Real-Time Estimation of Vehicle Counts on Signalized Intersection Approaches Using Probe Vehicle Data

This paper presents a novel method for estimating the number of vehicles traveling along signalized approaches using probe vehicle data only. The proposed uses Kalman Filtering technique to produce reliable count estimates real-time expected travel times. introduces variable estimation interval that allows higher precision, as updating time always contains fixed vehicles. is evaluated empirical and simulated data, former which were collected roadway in downtown Blacksburg, VA. Results indicate produced by are accurate. also examines model's accuracy when installing single stationary sensor (e.g., loop detector), producing slight improvements especially market penetration rate low. Finally, investigates sensitivity model traffic demand levels, showing works better at levels given more exist same rate.