Use of Multisensor Data in Reliable Short-Term Travel Time Forecasting for Urban Roads

As multiple traffic data sources have become available recently, a new opportunity has been provided for improving the accuracy of short-term travel time forecasting by fusing different but valid data sources. However, previous studies seldom quantified and integrated the reliability of data sources into model development to achieve the potential promised by data fusion. This paper proposes a combined method for short-term travel time forecasting for urban road links that uses travel time extracted from fixed vehicle detectors and probe vehicle data. The method uses the generalized autoregressive conditional heteroscedasticity model to forecast the mean and variance of each type of travel time data source, and the Dempster–Shafer model is used to calculate the fusion weights iteratively. Real-world data collected on urban roads in Kunshan, China, were used to validate and evaluate the proposed method. Empirical results show that the proposed method can effectively capture the variance of each type of travel time data source for iteratively calculating the fusion weights and hence can produce accurate travel time forecasts. Moreover, through a comparison with the alternative methods, the proposed method is shown to be able to consistently generate improved performance under varying traffic conditions.

Trip Travel Time Forecasting Based on Selective Forgetting Extreme Learning Machine

Travel time estimation on road networks is a valuable traffic metric. In this paper, we propose a machine learning based method for trip travel time estimation in road networks. The method uses the historical trip information extracted from taxis trace data as the training data. An optimized online sequential extreme machine, selective forgetting extreme learning machine, is adopted to make the prediction. Its selective forgetting learning ability enables the prediction algorithm to adapt to trip conditions changes well. Experimental results using real-life taxis trace data show that the forecasting model provides an effective and practical way for the travel time forecasting.