A bus stop seems like a simple structure: a pole with a sign on top. But when the Swedish Road Administration (SRA) decided to assess and update their roadside bus stops throughout the 450,000-sq-km country they realized that a seemingly easy project could easily become an overwhelming task.
First off, SRA needed on-site field surveys to assess conditions at each location The scale of work posed serious issues for teams charged with collecting, managing and utilizing the survey data.
To solve the challenges, a team of geospatial professionals has combined multiple technologies with customized workflows to collect, distribute and utilize the information. Their efforts reduced field time by 50% and cut overall surveying costs by about 30 percent.
Plan and Proceed
The initial work is taking place north of Stockholm in Uppsala County. Initiated in 2018, the three-year project involves reconstruction of stops at 100 locations in the county. With the simple title of “100 Bus Stops,” the project will redesign and rebuild bus stops along two rural roads. The stops will be reconfigured to better serve blind and disabled passengers and improve traffic safety. For example, bus pullouts are enlarged and some trees may be removed to improve sight lines for drivers and pedestrians. Other changes may include pavement marking, adjusting slopes and installing guardrails.
For the planning and design, SRA called on ÅF, a multinational engineering and design company. Founded in Sweden in 1895, ÅF has played a key role in Sweden’s development as an industrial and socioeconomic powerhouse. Lennart Gimring, surveying and mapping manager for ÅF – Division Infrastructure in Stockholm, said that ÅF has a long relationship with SRA on road and transportation projects. The value of that relationship was apparent during project planning meetings where SRA’s confidence in ÅF enabled Gimring to propose new approaches to data collection.
After reviewing SRA’s proposed methods for surveying the bus stops, Gimring suggested alternative techniques that would reduce field time and increase safety for the surveyors. For example, SRA had suggested using total stations for the work, including traversing between benchmarks. Instead of time-consuming traverses, Gimring pointed out that ÅF teams could leverage Sweden’s national GNSS network to collect data at bus stops using total stations, scanning and GNSS. They named the technique “RUFRIS” (an acronym for “Real-Time Updated Free Station”), which enables surveyors to establish precise georeferenced positions in areas where no control points or benchmarks exist.
Gimring assigned the work to two engineers, Annalena Hellström and Bridget Coulter. “They were given a free hand to plan and execute the project,” Gimring said. “My role was to convince the client that RUFRIS was the best approach.” With the green light from SRA, Hellström dug into the budget and planning. She used a simple project management tool that could be shared with stakeholders. To help with the planning, the bus stops were divided into groups of different priorities. The data was also converted into a KMZ file for use in Google Earth.
Because the work covered a large area, Hellström and Coulter needed to plan the fieldwork in detail. In addition to selecting driving routes, they planned their activities at each site. “Google Earth Street View was very helpful in this process,” Hellström said. “In addition to getting ideas on where we could establish our instrument stations, we could also decide where to safely park our vehicle.”
Hellström and Coulter used a Trimble R10 GNSS receiver and a Trimble SX10 Scanning Total Station to conduct the fieldwork. At each site, they first confirmed that they could work safely and stay clear of traffic. Next, they performed the RUFRIS procedure. Setting the SX10 in a suitable location, they used Integrated Surveying, in which GNSS receiver and prism target on a single pole occupy a point. With the R10 connected to the SWEPOS real-time GNSS network, they obtained positions based on the national coordinate grid and vertical datum. At the same time, the SX10 measured to the point and added the data to the free station solution in Trimble Access software running on a Trimble Tablet controller.
The RUFRIS technique allowed Hellström and Coulter to establish good geometry for the free stationing and enabled them to quickly orient the SX10 into the coordinate system. “The method was especially useful for us since no existing control points or benchmarks were available and high relative accuracy was required,” Hellström said. “Because we were in the countryside, we had no problems working with SWEPOS and RTK. We had previously compared existing benchmarks against RUFRIS in other locations. Those comparisons demonstrated that surveying with RUFRIS was efficient and provided high accuracy.”
Next, they split up the tasks, with Hellström using the R10 with a Trimble TSC3 controller to capture points for terrain modeling while Coulter worked independently using the SX10 as a robotic total station to collect details on pavement, structures and other features. In addition to capturing discrete point data, they used the SX10 to scan and photograph the entire site. “Depending on conditions we moved the SX10 three times at each location,” Hellström said. “We also completed one to three full-dome scans and captured panoramic photos.” Before leaving a site, the team established a benchmark point for use in subsequent visits.
An Efficient Path to Deliverables
Back in the office, Coulter downloaded and processed GNSS and total station data in Trimble Business Center software (TBC). Because the SX10 automatically combines the multiple scans in the field, she needed only to import the resulting single point cloud into TBC. With information captured in one georeferenced coordinate system, the different data types fit together seamlessly. After completing quality checks, Coulter prepared files for delivery to ÅF’s engineering team. Output from TBC included drawing files for use in AutoCAD as well as LAS files for the point clouds.
Gimring said they coordinated with the engineering teams to help them utilize the full potential of the multi-sensor field data. In addition to assisting with processes to create digital terrain models from the field data, the surveyors also showed engineers how to work in the point cloud. “By operating in the point cloud, we can bypass the work of modeling 3D objects,” Gimring said. With the complete data in hand, the engineers then developed plans for updating the bus stops.
The value of the scanning data became most apparent during the engineering phase. “It saved a number of revisits,” Hellström explained. “If the client wanted to add lines or objects, it´s done from the point clouds. For example, the client wanted big trees near the bus stops to be included, which was not in the original order. But it was possible to add the trees without returning to the site.”
The finished designs were provided to SRA for construction. Design data from ÅF could be viewed in Google Earth to aid stakeholders in visualizing the planned upgrades. ÅF also exported information on benchmarks from TBC to Google Earth to provide access to control information at each site.
The approach developed by ÅF produced both immediate and long-term benefits. In addition to speeding the work at each site, the RUFRIS approach eliminated the need for traversing and leveling between existing benchmarks (if located) along with the associated issues for safety and guard vehicles. Other safety benefits included the ability to capture points on the road from the point cloud instead of walking on the road. And reflectorless measurement enabled Hellström and Coulter to collect points on the road without being close to traffic.
The most important benefit came on the bottom line. “We were able to save time and money while working in a safe environment,” Gimring said. “It’s cutting costs without cutting corners.”