On a French vineyard in the Loire Valley, a tractor is driving between the grape vines with no one behind the wheel. Meet TREKTOR, the autonomous hybrid robot that works tirelessly to weed the organic vineyard producing some of the finest Gamay wine, called Anjou Gamay Village.
After TREKTOR worked the land for a month, its developer, a company called Sitia, reviewed the quality of their autonomous robot’s work. They counted grape vines damaged during operation — two in one month — and approached the farmer to reconcile the liability. To Sitia’s surprise, he responded, “When I use my manual tractor to get the same job done, I damage at least two vines a day! How did your tractor manage to be so careful?” Sitia’s developers thought for a while and then replied, “It’s thanks to the high quality and accuracy of the components that are inside.”
“Despite the strong magnetic field emitted by the generator on the TREKTOR, the AsteRx SB ProDirect receiver did not have any issues,” said Clément Aubry-Tardif, Sitia’s R&D manager. “The spectrum analyzer in its web interface showed other small radio interferences aboard the robot, but everything was still working fine.”
Integrated into the TREKTOR is an AsteRx SB ProDirect dual-antenna receiver, which provides the reliable high-accuracy positioning and heading needed for autonomous operation. Sitia chose the receiver for the following reasons.
It has centimeter-level accuracy with RTK, which reduces crop damage and increases yields.
Its heading helps point implements in the right direction. Unlike inertial systems, it’s reliable and accurate even in static or slow-moving applications.
Built-in advanced interference mitigation (AIM+) technology makes it resistant to radio interference, while its LOCK+ technology ensures robust satellite tracking even under intense vibrations or shocks.
It includes an intuitive web interface for fast prototyping and easy real-time testing.
Sitia is a French company specializing in autonomous robots. Its TREKTOR helps compensate for the current farmer shortage, which is especially felt on organic farms, where weeding is seven times more labor intensive due to the use of few (if any) herbicides. TREKTOR is a flexible solution that can adjust its height and width on the fly, adapting to various working environments. It can also change implements to perform various functions. Depending on TREKTOR’s dimensions and implements, the distance from the crop to the robot changes, making high-accuracy positioning crucial to minimize damage to any of the crops.
Approaches to providing real-time kinematic (RTK) solutions at high rates have existed in various forms for decades, providing value for high precision applications. This technique is nearly universally adopted in the industry, and many surveyors may have been using it for years without realizing it. Yet there are persistent misconceptions about the subject.
By Gavin Schrock, PLS
For many on the development side of high-precision real-time kinematic (RTK) GNSS, like those we interviewed for this article, the incorporation of high-rate solutions into their RTK products is a given — and has been for a very long time. Yet, in some end-user communities there may still be many question marks: Does my gear do it? Does other gear do it? What can it do for me? What are the pluses and minuses?
We asked for insights from 10 prominent firms that develop and manufacture RTK-enabled high-precision GNSS solutions and equipment, spanning multiple applications:
By high rate, we mean higher than 1 second (1 Hz) increments, such as 0.2 second (5 Hz), 0.1 second (10 Hz), etc. Part of the confusion about high-rate RTK is that there are two scenarios. One is transmitting corrections from a base or network at high rate, receiving and solving on-the-field sensors or rovers at a high rate (for example, 5 Hz base + 5 Hz rover).
The other is base transmission of corrections at a lower rate and receiving/solving on the rover at a higher rate (for example, 1 Hz on the base + 5 Hz or more on the sensor/rover).
While both can be valuable for different applications, what has been adopted as standard for most surveying, construction, agriculture and mapping applications is the latter.
What are applications that would run the base and rover at higher than 1 Hz? “Moving Base” applications are prime examples, where you are seeking to resolve positions for one or more sensors relative to a base that is also on a moving platform. Think of a barge on the ocean where a helicopter (or rocket) might be landing. Here is a definition from the user manual for a popular OEM receiver that has been in many makes and models since 2003:
“Moving Baseline RTK is an RTK positioning technique in which both reference and rover receivers can move. Moving Baseline RTK is useful for GPS applications that require vessel orientation. [For example, the] reference receiver broadcasts [correction] data at 10Hz, while the rover receiver performs a synchronized baseline solution at 10Hz. The resulting baseline solution has centimeter-level accuracy. To increase the accuracy of the absolute location of the two antennas, the Moving Reference receiver can use differential corrections from a static source, such as a shore-based RTK reference station.”
Beyond such specialized applications, running the base at a high rate is a burden on radios or bandwidth. Additionally, as industry experts explain below, it is of little (or no) value and may only unnecessarily use excess bandwidth and burden broadcast radios.
When would you run the base at 1 Hz and the rover at higher than 1Hz, such as 5Hz, 10Hz, or more? When the base is static. That pretty much covers nearly all surveying, mapping, precision agriculture and construction applications. What is meant by high rate in the sensor/rover receiver and its RTK engine, in the context of such applications? As one of the firms interviewed stated:
“The number of RTK position fixes generated per second defines the update rate.”
For most of the surveying, mapping, precision agriculture and construction applications, that means base 1 Hz + rover 5 Hz or 10 Hz. Then there are specialized applications, such as structural monitoring and geophysical studies, that may run sensors/rovers at 20 Hz, 50 Hz or (though rare) as high as 100 Hz. Whether a higher rate is a default, or 1 Hz is the default, changing the rate is almost always a user-configurable option.
A general perception is that base-rover gear defaults to base 1 Hz + rover 1 Hz. However, as the experts below note, that is not necessarily the case — often the rover rate is higher by default.
By any other name…
The respective approaches, and their appropriateness for different end-use applications, may seem fairly straight forward. However, part of the confusion about the subject for end users comes from the wide range of terminology used to describe how high rate is applied across the industry.
The understanding of processing approaches is clear among GNSS engineers, and in specific terminology, but this rarely gets translated well or consistently in terms meaningful to end users in documentation or marketing.
Developers might have different approaches to achieving high-rate solutions and would of course not wish to completely reveal their cards, but many of the fundamentals are the same. A mutual recognition of parallel development among GNSS engineers, and the manufacturers they develop for, in that each strives to continually improve solutions, means that the high-rate element of RTK generally does not get much marketing hype.
Often, when high-rate RTK does get laterally mentioned — in manuals, marketing or labeled as configuration options in GNSS field software — the mix of terms can confuse the user. Such terms as extrapolation, prediction, update rate and solution rate could evoke a negative connotation to an end user who is used to hearing one set of terms, and they might view otherwise like terms as contrasting terms.
GNSS engineers do not have issues with mixed terms. As some indicated in their respective interviews, they seem a bit puzzled as to why anyone would misunderstand the subject, and how marketing spin might lead users to be confused.
In recent years, the subject seemed to get discussed a lot more than usual in various high-precision end-user social media platforms. Perhaps this was a natural progression in growth of understanding of the nature of GNSS among these constituencies, and a desire to know more about what goes on in those black boxes — a positive thing. There may also have been some instances of marketing nudge.
For whatever reason it became a subject of discussion, we heard from readers who asked us to look into it. So here, in alphabetical order, are insights from of the experts in this field. You can jump ahead to the specific section for your equipment vendor, but we encourage you to read through each; combined, they provide a more complete picture of the subject.
Bad Elf
With Larry Fox, VP for Marketing and Business Development
Larry Fox uses the Bad Elf Flex. (Photo: Bad Elf)
Bad Elf has long provided GNSS solutions for aviation- and mapping-grade field applications. Several years ago, the company introduced a survey-grade-precision system, Flex. It is offered with an option for a modest initial investment in the hardware, and an innovative token system for enabling and operating at centimeter precision.
Larry Fox has been in the industry for a long time and has seen the evolution of real-time GNSS. He is Bad Elf’s vice president for marketing and business development, but he also had a key role in the development of the Flex system. Fox said that, of course, high-rate RTK is supported. “We allow options up to 20 Hz on the rover if the user has this enabled.”
For the approach of 1-Hz base and higher rates on the rover, he said that Bad Elf does not have a specific term for this. “For purposes of description, I could refer to it as high update rate, but I suspect high solution rate is pretty much synonymous.”
Fox explained how the standard approach works. “The rover knows the location of the fixed base and therefore applies the same processing techniques by simply reusing the last received data.”
He also mused about various hypothetical scenarios. “Given that the converse is also possible — a slow data rate from the base, say, 0.2 Hz at the base and 1 Hz at the rover — is there fundamentally any difference?”
For many applications, Fox does not see a substantial advantage in running at higher rates: “I see no benefit for higher data rates in a static situation such as a survey. I would argue that in a survey workflow, one should allow the RTK algorithm to settle over the static shot being taken, as the RTK algorithm likely benefits from aging out some of the data it used while moving.”
He adds, “I would suggest that once you have occupied a point for a modest amount of time and you remained fixed, I can’t see any benefit. My argument here is that by the time you have leveled and prepared your collector of choice, any decent RTK receiver with a good sky portrait and good corrections will not observe any benefit.”
As for disadvantages and trade-offs, “More and faster data,” Fox said, “must be better, correct? Sarcasm included. Unless there is a tangible need for more samples, what is one going to do with all the extra data? I could have seen a possible argument that a single constellation receiver may benefit from averaging, but that could be a be a whole different subject as multi-constellation is now standard. Arguably, at a higher data rate one could capture more epochs and reduce the time on station. With multi-constellation receivers I am just not convinced that these techniques have the same merit they may have had in the past.”
Bad Elf doesn’t support higher correction transmission rates from the radio. “The current module only supports RTCM3 at a 1Hz rate,” Fox said. “Even if we could transmit faster, the payload required would exceed the capability of the message transmission rate of the radio. The battery life of a radio is directly correlated to the transmission duty cycle. The more you are transmitting, the less battery life you will have. I would argue this would impact the useful field time you would have without an external battery solution.”
Fox notes that any application where a rover is moving — such as on a vehicle or for machine control — could benefit from high rate. “I could see a potential application for drones,” he added. “I would want to have the epoch of an image recording very tightly coupled to the image captured. Fundamentally, an RTK drone’s imagery is only as good as that. If one was taking video at any reasonable framerate, a higher frequency RTK GNSS may benefit the geolocation of more individual frames with less extrapolation.”
What about rates higher than 20 Hz? “We have run our receiver up to 20 Hz on the rover side. Although there are units capable of even higher rates, I don’t have any data that would convince me that this is viable, for mapping or surveying.”
I asked about some of the misunderstanding out there about high-rate RTK, and Fox replied, “We can be creatures of habit and tie ourselves to beliefs that ‘this is the way I did it and it worked then.’ People should always ask themselves the question, ‘do I still need to do it this way?’ Again, there is the premise that more is better. I can’t tell you how many times I have seen people collect very high-rate data for lines and poly features only to decimate the data because it reduced performance, increased storage, or lowered the performance of the apps rendering the data.”
Emlid
With Svetlana Nikolenko, Lead Application Engineer
Photo:Svetlana Nikolenko with an Emlid GNSS receiver. (Photo: Emlid)
Emlid, a relatively new entrant to the market for high-precision GNSS, has made a splash with their line of affordable systems, such as the Reach RS2 rover and base-rover kits, and RTK systems for UAVs.
“All our devices support this,” said Svetlana Nikolenko, lead application engineer. “We do not have a special term for this, as it is simply a standard. We recommend 5 Hz and higher for a moving rover, but it can be overkill for a stationary one.”
Asked why one would want to run at high rate, Nikolenko explained, “The need to set a higher update rate depends on the rover’s velocity and acceleration. The higher the update rate, the more solutions per second are calculated. So, if you’re moving fast, the higher update rate simply allows you to keep your position current. If the rover is stationary, there are no issues with working at 1 Hz. Still, there is nothing wrong with running a stationary rover at 5 Hz or higher: it is excessive, but produces more samples with different satellite geometries.”
For moving applications such as UAVs, higher rates are of value. “It really depends on velocity,” Nikolenko said. “For example, if the rover is on a drone flying at a speed of 5-20 m/s and the update rate is set to 1 Hz, you won’t have the actual positions of the images. The higher update rate our devices have is 10 Hz, and at a drone speed of 20 m/s, even if you take photos each second (which might be a bit excessive), you’ll get accurate positions.”
Using an Emlid receiver in harsh conditions. (Photo: Emlid)
Emlid does not support a moving base. However, if there is a strong demand from users, they will consider adding this. For non-moving applications, Nikolenko said, an approach of broadcasting from the base at a high rate is excessive. “This increases the load on the radio (or any other connection link) because the base sends its position and corrections to the rover as often as it calculates it. Anything excessive simply adds load to processors and batteries.”
CHC Navigation
With Carlos Cao, Technical Manager for the Asia-Pacific region
CHC Navigation, or CHCNAV, has steadily grown as a recognizable brand of GNSS and other geospatial products internationally. While the brand might be new to some in North America, in some regions of the world CHC has a substantial share of the market, selling hundreds of thousands of units over the past 15 years. The company develops its own solutions, but also incorporates OEM components. In all cases, CHCNAV has provided high rate as standard from its earliest days.
Multi-constellation rover with tilt compensation. (Photo: Schrock)
Carlos Cao, technical manager for the Asia-Pacific region, said that his company supports the approach of broadcasting at 1 Hz and solving at higher rates on the rover. “For example, you can get coordinates every 0.2 seconds in the Landstar 7 Topo Survey software,” said Cao. “Meanwhile, with different OEM boards, RTK models and supported software, [the equipment] can also reach 10-Hz or 20-Hz static data recording and NMEA data output (including GNGGA coordinate data).” Their term for solving RTK solutions at a high rate on the rover is “high update rate.”
This can bring advantages, specifically for moving applications, Cao said. “When you stake out, the 5-Hz update rate brings faster coordinate updates, especially when surveyors walk quickly. When you survey by time during movement, you can get denser points; while you survey by distance, the accuracy will be better if you are at high speed. For example, speed is 6 m/s, and you want to survey a point every 5 meters; 1 Hz update rate cannot do this with high accuracy.”
When would 1Hz be sufficient? “Normally,” Cao said, “a 1 Hz update rate is enough for a topography survey because users won’t survey at a high speed, so our default setting is 1 Hz, though you can choose higher rates if enabled and as needed. Unless you are moving, however, such as when some surveyors mount a rover on a vehicle, there is no significant difference in the final results.” He added that running at high rates can drain the battery faster.
Broadcasting at higher rates has several major issues. “With more satellites launched, especially BeiDou, correction data becomes much larger,” Cao said. “It means that network RTK requires more data flow, and UHF radio RTK needs a UHF modem that can send data at a high rate. It is a very big challenge for base RTK.”
Meanwhile, notes Cao, “The rover could even have a correction age of 5 or 10 seconds, and it will use the previous package to calculate the position. Since 1-Hz base and 5-Hz rover can work without degradation of precision, there’s no need to change the base to 5 Hz.”
Other applications CHC supports often use higher rates. “Navigation, machine control and precision agriculture normally use a 10-Hz, 20-Hz or 50-Hz update rate,” Cao said, “because these devices work under high-speed movement status, especially navigation. Also, they need to combine with high-update inertial measurement unit (IMU) data. The max update rate is 50 Hz. Normally the application data for these uses is NMEA data output by COM port or TCP/IP protocol. For surveying applications, such as topography, 1-Hz base and 5-Hz rover is enough. For other applications that need higher rates, we also provide such devices.”
Hemisphere GNSS
With Kirk Burnell, Senior Product Manager
Kirk Burnell
“At Hemisphere, we simply refer to this as RTK,” said Kirk Burnell, senior product manager for Hemisphere GNSS. Burnell added that they do not have any special term for this — it is simply a standard.
We were discussing specifically the approach of solving on the rover at higher rates than the base corrections. “All Hemisphere RTK products can work in this way, meaning corrections can come in at 1 Hz or slower, and rover output can be at 1 Hz, 5 Hz or 10 Hz as the user sees fit and as the application demands.”
Hemisphere develops GNSS and multi-sensor solutions for many industries: surveying, construction, agriculture and more. While Hemisphere has its own branded survey rovers, its OEM boards are in many other popular rover brands, makes and models. So, whichever you are running, you get high rate as a standard option.
Hemisphere’s receivers are frequently used in construction applications. (Photo: Hemisphere GNSS)
Burnell explained further that this is a given in the industry. “This is the standard expectation for RTK amongst our competitors, based on their product offerings, documentation, and standard operation. When describing RTK, the expectation is for 1-Hz base-station corrections, and a user-selectable rover output rate. Understandably, when people discuss RTK in technical terms, they may use different phrases to help distinguish between different techniques, which is why there might be different phrases out there. For us, it is simply RTK.”
As for the benefits of high rate, Burnell explained that inside the receiver, the measurement engine and RTK algorithms are typically running at 10 Hz or 20 Hz, and the selected output rate of the solution does not impact the RTK engine’s performance. The receiver will fix as fast and as accurately as possible given the quality of the RTK correction stream. Survey users could see a smoother update rate on their screen using 5 Hz compared to 1 Hz. This makes such tasks as leveling the rod or watching the change in height on screen while moving from the bottom to the top of a curb feel more natural. The user is not waiting an extra second each time to see the stability of the output. “A 5-Hz update rate is a good tradeoff for smooth workflows versus consuming CPU and battery power, compared to 10 Hz or 20 Hz,” he explained.
Would there be a disadvantage to simply running the rover at 1 Hz? “When using a 1-Hz update rate to the data collector, there will be fractions of a second spent waiting for the screen to update,” Burnell said. “Over the course of a day’s work, this could add up to a few minutes of extra time spent. In reality, this does not impact the ability to deliver a job on time. If the user does not feel impeded by the slower update rate of the screen, there is not a significant difference between the quality of the data, comparing 1 Hz and 5 Hz.”
Addressing one misconception that some users have about high rate, that it might significantly improve precisions, Burnell clarified, “For classic RTK surveying, outside of the workflow differences for the surveyor, the same quality of data is produced.”
Disadvantages? “Once you move beyond 5 Hz you start to exceed people’s hand-eye coordination ability, and the benefits diminish,” said Burnell. “Additionally, the data collector has a lot of communication to process, data to unpack, calculations to do, and screen refreshes to accomplish. Faster than 5 Hz leads to stresses in these aspects of the user experience, and ultimately can consume the data collector’s batteries at a faster rate.”
There have been instances of high rate being marketed as enabling users to save a lot of time, but as Burnell noted, this might actually be a potential problem. “There could be a false sense of having no latency, which could lead to rushing through a job, increasing the chances of making a mistake. A surveyor’s observations and measurements are the currency of their trade, and they should be made with care and attention to the work being done. Most surveyors take pride in a job well done.”
Regarding the other scenario, broadcasting at a high-rate and solving on the rover at the same high rate, “This mode of RTK operation has little or no benefit and a host of drawbacks,” Burnell said. “The biggest issue is the volume of data. For a multi-frequency multi-GNSS solution, there is an immense amount of data to be transmitted from the base to the rover. Running a link at 5 Hz requires huge data bandwidth generally only possible using an internet link as compared to a 450-MHz or 900-MHz radio link. Drawbacks for internet links are data volume costs. For dedicated radio links, the issue is most likely to impact radio range. To send five times as much data, the over-the-air baud rate needs to be five times greater. This means that the energy per bit of data is five times less when at high speed. The signal will lack the ability to punch through obstacles. While some may suggest that having five times as many corrections reach the rover compensates for this, some radio protocols can be configured to transmit multiple retries with 1-Hz data.”
However, there are advantages to running at higher rates for specific applications, Burnell said. “If data is being collected in a kinematic fashion as compared to shooting individual points, there will be more detail when collecting at 5 Hz. For example, driving along a road with a receiver mounted to the roof, in 1 minute of driving there will either be 60 measurements at 1 Hz or 300 measurements at 5 Hz. For many non-survey applications, this is critical. For example, at highway speed, 1-Hz data means 1 point every 30 meters (100 feet) or so. In machine control, the systems are not relying on hand-eye coordination and reaction time, and 20 Hz or 50 Hz are common speeds. Autonomous applications also typically use between 10 Hz and 50Hz for GNSS, and often combine this with 100-Hz or 200-Hz IMU data. Aerospace and defense applications have demanding conditions and use 100-Hz to 200-Hz IMU data to navigate, often combined with 1-Hz, 10-Hz or 20-Hz GNSS data.
There are even some applications for which it is warranted to broadcast corrections at rates slower than 1 Hz. “One example was a user in Japan, where radio links are often throttled to 4800 baud,” said Burnell. “They were looking to see how to slow down corrections to less than 1 Hz so that they could take advantage of multifrequency multi-GNSS RTK. Another example: I recently asked for some 10-Hz rover data for analysis. With very large files, analysis took much longer — I wished I had asked for 1-Hz data!”
Hexagon | NovAtel
Hexagon | NovAtel is a prominent tech firm providing positioning, navigation and timing (PNT) solutions for multiple industry segments, including defense, surveying, construction, agriculture, autonomy and more. While GNSS is a core technology, NovAtel develops multi-sensor systems (including inertial) and has a broad reach with its OEM products. Surveyors, for instance, might not be familiar with NovAtel first-hand, but have likely used its technology via NovAtel’s many OEM customers.
Iain Webster
Iain Webster, senior director of Geomatics and Software Engineering for NovAtel, said that not only does NovAtel support high-rate RTK, but the customer can choose the position output rate desired — 1 Hz, 5 hz, 10 Hz, 20 Hz, etc. — and the receiver will output RTK positions at that rate.
“We distinguish between a matched solution (where a correction is matched with a rover observation at the same time tag), and a low-latency solution, where base observations are extrapolated for position computation at the rover,” Webster said. He provided a description from a company manual:
“The RTK system in the receiver provides two kinds of position solutions. The Matched RTK position is computed with buffered observations, so there is no error due to the extrapolation of base station measurements. This provides the highest accuracy solution possible at the expense of some latency, which is affected primarily by the speed of the differential data link. The MATCHEDPOS log contains the matched RTK solution and can be generated for each processed set of base station observations.
The Low-Latency RTK position is computed from the latest local observations and extrapolated base station observations. This supplies a valid RTK position with the lowest latency possible at the expense of some accuracy. The degradation in accuracy is reflected in the standard deviation. The amount of time that the base station observations are extrapolated is in the “differential age” field of the position log. The Low-Latency RTK system extrapolates for 60 seconds. The RTKPOS log contains the Low-Latency RTK position when valid, and an “invalid” status when a Low-Latency RTK solution could not be computed. The BESTPOS log contains either the low-latency RTK, PPP or pseudo range-based position, whichever has the smallest standard deviation.”
NovAtel does not brand this as a specific feature — it is just a standard part of its RTK solutions, but the company refers to it in their documentation as a “low-latency” solution.
The main benefit of this solution, Webster explained, is for kinematic users to allow better representation of their actual trajectory (such as in applications on moving vehicles). “The higher the dynamics, the more impact the latency of the matched solution will have to the point that we recommend the low-latency solution to all but specialist customers with known static positioning needs. For surveyors, there may be improved workflow with the low-latency solution as they will be able to move from point to point more quickly.”
NovAtel produces GNSS and inertial hardware and software, including OEM boards, for multiple applications. (Photo: NovAtel)
Webster noted that for applications where the rover is static for observations, 1 Hz can be fine, but for moving rover applications — kinematic — running at 1 Hz is probably unacceptable, so low latency is quite standard.
Additionally, he pointed out, there are applications where longer periods between corrections may not necessarily be detrimental. “Note that some manufacturers, including NovAtel and Leica, offer the possibility of using PPP corrections to extend RTK solutions beyond, for example, a 60-second timeout,” Webster said. “There are various proprietary methods to achieve this, but ultimately the RTK solution could be extended without limit in this way.”
Are there tradeoffs to using extrapolation or other high-rate approaches? “With corrections coming in at 1 Hz,” Webster said, “there is very little error over that period, so for most users, there is little disadvantage and perhaps some productivity advantage with a higher rate. If there is any trade-off, it is between getting the highest accuracy possible versus the lowest latency solution.”
As for the other scenario — the base broadcasting at greater than 1 Hz and the rover solving at greater than 1 Hz — “There is little advantage,” Webster said, “except in some specialized applications such as when the base is moving (called moving baseline) to provide a cm-level baseline between the base and the rover for relative positioning. For typical surveying applications with a static base, the rover would have to wait until the corrections arrived before outputting a solution. Other downsides include increased bandwidth on the communication link and more loading on the rover CPU, meaning lower battery life.”
What are the non-surveying applications where a high rate (in either scenario) can yield a specific benefit? Webster noted that, in fact, they deal mostly with non-surveying applications. “Most use cases need 10 Hz or 20 Hz for machine control or precision ag. We do have some very specialist applications that have required up to or beyond 100 Hz — but it is often best in those cases to do a GNSS/inertial navigation system (INS) solution and use the IMU to output at that a high rate. As previously mentioned, there are other specialist applications where the base is moving. In this case, we run a matched solution at a high rate between the base and the rover.”
Leica GeoSystems
With Xiaoguang Luo, Senior Product Engineer, GNSS Product Management Group
Rover with calibration-free tilt compensation and camera-based offset point capabilities. (Photo: Schrock)
Leica Geosystems (part of Hexagon) has been a major global developer and manufacturer of GNSS systems for multiple disciplines for several decades, introducing its first GPS receiver, WM101, in 1985. Since then, Leica has been among the leaders in GNSS receiver innovation, including integrated systems such as a rover that incorporates calibration-free tilt compensation and an image-point capture feature (GS18 I). Therefore, it is no surprise that for Leica Geosystems equipment features high-rate RTK as standard.
Xiaoguang Luo is a senior product engineer in the GNSS Product Management group at Leica Geosystems. He confirms that this option is supported in all Leica Geosystems RTK rovers of the current product portfolio, and this option is enabled by default in the Leica Captivate (surveying field) software. A term Leica Geosystems uses is prediction for its high-rate RTK approach.
Xiaoguang Luo
The standard positioning rate is 5 Hz on the rover. “As far as GNSS processing is concerned, there is no fundamental need to go to higher positioning rates,” Luo said. “The need for high rates is mainly driven by applications. For example, we are using the 5-Hz position update rate at the rover by default for an improved staking workflow and user experience. The 10-Hz rate is also supported in Captivate, for example, when streaming NMEA messages.” He added that 10 Hz is supported for other applications, such as structural monitoring, and 20 Hz for machine control.
As for the advantages of a rate higher than 1 Hz, Luo said that working at high observation and solution rates enables the possibility of modeling fast-changing error effects with a period below 1 second, and allows for high-rate non-surveying applications such as bridge monitoring. Does a high rate have any significant effect on the final results? He said that it strongly depends on the use case where high-rate observations and positions are involved. In addition, the quality of prediction also affects the final results.
Bernhard Richter
By this he means that while the standard approach for applications where the base is stationary, such as surveying, can work so well with a base data rate at 1 Hz and rover at 5 Hz, the key conditions do not change much over a single second.
Luo’s colleague Bernhard Richter, vice president of geomatics, explained it. “To understand this, you need to separate the elements of corrections into those that are fast changing and range dependent (see the graphic below). If the errors change slowly, then they can be estimated and predicted very well. Or, if the range dependency is low, errors could come from a different source than the base station. If the range dependency is medium or high, then the corrections are more difficult to estimate on the rover side, but if such errors change very slowly, they can still be predicted very well with the precondition that corrections have been received at least once.”
The rate of change and dependencies for the elements of corrections. (Source: Leica GeoSystems)
You’ll notice that multipath is high in both regards. This brings up another misconception about high-rate RTK — some users have an expectation that it will improve their performance in limited sky-view situations (like thick tree canopy) or high multipath environments. This is not so. Any improvements in such environments come from having more satellites, more observations, and more modernized signals. With regard to high-rate and multipath, Richter said, “It is anyway futile, since multipath decorrelates so quickly that the advanced mitigation has to happen both in an analog and a digital way on the rover.”
While there are benefits to running at high rate, such as for staking, a balance has to be struck — for instance, in not running it at too high a rate. Luo outlined disadvantages that must be considered when performing high-rate RTK.
High processing load and battery drain, particularly with multi-constellation and multi-frequency RTK.
High temporal correlations between observations, which may not be considered in a sophisticated manner in the RTK algorithms.
High base rates provide challenges for the RTK data link devices, such as radios.
In addition, he noted that while any kind of predictive solution will introduce some amount of error, that would be so small in, for instance, a base data rate at 1 Hz and rover at 5 Hz solution, as to not even be noticeable in the positioning results.
Septentrio
With Bruno Bougard, Research and Development Director
Bruno Bougard
“Our rover solution computes RTK up to 100 Hz,” said Bruno Bougard, R&D director at Septentrio. “Update rate requirements for industrial machine control applications are typically 20 Hz. This is necessary to capture the motion dynamics. Also, it is not only the update rate that matters in those applications, but also the latency, which should be low (<20 ms typically) and constant.”
Septentrio NV is a designer and manufacturer of high-end multi-frequency GNSS receivers and integrated solutions. Markets they serve include surveying, mapping, construction, science, timing, agriculture, marine, autonomy, and more — all with specific applications where high-rate RTK may be employed They also provide OEM boards and modules for further integration by others.
Surveying users for instance may be familiar with their Altus line of rovers, such as the NR3, where high rate is a standard option. “There are new applications where a higher update rate is required,” said Bougard. “Surveying with UAV, using photogrammetry or lidar scanning requires at least 10Hz. In mobile mapping in general, RTK-INS solutions such as SPAN, Applanix or Septentrio SBi, require update rates up to 200Hz.”
Bougard acknowledged that manufacturers use many terms for their high-rate solutions. “Some may be used to masquerading a low-rate solution as a high-rate one. This is not what we do. The rover observables are captured at high rate and can be up to 100 Hz. The rover RTK filter is also run on high rate. Fixed base-station data does not have to be high rate. 1 Hz is typically enough. For moving base applications — for example, when the base station is on another vehicle, and we want to compute the baseline between the moving base and the rover — 10 Hz is required.”
Bougard said that the benefit is to track the motion of the rover. This is critical in machine control, but also relevant for new survey flows (such as UAV-based and mobile mapping). The disadvantage, he explained, is that it requires higher CPU loads. “Suppliers, who focus on cost, tend to compromise on this, notably running higher rate only for a subset of the constellation or signals. We use them all.”
Is running the base station at a higher rate advantageous? “It is possible to increase the output rate of our base station correction stream but, as explained, this is not needed if the base is static,” Bougard said. “This is applicable to moving base scenarios as explained above. Indeed, if you increase the base-station correction rate, the bottleneck becomes the datalink.”
Tersus GNSS
With Xiaohua Wen, Founder and CEO, Tersus GNSS
Xiaohua Wen with a Tersus GNSS receiver.
Xiaohua Wen, based in Melbourne Australia, is the founder and CEO of Tersus GNSS, another new entrant in the centimeter-grade GNSS market. One distinction about Tersus is that the company has developed and produces its own GNSS boards, instead of using OEM boards from other companies. Tersus implements its own tech, including GNSS receivers and IMUs in its own survey rovers, such as the Oscar, and for other high-precision applications. Additionally, it produces OEM boards for integration by others. Tersus entered the market with full multi-constellation support and, of course, high-rate RTK options, and has recently announced a PPP (precise point positioning) service.
“Our RTK boards support up to 20 Hz,” said Wen. “Often, surveyor will choose 5 Hz. We do a 5-Hz solution in this manner: the baseband takes raw measurements at a wanted moment, say at 1.2 s or 1.4 s, and RTK calculates solutions with the raw measurements. We understand that some older solutions might simply extrapolate or interpolate based on a position and velocity sequence, which is sometimes called predicted RTK or extrapolated RTK (though those terms get used in different ways by different developers). That is not how we approach our RTK solution updates. All Tersus RTK boards also support a maximum 20 Hz raw measurements outputs.”
Multi-constellation rover with calibration-free tilt compensation. (Photo: Schrock)
We asked about some of the advantages users may envision of high-rate RTK in general. Wen said there may be little or no gain with regard to faster initializations. Likewise, there is no significant gain with precision and accuracy. However, Wen said that higher rates can sometimes improve staking workflows. “For example, in the case of our Oscar rover with tilt compensation, the RTK outputs solutions at 10 Hz, while the IMU samples at 100 Hz. Oscar calculates the pole tip’s position at 10 Hz, aligned with the RTK solutions, and the data controller or tablet displays the point of the pole tip on the screen. We find that the point better refreshes at 2 Hz or higher to respond to the pole tip movements without noticeable lagging.”
That movement is an example of a key value of high rate,“Speed or movement,” Wen said. “For surveying applications, I would say that 1 Hz could suffice, considering the characteristic very low speed. Usually, applications like machine control and precision agriculture require an RTK update rate at 5 Hz or higher. Some UAV applications may use a 100-Hz position update. Most of these applications use an INS+RTK solution. With INS, it’s easy to get a 100-Hz position update, while for an RTK solution, a rate of 20 Hz is probably enough.”
Wen said that broadcasting corrections at a higher rate is pointless for most applications, “because the base data is highly correlated in the short term. If it’s a moving base, the high-rate base data would make some sense. Otherwise, it just imposes a greater load on communications and computation, with almost no gain.”
Topcon Positioning Systems
With Alok Srivastava, Director of Product Management
Alok Srivastava
“It is a standard option in our rovers,” said Alok Srivastava, senior director of Product Management (PM) at Topcon. “Around the time I joined the PM team, in 2010, the decision was made to make 10 Hz the standard, though this is user configurable and can be 5 Hz, 20 Hz, up to 100 Hz.” He explained that faster rates have been available through several generations of their receivers.
Typical applications consist of a static base and a moving rover. Fast-moving applications can benefit from higher rover position update rates since the RTK engine is computing real positions at a faster rate. Higher rates on the rover side provide accurate changes in position that can be missed by interpolating between positions computed at a slower rate.
A Topcon multi-constellation rover with tilt compensation. (Photo: Schrock)
High update rates on a base station do not provide advantages except in rare cases where the base is moving. While rovers are computing movements of the rover antenna, base stations are providing GNSS satellite corrections. A rate of more than 1 Hz for a static base station does not benefit rover accuracy; it only creates a burden on the communication between base and rover. Base and rover communication needs to be optimized to reduce bandwidth requirements. This is especially true as we continue to add constellations and signals to GNSS solutions.
Sufficiently high rates have been standard on Topcon rovers for a long time. Srivastava would rather see more focus put on other aspects of GNSS — such as interference, spoofing, the impacts of 5G, precise point positioning (which Topcon provides through its Topnet Live service) and sensor integration. “In many of our construction applications, we have IMUs,” Srivastava said. “When an application has an IMU for tilt compensation or for machine control, the IMU and GNSS complement each other. In kinematic mode, the IMU can help reject outliers.”
“High rate can be considered a common default mode of operation,” said Stuart Riley, vice president, Technology – GNSS, Trimble. “Typical rover position solution rates are 5 Hz, 10 Hz and 20 Hz.”
Trimble is one of the pioneering companies in GPS and GNSS, and Riley has been directly involved in the evolution of the company’s GNSS solutions for more than two decades. He has seen a lot of change, and in noting the nature of key technological advances, offered this intriguing observation about high rate: in many ways it has become less relevant.
“There have been considerable advances in RTK technology in recent years that make many of the earlier concepts related to how base and rover data should be combined for baseline processing largely irrelevant,” said Riley. “Most recently, survey receivers have included INS support for tilt compensation applications, and these receivers have available high-rate IMU data — at a much higher rate than GNSS observables — which drive the final GNSS/INS integrated solution. Thus, the rover GNSS data rate is not so important.”
Riley noted another relevant technology that Trimble has implemented: the use of precise satellite clock and orbit corrections — such as from the Trimble RTX precise point positioning (PPP) service — to augment RTK when there is a loss of the base correction stream. The implementation of PPP is broadening across the industry, and the company was an early implementer of a global service. It has the RTX-based xFIll feature that runs on and high-end survey receivers. One of the misconceptions about PPP services such as xFill is that it is just there to “take over” should the RTK or NRTK corrections be interrupted. Yes, it does that as well, but to be able to do that, it is running all the time, simultaneously with the RTK, so the rover is getting these enhanced PPP service clock, orbit and other data. This improves what the rover can do. “The emphasis in modern survey receivers,” Riley said, “is based more on the availability of rover data, and a fundamental base data rate of, say, 1 Hz, is all that is required.”
Along with various advances in the rover RTK engine, the GNSS constellations have expanded considerably, requiring increased bandwidth for the corrections from base to rover. “Our products can use various communication technologies to transmit corrections, such as Wi-Fi, cellular, and UHF (450 MHz or 900 MHz) radios,” Riley said. “Maintaining a 1-Hz correction rate enables all the GNSS observables to be broadcast from the base, providing a suitable highly compressed data format such as when Trimble’s proprietary CMRx format is selected.”
Many terms are used in the industry, and they typically refer to some proprietary aspect of an RTK engine. Riley said that a generic term would simply be high update rate. “Providing the position is based on the most current phase observables at the rover, a low latency solution is possible,” he said. “Thus low-latency solution goes hand-in-hand with a high update rate. Predicted RTK may refer to an old method where the static base corrections are propagated forwarded to account for radio latency and thus synchronize base/rover data. This is not used in modern PVT (position, velocity, time) RTK engines.”
High rate on the rover is standard, but what benefits should the user expect from it? “A fast update rate provides the best user interface experience in the field, in particular for stakeout,” Riley said. “Quite simply, nobody wants to be working with a laggy display. For survey field work, 5 Hz is typical. Other applications, such as machine control, benefit from higher update rates where a default of 10 Hz would be used, with options for higher rates.”
If the user chooses 1 Hz on the rover, what would be the downside? “Running at a 1-Hz rate is not really suitable for stake out,” Riley said. “For occupying static points, 1-Hz updates would suffice, as a typical occupation has a minimum time of 1 or 2 seconds. Very high rates for survey applications do not really buy anything in terms of field look and feel or performance.” I asked him about any points of diminishing returns, and he responded, “The higher the rate, the wider the measurement bandwidth (that is, the noise increases — you cannot get something for nothing), so in fact going for an unnecessarily high rate would start to be a disadvantage. For example, there would be no advantage to using a 50-Hz or 100-Hz rate for a land survey application. There is a relationship between measurement bandwidth and position noise.”
When is a high base rate a good idea? High rates are supported for some machine control and “moving base” applications where the reference frame has to move with the moving base, Riley said. In this case, the base and rover observables must be synchronized and the final solution has a fundamental latency depending on the base rate. For this reason, moving base rates are more typically 10 Hz or 20 Hz. For a static base, it is possible to use a higher rate. However, as Riley noted, “It’s more likely that a lower rate such as 0.5 Hz might be desirable to accommodate the radio when using repeaters (time multiplexing the data) or low data rates. There are disadvantages to high base rates, mostly related to radio bandwidth. Other factors, such as ‘high rate = more radio transmit power’, may need to be considered (affecting battery life).”
Are there other cases for even higher rover rates? “As mentioned, machine control applications use higher rates — necessary to reduce position latency in control loops,” Riley said. “Other applications such as UAVs and autonomous driving clearly benefit simply because of the speed of the platforms (higher dynamics). Precision agriculture is an excellent example of machine control, where auto guidance is used. Although high rates are possible, nearly all applications manage perfectly fine at rates up to 20 Hz. A more important consideration is system performance in terms of positioning accuracy and convergence times, which is dependent on the technology used in the PVT engine, such as Trimble ProPoint technology, rather than the correction stream data rate. ProPoint also includes xFill, as mentioned earlier, which provides centimeter-level backup for continuous operation when RTK or VRS correction streams are interrupted.”
Other Manufacturers
This was only a sampling of the developers and manufacturers, but it should be noted that several of the above firms produce OEM boards featured in dozens of other brands and models, such as Carlson and GeoMax. To try to list them all would be a challenge and might be missing a key point: high rate is quite standard, is not big news anymore, and you probably have it by default (or optional) no matter what system you are using.
Hypeful
As the insights the from industry experts above show: high rate can be essential for many applications, but unnecessary for others. It seems more about user experience (staking workflows or moving rover) than some way to seek higher precision.
Additionally, to borrow the gaming term hypeful, some users believe (or have been led to believe) that running at high rate will yield higher precision or work some kind of magic in dense tree cover or high multipath environments. Some may argue that it could get a result faster, but in practical terms even that might not be the case.
High rate has been around for a long time. And like any tech, has gone through different development and adoption phases. Think about automatic transmissions for motor vehicles; they have been around in one form or another for more than a century. There was a period in the mid-20th century where the development of different approaches was promoted in marketing campaigns with fanciful product names, like Durashift, Presto-Matic, Geartronic and Torque-Flite. But rarely do you see auto transmissions highlighted with such marketing flourish since then.
High-rate RTK was never singled out like that; it is common, and any differences are mostly in how it has been adapted for different applications. I suppose a firm could choose to emphasize it for marketing purposes and give it a buzz name like “Turbo Thrusted RTK”, which his fine for marketing purposes (albeit a bit “cheugy”). Every developer and manufacturer will have slightly different approaches, but if you believe, or are led to believe, that any represent high-rate fundamentals exclusively, that would be inadvertently misleading, if not subtle gaslighting.
As one of the experts said, “It does not really matter what manufacturers claim or don’t claim. You cannot beat physics. You can only understand and manage the physics.”
Coolness Ahead
While high-rate might seem a bit old hat, where GNSS development is going is not. The developers we interviewed are more interested in highlighting their complete high-precision solutions. For example, adding inertial measurement units (IMUs) for no-calibration tilt compensation, additional sensors for imaging (and likely soon, lidar), and multiple real-time GNSS solutions complimenting RTK, such as L-band precise point positioning (PPP).
The “high-rate” that is truly exciting is that of R&D, multi-sensor integration, automation of certain elements of workflows, artificial intelligence and multi-constellation/multi-signals.
For navigation and control of any robotic or autonomous outdoor system, GNSS and inertial navigation systems (INS) are key components. Inevitably, the question arises: Should you build your own custom solution or integrate an available GNSS/INS combined solution? What would give you the best performance, while keeping the total cost of ownership (TCO) to a minimum? The TCO is also known as the “long-term price” and is defined as the purchase price plus the costs of operation over time.
Xenomatix is a company offering automotive solutions based on lidar technology. With eight years of innovative experience, Xenomatix has installed a pre-integrated GNSS/INS receiver on its latest lidar product, achieving high GNSS/INS performance with minimal TCO.
In an integrated INS/GNSS receiver, the GNSS receiver provides positioning with centimeter-level accuracy. The other component is a micro-electromechanical inertial measurement unit (MEMS IMU), which measures 3D orientation in terms of heading, pitch and roll angles with sub-degree precision. For its latest product XenoTrack, Xenomatix chose an INS called XenoAsterx based on the AsteRx SBi3 from Septentrio, which it integrated alongside its lidar to collect road-quality data to the smallest detail.
From an in-house solution to a pre-integrated system
Three years ago, when Xenomatix started developing its new lidar road-inspection system, the company had a GPS receiver, an IMU and an odometer as accompanying sensors. The company wanted to expand into new markets of road inspection in accordance with international standards, and so it needed to improve its components to take the overall performance of its system to the next level with RTK high-accuracy positioning.
To achieve this, while saving time and costs, Xenomatix acquired an AsteRx SBi3 INS/GNSS receiver, which allowed it to focus on its core lidar technology and sensor-fusion algorithms.
This off-the-shelf INS/GNSS solution provided all the high-accuracy positioning and orientation information Xenomatix needed, while eliminating most costs of development, maintenance and support. The new receiver allowed them to drive for miles, without any offset in positioning, something impossible with the previous GPS receiver.
The unique technology from Xenomatix stitches images by using lidar point-cloud overlays. However, when the car is moving fast, this overlay is smaller. The pre-calibrated GNSS/INS extends system performance by allowing stitching even when driving at higher speeds.
“If we start driving and we stitch the road for tens of kilometers and we come back to the same starting point, then we see an offset of only a few millimeters,” said Filip Geuens, CEO, Xenomatix. “This is for us the strongest proof of accuracy and reliability of the GNSS sensor.“
Why pre-integrated GNSS/INS offers better value
The pre-integrated GNSS/INS allows XenoTrack to collect road data even at higher speeds. (Credit: Septentrio)
A pre-integrated GNSS/INS solution — versatile enough to fit into virtually any autonomous or mapping system — offers the best value in the long run for the following reasons.
Better performance. The manufacturer of a GNSS/INS solution specializes in fusing the GNSS receiver and the INS in an optimal way. To accomplish this, the sensors are synchronized and their output run through a sophisticated Kalman filter algorithm. The fused device is then fine-tuned for optimal operation under various conditions. Finally, it is extensively tested and validated.
While being used by numerous customers and in varying applications, the GNSS/INS solution proves itself on various levels such as accuracy and robustness. This results in superior performance, even in the most demanding environments.
After installing the AsteRx SBi3 GNSS/INS system, XenoTrack was able to extend its functionality to inspect longer distances of roads at higher speeds. The AsteRx SBi3 operates reliably, even in challenging environments, such as when driving near high cliffs or under bridges.
Less development time and lower costs. When building a system, the development time is usually about one year employing two full-time GNSS/INS specialists. Hardware components need to be integrated and synchronized, while various interfaces and the Kalman filter need to be implemented. Additional features may be developed, such as velocity input as well as tools for validation, before the intricate step of performance fine-tuning. Finally, additional testing efforts are needed for verification and validation of the device.
On the other hand, a pre-integrated GNSS/INS system with easily accessible interfaces and flexible configuration ensures quick installation, meaning the product is ready within weeks.
Lower maintenance costs and support. Certain high quality pre-integrated GNSS/INS receivers are future-proof — ready to use new GNSS satellite signals and services as soon as they become available. An example of such upcoming service is the Galileo OSNMA anti-spoofing authentication.
Some receiver manufacturers such as Septentrio also offer continuous product improvement in the form of free firmware updates. A system developed in-house, on the other hand, needs continuous investment to maintain its competitive edge.
When issues occur, Septentrio also offers local worldwide support, with experienced application engineers ready to solve GNSS, INS or coupling issues that could halt the production process. For example, when Xenomatix discovered that its GNSS/INS was not working optimally in a certain environment, the company called Septentrio. Within days application engineering experts who analyzed the logged data found the source of the issue and proposed a solution.
Focus on core technology. When the budget is limited, choices need to be made about where to focus the efforts. When a company saves on GNSS/INS development, more can be invested in core technology. This means avoiding any lost-opportunity costs and optimizing margins.
Building your own is not always the best option
Acquiring a pre-integrated GNSS/INS receiver allowed Xenomatix to have a superior and affordable product with a competitive edge. AsteRx SBi3 increased the performance of the XenoTrack mapping system, while a short integration period allowed a faster time-to-market.
Xenomatix also benefited from low maintenance costs, keeping overall TCO to a minimum. Since the company was not spending time developing a custom GNSS/INS system, it could focus fully on its core technology. This allowed Xenomatix to take its business to the next level at a high pace.
Award-winning technology
In November 2021, the XenoTrack road scanner, with AsteRx SBi3 inside, was announced a winner of the IRF Global Road Achievement Award for its innovative road scanning and surveying solutions.
“Original equipment manufacturer (OEM)” is a widely used but poorly defined term. In general, it refers to a manufacturer that provides components or sub-assemblies to another one for use in the latter’s end products. In the GNSS industry, the purchasers of OEM boards typically are manufacturers of products that require positioning or navigation capabilities, such as guidance systems for tractors, UAVs or automobiles. Sometimes, such manufacturers integrate the OEM GNSS receivers with other sensors, such as inertial measurement units and lidar devices. Often, the OEM also will provide technical support to the integrator.
Much of the OEM business is not visible to the end user of the equipment that contains OEM components, let alone to the casual observer, because those components are “inside the box,” such as a guidance system, and “the box,” in turn, is under the hood or in some other hidden place. There is almost never a sticker on the outside analogous to the one that says “Intel inside” on many computers to distinguish the Intel CPU inside from, say, an AMD processor. Furthermore, OEM sales are typically obscured by confidentiality provisions in OEM licensing agreements that also address issues of branding, payment, quality assurance, and the timing of deliveries.
Integrators can choose from a wide variety of OEM GNSS boards depending on their intended use; the environment in which they will operate; their performance requirements; and their size, weight, power consumption and (of course) cost. OEM GNSS boards range from development kits that assist users to integrate GNSS into their product design to differential, multi-frequency, and, increasingly, multi-constellation boards.
In the following pages, six GNSS OEM manufacturers address these questions:
How do you define OEM?
What distinguishes your latest generation of OEM receiver boards from previous ones?
What are the markets for your GNSS OEM receiver boards? Which ones are growing the most?
Additionally, each one showcases a product.
JAVAD GNSS Ready for Lift-Off
JAVAD GNSS has been transitioning to a new position in the market since the passing in May 2020 of its founder, president and CEO Javad Ashjaee, a giant of the GNSS industry. For several decades, the company eschewed mass production for such markets as the automobile industry and cellular phones, choosing instead to focus primarily on high-accuracy surveying applications.
“Our founder really loved the surveying market, created a lot of technology, and drove the rest of the industry through the evolutionary process to where it is today,” said Tom Hunter, the company’s chief sales officer. “You can see a little bit of JAVAD GNSS in just about any GNSS-based land survey product on the market today.”
At the heart of each of JAVAD GNSS’ OEM boards is a proprietary ASIC. The boards it sells are the same ones it uses in its own reference stations, land survey products and marine systems, Hunter said. Aerospace is a key focus, an industry that requires very high accuracy, precision and reliability despite operating in environments of extreme shock, vibration, acceleration and temperatures.
Photo: Javad GNSS
“Our successes have been in working with many of the companies that build these very large launch vehicles used to carry heavy payloads into orbit,” Hunter said. “Our customers are companies such as Orbital, Northrop Grumman and SpaceX.” Those heavy-duty launch vehicles, he pointed out, must also follow a pre-described flight path. “You don’t want to start another world war because another country sees something heading its way.”
Tracing All Components. JAVAD GNSS’ boards “have complete component traceability,” Hunter said. The company does not buy any of its components from brokers. “We have to buy either directly from the manufacturer or from the manufacturers’ designated distributor, and it has full part traceability in our own factory in San Jose, California.” Should a component ever fail, the company could quickly trace when and where it was made. “That’s very important when we’re dealing with customers such as NASA, the Air Force or Boeing, because the safety of flight depends upon the performance and the quality of the product.”
The company will soon supply a receiver that will spend about four and a half to five years in orbit on a cluster of small low-Earth-orbit satellites, Hunter said. (See “JAVAD Board Guides ESA Vega Mission” below.)
To make sure none of its products are exported illegally from the United States, JAVAD GNSS also traces where each one ends up. “We know where every one of those boards is.”
JAVAD GNSS must guarantee its aerospace customers, which have invested millions of dollars in designing their systems, that each model of its devices will remain exactly the same. Hence, it bought from some manufacturers their entire inventory of certain components, in case they discontinued making them, and certifies each
JAVAD GNSS’ products are more expensive than those from other manufacturers because they are better, Hunter claimed. “We use really high-performance, temperature-compensated oscillators in our boards to make sure we have precise timing. We use a custom ASIC that we designed and built. Our receivers have 864 channels, so they can receive just about anything broadcast in the L-band.” The company constantly upgrades its devices to match modernization of the signal structures.
“I can remember when the rest of the industry was saying, ‘You have a 12-channel GPS receiver? You’re nuts! I mean, who uses that much information?’,” Hunter recalled. “Today, we’re using every signal that comes out of GPS, whether it be L1, L2, L5, L1C, and the same thing with all of the GNSS constellations.” For example, when Japan will begin to broadcast its new QZSS signal soon, “we’re ready not only to find it, but to track it, decode it, and utilize it for position and timing solutions.” Anti-jamming and in-band interference rejection are standard in JAVAD GNSS’ products, while those from other manufacturers require external filtering or different types of antennas, Hunter pointed out.
New Leaders and Markets
After Javad Ashjaee — JAVAD GNSS’ founder, president and CEO — died in May 2020, Tom Hunter, who co-founded Ashtech with Ashjaee in 1987, returned to the company after a five-year retirement.
“He left the company with an awful lot of technology, a lot of patents, and a lot of people who knew how to design and build products, not only for today, but for the future,” Hunter explained. “They needed some guidance.”
So, in January, Nedda Ashjaee — Javad Ashjaee’s daughter and his close collaborator for the previous 25 years — and the board of directors asked Hunter to rejoin the company. “They said that they wanted me to help them make sure that we can be on a path where we can use our core technologies and enter into new market segments and new marketplaces.”
Hunter added, “We made some changes to how we introduce surveying products into the marketplace.” The company no longer sells its products directly to end users. Rather, it goes through a new process and channel for getting products into the marketplace. It also brought on board a new chief technology officer this summer who will be driving engineering efforts. “We are becoming market driven. And to do that we needed to expand our marketing, sales and engineering capabilities. We are changing every aspect of the company,” Hunter said.
JAVAD GNSS actually consists of two companies in San Jose: JAVAD GNSS, which designs, markets and sells products, and JAVAD GNSS EMS, which manufactures them. It also has a presence in Moscow — the company hired many engineers following the collapse of the Soviet Union, many of whom had worked on GLONASS. “Javad looked at that as an opportunity to hire them and use them to develop a multiple constellation receiver,” Hunter recalled. However, as a subcontractor for U.S. government projects, it is much easier for JAVAD GNSS to operate on U.S. soil with engineers who are U.S. citizens. “We’re expanding our San Jose operation to include on-site engineering development, not only in RF, but also in digital signal-processing software.” The company will continue to receive schematics from its Russian subsidiary. “Instead of exporting technology, we’re importing it.”
JAVAD GNSS is now moving into markets that did not interest Javad Ashjaee. It recently launched new products in the machine control, marine navigation and accurate heading markets, as well as the agricultural and construction markets, with integrated sensors that can be readily installed on various machines. Other GNSS manufacturers have been producing such devices for decades, Hunter acknowledges. However, he adds, “ours will be able to use multiple sources not only for satellite- and terrestrial-based corrections, but a combination of those.”
A JAVAD OEM GNSS board is at the heart of the navigation system of the Vega space vehicle developed by the European Space Agency to launch small satellites into low Earth orbit. It provides great flexibility of mission at an affordable cost and represents the European solution for space accessibility. (Photo: Avio, Italy)
JAVAD Board Guides ESA Vega Mission
AJAVAD OEM GNSS board is at the heart of the navigation system of the Vega space vehicle developed by the European Space Agency (ESA). ESA developed Vega to launch small satellites into low Earth orbit. It provides great flexibility of mission at an affordable cost and represents the European solution for space accessibility.
The JAVAD OEM GNSS board is embedded in the gle/RGU/G2T/HDA/MB1 for space missions. (Photo: GreenLake Engineering)
The JAVAD OEM GNSS board is embedded in the gle/RGU/G2T/HDA/MB1 — a cost-effective, high-performance, compact and rugged GNSS receiver specifically designed and environmentally qualified. Installed on the upper stage of the VEGA launcher, it allows accurate trajectory verification during the entire flight mission.
ESA’s initial request was for a GNSS unit built with commercial off-the-shelf components, thus maintaining low costs, but which could still operate in the extreme vibration and shock conditions typical of a space launcher. After an initial feasibility analysis, GreenLake Engineering — a subsidiary of Instrumentation Devices — developed the unit mechanically and electronically to satisfy ESA technical specifications. Its biggest challenge was to pass ESA’s extensive qualification and quality process.
For many years, Instrumentation Devices (based in Como, Italy) and JAVAD GNSS have been partners. Instrumentation Devices sub-contracted for the Vega project with Avio (based in Colleferro, near Rome), which is the prime contractor with ESA. Avio is an international group that designs and produces space launchers and both liquid and solid propulsion systems for space transportation.
ESA supervised the project and is responsible for all activities relating to flight safety and qualification of the equipment installed on board. JAVAD GNSS supported GreenLake Engineering with the integration and low-level configuration of the OEM board for this challenging application.
A Massey Ferguson tractor guided by a NovAtel GNSS OEM receiver. (Photo: Hexagon | NovAtel)
GNSS Makers Share Insights
OEMs Discuss Their Boards, Markets and Company Growth
Five prominent GNSS original equipment manufacturers discuss their current products and future markets.
How do you define OEM?
While all six manufacturers agree on the general definition of OEM given above, they focus on different aspects. OEM customers of JAVAD GNSS “require reliable, accurate and stable high precision measurements for positioning and timing,” Hunter said.
The performance of OEM products from Hexagon | NovAtel reflects on its customers and itself, Gerein said. “Our OEM receiver cards are selected, valued and relied upon as the core positioning elements in many applications across vertical markets. We offer full rebranding options with custom logos, colors and industrial designs to seamlessly integrate our technology into their offerings.”
At Trimble, OEM customers “combine Trimble’s GNSS technology with their domain expertise to deliver solutions to the end customer,” Norse said.
For Hemisphere GNSS, OEM clients can range “from a tinker/maker hobbyist working with GNSS, to a large multinational organization designing navigation solutions for global clients,” Burnell said, but the company looks at all of them “in the same light.” Additionally, “Some OEM clients have all the tools they need already built into the Hemisphere products, while others come to us looking for advanced or custom features to help set their products apart in the market.”
Septentrio has a worldwide support team that assists its OEM clients “in all the stages of their integration process, from validation to product release,” Freulon said.
What distinguishes your latest generation of OEM receiver boards from previous ones?
Septentrio’s most recent OEM receiver boards integrate the latest Septentrio GNSS and INS technology and algorithms. AsteRx-m3 OEM receiver boards use all GNSS constellations, can track all available satellites, and can be used as a base station to deliver RTK corrections or as a rover with a single or dual antenna.
Improvements include lower power consumption, increased security with secure boot, and greater resilience with anti-jamming and anti-spoofing. Its new receiver boards, Freulon said, “are backward compatible with extended capabilities of the latest GNSS signals and several variants of the inertial navigation system.” Upcoming software releases will include Galileo’s free High Accuracy Service (HAS) as well as OSNMA, the latest anti-spoofing mechanism.
Trimble’s latest generation of OEM GNSS boards are based on Trimble Maxwell 7 technology, which features the company’s seventh-generation baseband GNSS ASIC (application-specific integrated circuit). Trimble designed the Maxwell family of products to maximize the quality of observables derived from available signals transmitted from all GNSS constellations as well as satellite-based augmentation systems, Norse explained. This results in stronger signals, greater availability, reduced power consumption, advanced multipath mitigation and protection against spoofing.
The boards also run Trimble’s ProPoint positioning engine, which improves performance in challenging environments such as tunnels, urban canyons and tree canopies and provides continuous RTK using a base station or Trimble RTX correction services delivered via cellular or satellite connections.
JAVAD GNSS’ latest OEM products are “more cost effective” and integrate an IMU with an 874-channel multi-GNSS band module with up to 200Hz positioning and data output. “All are still proudly made in the United States,” Hunter said.
NovAtel’s OEM7 receiver boards feature added options for interference robustness and situational awareness “to help protect the user’s GNSS signals from an increasingly crowded RF spectrum and growing jamming and spoofing threats,” Gerein said. The company enhanced the sensor fusion capabilities with SPAN GNSS+INS technology, enabling a deeply coupled integration with IMUs that strengthens positioning through GNSS interruptions and allows the rapid reacquisition of signals post-outages. The boards are compatible with PPP TerraStar Correction Services “for precise positioning anywhere in the world.”
Hemisphere GNSS’ Phantom and Vega series of OEM board products can track all L-band GNSS signals, enabling the company’s OEM clients to upgrade the capabilities of their integrations and “tap into the performance of multi-GNSS, multi-frequency RTK and Atlas PPP solutions,” Burnell said.
The boards consume less power than the previous generation and introduce Hemisphere’s Cygnus automatic interference mitigation technology, which monitors the GNSS signal bands for interference and automatically deploys filters “with no need for integrators or users to understand signal theory,” Burnell explained. Cygnus, which turns off the filters when the interference fades away, is “automatic interference mitigation for the masses.”
What are your markets for your GNSS OEM receiver boards? Which ones are growing the most?
NovAtel said its receiver cards are highly configurable and integrate easily across a wide range of markets, including survey, mobile mapping, agriculture, defense, marine and autonomous platforms for both on- and off-road applications.
In particular, the company’s OEM7 cards “uniquely support the defense market and their requirements for increased protection against jamming and spoofing in mission-critical applications.” The cards also “meet the positioning availability and increasingly rigid product quality standards required in agriculture, automotive and autonomous system markets.”
Trimble lists precision agriculture, construction, mining, forestry, autonomous vehicles, port automation, distribution centers and mobile mapping among the uses of its GNSS OEM receiver boards. “We are seeing growth in markets where reliable, robust and high-precision positioning is required for a solution such as autonomous platforms,” Norse said.
Septentrio reports growing demand for its mosaic GNSS modules “due to their small footprint and impressive performance.” OEM boards, Freulon said, “remain very popular for applications where a quick integration is needed or where ultimate performance is expected.”
However, the most important markets for its OEM boards remain “UAV, together with industrial-grade automations in agriculture, construction or logistics.”
Septentrio sees an increase in “the number of positioning and mapping systems that require the ultimate performance of our receivers, especially when combined with other sensors,” Freulon said. In particular, he cites the performance of its single- and dual-antenna AsteRx-m3 receiver boards and of the AsteRx3i INS boards, which “provide a solution which combines industrial-grade IMU and GNSS all on a single OEM board, greatly simplifying the integration process in systems where both positioning and orientation are needed.”
Hemisphere GNSS, which has a significant OEM presence in the agriculture, marine, survey and GIS markets, reports seeing growth in several markets. “We have seen significant growth in all aspects of autonomous integrations, from ground vehicles for on-road or off-road, to in-flight applications with UAVs, to maritime applications focusing on dynamic positioning in both nearshore and offshore environments,” Burnell said. “There is a recognition that using precision navigation equipment benefits everyone and protects our environment through efficiencies of operation, either in resource management or by improved operational capacity.”
JAVAD GNSS lists maritime positioning and docking, timing, launch vehicle positioning and range safety, autonomous vehicle testing, in orbit positioning and drone guidance among the markets for its OEM receiver boards, with space-related applications the fastest growing market.
OEM7700. (Photo: Hemisphere GNSS)
Briefly describe one of your GNSS OEM receiver boards.
The OEM7700 receiver card from NovAtel is used in agricultural auto-steering applications. “The OEM7700 can receive all GNSS constellations across all frequencies, enabling a highly available position,” Gerein said. “When combined with TerraStar corrections and our SPAN GNSS+INS technology for sensor fusion applications, the OEM7700 ensures highly precise positioning scalable from meter- to centimeter-level accuracy.”
OEM7700 receiver boards help the company’s agriculture customers “solve the positioning challenge of repeatable pass-to-pass accuracy for auto-steering,” Gerein said. Plus, the card meets their strict environmental requirements for agriculture vehicles.
Photo: iXblue
Septentrio’s OEM client iXblue uses the company’s AsteRx OEM boards inside its Atlans A7 positioning and orientation system. “Atlans A7 was developed in close cooperation with Septentrio and is designed to provide continuous and accurate positioning in urban environments,” Freulon said.
Atlans A7 combines iXblue’s inertial navigation system (INS), which is based on a fiber-optic gyroscope (FOG), with Septentrio’s multi-frequency GNSS receiver technologies. To develop this INS-GNSS mobile mapping solution, experts from iXblue and Septentrio worked closely with the aim to develop a smart coupling method that combines the advantages of the two companies’ technologies. The same smart coupling technique is also applied in the post-processing software for an optimal result. The main advantage of Atlans A7 is to maintain a high heading precision in any circumstance, which “allows precise georeferencing for both land and air applications and drastically limits the drift during GNSS outages,” Freulon said.
AX940. (Photo: Trimble)
At Trimble, Norse cites the case of an agribusiness company that wanted to make its robotic tractors able to drive autonomously, requiring centimeter-level positioning and orientation at high update rates in challenging environments. The company chose the Trimble AX940i because of its “combination of GNSS and inertial technology in an easy-to-install smart antenna.” The Trimble ProPoint engine tightly couples the onboard IMU sensor data with the GNSS observations to provide up to 100-Hz outputs utilizing the NMEA-2000 standard or other interfaces. Additionally, Trimble VRS Now service provides instant access to RTK corrections and an operator can use the built-in Wi-Fi to configure and monitor the receiver from nearby.
The HydroBoard II flotation platform contains the RiverSurveyor M9 acoustic device, which measures the flow rates of rivers, streams and irrigation canals. (Image: Hemisphere GNSS)
Hemisphere GNSS’ Phantom 34 RTK receiver and antenna is employed by SonTek in its RiverSurveyor M9 product used by water districts and the U.S. Geological Survey to help monitor and manage water resources. The M9 is one in a series of SonTek products focused on determining flow rates for rivers, streams and irrigation canals. It consists of a small flotation platform with an acoustic doppler current profiler that measures the flow rate of the water column underneath it, a data telemetry system, and the Phantom 34 RTK to pinpoint the data collected.
The platform is floated from shore to shore across a channel using a tether, measuring along the way. “Using RTK simplifies collecting measurements as the survey will have continuous velocity profile measurements the entire way across the waterway,” Burnell said.
The mosaic-X5 and mosaic-H modules are being integrated into ArduSimple’s new evaluation kits, making resilient cm-level positioning easily accessible for testing and prototyping
With these new kits, ArduSimple brings to market triple-band real-time kinematic (RTK) GPS/GNSS as a plug-and-play solution for the most popular development platforms such as Arduino, STM Nucleo, Raspberry Pi, Ardupilot and Nvidia Jetson.
ArduSimple enables developers of robotics, UAVs and autonomous systems to easily try out mosaic, a unique module offering the latest high-performance GNSS positioning technology.
“The mosaic module complements the ArduSimple RTK product portfolio with a higher-end solution for the most demanding applications,” said Marc Castillo, senior consultant at ArduSimple. “Triple-band GNSS brings extra reliability to the RTK solution and removes the headache of transitioning from L2 to L5 band. This, combined with its feature-rich software, will allow our customers to accelerate even more their time-to-market.”
In addition to triple-band GNSS, mosaic module offers unmatched resilience to radio interference. This is especially important in robotic devices where electronic components, such as cameras and servos, are located close to the GPS/GNSS receiver, often interfering with GPS signals, which are weak, and causing positioning degradation. High-accuracy positioning is delivered at a uniquely high update-rate by mosaic-X5 in single antenna mode. Meanwhile, the board which mounts mosaic-H offers all-in functionality with dual-antenna mode for accurate GNSS heading.
“By partnering with ArduSimple we are bringing mosaic to emerging markets where its outstanding performance makes a difference. Mosaic makes accurate positioning so much easier to integrate and use, while giving a competitive edge to new products,” said Gustavo Lopez, market access manager at Septentrio. “ArduSimple is a great partner because they are known in the industry for offering user-friendly and affordable evaluation kits for RTK positioning, complemented by software tools, making integration and rapid prototyping easy.”
The SimpleRTK3B board, which allows evaluation of the mosaic GNSS module, is now available for purchase via the ArduSimple web shop. For more information about mosaic or other Septentrio products visit septentrio.com or contact Septentrio.
Septentrio now offers Qinertia post-processing software from SBG Systems on AsteRx-i3 D Pro+, AsteRx-i3 S Pro+ and AsteRx SBi3 Pro+ receivers
Septentrio will now offer post-processing solutions for its GNSS/INS (inertial navigation system) receivers, using SBG Systems’ Qinertia software.
The AsteRx-i3 Pro+ receivers are fully compatible with Qinertia post-processing software, so no data manipulation is required before the post-processing step.
Land or aerial mapping applications, which do not have access to real-time GNSS corrections, benefit from post-processing software for higher positioning and orientation (heading, pitch and roll) accuracy. With the addition of post-processing, Septentrio GNSS/INS products cover the full mapping workflow.
“As a result of our cooperation with SBG Systems, Septentrio’s mapping customers who use GNSS/INS are benefiting from a quicker and more reliable workflow,” said Danilo Sabbatini, product manager at Septentrio. “The intuitive user interface of Qinertia software makes it easy for users to further improve their positioning and orientation accuracy in the post-processing step.”
In case of GNSS outage or correction link failure, post-processing recovers accuracy for recorded positioning and inertial data.
After the mission, Qinertia gives access to real-time kinematic (RTK) corrections from more than 8,000 base stations to deliver centimeter level accuracy. Trajectory and orientation are greatly improved by post processing GNSS and IMU data forward and backward. The Qinertia GNSS/INS post-processed kinematic (PPK) solution provides accuracy, reliability, advanced quality-control indicators, and a modern application programming interface (API).
Qinertia recently added an image geotagging feature, and specific outputs dedicated to photogrammetry.
Qinertia post-processing software will be used on Septentrio receivers. (Photo: SBG Systems)
Mowi is an open-source reference design for Septentrio’s highly accurate GNSS module mosaic. It offers Wi-Fi and Bluetooth communication, which can easily be programmed for custom applications.
Septentrio, a manufacturer of high-precision GNSS positioning solutions, has added to its open-source resources for GPS/GNSS module receivers with mosaic wireless, which it calls mowi.
Mowi combines the Septentrio mosaic-X5 or mosaic-H module receiver with a dual-mode Bluetooth and integrated Wi-Fi from the well-known ESP32-WROVER programmable module by Espressif Systems. It is an addition to the already existing mosaicHAT board, designed on the Raspberry Pi platform.
“We are excited about the mowi project being part of the GitHub and prototyping community,” said Gustavo Lopez, market access manager at Septentrio. “The project is available as open-source, thus empowering the community to easily fit autonomous or robotic systems with communication and highly accurate and reliable GNSS positioning technology. Mowi empowers the native Ethernet features of the mosaic module, the perfect tool for fast prototyping and developing proof-of-concept projects in a simple and connected way.”
The mowi project facilitates accurate and reliable GNSS positioning for robotic and autonomous devices, on a hardware level. Numerous engineers today use the ESP32 and the multiple libraries available for internet-of-things (IoT) prototyping. The mowi board is an easy way for integrators to get started with Septentrio’s mosaic-X5 or mosaic-H heading module receivers.
The mowi board can be used on its own or plugged into a mobile computer such as Raspberry Pi or Arduino to deliver high-accuracy positioning with high update rates, suitable for machine navigation, monitoring or control. The internet connection via Wi-Fi or Bluetooth enables numerous industrial IoT applications, simplifying the connectivity to mobile data for the delivery of GNSS corrections needed for centimeter-level RTK positioning.
On top of the wireless communication, the 47.5 x 70 mm board can host IoT applications in its internal memory. It has onboard logging and exposes interfaces such as USB, serial communication and general-purpose pins. The schematic’s reference design, PCB layout and documentation are openly available for prototyping or further customization.
Septentrio, a leader in high-precision GNSS positioning solutions, has launched the AsteRx SB3 receiver family, enclosed in an IP68 housing. The receiver offers superior availability of RTK high-accuracy positioning because of its ability to track a wide variety of signals from all currently operating GNSS (GPS, GLONASS, Galileo and BeiDou).
Even in dual-antenna mode, AsteRx SB3 uses triple-frequency tracking to maximize robustness and availability of its heading angles.
“The AsteRx SB3 brings state-of-the-art GNSS positioning and heading performance in a very compact and rugged enclosure that is fully certified and ready to use,” said Silviu Taujan, product manager at Septentrio. “Its simplicity and ease of use make it a truly plug-and-play device, allowing customers to have a fully operational system within minutes.”
The new line includes two types of receivers, both offering unique triple-band sub-degree GNSS heading.
AsteRx SB3 Pro is a high-performance rover receiver featuring the latest core GNSS+ algorithms for maximal reliability and availability in challenging environments, such as near high structures or under foliage.
AsteRx SB3 Pro+ adds value to the Pro version with base station functionality and internal logging. It also offers higher update rates and ultra-low latency, both important factors for fast-moving vehicles or mechanical components in automation or guidance systems.
AsteRx SB3 products are pin-to-pin compatible with Septentrio’s AsteRx SB ProDirect receiver and with the recently released AsteRx SBi3 GNSS/INS system, making it simple to change receivers.
The Septentrio AsteRx SB3 Pro is ruggedized for tough environments. (Image: Septentrio)
A roundup of recent products in the GNSS and inertial positioning industry from the July 2021 issue of GPS World magazine.
OEM
GNSS/INS Sensors
Photo: Septentrio
Five receivers in new product line
The AsteRx-i3 product family provides an array of next-generation receivers, from plug-and-play navigation solutions to feature-rich receivers with raw measurement access. Included are OEM boards and ruggedized receivers enclosed in a waterproof IP68 housing. The Pro receivers offer high accuracy positioning with 3D orientation and dead-reckoning for plug-and-play integrations. Pro+ receivers provide integrated positioning and orientation along with raw measurements, in single- or dual-antenna configurations, suitable for applications with sensor fusion. One of the receivers offers an off-board inertial measurement unit (IMU), which can be mounted exactly at the alignment point of interest.
The RES 720 GNSS dual-frequency embedded timing module provides next-generation networks with 5-nanosecond accuracy. It uses L1 and L5 GNSS signals to provide superior protection to jamming and spoofing, mitigates multipath in harsh environments, and adds security features to make it suitable for resilient networks. At 19 x 19 millimeters, the RES 720 is suitable for 5G Open Radio Access Networks (RAN)/XHaul, smart grids, data centers, industrial automation and satellite communication networks, as well as calibration services and perimeter monitoring applications.
The new HG1125 and HG1126 IMUs are low-cost inertial measurement units that serve both commercial and military applications. They use sensors based on micro-electromechanical systems (MEMS) technology to precisely measure motion. They can survive shocks up to 40,000 G-force. The HG1125 and HG1126 can be used for a variety of defense and commercial applications, such as tactical military needs, drilling, unmanned aerial vehicles or navigation systems for general aviation aircraft.
The SDI170 quartz MEMS tactical-grade IMU is designed as a form-, fit- and function-compatible replacement for the HG1700-AG58 ring-laser gyroscope (RLG) IMU, but with superior overall performance, versatility and a significantly higher mean time between failures (MTBF) rating over ruggedized environments. The SDI170 IMU delivers highly linear accelerometer performance and longer life compared to the HG1700 IMU.
The OSA 5405-MB is a compact outdoor precision time protocol (PTP) grandmaster clock with a multi-band GNSS receiver and an integrated antenna. It ensures timing accuracy by eliminating the impact of ionospheric delay variation, enabling communication service providers and enterprises to deliver the nanosecond precision needed for 5G fronthaul and other time-sensitive applications. A multi-constellation GNSS receiver and antenna enable the OSA 5405-MB to meet PRTC-B accuracy requirements (+/–40 nanoseconds) even in challenging conditions. It receives GNSS signals in two frequency bands, using the differences between them to calculate and compensate for ionospheric delay variation. The OSA 5405-MB is resilient against jamming and spoofing, considered critical for 5G synchronization. It can work with up to four GNSS constellations concurrently (GPS, Galileo, GLONASS and BeiDou).
The Toughbook S1 is a fully rugged 7-inch Android tablet for capturing and accessing critical information in the field. GPS and LTE come optionally. The tablet is supported by Productivity+, a comprehensive Android ecosystem that enables customers to develop, deploy and sustain the Android OS environment in the enterprise. The Toughbook S1 tablet’s compact, rugged and lightweight body is made for portability and reliability for field workers. It has a 14-hour battery life and warm swap battery. Features include a sleek outdoor-readable anti-reflection screen, patented rain mode and multi-touch performance whether using a stylus, finger or glove.
The AGS-2 and AGM-1 are manual guidance and autosteering receivers. Location data enable crop optimization, including soil preparation, seeding, crop care and harvesting. Designed to suit virtually any agricultural machine type, make and model, the AGS-2 receiver and steering controller combines steering with network reception and tracking. It comes standard with DGNSS correction services and is upgradeable using NTRIP and an optional RTK radio in the Topcon CL-55 cloud connectivity device. The AGM-1 is offered as an economical entry-level manual-guidance receiver.
Provides fast data processing for construction and surveying
The Trimble T100 high-performance tablet is suitable for both experienced and novice users. It is optimized for Trimble Siteworks Software and supporting office applications such as Trimble Business Center. Its accessories are designed to complement user workflows, enabling users to complete quality assurance and quality control before leaving the field. The tablet is designed to be flexible and usable in a variety of configurations and job sites. It is engineered to be ergonomic and portable on and off the pole. Features include a 10-inch (25.4 cm) sunlight-readable touchscreen display, a directional keypad with programmable function keys and a 92-Wh internal battery.
Surfer has new gridding, contouring and surface mapping software to make it easier for users to visualize, display and analyze complex 3D data. Surfer enables users to model data sets, apply an array of advanced analytics tools, and graphically communicate the results. The scientific modeling package is used in oil and gas exploration, environmental consulting, mining, engineering and geospatial projects. Enhancements have been made to 3D Base Maps, Contour Volume/Area Calculation, 3D PDF Exporting Options and automated features for creation of scripts and workflows.
A Catalyst-AWS collaboration delivers actionable geoscience analytics to users, providing satellite-based Earth observation intelligence. The data and analytics are delivered via Amazon Web Services (AWS) Cloud. Catalyst is a PCI Geomatics brand. The initial solution, available through AWS Data Exchange, is an infrastructure risk-assessment service that uses satellite data to continuously monitor millimeter-level ground displacement over a subscriber’s area of interest anywhere on Earth. Catalyst is exploring additional risk mitigation solutions and monitoring services using AWS. Having image processing science and imagery on the cloud reduces latency and costly data transfers.
The GPS-aided INS-U is a fully integrated attitude and heading reference system (AHRS), IMU and air-data computer high-performance strapdown system that determines position, navigation and timing information for any device on which it is mounted. The INS-U uses a single antenna, multi-constellation u-blox GNSS receiver. With access to GPS, GLONASS, Galileo, QZSS and BeiDou, the INS-U can be used in a variety of GPS-enabled environments and is protected against spoofing and jamming. The INS-U has two barometers, a miniature gyro-compensated fluxgate compass, and tri-axis temperature-calibrated advanced MEMS accelerometers and gyroscopes. These high-performance sensors, along with Inertial Labs’ new on-board sensor-fusion filter, and state-of-the-art guidance and navigation algorithms, provide accurate position, velocity and orientation of the device under measurement.
The Reach M+ and Reach M2 positioning modules for UAV mapping provide centimeter-level accuracy in real-time kinematic (RTK) and post-processed kinematic (PPK) modes, enabling precise UAV mapping with fewer ground control points. The Reach M+ single-band receiver has a baseline up to 20 kilometers in PPK. The Reach M2 is a multi-band receiver with a baseline up to 100 kilometers in PPK. Reach connects directly to a camera’s hot-shoe port and is synced with the shutter. Time and coordinates of each photo are logged with a resolution of less than a microsecond. Reach captures flash sync pulses with sub-microsecond resolution and stores them in a raw data RINEX log in the internal memory. This method allows ground control points to be used only to check accuracy.
System simplifies drone inspection missions, deliveries
The Dronehub is an automated solution that can provide undisturbed drone service 24/7 in nearly any weather condition. With the integration of IBM artificial intelligence technology, Dronehub solutions can operate and automatically provide information with little human interaction. The system includes the drone and a docking station with automatic battery replacement. It can fly for 45 minutes in +/–45° C weather, up to 35 km in winds up to 15 m/s. It can carry a payload up to 5 kg as far as 15 km. It can be used for monitoring, inspection and measurement; cargo transport and parcel deliveries; and mobile ground infrastructure; and security.
The Propeller Platform and WingtraOne drone package enable construction professionals to collect survey-grade data across the worksite consistently and accurately. To operate, surveyors place Propeller AeroPoints (smart ground-control points) on their worksite, and then fly the WingtraOne drone to collect worksite survey data. Survey images are uploaded to Propeller’s cloud-based platform, where the fully automated geotagging and photogrammetry processing is completed within 24 hours of submission on the platform. Uses include mines, road and railway projects, highways and industrial complexes. Data collection with AeroPoints and Propeller PPK can serve as a reliable, single source for survey data and progress. Teams across the entire worksite can view geographically accurate and realistic 3D site models and track, inspect and report on job progress and productivity safely and accurately.
Stamp-sized module provides RTK positioning and heading
The PX1122R is a high-performance multi-band quad-GNSS real-time kinematic (RTK) receiver featuring 1 cm + 1 ppm position accuracy and under 10-second RTK convergence. It has a 12 x 16 millimeter form factor, about the size of a postage stamp. It can be configured for a base or a rover, and supports a moving base RTK for precision heading applications. The PX1122R has a maximum quad-GNSS RTK update rate of 10 Hz, providing a quick response time and more stable performance for fast-moving precision guidance applications.
Utilizing both L1 and L5 GPS frequencies, along with multi-constellation support (GPS, Galileo, GLONASS and BeiDou), the MSC 10 marine satellite compass provides precise positioning and heading accuracy within 2 degrees. Its 10-Hz position update rate delivers detailed tracking information. It eliminates magnetic interference, which can degrade heading accuracy. The MSC 10 is easy to install and can be used as the primary position and heading sensor across multiple systems, including autopilots. If the satellite signal is lost, it will transition from GPS-based to a backup magnetometer-based heading.
AsteRx SBi3 enables highly accurate and reliable 3D mapping solutions based on sensor fusion with lidar
The AsteRx-SBi3 Pro+ in a rugged housing. (Photo: Septentrio)
Septentrio, a leader in high-precision GNSS/INS positioning solutions, and XenomatiX, provider of true solid-state lidar technology for autonomous applications and road management solutions, are starting a partnership enabling high-quality lidar solutions.
XenomatiX will be using the compact and robust GNSS/INS receiver from Septentrio, AsteRx SBi3 Pro+, to provide analysis of pavement conditions geolocated with millimeter accuracy.
Septentrio’s high-quality GNSS/INS will be a part of Xenomatix’ road lidar XenoTrack, used by road surveyors and road management companies. Septentrio’s AsteRx SBi3, a high-performance RTK GNSS/INS receiver with a dual-antenna setup, ensures centimeter-accurate geotagging of the XenoTrack point-cloud frames for relative and global millimeter accuracy over large distances.
Moreover, the GNSS/inertial measurement unit (IMU) integration algorithm enables dead reckoning — continuous positioning in environments of low satellite visibility where GNSS outages occur.
While traditional road scanning based on laser profilers rely solely on extremely accurate GNSS/inertial navigation system (INS) to stitch consecutive profiles together, the XenoTrack captures a 3D topography of an entire area in a single shot. The XenomatiX sensor-fusion algorithms combine visual SLAM techniques with GNSS, IMU and CAN to obtain a seamless map of the road shape.
Septentrio’s proprietary GNSS+ technology plays a key role in delivering the accuracy and reliability needed for XenoTrack. The company’s AIM+ advanced interference mitigation technology ensures robust positioning even in the presence of jammers, which may be aboard vehicles trying to avoid road tolling. When the sky is obstructed for an extended period, the built-in RAIM+ integrity algorithm serves as an indicator of when it is best to give priority to other sensor inputs to maintain a high-quality solution.
XenomatiX offers a complete mapping system as well as services including data from the XenoTrack sensor, camera and real-time kinematic GNSS/INS receiver with dual antenna in an easy-to-install solution on a standard vehicle.
By expanding its GNSS/INS product family, Septentrio starts offering more application-specific positioning and orientation solutions
Septentrio, a leader in high-precision GNSS positioning solutions, has launched the AsteRx-i3 — a new product line of high-performance GNSS plus inertial navigation system (INS) receivers.
The AsteRx-i3 product family brings to market an array of next-generation receivers from plug-and-play navigation solutions to feature-rich receivers with raw measurement access. OEM boards are available for rapid integration as well as ruggedized receivers enclosed in a waterproof IP68 housing.
The variety of products in the AsteRx-i3 line accommodates the specific needs of applications that require high-accuracy positioning together with 3D orientation, heading, pitch and roll angles.
The AsteRx-i3 Pro+ in a rugged housing. (Photo: Septentrio)
“Using our off-the-shelf GNSS/INS systems allows our customers to focus their efforts on core technology and to reduce their products’ time-to-market,” said Danilo Sabbatini, product manager at Septentrio. “With this new generation of products, we aim to satisfy specific needs of various customers. Instead of releasing a single general-purpose product, we bring several dedicated solutions for fastest and easiest integration into systems that require robotic navigation or sensor fusion.”
AsteRx-i3 Pro+ receivers support either single- or dual-antenna modes. The single-antenna mode is suitable for compact and light-weight configurations. The dual-antenna mode reduces the need for movement during IMU initialization, allowing fully informed navigation from mission start.
The AsteRx-i3 product line includes five new GNSS/INS receivers.
The Pro receivers offer high accuracy positioning with 3D orientation and dead-reckoning functionality for fastest and easiest plug-and-play integrations.
The Pro+ are the most versatile receivers providing integrated positioning and orientation along with raw measurements, in single- or dual-antenna configurations, suitable for applications with sensor fusion. One of the receivers offers an off-board inertial measurement unit (IMU), which can be mounted exactly at the alignment point of interest.
Avalanches can be a danger for skiers as well as for the resort towns that welcome them. For protection, towns erect steel fences to act as barriers along the ski slopes. But before these snow barriers can be built, steep rock faces and cliffs need to be surveyed.
Darnuzer Ingenieure AG, a Swiss-based surveying and mapping company, uses a drone with a built-in high-performance GPS receiver to survey these harsh areas in hours.
“Drones have made mapping workflows faster, safer and more efficient,” said Septentrio’s senior market access manager Gustavo Lopez. “GNSS technology has led to the evolution of post-processing kinematic (PPK) methods, which help make the photogrammetry process efficient and accurate.”
Every year, thousands of tourists visit Davos in the Swiss Alps. To protect the town and the skiers, avalanche barriers were built along the steep slopes. To plan the work along the uneven rock face, a detailed 3D reconstruction of the area was needed, but getting to the survey site would be a rock-climbing feat.
Darnuzer Ingenieure used the WingtraOne fixed-winged drone, which features a top-quality camera and a Septentrio high-performance GNSS receiver. A single surveyor took the drone to the rocky Parsenn slope during the summer season, capturing ground images — without snow — that were needed for the 3D model.
WingtraOne PPK enabled high-precision mapping without the need for ground control points. During the flight, each image was accurately time-stamped and raw GNSS data from the Septentrio AsteRx-m2 receiver was carefully logged.
Even in this mountainous region, where peaks obstruct the sky, the receiver delivered continuous positioning. The GNSS data was processed by the GeoTagZ software library, which used corrections from a nearby base station to generate real-time kinematic (RTK) centimeter-level positioning.
The GeoTagZ software library incorporated Septentrio’s core GNSS algorithms to assure the best positioning performance. Accurate positioning was then synchronized with the images in the next step of the post-processing workflow.
It only took Wingtra a few days to integrate the GeoTagZ library into its WingtraHub software package. Integration of GeoTagZ into WingtraHub simplifies mapping jobs for customers like Darnuzer Ingenieure.
“The beauty of this solution is that the drone benefits from the receiver’s high-quality raw measurements without the need for a real-time corrections link for accuracy,” Lopez said. “The quality of the measurements comes from the technology built into Septentrio’s receivers, which is designed to be resilient to radio frequency interference and multipath. In post-processing, the GeoTagZ software enables the most accurate positioning, thanks to its high-performance RTK engine.”
