Montasir Abbas

A Heuristic Approach for Selecting Highway Investment Alternatives

A heuristic approach is developed for systemwide highway project selection. It can assess changes in total project benefits using different project implementation options under budget uncertainty and identify the best option to achieve maximized total benefits. The proposed approach consists of a stochastic model formulated as the zero/one integer doubly constrained multidimensional knapsack problem and an efficient heuristic solution algorithm developed using the Lagrange relaxation technique. A method is also introduced to improve the upper bound for the objective function by simultaneously changing multiple Lagrange multipliers. The approach is applied in a computational study to obtain a comprehensive highway investment plan for a State-maintained highway system in the United States.

IMPROVED DETECTION AND CONTROL SYSTEM FOR ISOLATED HIGH-SPEED SIGNALIZED INTERSECTIONS

High-speed signalized intersections have both safety and efficiency problems. The safety problems traditionally have been overcome by using advance detectors. However, at relatively moderate flow rates, multiple advance detector systems tend to extend the green phase to its maximum allowable duration (i.e., max-out). At max-out, drivers on the intersection approach may be faced with the decision to proceed or stop. This creates the safety problem the system was intended to prevent. Also, extending phases to their maximums increases delays to other movements. A new approach to high-speed intersection control is outlined—a dynamic dilemma zone allocation system that uses measured vehicle speeds with a control algorithm to decide when to end the signal phases. The new system was evaluated by simulation and field installation. In both instances, the new system indicated it could improve safety and maintain efficient operations at high-speed intersections.

Self-organizing control framework for driverless vehicles

Development of in-vehicle computer and sensing technology, along with short-range vehicle-to-vehicle communication has provided technological potential for large-scale deployment of autonomous vehicles. The issue of intersection control for these future driverless vehicles is one of the emerging research issues. Contrary to some of the previous research approaches, this paper is proposing a paradigm shift based upon self-organizing and cooperative control framework. Distributed vehicle intelligence has been used to calculate each vehicles approaching velocity. The control mechanism has been developed in an agent-based environment. Self-organizing agents trajectory adjustment bases upon a proposed priority principle. Testing of the system has proved its safety, user comfort, and efficiency functional requirements. Several recommendations for further research are presented.

Simulation-Based Optimization of Maximum Green Setting Under Retrospective Approximation Framework

Most traffic signal systems work under highly dynamic traffic conditions, and they can be studied adequately only through simulation. As a result, how to optimize traffic signal system parameters in a stochastic framework has become increasingly important. Retrospective approximation (RA) represents the latest theoretical development in stochastic simulation. Under the RA framework, the solution to a simulation-based optimization problem can be approached with a sequence of approximate optimization problems. Each of these problems has a specific sample size and is solved to a specific error tolerance. This research applied the RA concept to the optimal design of the maximum green setting of the multidetector green extension system. It also designed a variant of the Markov monotonic search algorithm that can accommodate the requirements of the RA framework, namely, the inheritable Markov monotonic search algorithm, and implemented the RA-based optimization engine within VISSIM. The results show that the optimized maximum green can considerably increase composite performance (reducing delay and increasing safety) compared with traditional designs. The optimization methodology presented in this paper can easily be expanded to other signal parameters.

Simulation of Driver Behavior with Agent-Based Back-Propagation Neural Network

Two microscopic simulation methods are compared for driver behavior: the Gazis–Herman–Rothery (GHR) car-following model and a proposed agent-based neural network model. To analyze individual driver characteristics, a back-propagation neural network is trained with car-following episodes from the data of one driver in the naturalistic driving database to establish action rules for a neural agent driver to follow under perceived traffic conditions during car-following episodes. The GHR car-following model is calibrated with the same data set, using a genetic algorithm. The car-following episodes are carefully extracted and selected for model calibration and training as well as validation of the calibration rules. Performances of the two models are compared, with the results showing that at less than 10-Hz data resolution the neural agent approach outperforms the GHR model significantly and captures individual driver behavior with 95% accuracy in driving trajectory.

Development and Field Testing of Platoon Identification and Accommodation System

This paper describes the development and initial field testing of an intelligent traffic control system for identifying and progressing platoons at isolated traffic signals on signalized arterials. This system uses advance detection to obtain real-time information about the presence and speeds of individual vehicles. Its algorithm identifies whether a platoon—of a userspecified minimum size and density—is approaching the signal and estimates platoon arrival time at the stop bar. When warranted, the system issues a low-priority preemption signal to progress the detected platoon. Duration of the initial preemption signal is based on estimated arrival and departure times for the smallest acceptable platoon and the estimated time needed to clear any queues present at the signal. Then, the system switches to an extension mode and provides progression to any additional vehicles determined to be in the platoon. It accomplishes that by increasing preemption time until no more vehicles are determined to be in the ...

Modeling ring-barrier traffic controllers using Colored Timed Stochastic Petri Nets

As one of the many techniques used in modeling traffic processes and systems, Petri Nets are recognized as a tool for modeling in traffic signal control. In this paper, ring-barrier traffic signal control structure is modeled using Petri Nets. Colored Timed Stochastic Petri Nets is used in this paper to provide additional modeling capabilities. The proposed model incorporates all the main features of ring-barrier structure and includes the modeling of left-turning vehicles. We also describe and discuss possible control structures, previously developed Petri Net models and implementation issues.

A two-step segmentation algorithm for behavioral clustering of naturalistic driving styles

This research effort aims to investigate the hypothesis that drivers apply different driving styles in their daily driving tasks. A two-step algorithm is used for segmentation and clustering. First, a car-following period is broken into different duration segments that account for their temporal distribution. Second, the segments produced by the previous step are clustered based on similarity. Variations of k-means clustering and optimization techniques are used in this process. The segments centroids, used for clustering, are 8-dimensional and are produced by taking the average of the data points in each segment based on longitudinal acceleration, lateral acceleration, gyro (yaw rate), vehicle speed, lane offset, gamma (yaw angle), range, and range rate. The results of this methodology are continuous segments of car-following behavior as well as clusters of segments that show similar data and thus similar behaviors. The sample used in this paper included three different truck drivers that are representative of a high-risk driver, a medium-risk driver, and a low-risk driver. . In summary, the results revealed behavior that changed within a car-following period, between car-following periods, and between drivers. Each driver showed a unique distribution of behavior, but some of the behaviors existed in more than one driver but at different frequencies.

Life-Cycle Cost Analysis of Highway Intersection Safety Hardware Improvements

A methodology is proposed for project-level life-cycle cost analysis of highway intersection safety hardware improvements. It incorporates a disaggregated risk-based approach for computing the safety index that could be used to assess intersection vehicle crash risks affected by the conditions of intersection safety hardware such as signs, signals, lighting, pavement markings, and guardrails. With safety indices estimated before and after implementation of an intersection safety hardware project, the annual potential for safety improvements (PSI) could be computed using the concept of consumer surplus. The annual PSI is further converted into dollar values and extrapolated to the overall intersection safety hardware service life cycle, defined as the useful service life of longest-lasting intersection safety hardware, to determine the maximum project-level life-cycle benefits of intersection safety hardware improvements. A computational study is conducted for methodology application and validation using 5-year data on 226 intersections in Ozaukee County, Wis. The proposed methodology could be adopted by state and large-scale local transportation agencies for intersection safety hardware project evaluation and investment decision making.

Evaluating Green-Extension Policies with Reinforcement Learning and Markovian Traffic State Estimation

Several protection algorithms strive to reduce the number of vehicles trapped in the dilemma zone. These algorithms use some arbitrary policies such as terminating the green when only one vehicle is present in the dilemma zone and the dilemma zone has not cleared after a certain period of time. The research proposes a control agent that is able to develop and adapt an optimal policy by learning from the environment. The agent incorporates a Markovian traffic state estimation into its learning process. A novel approach is presented for controlling traffic signals so that the number of vehicles trapped in the dilemma zone is reduced in an optimal fashion according to changes in traffic states. A comparison between the proposed optimal policy and the emerging detection-control system two-stage policy was conducted, and it was found that the policy based on reinforcement learning reduced the number of vehicles caught in the dilemma zone by up to 32%.