The Arkansas and Missouri (A&M) Railroad is a short-line railroad that operates in a 139-mile corridor from Fort Smith, Ark., to Monett, Mo. Along the way, it interchanges with Burlington Northern Santa Fe, Union Pacific, and Kansas City Southern. Ensuring that the railroad remains viable requires a constant assessment of assets, rail conditions, safety and support services.
When Aerial Data Service Inc. (ADS), a geospatial services firm based in Tulsa, Okla., approached Jim Seratt, A&M Railroad’s general superintendent, and J. Reilly McCarren, principal owner/chairman of the board, in early 2009 about conducting a proof-of- concept investigation of mobile LiDAR to survey a section of the railroad infrastructure, the men were intrigued. Of particular interest was the condition of a quarter-mile-long tunnel that the trains pass through on their journey. However, this application of mobile LiDAR was still untested. Could LiDAR collection on a mobile platform work through tunnels, and would it be a cost-effective means of gathering the data?
ADS was eager to explore the full capabilities of the technology. “We had seen demonstrations of the LYNX Mobile Mapper system [from Optech Inc.] and were impressed with its ability to accurately capture georeferenced 3D spatial data from a variety of mobile platforms,” says Brian Falls, ADS Tulsa project manager. “We wanted to see if the technology could provide enough data with high enough accuracy to be used for railway maintenance and inspection. We also wanted to determine whether the LiDAR data could be used to monitor track and switch conditions, inventory, signage, and obstructions such as vegetation encroaching on the rail right-of-way.”
Proving the Concept
Working closely with Optech, ADS first evaluated the technology on a 3.4-mile section of railway at the Port of Muskogee (near the firm’s Tulsa offices) in late winter of 2009. The team mounted the LYNX Mobile Mapper V100 dual sensor mobile LiDAR system to a track maintenance car (commonly called a speeder) and used the system’s LiDAR sensors to scan the area surrounding the rail line. The topographic contour maps and models generated from the scan data provided a wealth of useful information for rail corridor design and monitoring. Combined with the system’s ability to handle road corridor mapping and building exterior scanning for building information modeling (BIM) applications, these results were enough for ADS to justify purchasing Optech’s new LYNX Mobile Mapper V200 scanner when it was released in March.
The next step was to find out what would happen inside a tunnel. A&M Railroad would provide an ideal testing ground.
The railroad managers were enthusiastic about the project, but the economics of railroading had to be considered: The survey would have to take place without delaying the trains, which operate six days per week. While the LYNX technology is mobile and can be operated on an active railway, the team decided to work around the railroad’s schedule and perform the scans on a day when the track wasn’t operational to avoid any potential for service disruptions. Since ADS would use GPS positioning for control, the position dilution of precision (PDOP) window also had to be considered. ADS had roughly a three-hour window of opportunity to gather the required scans for the A&M Railroad proof-of-concept project.
On July 27, 2009, ADS deployed its LYNX Mobile Mapper to West Fork, Ark., to collect rail and tunnel data along 10 miles of the A&M Railroad. Through a prior relationship, ADS was able to access a speeder and coordinate its use as the mobile platform. To attach the equipment to the speeder, the firm used the mount fabricated for the Port of Muskogee rail test earlier in the year. Calibration and bore sighting were performed while the speeder was on the trailer. The two-person survey crew also set a base station in the vicinity to aid in data collection.
Once these tasks were complete, the speeder was rolled onto the track with the assistance of a three-person crew provided by A&M Railroad, and ADS began collecting data heading south toward the tunnel in Winslow, Ark., at about 25 mph.
Before entering the tunnel, the crew stopped and set a control panel at the tunnel’s entrance. Crew members verified good PDOP and then began collecting data inside the tunnel, relying on an inertial measurement unit (IMU) for guidance. “As expected, accuracy degraded without positional updates from the GPS as the speeder proceeded through the tunnel,” says Scott Dunham, ADS Austin project manager, who was onsite for the data acquisition. “However, when the GPS lock was re-established, the accuracy of the system was restored.” Any errors were smoothed and minimized through post-processing techniques.
As the surveyors passed through the tunnel, they placed five additional control panels at evenly spaced intervals and a seventh control panel at the tunnel’s exit. These panels would be surveyed later to verify accuracies.
The crew ran the speeder down to the next access point in Chester, Ark., where the system was removed from the track and dismantled. “Because the LYNX system uses two opposing LiDAR sensors to collect data in an X-shaped bidirectional pattern, only one pass was needed to cover the area and fill all of the voids,” Dunham says.
The entire process, including moving the speeder on and off the track, took about four hours. The scan work was completed in about 1.5 hours. “There were a lot of stops going down to the tunnel as tree canopy and steep hillsides interfered with the GPS signal,” Dunham says. “Otherwise, we would have been able to complete the scan work even faster.”
Post Processing and Verification
Using proprietary software associated with the LYNX system, the ADS crew processed the data onsite to ensure accuracy. The following day, the surveyors returned to the tunnel and ran control on the seven targets using Trimble receivers and differential levels. The final results were 7 millimeters relative (exceeding specifications), and an absolute accuracy of 1.57 inches was achieved through the tunnel. “The tunnel dropped 25 feet from one end to the other,” Dunham says. “Considering the terrain and other factors, we were very satisfied with the results.”
The control data were given to the ADS LiDAR analyst for coordinate transformation and verification, which were accomplished using the firm’s proprietary software. Final processing and verification were completed with Terrasolid, which works on top of Bentley Systems’ MicroStation software.
Within three days, ADS was able to convert the point clouds to LAS files (a commonly accepted format for storing and exchanging LiDAR point data) and an AutoCAD file with planimetrics and contours. “The data were extremely dense, and no points were lost due to the dimensions of the tunnel and proximity to the sensors,” Falls says. “Acquisition to processing the final deliverables took about 40 percent less time than if the project had been done with conventional survey techniques.”
What’s more, the LiDAR data set provides a permanently searchable database, which ensures that the data remain valuable long after the initial scan work is complete. Height differences can be extracted repeatedly from a single point cloud, and topographic maps generated by the point cloud data can serve as the basis for analyzing rail corridor water-flow patterns, vegetation growth and other maintenance and safety concerns.
While economic conditions have limited the ability of A&M Railroad to move beyond the proof of concept, ADS has continued working with railroad management to further evaluate the mobile LiDAR system and is exploring the integration of the scan data into a future GIS for the railroad.
“Mobile LiDAR allows us to quickly collect highly accurate, very dense data, and as a result, it gives surveyors less exposure to potentially hazardous conditions on site,” Falls says. “The density of the data allows us to create cross sections wherever they are needed within the tunnel, including the track itself.” In addition, the intensity values returned from the laser points allows moisture to be detected along the ceiling of the tunnel. From this information, environmental conditions such as wind direction and pooling of water can be determined.
“These are just some of the things we can currently do with the LiDAR data,” Falls adds. “New algorithms and processes are constantly being developed that may allow even further exploitation of today’s point clouds tomorrow. It’s a promising technology for the railroad industry.”
Sidebar 1: In Search of PowerWhen mounted to an automobile, the LYNX system is powered by the vehicle’s electrical system. A similar source of energy was needed for the railway speeder in the Port of Muskogee and A&M Railroad trials. “We decided to use deep cycle batteries mounted in the speeder,” says Carlyle Witherell, ADS mobile operations manager.
While this approach worked, Witherell admits that it wouldn’t be the ideal solution for large projects. “The system consumes a lot of power, and there usually aren’t a lot of places along the rail with access to swap out batteries,” he says. “On future railway projects, we plan to beef up the alternator on the speeder and use that to power the LYNX system.”
Sidebar 2: Positive Train ControlShort-line and regional railroads own, maintain and operate approximately 40,000 miles of track in the United States--nearly one-third of the entire rail system. These railways are a vital link in North America’s rail transportation infrastructure. However, according to Richard F. Timmons, president of the American Short Line and Regional Railroad Association (ASLRRA), the industry is facing unprecedented changes due to the economy and new legislation.
One of the most challenging pieces of recent legislation is the Rail Safety Improvement Act of 2008, signed into law on Oct. 16, 2008. This act requires, among other obligations, that every railroad providing intercity or commuter rail passenger transportation develop and submit a plan for implementing a positive train control system to the U.S. secretary of transportation by Dec. 31, 2015, or risk facing severe penalties. While the legislation primarily affects large Class I railroads, it is likely that all lines will eventually need to monitor their tracks with some sort of intelligent monitoring system. A big part of this effort includes a high-accuracy (survey-grade) collection of assets within the industry’s rights-of-way. Mobile LiDAR provides a cost-effective way to achieve this goal.
For more information about short-line and regional railroads, visit the American Short Line and Regional Railroad Association’s Web site at www.aslrra.org.