Baher Abdulhai

Baher Abdulhai

Auto-generated profile, based on publicly available information

Baher Abdulhai

Department of Civil and Mineral Engineering, University of Toronto, Toronto, Ontario, Canada

Research Focus & Metrics

Academic Career Length

23Years (2004-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
8 October 2021 in Processes
Verified
Journal Article
Open access
PDF file
MUMuhammad Umair
Muhammad Umair
MFMuhammad Umar Farooq
Muhammad Umar Farooq
RRRana Raza
Rana Raza
QCQian Chen
Qian Chen
BABaher Abdulhai
Baher Abdulhai
In the Intelligent Transportation System (ITS) realm, queue length estimation is one of an essential yet a challenging task. Queue lengths are important for determining traffic density in traffic lanes so that possible congestion in any lane can be minimized. Smart roadside sensors such as loop detectors, radars and pneumatic road tubes etc. are promising for such tasks though they have a very high installation and maintenance cost. Large scale deployment of surveillance cameras have shown a great potential in the collection of vehicular data in a flexible way and are also cost effective. Similarly, vision-based sensors can be used independently or if required can also augment the functionality of other roadside sensors to effectively process queue length at prescribed traffic lanes. In this research, a CNN-based approach for estimation of vehicle queue length in an urban traffic scenario using low-resolution traffic videos is proposed. The queue length is estimated based on count of total vehicles waiting on a signal. The proposed approach calculates queue length without the knowledge of any onsite camera calibration information. Average vehicle length is approximated to be 5 m. This caters for the vehicles at the far end of the traffic lane that appear smaller in the camera view. Identification of stopped vehicles is done using Deep SORT based object tracking. Due to robust and accurate CNN-based detection and tracking, the queue length estimated by using only the cameras has been very effective. This mostly eliminates the need for fusion with any roadside or in-vehicle sensors. A detailed comparative analysis of vehicle detection models including YOLOv3, YOLOv4, YOLOv5, SSD, ResNet101, and InceptionV3 was performed. Based on this analysis, YOLOv4 was selected as a baseline model for queue length estimation. Using the pre-trained 80-classes YOLOv4 model, an overall accuracy of 73% and 88% was achieved for vehicle count and vehicle count-based queue length estimation, respectively. After fine-tuning of model and narrowing the output classes to vehicle class only, an average accuracy of 83% and 93% was achieved, respectively. This shows the efficiency and robustness of the proposed approach.

14 June 2022 in Transportation Research Record: Journal of the Transportation Research Board
Verified
Journal Article
Open access
KOKareem Othman
Kareem Othman
ASAmer Shalaby
Amer Shalaby
BABaher Abdulhai
Baher Abdulhai
Exclusive bus lane (EBL) is one of the most common transit prioritization strategies implemented to improve transit speed. However, one major drawback of implementing EBLs is the associated reduction in road capacity left for other road users. In corridors with EBLs and infrequent bus service, the lanes are underutilized for extended periods of time. Dynamic bus lane (DBL), a new priority strategy enabled by vehicle connectivity, can provide buses with priority while allowing the general traffic to access the bus lane when buses are not present. Although the DBL concept is promising, a limited number of studies have explored its effectiveness under various conditions. Thus, this paper investigates the impacts of DBLs through a comparison with EBLs and mixed traffic operation under different levels of traffic demand and transit frequency. As a case study, the Eglinton East corridor in Toronto, Canada, was simulated using Aimsun Next, and different scenarios of behavioral impacts were considered in the analysis. The results reveal that DBL is a promising strategy with potential to improve the overall corridor performance over a wide range of traffic and transit service conditions, especially under intermediate traffic demand levels. On the other hand, EBL can be an efficient prioritization strategy that improves the overall corridor performance under high traffic demand and high transit frequency levels, but only if accompanied by a major mode shift from auto to transit.

28 June 2023 in Transportation Research Record: Journal of the Transportation Research Board
Verified
Journal Article
Open access
OEOmar ElSamadisy
Omar ElSamadisy
TSTianyu Shi
Tianyu Shi
ISIlia Smirnov
Ilia Smirnov
BABaher Abdulhai
Baher Abdulhai
Over the last decade, there has been rising interest in automated driving systems and adaptive cruise control (ACC). Controllers based on reinforcement learning (RL) are particularly promising for autonomous driving, being able to optimize a combination of criteria such as efficiency, stability, and comfort. However, RL-based controllers typically offer no safety guarantees. In this paper, we propose SECRM (the Safe, Efficient, and Comfortable RL-based car-following Model) for autonomous car-following that balances traffic efficiency maximization and jerk minimization, subject to a hard analytic safety constraint on acceleration. The acceleration constraint is derived from the criterion that the follower vehicle must have sufficient headway to be able to avoid a crash if the leader vehicle brakes suddenly. We critique safety criteria based on the time-to-collision (TTC) threshold (commonly used for RL controllers), and confirm in simulator experiments that a representative previous TTC-threshold-based RL autonomous-vehicle controller may crash (in both training and testing). In contrast, we verify that our controller SECRM is safe, in training scenarios with a wide range of leader behaviors, and in both regular-driving and emergency-braking test scenarios. We find that SECRM compares favorably in efficiency, comfort, and speed-following to both classical (non-learned) car-following controllers (intelligent driver model, Shladover, Gipps) and a representative RL-based car-following controller.

1 August 2011 in Journal of Transportation Engineering
Verified
Journal Article
Closed access
HAHossam Abdelgawad
Hossam Abdelgawad
BABaher Abdulhai
Baher Abdulhai
GAGlareh Amirjamshidi
Glareh Amirjamshidi
MWMohamed Wahba
Mohamed Wahba
CWClarence Woudsma
Clarence Woudsma
MRMatthew J. Roorda
Matthew J. Roorda
This paper assesses the impact of exclusive truck facilities on urban freeway performance. A large-scale regional microscopic traffic simulation model is developed for morning and afternoon peak hours and is used to model two alternative truckway configurations in the Greater Toronto Area (GTA), including a truck-only highway and a truck lane conversion on Highway 401. Demand inputs to the microsimulation model are generated by a regional transportation demand model that provides origin-destination (OD) matrices for light, medium, and heavy trucks and passenger cars. Plug-ins are developed to represent a truck-only lane in the simulation environment. The simulation model is successfully calibrated to reflect observed road counts and to produce realistic congestion patterns. The effect of infrastructure changes on travel distances, travel times, exclusive truck lane usage, and travel speeds is assessed. Microscopic simulation allows queuing formation/dissipation and bottlenecks to be represented and allows...

22 June 2012 in Neural Computing and Applications
Verified
Journal Article
Closed access
PDF file
JAJavad Abdi
Javad Abdi
BMBehzad Moshiri
Behzad Moshiri
BABaher Abdulhai
Baher Abdulhai
ASAli Khaki Sedigh
Ali Khaki Sedigh

1 October 2019
Verified
Conference Paper
Closed access
SSSoheil Mohamad Alizadeh Shabestray
Soheil Mohamad Alizadeh Shabestray
BABaher Abdulhai
Baher Abdulhai