Browsing by Subject "Traffic signal"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
Item Improving traffic signal performance using high-resolution data(2013-12) Hu, HengTraffic congestion has become a more and more severe problem for metropolitan areas all over the world. Although lots of effort has been devoted to improve the operation on major corridors, how to efficiently and effectively manage traffic based on the existing infrastructure is still a challenging task, and the task seems even more challenging for signalized arterials due to lack of traffic monitoring and data collection system. This research aims to improve the traffic signal performance based on the collected high-resolution traffic signal data and the derived performance measures by the SMART-Signal system developed at the University of Minnesota. In particular, this research focuses on the following three areas: 1) Optimize offsets to reduce congestion Traditionally, offset optimization for coordinated traffic signals fails to consider the stochastic nature of field traffic. Using the archived high-resolution traffic signal data, in this research, we develop an arterial offset optimization model which will take two well-known problems with vehicle-actuated signal coordination into consideration: the early return to green problem and the uncertain intersection queue length problem. To account for the early return to green problem, we introduce the concept of conditional distribution of the green start times for the coordinated phase. To handle the uncertainty of intersection queue length, we adopt a scenario-based approach that generates optimal offsets using a series of traffic demand scenarios as the input to the optimization model. Both the conditional distributions of the green start times and traffic demand scenarios can be obtained from the archived high-resolution traffic signal data. Under different traffic conditions, queues formed by side-street and main-street traffic are explicitly considered in the derivation of intersection delay. The objective of this model is to minimize total delay for the main coordinated direction and at the same time it considers the performance of the opposite direction. The results from the field implementation show that the proposed model can reduce travel delay of coordinated direction significantly without compromising the performance of the opposite approach. 2) Manage oversaturated signal arterials.Under oversaturated traffic conditions, signal timings need to be adjusted accordingly in order to alleviate the detrimental impacts caused by oversaturation. In this research, our focus is to mitigate two types of detrimental effects, signal phase failure with residual queue and downstream queue spillover. Building upon the previous work on the oversaturation severity indices, a maximum-flow based approach to manage oversaturated intersections is developed. The proposed model maximizes the discharging capacity along oversaturated routes, while satisfying the constraints on available green times. We show that a simple forward-backward procedure (FBP) can be used to obtain the optimal solution to the maximum flow model. The forward process aims to increase green time to mitigate oversaturation, therefore improve the throughput for the oversaturated approach; and the backward process aims to gate the traffic at some intersections to prevent residual queues and downstream queue spill-back when the available green time is insufficient. The algorithm is tested using a microscopic traffic simulation model for an arterial network in the City of Pasadena, CA. The results indicate the model can effectively and efficiently reduce oversaturation and improve system performance.3) Mange integrated corridors An integrated control model is proposed to manage traffic congestion along a freeway and a parallel signalized arterial. This model focuses on freeway diversion, which aims to utilize available capacities along parallel arterial routes to reduce network congestion. The potential impact of the diverting traffic to the performance of the arterial route is considered in this research and the maximum flow based signal control model is utilized to manage congestion on the arterial route. The integrated control model does not need the time-dependent traffic demand information as most of previous approaches do and it is suitable for online applications because of its low computation burden. The model is tested using the microscopic traffic simulation in the I-394 and TH 55 corridor in Minneapolis, MN. The results indicate that the model can effectively and efficiently reduce network congestion.Item Vehicle routing problems in signalized traffic networks(2014-08) Sun, JieThis dissertation studied various path search problems when traffic signal information and traffic state is explicitly considered. The research is motivated by the increasing availability of high-resolution traffic data including signal information, which is seldom available in the past. In order to properly account for the randomness resulting from vehicle-actuated traffic signals and the correlation from signal coordination, the theory of Markov decision process (MDP) is used. By taking advantage of the cyclic property of traffic signals, the problem is formulated as an infinite horizon and finite state space MDP with absorbing state set. The objective is to find the optimal policy that gives the minimum expected total cost to the destination.The state space of the problem is generated based on underlying traffic network geometry and signal control information. Delay distributions at intersections together with signal control parameters, such as cycle length and offset, are used to construct the transition probabilities between states. It will be shown that the required delay distributions can be estimated from readily available field traffic data. The problem where the cost is travel time is first studied. When the cost of concern is the travel time, it includes intersection delays and link travel times. Value iteration method is used to solve the MDP problem when there is only one cost of concern.In addition, the problem whose cost of concern is environmentally related is also studied. Vehicle trajectories are estimated based on traffic signal information and queuing dynamics at intersections, and put into microscopic vehicle emission models, the results from which are used to calculate the environmental costs for the path search problem. When multiple costs of concern present, the problem is formulated as a constrained MDP problem. Linear programming formulation of MDP is introduced to solve constrained MDP problem. The proposed methods are tested in a hypothetical traffic network, as well as a real world traffic network in the City of Pasadena, CA.