Vehicle Driving Behavior Recognition and Optimization Strategies Based on Cloud Computing and SSA-BP Algorithm

Xing YANG, Ke XIANG, Shan YUAN, Jilan HUANG

Abstract

As the number of cars increases annually, driving behavior analysis has become a direction worthy of in-depth research in the field of public transportation. Traditional vehicle brakes generate torque during non-braking high-speed rotation, which has a negative impact on transmission efficiency. In this paper, the sparrow search algorithm was combined with the backpropagation neural network algorithm in the context of collecting and analysing multi-source data during vehicle driving, and the backpropagation neural network algorithm was improved by using the sparrow search algorithm`s global search ability. A driving behavior recognition model was built in order to accurately distinguish various driving behaviors and evaluate their impact on fuel consumption and driving safety. The experimental outcomes showed that the vehicle driving behavior recognition method based on cloud computing and the backpropagation neural network algorithm combined with the sparrow search algorithm featured a high driving behavior recognition efficiency, and the research method featured a small error in the identification of vehicle driving behavior. The prior accuracy of the driving behavior-based fuel consumption evaluation model was as high as 89.216%, and the regression fit for the employed model after training reached around 98%. To sum up, the obtained results not only allow the implementation of new ideas and methods for the intelligent vehicle driving and traffic safety management, but they also provide an important theoretical support and practical guidance for the design and optimization of intelligent transportation systems.

Keywords

Sparrow search algorithm, BP neural network, Cloud computing, Vehicle driving behavior, Cloud platform.

View full article