Shoufeng Lu

Changsha University Of Sceince And Technology, Traffic & Transportation Engineering, Faculty Member

Followers

Following

Public Views

Interests

Uploads

Papers by Shoufeng Lu

Adaptive and Coordinated Traffic Signal Control Based on Q-Learning and MULTIBAND Model

Abstract-Adaptive and coordinated signal setting has been research emphasis. Cycle, split, and of... more Abstract-Adaptive and coordinated signal setting has been research emphasis. Cycle, split, and offset are three important parameters. Lessons and experiments about adaptive signal control have shown that cycle and phase sequence renewal interval should be less than 20 minutes. So frequently optimized parameters are split and offset. For Webster signal setting theory, the ratio between flow rate to saturation flow rate is the determining parameter. But this ratio is not sensitive to small flow rate change. Therefore, the paper integrates Q-learning with Multiband model to realize adaptive and coordinated signal setting, in which the former optimizes split, the latter optimizes offset. Based on this integrated model, adaptive and coordinated signal setting for the three-intersections artery is done.

Download

CombinedTraffic Signal Control andRouteGuidance: Multiple UserClass Traffic Assignment ModelversusDiscrete Choice Model

Intelligent Transportation Systems (ITS) arethe majorsolutions torelieving traffic congestion. Tr... more Intelligent Transportation Systems (ITS) arethe majorsolutions torelieving traffic congestion. Traffic signal control androuteguidance, twokeysubsystems of ITS,jointly influence traffic flowintimeandspace. Over thelast threedecades, combined traffic signal control and routeguidance (CTSCRG)hasbeenresearch emphasis. Therearetwokindsofmodelfordescribing CTSCRG: Multiple UserClass Traffic Assignment ModelandDiscrete ChoiceModel.Theintent ofthepaperistocomparethe traffic equilibrium results ofthesetwo models.For describing routeguidance systemwhichiscomposedof guideddriverand non-guided driver, theconceptof multiple userclass (MUC)isintroduced. InMUC traffic assignment model, guided drivers aremodeled ascomplete information userequilibrium traffic assignment, andnon- guided drivers aremodeledasstochastic userequilibrium traffic assignment. Indiscrete choice model, guided drivers andnon-guided drivers aremodeled asstochastic dynamic useroptimumandroutechoice modelislogit model.The percep...

A Discrete Velocity Traffic Kinetic Model Including Desired Speed

Research Journal of Applied Sciences, Engineering and Technology

We introduce the desired speed variable into the table of games and formulate a new table of game... more We introduce the desired speed variable into the table of games and formulate a new table of games and the corresponding discrete traffic kinetic model. We use the hybrid programming technique of VB and MATLAB to develop the program. Lastly, we compared the proposed model result and the detector data. The results show that the proposed model can describe the traffic flow evolution.

Download

Using taxi GPS data for macroscopic traffic monitoring in large scale urban networks: calibration and MFD derivation

Transportation Research Procedia

В статье рассматривается художественное своеобразие фильма Фрица Ланга "Завещание доктора Мабузе"... more В статье рассматривается художественное своеобразие фильма Фрица Ланга "Завещание доктора Мабузе" в соотнесении с ранними работами режиссёра. Делается вывод о том, что именно в этой картине автор, формально прибегая к техническим приёмам, свойственным немецкому экспрессионизму, отказывается от способов восприятия мира, характерных для этого художественного течения, и реализуется в жанре реалистического детектива. В работе обосновывается отказ от традиционного понимания киноленты как аллегории, направленной на разоблачение политики национал-социализма.

Download

Mesoscopic Traffic Flow Model Considering Overtaking Requirements

Journal of Highway and Transportation Research and Development (English Edition)

AbstractOvertaking probability formula in the Prigogine—Herman traffic flow mesoscopic model is t... more AbstractOvertaking probability formula in the Prigogine—Herman traffic flow mesoscopic model is the function of traffic density, which is a linear function and does not consider overtaking requirements. In this paper, we use desired speed to improve the overtaking probability formula and propose a new overtaking probability formula that is nonlinear and a corresponding traffic flow mesoscopic model. The improved model can simultaneously consider traffic density and overtaking requirements, thus reflecting traffic flow operation realistically. We use the proposed model to simulate the diffusion process of vehicles from a high-density section to a low-density section. The example analyzes the difference between the linear overtaking probability formula and the parabolic Greenshield’s overtaking probability formula. Results show that the linear overtaking probability formula evolves quickly and converges to three speed classes. The parabolic Greenshield’s overtaking probability formula converges to six speed...

Sensitivity Analysis of Combined Boundedly Rational System Optimal Traffic Assignment and Arterial Green-Wave Signal Control Model

Information Computing and Automation, 2008

Platoon Dispersion Prediction Under the Condition of Adjacent Cycle Traffic Flow Overlapping Based on Least Square SVM

Information Computing and Automation, 2008

A Macroscopic Timing Optimization Model for Signalized Intersections Based on an Extended Internal State Node Model

Communications in Computer and Information Science, 2010

This paper aims to construct the macroscopic timing optimization model for signalized intersectio... more This paper aims to construct the macroscopic timing optimization model for signalized intersections with considering traffic flow dynamics. First, Lebacque’s internal state node model is extended to the situation of signalized intersections. The formulas of travel time are derived respectively for one section and multiple sections by using cumulative curve method. Then a macroscopic timing optimization model for signalized intersections is presented. Finally, some examples concerning under-saturated and over-saturated intersections are presented, which give support to the proposed model.

Platoon Dispersion Prediction Under the Condition of Adjacent Cycle Traffic Flow Overlapping Based on Support Vector Regression

Cisw, 2007

Deriving the Macroscopic Fundamental Diagram for an urban area using counted flows and taxi GPS

This paper applies the Macroscopic Fundamental Diagram (MFD) and the Generalized Macroscopic Fund... more This paper applies the Macroscopic Fundamental Diagram (MFD) and the Generalized Macroscopic Fundamental (GMFD) Diagram to estimate traffic characters for an urban area. We used traffic flow data manually counted from intersection traffic video and taxi GPS data to derive MFD and GMFD. The method is suitable to recognize traffic situation. These figures show the property of the road network, which consists of infrastructure and control. Mastering the properties of the network supply side, we can better manage and control traffic demand. In the process of data analysis, we found that the time step for data processing is a determinant for the MFD and GMFD shapes. That means different time step produces different results.

Macroscopic model for interrupted traffic flow of signal controlled intersection

2009 ISECS International Colloquium on Computing, Communication, Control, and Management, 2009

For urban road network, traffic flow at the intersections is the interrupted traffic flow. Recent... more For urban road network, traffic flow at the intersections is the interrupted traffic flow. Recently, applied mathematics developed hydrodynamic traffic flow models on networks, in which signal timing was not considered. The paper put forward hydrodynamic model for interrupted traffic flow at the intersection, in which entropy condition was delay minimization. Based on this proposed model, we presented delay calculation method with physical queue.

Stochastic prospect value model for simultaneous route choice

2009 ISECS International Colloquium on Computing, Communication, Control, and Management, 2009

Following the studies of human decision making under risk and uncertainty, traffic scientists had... more Following the studies of human decision making under risk and uncertainty, traffic scientists had applied prospect theory to route choice modeling, and studied prospect value user equilibrium. But most papers focused on that the travelers choose route one by one basing on prospect theory, did not study collective effect of simultaneous route choice. The paper analyzed the influence factors on travel time uncertainty. Following the modeling method of stochastic expected utility theory, we put forward stochastic prospect value model. For one OD pair toy network, we calculated stochastic prospect value traffic assignment for different reference point for day-to-day route choice, in which disseminated information is the latest day travel time. Lastly, we compared the flow pattern between stochastic prospect value traffic assignment and stochastic expected utility traffic assignment. Simultaneous route choice don't converge to equilibrium, and the larger the reference point is, the smaller flow fluctuation is.

Based on Hybrid Genetic Algorithm and Cellular Automata Combined Traffic Signal Control and Route Guidance

2007 Chinese Control Conference, 2006

Signal control and route guidance jointly influence traffic flow in time and space. Over the last... more Signal control and route guidance jointly influence traffic flow in time and space. Over the last three decades, combined traffic signal control and route guidance (CTSCRG) has been research emphasis. Firstly, the conceptual structure and definition of CTSCRG was analyzed. Then, the mathematical models of CTSCRG were summarized. Link travel time function and signal control policy have significant influence on solution uniqueness and convergence of CTSCRG model. Simulation-based method can allow more complex interactions, therefore win in reality value than travel time formula. The paper combines hybrid genetic algorithm with cellular automata simulation to calculate travel time and optimize signal setting plan. Iterative simulation and assignment procedure is built: road is discretized by cellular automata. Traffic flow dynamics is represented by cell transmission model; signal setting is optimized by hybrid genetic algorithm; vehicle agent can receive route guidance information and select suggested route. The simulation result is encouraging, when combined traffic signal control and route guidance equilibrium converge, the saving in total travel time is 54.4%.

Vehicular Traffic Flow Dispersion Prediction of Coordinated Signal Control Based on Support Vector Regression

2007 Chinese Control Conference, 2006

Signal control can reduce traffic delay and improve traffic safety. But when the vehicle meets wi... more Signal control can reduce traffic delay and improve traffic safety. But when the vehicle meets with red light, it must stop at intersection to wait green light. Coordinated signal control can improve the continuity of vehicular traffic flow movement and reduce delay. Coordinated signal setting is based on platoon dispersion prediction. Thus, improving prediction accuracy can obtain significant benefit for signal coordination. The paper bases on support vector regression to predict platoon dispersion and compares prediction accuracy with Robertson formula.

Adaptive and coordinated traffic signal control based on Q-learning and Multiband model

2008 IEEE Conference on Cybernetics and Intelligent Systems, 2008

Adaptive and coordinated signal setting has been research emphasis. Cycle, split, and offset are ... more Adaptive and coordinated signal setting has been research emphasis. Cycle, split, and offset are three important parameters. Lessons and experiments about adaptive signal control have shown that cycle and phase sequence renewal interval should be less than 20 minutes. So frequently optimized parameters are split and offset. For Webster signal setting theory, the ratio between flow rate to saturation flow rate is the determining parameter. But this ratio is not sensitive to small flow rate change. Therefore, the paper integrates Q-learning with Multiband model to realize adaptive and coordinated signal setting, in which the former optimizes split, the latter optimizes offset. Based on this integrated model, adaptive and coordinated signal setting for the three-intersections artery is done.

Download

Unfairness of System Optimal Traffic Assignment Strategy for Signal Controlled Network

International Conference on Transportation Engineering 2007, 2007

For System Optimal (SO) criteria, small fraction of the traffic is sacrificed to the higher cost ... more For System Optimal (SO) criteria, small fraction of the traffic is sacrificed to the higher cost length routes because this reduces the congestion experienced by the overwhelming majority of network users. Roughgarden studied how much worse off drivers are in SO assignment than in User Equilibrium (UE). For non-signal controlled network, he quantified this worse off with the unfairness concept. Unfairness of SO criteria for signal controlled network has not received much attention. For signal controlled network, unfairness is difficult to be derived from analytical formulation because there are complex interactions between signal control, traffic flow dynamics and traffic assignment. Travel time does not always increase with traffic counts. This is different from non-signal controlled network, in which, travel time monotonously increases with flow rate. Signal control policy can divert unfairness from one route to another route. The paper extends the unfairness concept to signal controlled network and adopts simulation-based method to compute unfairness.

Traffic Safety Oriented and User Centered Human Machine Interface Design of In-vehicle Information System

2006 IEEE International Conference on Vehicular Electronics and Safety, 2006

The need of in-vehicle information system was analyzed. The paper summarizes principles, existent... more The need of in-vehicle information system was analyzed. The paper summarizes principles, existent design standards and guidelines for human machine interface of in-vehicle information system. We comply with traffic safety-oriented and user-centered design concepts. Basing on existent design standards and guidelines, we design human machine interface of in-vehicle information system. Design contents consist of visual display, auditory display, and input control. From the perspective of traffic safety, we should relieve interruption of driving task as possible as we can. Meanwhile, we should complete the interaction with in-vehicle information system. The total design principle is to keep the drivers' two hands on the steering wheel, and minimize the interaction time of visual manipulation, auditory manipulation. Different drivers have different physiology-psychology characteristic, so design requires human factors engineering knowledge.

Human-computer interface design for an in-vehicle information system

International Conference on Intelligent Transportation, 2003

In-vehicle information system (IVIS) is an important component of urban traffic flow guidance sys... more In-vehicle information system (IVIS) is an important component of urban traffic flow guidance system (UTFGS). It consists of an input device, data storage, a driver information display monitor, processing modules, a voice synthesizer module and a communication module. Human-computer interface (HCI) is the comprehensive environment of manipulating IVIS. It is also the media of delivering and exchanging information between the

Monitoring travel times in an urban network using video, GPS and Bluetooth

The travel time is an important measure for the quality of traffic. This paper discusses a few me... more The travel time is an important measure for the quality of traffic. This paper discusses a few methods to measure or estimate the travel time in urban road networks. First of all, it is important to know that urban travel times display a large variation, so that the measurement of a single (average) travel time is not so meaningful. The travel time distribution is more relevant than the single value of the average. This distribution can be obtained from observations of travel times of individual vehicles, for instance by tracing probe vehicles with GPS. The distribution of travel time can also be measured by Bluetooth scanners that can recognize Bluetooth devices in a car. Travel time over a link can be estimated by comparing passing times at its beginning and end. Automated Number Plate Recognition cameras have a similar possibility to follow individual vehicles from point to point. In the study of travel time measured in urban areas, the quality of data from Bluetooth scanners appeared to be disappointing. The most important reason is that outliers cannot easily be eliminated. This is especially a problem in urban areas. Another reason is the uncertainty about the Bluetooth devices' carriers. It is difficult to identify whether the Bluetooth device belongs to a car or bus passenger, a cyclist or a pedestrian. Probe vehicles with GPS are very well appropriate for the measurements of travel times in urban areas.

Download

A discrete traffic kinetic model – integrating the lagged cell transmission and continuous traffic kinetic models

Continuous traffic kinetic models are difficult to solve because of the occurrence of integro-dif... more Continuous traffic kinetic models are difficult to solve because of the occurrence of integro-differential equations in the models. In this paper, we formulate a discrete traffic kinetic model by extending the cell transmission mechanism, which can capture not only the number of vehicles, but also the velocity probability distribution. The variation in the velocity probability distribution is modeled on the basis of vehicle conservation and cell transmission to avoid integro-differential terms. An example with a discontinuous initial density is analyzed to demonstrate the validity of the proposed discrete traffic kinetic model. From the evolution curve of the velocity probability distribution, we can see that the proposed model can be used to describe the diffusion process of vehicles from a high density to a low density section.

Shoufeng Lu

Uploads

Papers by Shoufeng Lu

Log In