![Photo: SBG Systems](https://www.gpsworld.com/wp-content/uploads/2021/08/Photo-SBG-Systems-300x249.png)
UNMANNED SOLUTION, a South-Korean company based in Seoul, develops autonomous vehicles, including driverless shuttles, autonomous agricultural equipment, robots, and educational platforms. (Image: SBG Systems)
The four global navigation satellite systems (GNSS), two regional navigation satellite systems and public and private augmentation services continue to provide exceptional levels of accuracy and reliability for positioning, navigation and timing (PNT). Yet their well-known vulnerabilities also continue to fuel the need for alternative/complementary sources of PNT data, especially for new and rapidly expanding user segments such as autonomous vehicles.
What constitutes a complementary service to GNSS for PNT and what constitutes a true alternative is partly a matter of definition and opinion. In a January report, the U.S. Department of Transportation stated
…suitable and mature technologies are available to owners and operators of critical infrastructure to access complementary PNT services as a backup to GPS. To achieve the parallel objective of resilience, as described in Executive Order (EO) 13905, that path should involve a plurality of diverse PNT technologies. Promoting critical infrastructure owner/operator use of those technologies that show strong performance, operational diversity, operational readiness, and cost-effectiveness is worthwhile. Based on this demonstration, those technologies are LF and UHF terrestrial and L-band satellite broadcasts for PNT functions with supporting fiber optic time services to transmitters/control segments. (Andrew Hansen et al., Complementary PNT and GPS Backup Technologies Demonstration Report, prepared for the Office of the Assistant Secretary for Research and Technology, Department of Transportation, January 2021, p. 195.)
A portion of the former USCG Loran Support Unit in Wildwood, New Jersey, with its iconic Loran antenna. (Image: UrsaNav, Inc)
For this year’s Q&A on complementary / alternative PNT, I asked five companies in the GNSS/PNT space to tell us how they define the issue, what solutions they prioritize, what markets they target, and which of their products specifically address the need to make PNT more resilient.
The participants are:
- Roger Hart, Director of Engineering, Spirent Federal Systems
- David Sohn, Solution Architect, Orolia
- Matthieu Noko, Product Manager, SBG Systems
- Jacob Amacker, Product Engineer, OxTS
- Charles Schue, Founder & CEO, UrsaNav
How do you define alternative PNT?
Roger Hart: The deep adoption of the state-sponsored, space-based global navigation satellite systems (GNSS) defines them as the primary PNT source at this time. Inertial navigation, long predating GNSS, does provide an independent navigation solution but does not provide time. In today’s conversation, alternative PNT generally refers to deriving position and timing from existing signals not purposed for navigation, to ground-based location systems, and also to emerging satellite systems that operate at higher power — or out of the GNSS band — to provide a diversity of PNT sources.
David Sohn: Simply put, alternative PNT is usually anything that is not GNSS. So, this includes PNT derived from low Earth orbit (LEO) satellites; vision, radar, lidar combined with inertial measurement units (IMUs) and map matching; positioning off cellular WiFi, digital TV signals and other signals of opportunity; legacy nav aids like VORTAC, ILS, DME and eLoran; and new dedicated infrastructure positioning systems like Locata, NextNav or RFID.
Matthieu Noko: Here at SBG Systems, for 15 years we have been developing navigation systems based on smart coupling of GNSS technology and inertial sensors. From our perspective, inertial sensors as well as sensors such as odometers or DVL, combined with high-end algorithms and RAIM, build consistent alternatives to GNSS-only systems in the vast majority of outdoor applications. Inertial sensors dramatically enhance GNSS-only navigation systems, making it possible to provide navigational data during GNSS outages in urban environments or to reject false GNSS measurements due to multipath effects.
As its entry into the driverless category of the Formula Student Germany car race, AMZ modified the car it has used in competition since 2015 to be driverless. (Image: SBG Systems)
The hybridization of several technologies makes it possible to reach a sufficient reliability level for the majority of commercial applications. However, in some indoor applications or in case of intentional jamming or spoofing, a higher resiliency is required. Using visual odometry can then complement quite effectively the navigation system, although this technology is still at a research level. Compared to infrastructure-based alternative PNTs — such as WiFi, Bluetooth or ultra wideband (UWB) ranging — visual odometry has the great advantage of not requiring any infrastructure.
Jacob Amacker: GNSS remains the dominant method of PNT in terms of market applicability and performance, but there are many rival technologies that have great potential and will compete with GNSS going into the future. The most important changes in PNT will be methods of position localization that are able to replace GNSS, and we will likely see these technologies integrated into huge systems, making use of ubiquitous technologies such as lidar. Whereas GNSS still provides the most accurate timing, as systems get more complex, time synchronization becomes a bigger issue, so different methods of this need to be explored.
There are many ways of improving the navigation data overall. Most commonly an IMU and a Kalman filter will be employed to stabilize any errors in the position localization method. A Kalman filer is a method of processing data from a range of sources—say, GNSS, an IMU, and a wheel speed sensor—and using them in such a way as to arrive at the position with a greater accuracy and precision than either source alone would be able to achieve. This process, however, requires precise timing for each data stream. Therefore, one area in which alternative PNT has to compete with GNSS is timing precision. GNSS makes use of atomic clocks used on satellites that are as accurate as you will get. There are also several ways of synchronizing time. A timing system can only be as precise as the most precise clock on the network, but there have been developments, such as Precision Time Protocol (PTP) that can synchronize timings across a network of clocks over Ethernet connections. Traditionally, PPS has been used and whereas this is still very precise it is not able to compete with PTP on convenience or sophistication.
Charles Schue: The common definition these days for “alternative PNT” seems to be with respect to, or as compared to, GPS or GNSS. Even the U.S. DOT’s website speaks to PNT as related to GPS.
I used alternative, complementary and backup somewhat interchangeably during my entire career with the U.S. Coast Guard. In recent years, I injected “co-primary” into the conversation as well. Prior to GNSS becoming ubiquitous, alternative, complementary and backup were not technology-based terms, but were instead operationally based. For example, “the prudent mariner” or “the prudent aviator” should use all means at their disposal to safely navigate their platform. For the navigator, this would include visual, audible and electronic signals or aids. The solution of choice obviously was the one that provided the highest accuracy, availability, integrity and continuity. However, prudence required always checking the solution of choice against other readily available alternatives, preferably that complemented each other, to ensure safety and continuity of operations. At one time, shipboard navigators might have at their disposal Loran-C, OMEGA, GPS, INS, radar, sextant, visual bearings (such as lights and landmarks), beacons, and soundings. Similar alternatives were available on aircraft.
Although always in the mix, timing was often in the background until around 2000. Then it started to become as important as positioning and, in many areas, even more important than positioning. Today’s incredible dependence on technology, and interdependence between technologies, means that knowing your “when” has become as important as knowing your “where”.
Whatever the terminology, the definition of alternative PNT should include some key features. Firstly, we should accept that the solution of choice today is GNSS, and we should define it as primary or co-primary. Next, we should acknowledge that when the primary solution is available and trustworthy, it should always be used, or at least considered. Finally, the primary solution should continually be compared with alternatives to ensure safe and secure provision of PNT to the user. Thus, an alternative PNT solution is one that is readily available; provides an easy and seamless transition to/from the primary or other alternatives; allows continuity of operation at a possibly degraded, yet usable, level of accuracy, availability, integrity or continuity; and is dissimilar enough from alternatives to withstand the effects that might be affecting the primary solution.
Do you agree with the U.S. DOT’s assessment, cited above, of what it will take to make the national PNT much more resilient and reliable? If you do, how do your offerings fit into that framework?
RH: While there are intricate differences in the signals generated by the primary PNT systems, they are all quite similar in terms of frequency and power and are all vulnerable to the same types of interference. Achieving the most resilient solutions will require the use of alternative RF bands and non-RF sources. Having a variety of alternative PNT sources will allow users to integrate the method most applicable to their platform constraints. Integration across the various PNT sources will need time synchronization to take full advantage of the alternate PNT systems. Our offerings work concurrently with GNSS, providing simulation and testing of GNSS and alternative PNT as true complements, while also offering the ability to measure timing accuracy in real time.
DS: Yes, we agree with the DOT’s assessment. However, to be clear, the DOT does not require “LF and UHF terrestrial and L-band satellite broadcasts for PNT functions with supporting fiber-optic time services to transmitters/control segments.” It stated that to achieve resilience, systems “…should involve a plurality of diverse PNT technologies…that show strong performance, operational diversity, operational readiness and cost-effectiveness.” Their demonstrations showed that those technologies they called out meet these criteria. Our solutions have been leading this resilient approach by offering several diverse, alternative PNT references.
We have fielded time-server equipment that operates from both GNSS and eLoran. Our standard offering time servers are equipped with multiple references from GNSS, network-based time services from NTP, PTP and PTP WR; internal references from disciplined atomic clocks; wireline references from IRIG, 1PPS or ASCII time code; and LEO PNT reference from the STL signal.
(Image: SimonSkafar_E+_Getty Images)
L-band or more generally the use of geostationary satellites was until very recently the only communication link for PNT augmentation services, although these signals are weak and easily disturbed or masked, especially at high latitudes. Resilient navigation will clearly need to allow multiple downlinks for corrections such as terrestrial networks (4G/5G) or satellite-based internet. In the mid-term, we expect the correction delivery over IP to become the standard, and L-band corrections to be used as a backup only. All our high-performance products already include an NTRIP client able to handle the IP corrections very easily.
JA: This is certainly one option. Largely, it is borne out of a need to compensate for the disadvantages of GNSS. This larger range of frequencies would provide a range of satellite-borne signals that have different penetration characteristics and information carrying properties but the same core purpose. Therefore, somebody making use of such a system will be better able to receive these signals even when under obstructions. Of course, some obstructions will still be impenetrable to GNSS signals and there is a long way to go to developing a comprehensive solution that can deal with timing differences when the signals travel through objects. It is likely that some other source of timing information, for example through the proposed fiber-optic services, will be necessary to smooth out these issues. Although we will see this much-needed upgrade to cover the shortfalls of GNSS employed, many other alternatives will start to take prominence. It is difficult to say which solution will win out, and it is likely that an upgraded GNSS will continue to dominate for the next decade or two at least. In terms of our offerings, we are exploring all possibilities and keep our core technology open to any position localization method. Of course, we will welcome any new technology that is a viable and improved method of PNT.
CS: I have long been an advocate of a system-of-systems approach simply because there is no PNT solution available yet that works everywhere, under all conditions, for all users, all the time. Many solutions provide only a component of PNT: an INS provides position (the “PN”), and an atomic clock provides time (the “T”). However, an INS does not know “where” it is without initialization and updates, and an atomic clock does not know “when” it is without initialization and updates. Fiber is awesome but is not wireless. Many alternatives depend upon GPS/GNSS as a necessary input. Others are augmentations that depend upon GPS/GNSS as inputs and not direct alternatives, such as space-based or land-based augmentation systems. Some are mode-dependent — such as VOR, DME, ILS, and TACAN for aviators — and thus not useful to other modes: time/frequency, maritime, land-mobile or handheld.
So, yes, we agree with the government’s assessment that low-frequency (LF), generally referring to eLoran in the United States, is the best, very wide area, terrestrial, wireless alternative, and is an essential component of any resilient PNT framework. Irrespective of whether the implementation is Loran-C, eLoran or LFPhoenix, LF is the lowest cost terrestrial PNT solution per million square miles of coverage. All our offerings are focused on the LF portion of the resilience framework. Our offerings easily integrate with any existing PNT technology and have proven in real-world government testing their ability to survive heavy jamming and spoofing environments.
What markets and applications do you target?
RH: Spirent Federal provides simulation test solutions to U.S. government and affiliated organizations. Applications range from core GNSS receiver development to real-time, hardware-in-the-loop system integrations. We have a long history of supplying the U.S. government and contractors with first-to-market products, from Y-code, SAASM, inertial and M-code, to sensor fusion of the latest alternative signals and sensors. We provide test solutions to safety-critical applications that are expected to have the same level of operational performance both in GNSS-available and GNSS-denied environments. Providing a single test platform that can help validate performance in both environments has received positive responses from users in the autonomous vehicle industry.
DS: Aerospace and defense, data-center and communication networks, public safety, industrial control, search and rescue, and space.
Autonomous self-driving mobility solutions move people and goods at appropriate speeds in urban and campus environments. (Image: SBG Systems)
MN: SBG targets a large range of applications including from a relatively small BVLOS drone for remote operation to large hydrographic vessels or airborne survey. We divide the applications into two main categories:
- Surveying and mapping, where the inertial navigation system is used to stabilize the measurements from a lidar, sonar or camera to generate high-precision maps.
- Control applications, where the PNT and orientation solution is used in real time to feed autopilot or to stabilize a camera. These applications include unmanned vehicles, machine control, camera pointing and more. High resilience is then critical to ensure safe navigation.
JA: Two main applications we are targeting with alternative PNT are surveying and ADAS systems. Both of these applications often make extensive use of lidar systems. We are therefore looking at lidar-based simultaneous localization and mapping (SLAM) algorithms to aid PNT or to provide relative position localization without GNSS. In cases when GNSS is totally unavailable, it is usually possible to set up ground control points. Although these cases are limited, they give much more flexibility in options. Anticipating a future where autonomous driving is the norm and not the exception, new building projects will need to be planned with the adequate systems in place to allow for them, and this will include a system such as UWB.
CS: Our employees have been involved in the design, development, deployment or sustainment of every Loran-C and eLoran system site in the world (transmission, control or monitor) since the mid-1970s, including components of the Russian Chayka system. Our service provider and end-user technologies are operationally proven in commercial and military environments. We specifically target the maintenance and upgrade of existing systems, as well as the implementation of new systems, globally.
Which of your products directly address the need for alternative PNT?
RH: In a broad sense, Spirent offers a market-proven and innovation-driven solution portfolio for the simulation of inertial sensors through the SimINERTIAL and SimSENSOR product lines, seamlessly integrated with our GNSS simulation. Spirent is actively engaged with several alternative RF vendors to incorporate signal simulation capability and will offer an alternative RF navigation product in 2021 called SimAltNav Replay. This product will allow for concurrent GNSS and alternative RF signal simulation. Additionally, Spirent offers many other alternative PNT solutions for testing resilient systems for connected vehicles and sensor-fusion algorithms for tactical and military-grade systems. We are developing new products to incorporate an open Ethernet interface allowing for open-source Ethernet-based sensor simulation.
Remotely controlled rovers are used to test and practice complex tasks in Mars-like desert environments. (Artist’s Rendering: Stocktrek Images_Stocktrek Images_Getty Images)
DS: Our time servers are equipped with high-quality precise internal time references such as OCXOs or atomic clocks and then disciplined by external references such as GNSS. They are resilient because they can operate precisely for long periods in GNSS-denied situations as standalone devices in holdover mode or from multiple alternative references, such as:
- network-based NTP, PTP and PTP WR time services
- wireline references from IRIG, 1PPS or ASCII time code
- LEO PNT reference from the STL signal
- eLoran when available
They are also resilient because they detect and mitigate interference from the GNSS signal before it can corrupt the PNT solution.
Our GNSS simulators are adding alternative PNT features to provide a complete test and evaluation solution for resilient PNT systems. We have recently added INS/IMU test features and have integrated with Anritsu’s cellular test stations to evaluate and qualify combined GNSS/cellular location functions. Orolia GNSS simulators support generation of custom GNSS signals and playback of IQ waveforms, and provide complete toolsets for GNSS jamming and spoofing testing. This allows creation of the threat environment to allow evaluation of alternate PNT signals as backup or alternative to GNSS. Orolia offers an open-source framework allowing any end user to develop their own sensor plug-in leveraging the Skydel simulation engine.
Our Resilient PNT for Defense product line includes the VersaPNT, which uses alternate non-GNSS PNT sensors such as IMUs, barometers, wheel ticks, INS and non-GPS-based LEO satellites. Alternate RF navigation or non-GNSS sources of radio frequency (RF) are of interest in highly degraded or contested signal environments. Interest is focused on low-Earth-orbit (LEO) constellations. These systems offer high receiver signal power (relative to GNSS) and a secure and resilient link to augment GNSS.
MN: All our products are designed to answer to challenging GNSS conditions, starting with our Ellipse series, which includes an industrial-grade IMU capable of coping with short-term GNSS outages. Its miniature size allows integration in robotics and also makes it suitable for cost-sensitive applications. Our Apogee and Horizon series, with their navigation-grade IMUs, are the most resilient systems in the event of GNSS outages. These products reach very high-end performance in real time, but become exceptional when used with our post-processing software Qinertia. Tightly coupled algorithms make the solution capable of coping with long-term GNSS outages.
JA: We have previously created solutions using retroreflective strips for path following with driving robots. and we are also compatible with Locata’s system, a large infrastructure solution popular for automation in shipping ports. More recently, we have released an offering for UWB in an integration with Pozyx. This is perfect for GNSS-denied environments as a direct replacement for what GNSS can provide in terms of position information. We are also exploring alternative ways to synchronize clocks and get timing information. This year we have developed PTP functionality on all of our devices. Alternative PNT is going to be vital as we look to the future of navigation and thinking about how we can navigate flawlessly anywhere and address more complex environments, particularly urban areas.
CS: We are focused on the provision of terrestrial low-frequency equipment and systems for primary, co-primary, alternative, complementary and backup PNT. We provide all the products and services required to design, develop, install, certify, operate and maintain Loran-C, eLoran and LFPhoenix equipment and systems. We provide the technology to perform coverage diagrams and site surveys; all the equipment required at a transmission site; all the equipment required at a differential reference station or quality-of-service site; all the equipment required for a monitor and control site; ASF measurement and analysis equipment; and various models of end-user equipment (including receivers and antennas) for the timing/frequency, maritime, aviation, land-mobile and handheld markets.