Evaluation of Freeway Sensor Placement Based on the Aggregation of Cellular Probe System and Loop Detectors

摘要

Travel speed estimation is a fundamental task for traffic management centers and is often a critical elementof Intelligent Transportation Systems (ITS). For this purpose, various sensors are used to collect trafficinformation. As a newly effective and practical travel speed measurement tool, cellular probe system isgaining market penetration. In this paper, the handoff system, one of cellular probe technologies, is usedfor speed detection. However, the handoff coverage size is usually large on freeways, which meansconsecutive handoff points are remote and this may affect the accuracy of speed estimation. To improvethe accuracy of speed estimation, this paper proposes a method to aggregate the cellular probe and loopdetectors. For the purposes of a rigorous comparison, data are generated from microscopic simulationmodels of virtual one-direction freeway sections under different traffic conditions. The microscopicsimulation model includes loop detectors and cellular probe system, allowing for an automated method ofprobe vehicle data collection. Results show that the increasing number of loop detectors will improve theaccuracy of speed estimation by the aggregation of cellular probe system and loop detectors. Furthermore,the analysis shows that the estimation error is large under low-flow or high-flow traffic conditions, nomatter how the cellular probe system and loop detectors are aggregated. If the result can be validated usingreal world data, it can be an excellent reference for traffic engineers to determine the sensor locations.