Browsing by Author "Kwon, Taek Mu"

Now showing 1 - 13 of 13

Access to Destinations: Computation of Travel Time Data for Access to Destinations Study
(Intelligent Transportation Systems Institute, Center for Transportation Studies, University of Minnesota, 2008-11) Kwon, Taek Mu; Klar, Scott
Freeway travel time data is used as an important input for measuring travel reliability and accessibility. The goal of this project was to generate reliable travel time data using loop detector data from the Twin Cities’ freeway network collected over the past 14 years for the Access to Destinations projects. Several difficulties exist: one year of loop data can be missing up to 31.7%, loop data does not include vehicle lengths or classification to use in speed calculations, and link travel time must be computed without knowing speed variability within the link. The three basic approaches used in this project for imputing speed data are linear regression, spatial imputation, and week-toweek temporal imputation. A new method was developed to estimate average vehicle length using volume, occupancy, free-flow speed, and speed limit data. Link travel times were calculated by dividing the distance between stations into thirds and using speed data from both stations. Overall (for the last fourteen years), the imputation increased the average amount of valid data from 81.7% to 98.6%. Travel time for a selected route was verified by comparing the resulting calculated travel times to manual travel times measurement information.
Advanced LED Warning Signs for Rural Intersections Powered by Renewable Energy
(Minnesota Department of Transportation Research Services Section, 2010-12) Kwon, Taek Mu; Weidemann, Ryan
A majority of intersection-related fatal crashes occur at rural, through/stop intersections. At these intersections, sight restrictions caused by vertical and horizontal curves negatively affect a driver’s ability to safely accept a gap in the traffic stream. Static advanced warning signs are sometimes installed at these intersections to warn drivers on the main, through approaches that an intersection is ahead. These warning signs appear to be ineffective. A new Advanced Light-Emitting Diode (LED) Warning System was developed and deployed at a rural, through/stop intersection with limited intersection sight distance due to a severe vertical curve. This warning system actively detects vehicles on all approaches and activates LED blinker warning signs for the conflicting movements. The research project included analysis of driver behavior obtained through video data and a survey of local residents and frequent users of the intersection. This report describes the development, implementation, and evaluation results of this new warning system.
Atmospheric Visibility Measurements Using Video Cameras: Relative Visibility
(2004-07-01) Kwon, Taek Mu
Poor visibility conditions often lead to large-scale chain accidents causing human fatalities and property damage. Such visibility-related vehicle accidents could have been prevented if motorists were warned ahead of time to reduce speed and remain cautious before moving into the poor visibility zone. The objective of this research was to advance the visibility measurement technologies that compute visibility through processing images captured by video cameras. There are two fundamental difficulties in measuring visibility. The first is that visibility is a complex multivariable function of many parameters such as objects available, light sources, light scatter, light absorption, etc., so that measurements of one or two parameters (as in most of today's visibility meters) cannot accurately estimate the true human-perceived visibility. On the other hand, any attempt to measure every possible atmospheric parameter to derive human perceived visibility is simply too complex and costly. The second source of difficulty is contributed by an attempt to express the spatially variant nature of atmospheric visibility using a single representative value, distance. It works only if the atmosphere is uniform, which rarely happens. A solution presented in this report is to measure visibility using visual properties of video images (perceived information) instead of indirectly measuring physical properties of atmosphere and converting them to visibility. The spatial variance problem in visibility was solved by introducing a new concept of relative measurement of visual information referred to as the Relative Visibility (RV). This report also includes a study result on the limitation of CCD cameras in visibility measurement applications, and shows how to overcome them through spatially arranged multiple targets. In addition, we explored various apparatuses of Near Infrared (NIR) light source and cameras for measuring night visibility. This result is included in the report.
An Automatic Visibility Measurement System Based on Video Cameras
(Minnesota Department of Transportation, 1998-09) Kwon, Taek Mu
The Minnesota Department of Transportation (Mn/DOT) conducted a two-year study on visibility measurement methods using video cameras. This report describes the study's theoretical basis, practical methods, and experimental results. Among several methods and algorithms developed, the edge decay model along with a proper threshold technique worked best for evaluating daytime visibility. This approach estimates the distance where an object of specified size and shape is no longer distinguishable from the background in terms of edge information. For night time, a constant light source is required to evaluate visibility. Researchers developed a light diffusion model that follows an exponential decay curve. Researchers determined that the volume of light diffused out of the original source logarithmically correlates to visibility. Mn/DOT implemented day and night algorithms in the field and evaluated them using manual measurements. For daytime, visibilities measured using the edge decay model closely approximated the manual measurements on all types of weather. Unreliability of manual measurements at night made night-time evaluation very difficult. However, research verified that the trend of visibility change obtained by the proposed approach closely approximates the trend of manual measurements. 17. Document
Blind Deconvolution of Vehicle Inductance Signatures for Travel-Time Estimation
(2006-02-01) Kwon, Taek Mu
Travel-time data provides vital information for traffic monitoring, management, and planning. The objective of this research was to develop a new computational approach that could accurately measure travel time from two sets of spatially separated loop detectors using re-identification of vehicle inductance signatures generated by the loops. Although measuring travel time using loop inductance signatures is not new, all past approaches essentially relied on pattern matching of raw inductance waveforms without restoring the loss of detailed features caused by a large detection zone of inductive loops. The main effort in this research was to develop a new computational algorithm that restores the lost details from the raw inductance waveforms by modeling the output of loop detectors as a convolution of the original vehicle signature and the loop system function. This restoration problem was formulated as a blind deconvolution problem since we know neither the impulse response of the loop detectors, nor the original vehicle signature. To solve this blind problem, two basic blind deconvolution approaches were used, Godard deconvolution and constrained least squares. Experimental results showed that both methods performed well and significantly exposed the original signature information with unique vehicle characteristics.
Development of Dynamic Route Clearance Strategies for Emergency Vehicle Operations, Phase I
(2003-06-01) Kwon, Taek Mu; Kim, Sangho
A route-based signal preemption strategy is developed to provide the most efficient and safe route for an emergency vehicle under a given network and traffic conditions. It combines an on-line route selection procedure and a dynamic sequential preemption method. The on-line route selection module first quantifies the level of congestion for each link on a given network using a congestion index and finds the least congested route for a given origin/destination pair using the well-known Dijkstra's algorithm. Further it also selects the safest signal phase for each intersection for a given travel direction of an EV. Once an emergency route is selected, the dynamic preemption module starts the preemption of the signals on the emergency route sequentially considering the location of the EV and the state of signal phase for each intersection. By sequentially preempting the traffic signals on a route with advance activation, the proposed strategy tries to clear the traffic queue for an EV approaching each intersection. The evaluation results with pre-specified emergency routes show 10 - 16% reduction of the emergency vehicle travel time for relatively long and/or complicated routes compared with the existing intersection-by-intersection preemption method. Further, the network-wide performance measures with the proposed dynamic preemption method were very compatible with those from the existing intersection-by-intersection clearance method.
Development of Efficient Integrated Data Archival/Retrieval Model for R/WIS, RTMS, and Loop Traffic Data
(2006-06-01) Kwon, Taek Mu
This report describes a new data warehouse model developed for integrating Road Weather Information System (R/WIS) and traffic data and the prototype implemented. The building blocks of the prototype include data aggregation methods from sensors, a data archiving system, and multi-user data access and retrieval environments through a network. This new data warehouse model seamlessly integrates the heterogeneous nature of R/WIS and traffic data. The key to this data model was utilization of a network storage model referred to as a parallel First-In-First-Out (FIFO) data storage where various sensor data are deposited as they are aggregated while different types of data-consuming modules obtain data without an explicit protocol requirement. For the prototype implementation, four different data aggregation methods from traffic and R/WIS sources were used to demonstrate that diverse data types and collection methods could be seamlessly integrated together. As an application of this data warehouse, weather impact on traffic flow was studied by retrieving traffic data under various atmospheric and pavement conditions, and the results are included. It was noticed that R/WIS provides a significant advantage over the traditional National Weather Service data in learning detailed location specific weather and pavement conditions from which weather impact on traffic flow could be accurately analyzed.
Next Generation R/WIS: Concept and Prototype Implementation
(Minnesota Department of Transportation, 1999-03) Kwon, Taek Mu
Traditional R/WISs offer limited scope of data types, limited communication methods, and proprietary incompatibility problems. The R/WIS in Minnesota District-1 experienced the typical problems of the traditional R/WIS. This project looked at solutions to these problems--developing a new layered hierarchical architecture referred to as NG-R/WIS. Four layers--a sensor layer, a data integration layer, a database layer, and an application layer--for the basis of the NG-R/WIS. This architecture allowed integration of heterogeneous systems thorough the data integration layer and provided structured data to applications through a standard relational database and computer networks. This project developed three applications: web page service, automated voice service, and live stream-video service. The web pages and automated voice serviced provided a new class of information. The report concludes with the recommendations and future direction of the proposed NG-R/WIS.
Portable Cellular Wireless Mesh Sensor Network for Vehicle Tracking in an Intersection
(University of Minnesota Center for Transportation Studies, 2008-12) Kwon, Taek Mu; Weidemann, Ryan
This report describes the result of a research project that developed an automatic, portable vehicle tracking system that can count the vehicle travel trajectories in an intersection. Using a cellular wireless mesh sensor network (WMSN), wireless sensor nodes are placed in the middle of traffic lanes in the intersection to collect the data. Each node consists of an anisotropic magnetoresistance (AMR) circuit for detecting vehicles, a PAN4570 Radio Communications Module (RCM), and a lithium-ion rechargeable battery. When a vehicle travels over a wireless node, a detection occurs and a timestamp is recorded by the node and sent to the WMSN coordinator. The coordinator is responsible for logging the vehicle detections recorded by every node in the WMSN. From this logged data, a vehicle tracking algorithm that has been developed tracks the trajectories of the vehicles through the intersection and also records a total vehicle count of the intersection. The system performance was evaluated through both intersection simulations and real data collection. The details on the system component design, implementation, experimental results, and analysis are described.
TDRL Projects: Solar/Wind Hybrid Renewable Light Pole, Gravel-Road Traffic Counter, DLL-Based Traffic Software Development Kit
(University of Minnesota Center for Transportation Studies, 2008-09) Kwon, Taek Mu; Weidemann, Ryan; Cinnamon, Dan
This report describes the results of three projects initiated in Fiscal Year 2006 by the Transportation Data Research Laboratory (TDRL), using the funds provided by the Northland Advanced Transportation Systems Research Laboratories (NATSRL). They are (1) Development and field tests of solar/wind hybrid renewable light pole, (2) Development of a gravel road traffic counter, and (3) Development of a DLL-based software development package. Each project is relatively small and developed as a seed project for future expansion to a full independent project. Each of them is described as a chapter that includes its own introduction, the main body of work, and its own conclusion.
Technical Support for Implementation of Freeway Network Simulation Software
(2000-06-01) Kwon, Taek Mu; Kota, Ramesh
This report summarizes the results from the technical support and testing of the freeway traffic simulation software developed at the University of Minnesota. The University's civil engineering department developed and implemented an experimental course that addressed fundamental theories in traffic flow modeling and simulation. They also incorporated a set of an online help manual into simulation software. Software testing was conducted through two sets of case studies involving the high-occupancy vehicle (HOV) lane and ramp metering simulation modules. The HOV lane case study used a 10-mile section of the westbound I-94 freeway with an additional lane and evaluated the performance of alternative HOV lane strategies with different sets of demand and HOV proportions. For the same amount of total demand without assuming induced trips, the higher HOV proportion, the smaller the mainline delay. The ramp metering case study compared the performance of the current Mn/DOT metering policy with that of a no-metering option for a 16-mile section of the northbound 169 freeway with 20 percent higher demand than the current level. The metering case study showed that the total system delay, including both mainline and ramps, was significantly reduced when the ramps were controlled compared to a no-metering case for a given set of demand. The case studies indicate that the simulation software can be applicable in evaluating alternative design and operational strategies for a given set of demands.
TMC Traffic Data Automation for MnDOT's Traffic Monitoring Program
(2004-07-01) Kwon, Taek Mu
The Minnesota Department of Transportation (Mn/DOT) has been responsible for collecting, analyzing and publishing traffic count from the various roadway systems throughout the state. The traffic reporting system mainly developed by the Traffic Forecasting and Analysis Section (TFAS) of Mn/DOT has been used in several federal programs, internal Mn/DOT applications and many private sectors. The objective of this project was to continue the TFAS' automation efforts by automating the TMC portion of traffic data (ITS generated data) contributed to the Mn/DOT's Traffic Monitoring System. The focus was given to develop an Internet based system that produces computerized reports on continuous and short-duration count data. One of the challenges of utilizing ITS generated traffic data for computing continuous and short-duration count was in dealing with missing and incorrect data produced by faulty conditions of traffic data collection systems including detectors and communication links. This study found that data imputation techniques based on spatial and temporal inferences of traffic flow can overcome the difficulties and produce accurate statistical data. This report describes the details on actual implementation of the algorithms developed, analysis utilities and practical system integration examples. One unresolved issue in this project was dealing with the stations in which nearly no data is available for the entire year, which was observed from 2-3% of the short-duration count stations. This problem is left for future work.
Transportation Data Research Laboratory: Data Acquisition and Archiving of Large Scaled Transportation Data, Analysis Tool Developments, and On-Line Data Support
(Intelligent Transportation Systems Institute, Center for Transportation Studies, University of Minnesota, 2009-02) Kwon, Taek Mu
This report contains a collection of reports for projects completed in FY 2004 and 2005 at the Transportation Data Research Laboratory (TDRL). First, an archiving technique referred to as the Unified Transportation Sensor Data Format (UTSDF), which allows simple management of large scaled Intelligent Transportation Systems (ITS) sensor generated data, is described. UTSDF was used for the development of a Data Center (DC) at TDRL. Next, data imputation algorithms to estimate missing data are presented. These algorithms were developed during the process of developing an automated on-line Automatic Traffic Recorder (ATR) and short count data system for the Office of Traffic Data & Analysis (TDA) at MnDOT. Utilizing the archived loop data, TDRL also developed a detector fault identification algorithm and software. This algorithm and test results are reported. Another project report involves cross-utilization of Road Weather Information System (RWIS) and traffic data. Several analysis approaches were developed to analyze the actual data. The analysis approaches used and findings are reported. Another project report involves development of a Weigh-in-Motion (WIM) Probe. This tool was developed as a diagnostic tool for the MnDOT's current WIM systems, and is based on a MnDOT problem statement. It is used for identification of signal anomalies and data verification. The details of this project are reported.

University Digital Conservancy

Browse by Author

Browsing by Author "Kwon, Taek Mu"