Increasing surveying accuracies and productivity

Land surveying is an ancient practice, dating back at least 5,000 years to when Egyptian rulers used it to tax land plots. Over the centuries, it has been repeatedly transformed by new technologies — the compass (about 200 B.C), the theodolite (1550s), Gunter’s chain (1620), the sextant (1757), electronic distance measurement (1950s), and total stations (1970s). Then came GPS, followed by the other GNSS and corrections services.

Now comes sensor fusion, which aims to compensate for the limitations of GNSS — orbit and satellite clock errors, ionospheric and tropospheric delays, multipath, dilution of precision, urban canyons, jamming, extremely weak received signal, etc. — by integrating it with other sources of positioning data, including inertial measurement units (IMUs), lidar sensors and cameras. Even crowdsourced geolocation data collected with cell phones help expedite surveys by guiding surveyors to landmarks.

In the following article, representatives of five companies share their perspectives on recent advances in surveying and the remaining challenges.

Many More Satellites

Compared to just a few years ago, there are many more GNSS satellites, signals and options for correction services. Over the past decade, the average number of satellites in view has more than doubled to more than 40 today. Some parts of the world have more than 70 satellites in view, said Boris Skopljak, vice president, Surveying & Mapping Strategy and Product Marketing at Trimble Inc.

“The developments in GNSS field systems have always been geared toward simplifying workflows, improving accuracies and increasing productivity,” Skopljak said. “In the last few years, we’ve seen that on a massive scale. In some of our materials, we no longer even quote how many signals our GNSS receivers are tracking.”

The vast increase in the number of satellites has extended high-precision applications to the robotics and automotive markets. The challenge now is “position solution,” not just GNSS, said Simon Peng, director of the Overseas Department at ComNav Technology. The improvements in the satellite constellations, antenna technologies and algorithms also enable surveyors and other users to obtain results faster and to operate in environments previously impervious to GNSS, such as under heavy canopy and very close to buildings.

“Our customers can now operate in environments where there is no virtual reference station (VRS) infrastructure or real-time kinematic (RTK), by leveraging precise point positioning (PPP) solutions, such as the Trimble RTX corrections service,” Skopljak said.

“Additional satellite signals and constellations (like Beidou),” Skopljak said, “improved antenna technology and continuously evolving algorithms are contributing to improving the RTX accuracy while bringing the convergence times to almost instantaneous in normal conditions and making technology available in more regions.”

“When I first started surveying, if we had a 12-channel receiver, that was doing very well,” recalled Jesse Huff, head of Sales and Marketing, JAVAD GNSS. “Now, we’re tracking 36 birds in the sky at one time with an 874-channel receiver. That’s phenomenal.”

Huff described a patent-pending feature called real-time post-processed kinematic (RTPK). “It combines RTK, PPK and PP techniques, with multiple core processing engines and a single solution coming out of that. It is impressive standing underneath a giant oak tree and surveying that monument with GPS and knowing what your accuracies are. We’re not even chasing RMS values; we can report the actual positional uncertainties, which is amazing.”

“With so many signals and the new ways of how we compute positions based on PPP technology, we can almost globally get to centimeter-level positioning within a couple of minutes from just one global correction link,” said Bernhard Richter, vice president of Geomatics at Leica Geosystems AG, part of Hexagon. “Under optimum conditions, you can have almost an instantaneous global accuracy of a couple of centimeters.” In mature areas, he added, a local RTK network infrastructure enables achieving centimeter accuracy within a couple of seconds.

Galileo, Richter pointed out, will be fully operational in 2023 with great signals, though he’s “a bit skeptical” about the system’s target date for its high-accuracy service. “So, we will basically get global constellation corrections that allow us also centimeter-level positioning.” BeiDou has been fully operational since 2020. “GLONASS is more unpredictable,” Richter said. “It looks like modernization is slowing down a bit, in particular the CDMA developments.” Additionally, he pointed out, it is possible that one or more governments may decide not to use those signals, for military or political reasons. “It’s not the manufacturers who decide which signals to take.”

“In open-sky conditions, additional satellites have added redundancy — which is always good for position integrity — but it’s only when obstacles start to appear on the horizon, blocking out parts of the sky, that all-in-view RTK really comes into its own,” said François Freulon, Head of Product Management at Septentrio. When they did not have a full view of the sky, he recalled, GNSS users used to have to carefully schedule their work to coincide with times of high satellite visibility. “Nowadays, by using multiple constellations and signals, RTK can reach the parts that receivers in the past could not tread. More signals and constellations have also helped in easing the collection workflow for surveyors, making the capture of data in difficult conditions much quicker and more efficient.” New correction services are further simplifying the workflow “thanks to new positioning techniques, pricing business models and simplified network density.” However, corrections companies still face challenges in ensuring that centimeter accuracy can be uniformly achievable at a global scale.

Sensor Fusion

The ongoing evolution in computing power and communication technology “leads to many more sensor combinations,” Skopljak said. “We are not talking about GNSS alone anymore. We are talking about integrating a GNSS antenna, a receiver, an IMU, power and communications into a single compact housing.” The integration of inertial sensors makes it possible to localize the instrument rod tip when the pole instrument is tilted. “That allows our customers to measure more safely in dangerous environments.”

“We are reaching a maturity stage of what we can do only with GNSS,” said Richter. “It’s all about sensor fusion. The problem when signals are obstructed, that’s not solved, even though we can do positioning from Wi-Fi hotspots or from local pseudolites.” So, fusing data from cameras, lidar, GNSS and IMUs in better ways is the way to go and presents “a huge open research ground.”

For Richter, the challenge is not just positioning, the orientation of objects is almost as important as that, especially for such tasks as machine control. “It’s also about what you do with the data that you collect. Hexagon’s vision is of an autonomous future where we put data to work in connected ecosystems to boost efficiency.” However, he pointed out, this requires large amounts of data, such as those from aerial photogrammetry, lidar and mobile mapping systems used to create city models and digital twins of buildings. “If you really want a car to drive autonomously through a city with all the things that could happen, you must rely on a perfect replication of the real world,” he said. Other examples he cited are more efficient evacuation plans and flooding simulations. “GNSS will never be enough, but it will always be a very good enabler because it works.”

Classes of Receivers

Two decades ago, we would often group GNSS receivers by accuracy into three buckets: consumer grade, resource or mapping grade, and survey grade. As accuracy has increased for all GNSS receivers, the boundaries between those categories — especially between mapping and surveying — have blurred. “The performance of GNSS has increased so much that we are not using the traditional accuracy-based differentiation between surveying and GIS,” said Skopljak. “For mapping professionals, 10 years ago it was all about points, lines and polygons; now it is all about locating assets and adding the most accurate positions as attributes to those assets. For our survey and engineering customers, what matters is still geometry and working with the models to serve the connected construction in the field.” As for the pure GNSS technology stack, “we are seeing fewer differences between mapping and surveying receivers, but we are focusing on serving the customer in terms of product-as-a-service or as a productivity tool.”

Huff made two points. First, that “survey grade” does not necessarily equal RTK. “Some education needs to happen so that people understand RTK as a technique, not an accuracy. You can get poor accuracy and poor fixes with RTK, even when you’re using good techniques. So, when I say ‘survey grade’ I’m still talking about the full frequency receivers, using all available signals.” Second, that consumer-grade receivers, such as the chipsets in our phones and computers, do not require the same robustness as professional ones. “While they may be achieving the same precision, surveyors must be able to defend their position in a court of law.”

Crowdsourcing

Huff cited the “phenomenal” success of the simultaneous localization and mapping (SLAM) movement with all kinds of positioning challenges. “From a survey perspective,” he said, “we’re dealing with a much more feature-rich dataset than we were even just 10 years ago, with everybody having some type of GPS device on their phones. There are location tags on everything. That creates evidence for the surveyor to be able to go out and recreate things, reduce trips to the field, reduce rework times — all those things that make a surveyor’s life much easier.”

Surveyors now can fly aerial surveys of hundreds of acres in less than half an hour using drones with RTK, Huff said, instead of having to wait for the flying season with traditional airborne photos. If needed, they can pick a few ground-control points for ground truthing. “We’re able to do that with photogrammetry techniques, but using GNSS technology to position drones, whether it’s real time or post-processing, has definitely made surveying jobs easier.”

Correction Services

Correction services — such as satellite-based augmentation systems (SBAS), the ground-based Wide Area Augmentation System (WAAS) and the European Geostationary Navigation Overlay Service (EGNOS) — make a big difference along with PPP and similar techniques when base stations are not available. “We have the whole CORS network here in the United States,” Huff pointed out. “We also have services available from the National Geodetic Survey.”

Those who don’t want to have to fully engage in post-processing can upload their data to the Online Positioning User Service (OPUS), AUSPOS (a free online GPS data-processing facility provided by Geoscience Australia) or other corrections services that will post-process positioning data. “It has made it more accessible for all the surveyors all the way around, especially as the technology has improved and the cost barrier to entry into a survey-grade GPS receiver has come down significantly as well,” Huff said.

Growing Adoption of GNSS

The greater number of satellites in orbit significantly reduces convergence time and increases the accuracy of the solution, which makes the technology much more user-friendly for professionals and nonprofessionals alike.

For surveyors and mapping professionals, the increasing levels of GNSS performance means that “GNSS continues to be the dominant equipment and they can operate in challenging GNSS environments while still meeting the accuracy and precision requirements,” Skopljak said. GNSS usage is also growing in such industries as agriculture, construction, transportation and logistics. “Now, when farmers are on a combine, they don’t have to wait for an RTX or PPP solution to converge for 20 minutes. The solutions just work, and they can perform their task.”

Skopljak also pointed to “more flexible business models, such as pay-as-you-go or equipping seasonal workers or fleets of spatially enabled consumers to use GNSS,” that reduce the required upfront investment. “Surveyors now can go for longer and be productive in more areas where they could not use GNSS technology before. The non-surveying professionals — such as in natural resources, farming or construction — now can just turn on the machine and things work for them. They don’t have to worry about coordinate transformations and things like that.”

The Role of Total Stations

“Twenty years ago, when RTK and networks kicked in and then became popular, we were discussing whether it was the end of the automated total station,” Richter recalled. “Yet, the number of automated total stations has grown ever since.” To him, this is proof that GNSS alone will never solve all surveying problems. GNSS’ weak signal will always require surveyors to supplement it with other sensors, such as reflectorless total stations. “These instruments always need to work in harmony,” Richter said.

Success on both construction sites and in machine control require a very good robotic total station and a very good GNSS receiver, Richter said. “The simple problem of leveling a pole is actually solved, and we are using the technology that we developed for tilt-compensating GNSS receivers. We’re leveraging this now into the world of the total station.” This has solved one of the fundamental problems surveyors have long had, because they no longer need to level up and can measure tilted poles with a total station and with a GNSS receiver. “We have also made it very seamless for surveyors to switch between using GNSS receivers and total stations,” Richter said.

Featured Photo: Septentrio