Task 27: Prepare the Experiment. Center for Infrastructure and Transportation Studies Technical Report No: October 25, 2008

Transcription

1 Task 27: Prepare the Experiment Center for Infrastructure and Transportation Studies Technical Report No: October 25, 2008 Prepared by Jeffrey Wojtowicz Performed in conjunction with the Electronic Toll and Traffic Monitoring Project with Rensselaer Polytechnic Institute

2 Contents 1 INTRODUCTION WIRELESS SOLAR POWERED TAG READER DEPLOYMENT CAPITAL DISTRICT TESTBED DEPLOYMENT Testbed Deployment (July 2007) Testbed Re-Deployment (September 2007 January 2008) Prototype Reader versus Conventional Reader NYS FAIR AND I-690 WORK ZONE DEPLOYMENT CONCLUSIONS REFERENCES... 8

3 1 Introduction The objective of this report is to explain the process whereby the wireless solar powered tag reader field experiment was prepared and the lessons learned. The Task 26 report outlined the anticipated deployment plan; this report describes the actual deployment of the readers. In June and July of 2007 the wireless solar powered E-ZPass tag readers were deployed along NYS Route 4 and Jordan Road in North Greenbush, NY. These locations were selected as part of the Task 26 effort. In mid August 2007 the readers were then transported and deployed in Syracuse, NY for the NYS Fair. Upon completion of data collection at the NYS Fair, the six readers were again redeployed in North Greenbush, NY. The report is organized as follows. Section 2 discusses the deployment for the various experiments and Section 3 provides conclusions. Appendix A contains detailed photos of the Capital District testbed deployment and Appendix B contains detailed images of the NYS Fair deployments. 2 Wireless Solar Powered Tag Reader Deployment The Task 26 report documented the final controlled testing that was completed before the actual field experiment began. The Task 26 report also prescribed possible locations along NYS Route 4 and Jordan Road in North Greenbush, NY where the tag readers could be deployed. This section of the report discusses the refinements that were necessary to make to the devices as well as the logistics for deploying the readers alongside of the road. In addition to the plan to deploy the tag readers in the Capital District Intelligent Transportation Systems (ITS) Testbed as described in Task 26, the deployment of the readers for the NYS Fair and I-690 work zone in Syracuse, NY is discussed in Section Capital District Testbed Deployment The Task 26 report proposed several different experiments to be conducted. The testing that will be outlined in the subsequent sections are the deployments in and around North Greenbush, NY in the Capital District ITS Testbed. In addition to the experiments outlined in Task 26 there are details pertaining to the deployment at the NYS Fair Testbed Deployment (July 2007) In early July of 2007 the sites identified in the Task 26 report were used as deployment locations for the wireless solar powered E-ZPass tag readers. The readers were deployed until mid August. These locations can be seen in Figure 1. For the first deployment reader #4a and #4b were not deployed. At each one of these locations the team plans on collecting data for one lane in each direction via the side-fire application. For vehicles equipped with E-ZPass the through movements on Route 4 north and southbound can be captured as well as the turning movements into the RPI Tech Park (location #1). The sixth reader will not be deployed initially and will serve as a backup if needed (location #4a and 4b). If the data collection is successful a deployment location will be identified for the sixth reader.

4 Figure 1 Proposed deployment locations Appendix A is a PowerPoint presentation with photos showing each of the deployment locations in detail. Of the five sites two of them were deployed directly on the trailer mount (locations 1 & 6) while the remaining three were modified so the antenna was installed on an existing sign structure. At each of the locations the antennas were adjusted in the field to achieve maximum tag read reliability. Prior to deploying the readers along the road it was necessary to seek the approval of the NYSDOT; this was done over the course of several meetings and field visits with the Project Manager. On June 13, 2007 a field visit was conducted with the DOT Resident Engineer to ensure safety at the sites and make sure the devices would not interfere with traffic or maintenance operations. Once this was done the team contacted the property owners and got their permission to deploy the devices. During the first week of July 2007, the devices were delivered to their respective locations and were setup. Once the sign mounted antennas were installed and the trailer was parked in close proximity longer RF (radio frequency) cables were constructed. It was found that 60 cables would be necessary. The team was concerned that since the cable length was double what was originally specified there might be problems with RF loss. It turned out that as long as the connections were made with the standard straight connectors instead of

5 the right angle connectors necessary for the trailer mounted antennas the difference in RF loss was negligible. Approximately one week was devoted to setup the readers and see how they behaved with traffic at highway speeds. This helped the team understand the operating characteristics of the system, including the antenna and a pocket PC communication device (PPC). Several issues were quickly discovered, the most critical was that the RF cable should have weather tight connections. There were several instances where the team thought they were weather tight but after several days of rain the reader would not work as it previously did. Upon further review some moisture made its way into the connections and compromised the performance of the system. Another issue that was uncovered was the software installed on the PPC s. There were miscellaneous data handling issues that existed and caused the PPC to randomly shut down the encryption software. As these issues were discovered the software was modified to correct them Testbed Re-Deployment (September 2007 January 2008) The re-deployment in the Capital District in September 2007 was virtually the same as discussed in Section The main difference was that the sixth reader was also deployed. This reader was deployed twice at two different locations, one on Winter Street and the other on NYS Route 136. These locations can be seen as ETTM_4a and ETTM_4b respectively in Figure 1. Based on the lessons learned from the first deployment slight changes were made to the devices. The changes included such things such as antenna angle and placement. The team anticipates collecting some volume counts in conjunction with this deployment. These counts will likely be done with the Nu-Metrics count devices and the data can be used to estimate the tag density within the traffic stream. In September the team planned on leaving the devices out until the first sign of inclement weather. As the testing progressed the team received NYSDOT s permission to leave the devices deployed during the first few snow storms. This would allow the team to truly monitor the performance of the system Prototype Reader versus Conventional Reader As deployment discussions progressed the team found it difficult to arrange a site where testing could be done in conjunction with a traditional Transmit reader. The main reason for this is that obtaining permission to deploy this reader alongside the highway has been more difficult than expected. When the first prototype was built the team brought a unit to NY Route 7 and tested the device. The team anticipated getting data from either the NYSTA or Transcom for the nearby Transmit system since the two readers were located on the same stretch of road with no possibility for vehicles to enter or exit between the two locations. Unfortunately Transcom informed the team the reader was not functioning properly when the team performed the field tests. In lieu of these comparison tests the team collected data using a variety of data sources including the reader data, manual field counts using a laptop and video footage that was extracted.

6 The team has also been in contact with NYSDOT and the NYSTA to hook the mgate reader up to an existing Transmit cabinet. The purpose of this test will be to see how the mgate performs at collecting data with overhead antennas. The details pertaining to this deployment are not yet known but if completed the results will be shown in future task reports. Since there will be no way to compare the mgate data with that collected with the traditional system manual field counts are expected to be collected and compared to the counts received on the mgate reader. 2.2 NYS Fair and I-690 Work Zone Deployment The team was asked to participate with a data collection effort during the 2007 and 2008 New York State Fair s in Syracuse, NY. The Fair is held in late August and ends on Labor Day. During the 2007 and 2008 Fair, NYSDOT had an ongoing construction project in close proximity to the fairgrounds. In years past construction projects were not scheduled during the time the Fair was in operation, however this project was a major multiyear project that had to take place during the Fairs. The team deployed the wireless solar powered E-ZPass tag readers at six locations with the help of NYSDOT. In the Data Collection Plan for the New York State Fair (1), the team identified several locations where the readers antennas could have been mounted in an overhead application. NYSDOT preferred to deploy the readers as they were built in a sidefire application. The reason for this was to demonstrate the quick deployment for a special event or work zone. In the side-fire position an antenna is only capable of capturing one lane of traffic. For example on I-690 westbound, two readers were setup directly across from one another. If one antenna had been mounted overhead only one reader would have been needed and the second reader could have been used elsewhere. Therefore, the team decided to focus on collecting travel time data through a temporary detour that was put into place when certain parking lots at the fairgrounds reached capacity. In 2007 the readers were placed in such a way that travel times would be captured even when the detour was not in effect. The deployment locations for the 2007 Fair can be seen in Figure 2. Since the detour movement was only implemented a few times in 2007 the team decided to capture different travel times in In 2008 the readers were setup in such a way that the team could monitor parts of the detour and also the travel times on State Fair Blvd to and from some of the main parking lots. Most of this data was available in real time but the team had setup the devices so that when the data from the NYSTA and the parking readers was available the dataset would be much richer. The deployment locations for the 2008 Fair can be seen in Figure 3.

7 Legend FAIR_1 Bear St on ramp to I-690 WB FAIR_2 I-690 WB in work zone FAIR_3 I-690 EB in work zone FAIR_4 Hiawatha Blvd FAIR_5 I-690 WB in work zone FAIR_6 Base of exit 7 orange lot Figure NYS Fair deployments For future applications it would be advantageous to install overhead mounted antennas to increase the number of E-ZPass tag reads and to produce a richer dataset which could be used for real-time traffic management. The readers were in operation from August 20, 2007 until September 7, 2007 and August 20, 2008 until September 2, More detailed photos of the deployment locations can be seen in Appendix B. For both the 2007 and 2008 Fair s NYSDOT deployed the trailers with the help of the RPI team. The locations that were chosen were all on DOT right-of-way. In addition to the RPI tag readers, data was obtained from the NYS Thruway Authority. After the Fair, tag transactions from the three parking lot readers and Thruway Exits 34a, 36 and 39 were processed using RPI s encryption algorithm. This would allow the research team to have access to more detailed travel time data.

8 Figure NYS Fair deployments 3 Conclusions Based on earlier testing the team feels that the deployment locations that have been chosen will fulfill the needs of the project. The team has learned a great deal about the devices and how they should be placed for optimal performance. There were a number of issues that had to be resolved dealing with the site deployments of the readers. Besides choosing locations that were ideal in terms of measuring travel times, the team also had to take into account safety issues such as the devices proximity to the road and how to obtain permission to park the devices on private property. In addition to testing the prototype devices the data will be useful for modeling purposes. The locations that were chosen for both the Capital District testbed and the NYS Fair will provide invaluable data for microsimulation models that have recently been built. This data includes partial OD (Origin-Destination) pairs and travel time data throughout the day. Based on the results from the deployments outlined in this report the experiment will be conducted and the results assessed. 4 References (1) Bertoli, B., Lee, E., Wallace, W. and Wojtowicz, J. Task 32: Data Collection in Support of Developing a Traffic Management Plan for the 2008 New York State Fair. Technical Report Center for Infrastructure and Transportation Studies, Department of Civil and Environmental Engineering, Rensselaer Polytechnic Institute, 2007.