The most critical systems of our modern world rely on GPS, from air and road networks to emergency and disaster response, from precision agriculture and power grids to weather forecasting and military defense. This dependency is to become a problem.
“We have an increased threat from foreign adversaries who have demonstrated capabilities to jam, destroy, spoof GPS signals, which is scary,” said Shaun Moore, CEO and co-founder of Tern AI, a startup that wants to offer an alternative to GPS, told TechCrunch. “The economic impact on the United States, if that were to happen, would be devastating.”
Moore, who sold his last company Trueface to Pangiam in 2021, said the goal of Tern AI is to “take the bull’s eye off the back of GPS right now.”
Tern AI came out of stealth in February and just raised a $4.4 million round from Scout Ventures, Shadow Capital, Bravo Victor VC and Veteran Fund. The startup has created a so-called Independently Derived Positioning System (IDPS) that can identify the location of a vehicle or a person without relying on a satellite signal. This means there is no risk of jamming, construction or dead zones preventing an accurate position.
“We can do the work of GPS without having to call into space and say, ‘Where am I?’ Moore said.
The current system works by having GPS receivers in cars or phones receive signals from satellites orbiting the Earth. The signals contain satellite information and the exact time the signals were sent. GPS receivers then use the time it took each signal to travel to calculate the distance to each satellite. By triangulating the distances, the receiver can determine its exact location, which is then displayed in applications such as Google Maps or Waze.
This approach (shared by newer constellations such as Europe’s Galileo) creates a global and standardized positioning technology, but one that is prone to local disturbances. For example, GPS spoofing has been used to steer merchant ships in US Navy shipping lanes and some Shipping companies experienced jamming of signals that completely stop operations. While it is unlikely that your car’s GPS system will be targeted by agents of adversary states, there are many military and civilian assets that are at risk and would welcome a more secure solution.
Tern AI’s IDPS is powered by an AI model that ingests real-time vehicle and phone sensor data, as well as third-party map data, to provide a consistent and accurate location. This model can be downloaded as a software package to the infotainment system of a compatible vehicle. If the vehicle is not compatible, the system can be downloaded to a smartphone. Tern’s model can use phone sensor data to help determine a location, but in some cases, the smartphone will need to pull car sensor data from the vehicle’s diagnostic port.
Moore said the goal is for leading mapping products such as Apple Maps, Google Maps or Waze to derive location information from Tern’s AI model rather than GPS. The startup plans to approach these vendors in the future to negotiate direct integration with them.
Tern AI’s biggest selling point is that it uses artificial intelligence to analyze sensor data, such as speed and motion sensors, that are already available in smartphones and vehicles. Moore noted that Tern does not rely on computer vision and cameras to help it determine its position.
Using existing sensors is what prompted Diana Furchtgott-Roth, former DOT deputy assistant secretary for Research and Technology, to call Tern AI technology a “low-cost response to a problem that the [DOT] has worked for decades.”
Other alternatives to GPS satellites, such as low-Earth-orbiting satellites and ground-based beacons, while impressive, are very expensive to install and maintain.
“GPS technology has not changed substantially in 50 years and what we see presented as solutions to solve or mitigate the risk are only marginal improvements. It’s not true innovation,” Moore said.
He noted that while there have been improvements in GPS, the fundamental nature of the technology has not changed, nor have there been sufficient advances to protect GPS from becoming a target.
“One of our founding principles was ‘Unmarked’, meaning we’re not going to look outside to help us determine where we are. Whatever you have at your disposal, you can do now if you add intelligence to it.”
Tern has primarily validated its technology through internal testing against the US Department of Transportation framework. Moore says Tern’s IDPS can place vehicles within 4 meters of accuracy, which he says is in line with current DOT guidelines for emergency services.
The startup is set to begin testing in July with its go-to-market partner, a logistics, shipping and delivery giant that Moore could not name. Commercial release of Tern is scheduled for September.
The startup is also talking to emergency services and autonomous last-mile delivery companies, industries that “rely on good location information to complete their project or complete a job.” Going forward, Moore said he also expects the Department of Defense and DOT to be partners one day.
“When we first met Tern AI, what stood out the most was how diverse and scalable their approach was to solving a critical national security problem,” said Stephen DiBartolomeo, director of Scout Ventures. “Once we understood the practical commercial applications and compatibility with on-road autonomy, their vision to redefine the position resonated with us.”
Correction: This article has been updated to clarify that Tern AI has not yet reached out to mapping companies to incorporate its technology.
