Signal Timing Estimation Using Sample Intersection Travel Times

Abstract

Abstract-Signal timing information is important in signal operations and signal/arterial performance measurement. Such information, however, may not be available for wide areas. This imposes difficulty, particularly for real-time signal/arterial performance measurement and traffic information provisions that have received much attention recently. We study, in this paper, the possibility of using intersection travel times, i.e., those collected between upstream and downstream locations of an intersection, to estimate signal timing parameters. The method contains three steps: 1) cycle breaking that determines whether a new cycle starts; 2) exact cycle boundary detection that determines when exactly a cycle starts or ends; and 3) effective red (or green) time estimation that estimates the actual duration of the red (or green) time. The proposed method is a combination of traffic flow theory and learning/estimation algorithms and can be used to estimate the cycle-by-cycle signal timing parameters for a specific movement of a signal. The method is tested using data from microscopic simulation, field experiments, and next-generation simulation with promising results.

Towards fine-grained urban traffic knowledge extraction using mobile sensing

“…The results however are not reliable due to oscillation and noise in measurements, and especially low penetration rate of mobile data. In [9], the SVM model applies two features: the arrival time difference t i − t i−1 and the delay difference τ i − τ i−1 between two consecutively sampled vehicles. Here t i is the ith sample vehicle's arrival time at VTL1 and τ i is the intersection delay of the ith sample vehicle.…”

“…The second feature is exactly what was used in [2]. Figure 5 depicts these two features for a field test under 60% and 30% penetration rate of travel time data [9]. In the figure, dots are for NCBVs and plus signs are for CBVs.…”

“…SVM divides the data set into two groups: one for y i = 1 and the other for y i = −1. It can further produce two support planes (lines in the R The distance between these two lines can be shown as 2/||w|| with ||w|| denoting the norm of w. If we aim to maximize the distance between these two support planes, (w, b) can be determined by solving the following SVM problem [9]:…”

“…After solving the SVM model, the (w, b) pair, in particular, the two planes (lines): wx i −b = 1 and wx i −b = −1, can be used to identify whether a given data sample xj is a CBV or NCBV. Figure 6 shows the results of applying the SVM model to break cycles using the datasets of simulation and NGSIM (next generation simulation, widely used for traffic research) [9], based on which to estimate the cycle boundaries and lengths. Compared with the previous method in [2], the root mean square error (RMSE) of the results is much improved: from 9.6 seconds to 1.6 seconds for the simulation data, and from 8.2 seconds to 3.0 seconds for the NGSIM data.…”

“…Arterial traffic however is often disrupted, e.g., by traffic signals, resulting in discontinuities and kinks in arterial traffic flow states. Such unique features can actually be utilized to reconstruct arterial traffic flow patterns such as delay (or travel times) [2] and queue lengths [1,6], as well as signal timing information [9].…”