One of the more intriguing companies exhibiting at last month’s Farnborough International Airshow (FIA 2024) was Flare Bright: a UK SME based in Buckinghamshire that offers a unique software-enhanced navigation system for unmanned aerial vehicles (UAVs) operating in GPS- and radio-frequency-jammed environments.
In a one-on-one interview with ESD at FIA 2024 on 23 July, Founding CEO Kelvin Hamilton explained how, with a background in subsea autonomous systems, he established Flare Bright in 2015 to take artificial intelligence (AI)-based autonomous systems into the aerospace sector. Hamilton initially set out to make tactical UAVs, but soon found potential buyers were less interested in the company’s UAV design and much more interested in its GPS-free navigation system. Then, in 2020, combining revenues from a previous company spin-off with funding from the UK Research and Innovation body’s Future Flight Challenge and the UK government’s Defence and Security Accelerator (DASA), Flare Bright honed its focus on GPS-free navigation.
Regarding the company’s initial hurdles, Hamilton said Flare Bright firstly had to prove that its system would work on larger, operational UAVs. The company then had to extend the endurance of the technology to operationally significant ranges/durations, with its capabilities currently facilitating accurate navigation for up to one hour.
Flare Bright’s technology essentially provides software-enhanced dead reckoning, enhanced by machine learning, that minimises the error rate of a UAV’s inertial navigation system (INS). At any point where the UAV re-establishes a GPS signal, the INS can be reset to a zero-error baseline.
“It’s software-enhanced dead reckoning,” said Hamilton. “It’s all about taking extra sources of information that other people don’t use and using them for navigation. So we don’t use cameras, but we can work with them. We don’t need communications.”
The technology works in any terrain, including over deserts or the sea where machine vision stops working.
“Almost every drone has a little tiny accelerometer, an inertial navigation system, fitted into it already. It’s really small and cheap, and it doesn’t work very well,” said Hamilton. When a UAV’s machine vision system stops working over featureless terrain, he explained, it falls back to using its INS, “but the error on these grows very quickly, so what we do is reduce the way this field of error grows out so it makes it easier for these systems to work.”
Flare Bright’s software-enhanced navigation system thus allows UAVs to get close enough to their target for their terminal guidance systems to take over.
While UAVs can, of course, employ better sensors, these can be large, heavy and expensive with a large supply chain footprint. The focus for Flare Bright, on the other hand, is in providing what Hamilton called “non-exquisite” systems that can be produced in high volume at low cost while using very little computing power, thus enabling them to be added to a UAV with no performance penalties.
Meanwhile, Flare Bright is now working on a Mk2 GPS-free navigation system that will facilitate strike-grade accuracy over thousands of kilometres, with flight demonstrations set for 2025.
The company is also looking to make inroads into the commercial market, where jammed GPS signals, especially in urban areas, are increasingly likely to cause UAVs to crash.
In terms of bringing in customers from the UAV and antenna manufacturing sectors, Hamilton said, “We’re having all the right conversations to take it over to operational platforms, though we’re not there yet, but I think around Christmas or early next year we will be.”
The company’s initial marketing effort is focused on the United Kingdom and United States, with Hamilton noting, for example, that a US Air Force delegation was among the visitors to the Flare Bright stand at FIA 2024. Once a market is established in the United States, Hamilton said Flare Bright would then look to the Asia-Pacific market and intends to open an office in Southeast Asia in the next few months.
