The taxi service provides a convenient door-to-door way of transportation and plays a uniquely important role in our daily life. As taxis often run on the road most time of a day, their movement is a good indicator of the traffic status of a city. The advances in location-acquisition and mobile computing techniques has generated massive taxi trajectory data from the GPS devices installed on taxis or the mobile devices carried by passengers. The trajectory data contain rich temporal and spatial information of the behaviours of taxis, it is appealing to analyse such data to obtain informative insights for traffic management and short-term travel planning. Given the unstructured and large-volume features of GPS trajectory data, it is still a challenge to traffic status prediction.
This project aims to explore the problem of predicting traffic congestion through analysing the taxi GPS trajectory data. Taxi GPS data sets will be provided or the students can search online for their preferred data sets given many such data sets are available. The student needs to use the big data processing platform to deal with these data sets given the large volume. Then, the students need to implement appropriate algorithms to do the prediction. A web-based interface should also be developed to visualise the map, trajectories data and prediction results.
Upon completion of the project, the students are expected to achieve
• An application predicting and visualising the congestion of a city where the data set is given
• Scalable algorithms implemented based on cloud
Lab allocations have not been finalised