Abstract

Detecting hot spots of moving vehicles in an urban area is absolutely required to many smart city applications. Crowdedness spot is a crowded area with a number of irregular objects. The practical investigation on hot spots in smart city offerings many unique features, such as highly mobile environments, the non-uniform biased samples, and supremely limited size of sample objects. The traditional density-based clustering algorithms flop to capture the actual clustering property of objects, making the outputs meaningless. In this paper we recommend a novel, called mobility-based clustering which is non-density-based approach. The basic idea is that sample objects are hired as “sensors” to recognize the vehicle crowdedness in nearby areas using their instant mobility, rather than the “object representatives”. As such the mobility of samples is certainly incorporated. Several important factors beyond the vehicle crowdedness have been identified and techniques to remunerate these effects are proposed. Furthermore, taking the identified crowdedness spots as a label of the taxi, we can identify one individual taxi to be a crowdedness taxi that crosses a number of different crowdedness spots. We estimate the presentation of our methods and baseline approaches based on real traffic situations and real-life data sets.

Authors and Affiliations