A fallback companion computer to empower UAV navigation in airspace with compromised GNSS connection. The enemy’s electronic warfare won’t restrain UAV missions, and operators won’t lose valuable equipment – all without heavy investment.
A critical military edge in scenarios when failure is not an option.
The module is similar to standard GNSS add-ons you’ve seen – it’s designed to provide location information. The difference that makes it unique – it relies on captured visual information, not unstable satellite signals, which are oftentimes jammed or spoofed on battlefields.
Abto Software is your go-to partner to unlock new prospects in operating UAVs & FPV drones in combat.
The module will keep UAV systems on mission even when GNSS connection is jammed/spoofed, or unavailable. A weak satellite signal is not an issue – ML and VO algorithms will help to lock exact positioning with precision, thus providing reliable backup.
A smart decisive edge when facing electronic warfare – UAV systems that don’t lose their way when it matters.
The module is built to integrate with different UAV platforms and offer an upgrade without spending a fortune. An installation as simple as plugging in standard GNSS modules without buying custom airframes or retrofits – a reliable, battle-ready solution at scale.
Minimize investment, maximize compatibility & resilience.
The module is a bottom-facing camera that’s mounted like standard GNSS add-ons and runs on a Jetson SBC. The solution will perform visual calibration by using satellite signals and switch to automatic visual navigation the moment the connection is degraded – all without direct intervention.
The add-on doesn’t require specialized training:
The module instantly switches to autonomous visual navigation as soon as the GNSS connection is weak or lost. If necessary, the operator can manually override commands.
After landing, VO traces, ML records, and logs are available for debrief and analysis.
Visual odometry is used to evaluate UAV movement by locating and matching visual features between frames. It calculates displacement gradually over periods, which makes it reliable for short-range UAV trajectories.
Machine learning is used to adjust predicted locations from time to time and discard accumulated deviations. The frames are compared with preloaded map snippets to find the most similar tiles to the ones recognized.
The submodule will combine VO and ML outputs to narrow the search and distinguish similar-looking objects. This approach solves ambiguity in visually similar areas to maintain a precise UAV location.
By leveraging AI & CV techniques, our autonomous UAV modules are pushing beyond today’s GNSS limitations. Neither spoofing or jamming can’t stop mission success.
Among others, we deliver:
For over two years, we have been partnering with manufacturers of drones and specialized mission computers. We know the value of following NDAA and STANAG requirements and understand MSP and MAVLink protocols.
What’s more, we build:
We stack 20+ specialists who have proven knowledge and experience in both UAV excellence and engineering. Our goal – to secure mission success by automating dangerous tasks.
In particular, we master: