Spatiotemporal summarization of traffic data streams
Proceedings of the ACM SIGSPATIAL International Workshop on GeoStreaming
A data stream-based evaluation framework for traffic information systems
Proceedings of the ACM SIGSPATIAL International Workshop on GeoStreaming
Geospatial stream query processing using Microsoft SQL Server StreamInsight
Proceedings of the VLDB Endowment
Traffic jams detection using flock mining
ECML PKDD'11 Proceedings of the 2011 European conference on Machine learning and knowledge discovery in databases - Volume Part III
| Hi-index | 0.00 |
Trajectories based on GPS tracks have been studied for a number of years but only to a limited degree been used for analyzing and monitoring traffic. This paper shows how novel and important information about traffic can be computed from trajectories. Concretely the paper proposes to compute the central metric free-flow speed from trajectories, instead of using point-based measurements such as induction-loops. This free-flow speed is widely used to compute and monitor the congestion level. The paper argues that the actual travel-time is a more accurate metric. The paper suggests a novel approach to analyzing individual intersections that enables traffic analysts to compute queue lengths and estimated time to pass an intersection. Finally, the paper uses associative rule mining for evaluating green waves on road stretches. Such information can be used to verify that signalized intersections are correctly coordinated, and navigational device manufacturers to advice drivers in real-time on expected behavior of signalized intersections. The main conclusion is that trajectories can provide novel insight into the actual traffic situation that is not possible using existing approaches. Further, extracting this information requires no expensive changes to the road-network infrastructure, which is a problem with the technologies currently used.