Kalman Filter-based Vehicle Count Estimation Approach Using Probe Data: A Multi-lane Road Case Study

The paper presents a novel approach for estimating the number of vehicles on multi-lane signalized approaches using only probe vehicle data. A model-based estimation technique is developed using the Kalman Filtering technique. The model-based state equation employs the traffic flow continuity equation, while the measurement equation uses the hydrodynamic equation. This study also proposes a novel variable estimation interval, leading to estimation improvements. The paper evaluates the proposed estimation method on multi-lane traffic signal approaches using simulated data. Results show that vehicle-count estimates are robust. The estimation model is evaluated considering two traffic signal control methods: a fixed-time plan and an adaptive phase split optimizer. The results demonstrate a higher estimation accuracy for fixed traffic signal timings at low traffic demand levels, while the estimation accuracy is better when the adaptive phase split optimizer is activated for high traffic demand levels. Finally, the paper investigates the sensitivity of the estimation model to vehicle length, demonstrating that the presence of longer vehicles (e.g. trucks) in the traffic stream reduces the estimation accuracy.