Predicting Traffic Congestion Time Based on Kalman Filter Algorithm

This paper mainly solved the problem of predicting the time required for vehicles to pass through congested roads. In order to obtain more accurate prediction time, a Kalman prediction model based on multiple linear regression was established in this paper. Taking the 2008 Yanan elevated road in Shanghai as an example, the measured data in this section was collected from the traffic measured data sharing network, and the above model was used to obtain good prediction results. As an improvement, we used BP neural network instead of multiple linear regression to make the prediction result more in line with the actual situation.