Estimating traffic signal phases from turning movement counters

Gahrooei, Mostafa Reisi; Work, Daniel B.

doi:10.1109/itsc.2013.6728381

Cited by 4 publications

(8 citation statements)

References 16 publications

(18 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The first test is similar to the first numerical experiment in [12] and serves as a benchmark to highlight the significant performance improvement of the new Bayesian method proposed in this work. To synthetically generate traffic maneuver observations, a state sequence is constructed from alternating phases p 1 and p 5 [shown in Fig.…”

Section: ) Intersection Of a One-way Street With A Two-way Streetmentioning

confidence: 99%

“…First, a synthetic data set is generated and used to train an HMM [with initial parameters generated from Dirichlet distribution parameters as in (10)- (12)] using the Bayesian learning algorithm. The trained HMMs are then used to infer the phase sequence corresponding to the maneuver counts in the generated data set.…”

Section: ) Intersection Of a One-way Street With A Two-way Streetmentioning

confidence: 99%

“…This paper is an extended and improved version of our preliminary work [12], which tested the approach with synthetic numerical data and experimental data using the Baum-Welch learning algorithm on fixed-time intersections. When inferring on experimental data, the earlier algorithm was unable to correctly distinguish between phases when specific maneuvers were rarely observed.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Inferring Traffic Signal Phases From Turning Movement Counters Using Hidden Markov Models

Gahrooei

Work

2015

IEEE Trans. Intell. Transport. Syst.

Self Cite

View full text Add to dashboard Cite

This work poses the problem of estimating traffic signal phases from a sequence of maneuvers. We model the problem as an inference problem on a discrete-time hidden Markov model (HMM) in which maneuvers are observations and signal phases are hidden states. The model is calibrated from maneuver observations using either the classical Baum-Welch algorithm or a Bayesian learning algorithm. The trained model is then used to infer the traffic signal phases on the data set via the Viterbi algorithm. When training with the Bayesian learning algorithm, we set the prior distribution as a Dirichlet distribution. We identify the best parameters of the prior distribution for both fixed-time and sensor-actuated signals using numerical simulations and employ them in the field experiments. It is shown that when the model is trained by the Bayesian learning method with appropriate prior parameters from the Dirichlet distribution, the inferred phases are more accurate in both numerical and field experiments. Because the best set of prior parameters for a fixed-time intersection is different from those for sensor-actuated signals, a classification strategy to distinguish between these two types of signals is proposed. The supporting source code and data are available for download at https://github.com/reisiga2/TrafficSignalPhaseEstimation. Index Terms-HiddenMarkov model, traffic signal phase estimation, TrafficTurk.

show abstract

Section: ) Intersection Of a One-way Street With A Two-way Streetmentioning

confidence: 99%

Section: ) Intersection Of a One-way Street With A Two-way Streetmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Inferring Traffic Signal Phases From Turning Movement Counters Using Hidden Markov Models

Gahrooei

Work

2015

IEEE Trans. Intell. Transport. Syst.

Self Cite

View full text Add to dashboard Cite

show abstract

“…The system builds a database of fixed time signals, while adaptive signals are predicted with a support vector machine using a week long log of the adaptive signals. Our work differs from SignalGuru, both in the sensing (Traffic Turk measures vehicle maneuvers, from which the signal phase timing must be inferred [9], and not directly the traffic signal), and in the control objective estimation approach. The SMART-SIGNAL [15] system is another initiative that aims to collect high-resolution data from signalized intersections and use it to infer useful knowledge of the traffic system.…”

Section: B Related Workmentioning

confidence: 99%

“…For the purpose of inverse optimal control, we can com bine (6) and (8) for multi-objective optimization as (9) where…”

Section: Objective Fu Nctionsmentioning

confidence: 99%

Estimating traffic control strategies with inverse optimal control

Gowrishankar

Work

2013

16th International IEEE Conference on Intelligent Transportation Systems (ITSC 2013)

Self Cite

View full text Add to dashboard Cite

Ahstract-This article formulates the problem of estimat ing the traffic control strategy of a single intersection traf fic signal controller from observed data. Building on previ ous results in the literature on optimal traffic signal con trol and convex inverse optimal control problems, a method is proposed to efficiently solve the inverse optimal traffic signal control problem to recover the control objective of the observed traffic controller. The resulting program is tested in a numerical experiment with synthetically generated data. Supporting source code is available for download at https:llgithub.comlsgowris2IInverseOptimaIControl.

show abstract

Signal timing parameters inference method at intersections using license plate recognition data

Lin

Chen

et al. 2022

IET Intelligent Trans Sys

View full text Add to dashboard Cite

Signal timing parameters are essential components in traffic signal control (TSC). It affects not only traffic management, but also traffic safety. However, due to the confidential issues of the traffic management department, or lack of data integration from different signal manufacturers, it is intractable to obtain the city‐scale signal timing data. In the previous studies, some existing estimation methods focused on a single parameter and fixed‐timing scheme. To tackle this issue, this study attempts to develop an integrated parameters inference method based on license plate recognition (LPR) data, considering phase weight, average phase duration information and the overall phases of the intersections. In particular, the proposed method includes phase sequence inference model, cycle length inference model and phase duration inference model. To testify the performance of the proposed method, a real‐world LPR dataset from Guangzhou, China, is applied. Numerical results show that the proposed method performs more efficiently on parameter inferences than the state‐of‐the‐art (SOTA) approach. For instance, in the given research time period, the mean absolute error (MAE) of each phase duration is 2 s averagely (6.32 s in the SOTA approach), and the mean relative error (MRE) of cycle length is 0.91% (11.67% in the SOTA approach).

show abstract

Estimating traffic signal phases from turning movement counters

Cited by 4 publications

References 16 publications

Inferring Traffic Signal Phases From Turning Movement Counters Using Hidden Markov Models

Inferring Traffic Signal Phases From Turning Movement Counters Using Hidden Markov Models

Estimating traffic control strategies with inverse optimal control

Signal timing parameters inference method at intersections using license plate recognition data

Contact Info

Product

Resources

About