Roadway Measurement System Evaluation (2011)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

C H A P T E R 1 Introduction: The Research ApproachThe work for this assignment was completed under the fol- lowing tasks, as described. Task 1: Finalizing Work Plan and QA/QC Plan Task 1 involved reviewing the original Work Plan and Qual- ity Assurance/Quality Control (QA/QC) Plan provided to SHRP 2 as part of the S03 proposal package, updating these plans as necessary. Task 2: Determining Prioritization of Data Elements During Task 2, the research team developed a prioritized list of roadway safety data elements to be used in evaluating the per- formance of the S03 participants. This list was developed to reflect the importance of the roadway safety data elements in crash analysis and input from the Safety Technical Coordinating Committee and other SHRP 2 safety contractors. The list of data elements was developed for use in the Roadway Measurement System Evaluation Rodeo to evaluate automated data collec- tion firms. The data elements list and prioritization methodol- ogy were provided to SHRP 2, FHWA, and other safety research stakeholders for review and comment in April 2008. The data elements report was provided to SHRP 2 for review and com- ment on May 6, 2008. Comments received were incorporated or addressed, and the revised data elements list with definitions and target accuracies were provided to the rodeo participants for review and comment in August 2008. Comments received were addressed or incorporated, and the final data elements list was provided to the rodeo participants on September 15, 2008. Task 3: Developing the Test Site Selection Criteria Under Task 3, the S03 research team developed test site selec- tion criteria, combining the prioritized data elements list with4physical roadway characteristics and logistical considerations. These criteria were used in selecting challenging test sections for evaluating the participating vendors under conditions sim- ilar to those that may be encountered during the S04B project. Task 4: Test Site Evaluation and Recommendation In Task 4, the S03 research team used the criteria developed in Task 3 to evaluate three general locations around the country to host the Roadway Measurement System Evaluation Rodeo. Roadways in Northern Virginia were selected, and the same criteria were then used to select 10 possible test locations for detailed evaluation of the rodeo participants. From the 10 pos- sible test site locations, four locations were selected, with Vir- ginia Department of Transportation input, for use during the rodeo, as shown in Figures 1.1 through 1.4. The selected locations resulted in the following six test sites: • Site 1: SR 120, WB—Arlington, Virginia; • Site 6N: SR 287—Lovettsville, Virginia; • Site 6S: SR 287—Lovettsville, Virginia; • Site 7N: SR 15—Lucketts, Virginia; • Site 7S: SR 15—Lucketts, Virginia; and • Site 10: SR 15 SB—Leesburg, Virginia. These six test sites were contained within two rodeo survey routes (eastern and western) covering approximately 43 cen- terline miles and were used to evaluate the consistency of the participants’ data collection activities under a variety of real- world conditions. Each route was surveyed three times in both directions, for a total of 258 survey miles. The six test sites were blind tests and not marked in the field. Each was 2,500 ft long and included most of the road- way asset types contained in the Data Elements List. The test sites covered 8.5 lane miles of roadway (2,500 ft/site × 6 sites × 3 repetitions) and included a variety of land use, cover types, and roadway types.

5© 2011 Google, Map Data © 2011, Tele Atlas. Figure 1.1. Site 1: SR 120 westbound, Arlington, Virginia. © 2011 Google, Map Data © 2011, Tele Atlas. Figure 1.2. Site 6: SR 287 northbound and south- bound, Lovettsville, Virginia.© 2011 Google, Map Data © 2011, Tele Atlas. © 2011 Google, Map Data © 2011, Tele Atlas. Figure 1.4. Site 10: SR 15 southbound, Leesburg, Virginia. Figure 1.3. Site 7: SR 15 northbound and southbound, Lucketts, Virginia.Task 5: Test Site Mapping and Surveying In Task 5, the team used KCI Surveys, a professional land surveyor (PLS), to collect highly accurate positional infor- mation and asset attribute data for the data elements locatedalong the six test site locations. The data thus collected were then used as the basis of the reference data sets for the evaluation of the rodeo participants. The following list specifies the equipment and the order of deployment used in Task 5, as well as the primary data collection devices KCI Surveys used to obtain the highest degree of accuracy

6and mitigate the effects of the environmental positional degradation. 1. Global Navigation Satellite System (GNSS) instruments were used to provide primary control for each area. The project team followed the Federal Geodetic Control Sub- committee guidelines and the National Geodetic Survey draft report using real-time GNSS procedures meeting a positional tolerance of 1.5 cm at 20 PPM. The GNSS was used to compute the trajectories along the roadways so that satellite availability could be validated. 2. Conventional differential level techniques were used to establish the height difference between the control stations. 3. Robotic total station (RTS) was used to capture the asset information referenced to the control stations. The six selected rodeo test sections were surveyed between August 8, 2008, and September 11, 2008. The 1 sigma or 95% confidence of error for the primary control at each site was less than 2 cm horizontal and 1 cm vertical. For each site, RTSs were used to capture the asset information relative to the control stations. These instruments produced a positional accuracy at the 1-sigma level of less than 1 cm. KCI provided Applied Research Associates, Inc., (ARA) with the PLS certi- fication on January 29, 2009. The electronic data files, includ- ing DNG and DWG files, and geo-databases, were provided on February 28, 2009. ARA’s review of the delivered geo- databases revealed that some of the attribute data for some of the assets were missing. ARA used the CAD files and the dig- ital images collected by ARA to fill the attribute gaps and cre- ate the asset attribute data for the reference data set. Task 6: Organizing and Conducting the Rodeo During Task 6, the S03 research team planned, organized, and executed the Roadway Measurement System Rodeo inNorthern Virginia to prequalify the automated data collection vendors for participation in the follow-on Safety Project S04B. The rodeo participants were provided with the list of data ele- ments with the desired target accuracies developed in Task 2. The Roadway Measurement System Evaluation Rodeo was held from September 14 to 20, 2008, in Fairfax, Virginia, and included the following 11 firms or agencies: • Data Transfer Solutions (DTS); • eRoadInfo; • FHWA; • Fugro/Roadware, Inc.; • GeoSpan; • Mandli Communications, Inc.; • Michael Baker, Jr., Inc.; • Pathway Services, Inc.; • Sanborn; • Tele Atlas; and • Yotta. Task 7: Evaluating Rodeo Data In Task 7, the research team compared the data provided by the nine commercial rodeo participants with the reference data set for the six rodeo test sites (Team 10 did not provide any final data for the six test sections). The goal was to determine how well the participants were able to capture the various priority safety data elements of interest to the highway safety research community as determined in Task 2. The participants were evaluated for completeness of coverage, positional accu- racy and precision, and attribute accuracy and precision. Task 8: Producing the Final Report This task involved the creation of this final project report.