Establishment of driver's driving behavior recognition and evaluation model based on vehicle big data analysis
2022-01-0258
03/29/2022
- Event
- Content
- With the rapid development of intelligent networked vehicles, the data generated by intelligent networked vehicles is also increasing explosively. Automobile manufacturers have gradually begun to realize the importance of vehicle big data analysis to preventive maintenance, safety precautions and updates of enterprise products.In this paper, according to the requirements of the enterprise project, the big data analysis of the vehicle driving data uploaded to the supervision platform by the vehicle Telematics BOX (T-Box) is carried out. Starting from the level of the car driver, the basic historical data of the vehicle is used to analyze the driver's behavior. Constructed a driving style and driving condition recognition model and a comprehensive evaluation model of driving behavior for the big data platform. For the driving style and driving condition recognition model, use a large amount of historical vehicle driving data in the enterprise cloud platform to perform data preprocessing and divide the data set. Classify samples of the original data set through an unsupervised learning algorithm. Then select appropriate machine learning classification algorithms for model training on the sample data set with classification labels, and establish a recognition model for driver style and car driving conditions. For the comprehensive evaluation model of driver's driving behavior, define common bad driving behaviors and establish the identification and evaluation models of bad driving behaviors respectively, and select appropriate weighting methods to assign weights to each kind of bad driving behaviors. Finally, according to the single bad driving behavior score and its corresponding weight, the comprehensive score of the driver's driving behavior is calculated.
- Citation
- Shi, S., "Establishment of driver's driving behavior recognition and evaluation model based on vehicle big data analysis," SAE Technical Paper 2022-01-0258, 2022, .