insurance fee of the insurance companies according to this risk level. Support Vector Machine (SVM), CatBoost, k-NN (k-nearest neighbors), and Light Gradient Boosting Machine (LGBM) were used in the study. The best recognition was achieved with the SVM model.

Keywords

• Classification
• Recognition
• Machine learning
• Driver behaviors

References

1. Akhiladevi M, Anitha K, Amrutha K, Amrutha M, Chandanashree K. 2022. Accident prediction using KNN algorithm. Fourth International Conference on Emerging Research in Electronics, Computer Science and Technology (ICERECT): 26-27 December, Karnataka, India, pp: 1-5.
2. Bao N, Carballo A, Miyajima C, Takeuchi E, Takeda K. 2020. Personalized subjective driving risk: Analysis and prediction. J Robot Mechatron, 32(3): 503-519.
3. Chen H, Wu T. 2022. An improved CatBoost algorithm for red fox optimization in the field of anomaly detection. 2nd International Conference on Computer Science and Blockchain (CCSB), 28-30 October, Wuhan, China, pp: 148-153.
4. Escottá Á T, Beccaro W, Ramírez M A. 2022. Evaluation of 1D and 2D deep convolutional neural networks for driving event recognition. Sensors, 22(11): 4226.
5. Huang Y, Jafari M, Jin P. 2022. Driving safety prediction and safe route mapping using in-vehicle and roadside data. arXiv, 2022: 2209.05604.
6. Judson I R. 2015. Assigning driver risk. https://github.com/irjudson/AssigningDriverRisk (accesed date: May 12, 2023).
7. Ke G, Meng Q, Finley T, Wang T, Chen W, Ma W, Ye Q, Liu T-Y. 2017. Lightgbm: A highly efficient gradient boosting decision tree. Adv Neural Inform Process Systems, 30: 1-9.
8. Malik M, Nandal R. 2023. A framework on driving behavior and pattern using On-Board diagnostics (OBD-II) tool. Materials Today: Proc 80: 3762-3768.

9. Marafie Z, Lin K-J, Wang D, Lyu H, Liu Y, Meng Y, Ma J. 2021. AutoCoach: an intelligent driver behavior feedback agent with personality-based driver models. Electronics, 10(11): 1361.
10. Mohammed K, Abdelhafid M, Kamal K, Ismail N, Ilias A. 2023. Intelligent driver monitoring system: An Internet of Things-based system for tracking and identifying the driving behavior. Comput Stand Interfaces, 84: 103704.
11. Prokhorenkova L, Gusev G, Vorobev A, Dorogush A V, Gulin A. 2018. CatBoost: unbiased boosting with categorical features. Adv Neural Inform Process Systems, 31: 1-11.
12. Silva I, Eugenio Naranjo J. 2020. A systematic methodology to evaluate prediction models for driving style classification. Sensors, 20(6): 1692.
13. Warren J, Lipkowitz J, Sokolov V. 2019. Clusters of driving behavior from observational smartphone data. IEEE Intelligent Transport Systems Magazine, 11(3): 171-180.
14. Yuksel A, Atmaca S. 2021. Driver’s black box: A system for driver risk assessment using machine learning and fuzzy logic. J Intelligent Transport Systems, 25(5): 482-500.
15. Zantalis F, Koulouras G, Karabetsos S, Kandris D. 2019. A review of machine learning and IoT in smart transportation. Future Internet, 11(4): 94.
16. Zhang Y, Li J, Guo Y, Xu C, Bao J, Song Y. 2019. Vehicle driving behavior recognition based on multi-view convolutional neural network with joint data augmentation. IEEE Transact Vehicular Technol, 68(5): 4223-4234.