Ukrainian startup Strategy Force Solutions claim their drone motherships have carried out the first autonomous missions with attack drones in a trials against Russian targets.
“A $10,000 mission replaces what previously required $3-$5 million missile systems,” CTO Andrii (surname withheld), told me.
The StatForce approach has a reusable GOGOL-M mothership delivering two FPV-type attack drones to hit precision targets up to 300 kilometers away. This approach leverages the ability of small drones to have big effects on vulnerable targets such as parked aircraft, air defence sites or infrastructure.
Andrii says they cannot yet release imagery of the attacks. However it is a logical progression of previously seen drone motherships and FPVs with AI-enabled automated targeting. In fact, it might be seen as an operational version of the autonomous long range attack system which the Pentagon’s DIU is currently racing to build, or the CGI swarm mothership shown off by China.
“By pairing them [small FPV type drones] with AI mothership drones, we can guarantee precision strikes,” says Andrii.
Genesis: The Woman on The Tracks
Andrii says the underlying idea occurred to him many years ago. When he was a boy, his family were driving beside a railway when they saw a woman in unform walking down the line, miles from any settlement. Andrii’s father explained the woman was checking the tracks for signs of problems which might cause a derailment. The young Andrii’s first thought was this tedious job should to be done by a robot.
Years later, Andrii was following up this idea with drones for automated infrastructure inspection. A smart drone flies along power lines or oil pipelines and, where it sees a possible problem, drops lower and takes detailed images. On return the drone uploads data of possible issues with exact locations, so the maintainers decide whether they need to send an engineer to take action.
Andrii’s work on infrastructure inspection was interrupted by Russia’s full-scale invasion in 2022. He turned his project on AI drones towards defending Ukraine.
A Smarter Pilot
StratForce’s key product is a SmartPilot system which uses a combination of advanced sensors and AI. The roots in infrastructure inspection are clear. While other developers use basic drone cameras, the requirement for high precision calls for something more sophisticated.
“In some ways it is like a self-driving car,” says Andrii. “In order for the autopilot to work properly they need a lot of cameras. There are not so many obstacles in the air, but we he system as to be lightweight. That’s how we arrived at our system of cameras, LIDAR and communication to allow the AI to navigate, coordinate and maneuver around obstacles.”
LIDAR, effectively laser radar, builds up a 3D map of the surroundings, and works in all lighting and weather conditions. SmartPilot’s other main feature is AI which combines the inputs from multiple sensors.
“Smart Pilot uses a multi-sensor fusion approach, combining data for environment perception and target recognition,” says Andrii.
This world picture enables the AI to make decisions, plot a flight path and carry out the mission just like a human pilot.
“With SmartPilot it was a really serious challenge how to imitate a flight of the drone the way it would be remotely controlled by a pilot,” says Andrii. “How to maneuver around obstacles, with the limitations of the processor and in real time.”
Andrii says SmartPilot can carry out missions autonomously, finding its way to a specified location and engaging targets there.
“It enables autonomous flight, navigation, and engagement without GPS and without constant operator control,” says Andrii.
Attack Missions: Strike, Search, Or Ambush
In a typical application, the GOGOL-M mothership with a 20-foot wingspan flies to the target area and releases two
FPV-type drones. Each of these has its own lightweight version of SmartPilot. The mothership returns, leaving to FPV drones to find, identify and carry out attacks on predefined targets such as air bases, missile launch sites or air defence.
Andrii notes that the system is also effective against vulnerable infrastructure such as oil storage, power distribution and railways.
While a drone like the Shahed with its single warhead can only hit one target, the mothership can hit multiple targets, then return for re-use.
Unlike the Shaheds, SmartPilot is not limited to static targets.
“It supports ambush missions, landing and waiting for targets, and autonomous search in real time,” says Andrii.
The drones can land and wait at an airbase for aircraft to arrive or emerge from their hardened shelters. Or they can be left in the path of a convoy, ready to attack autonomously when it appears, a tactic already widely used.
Andrii says StatForce can currently produce 50 GOLGOL-M mothership drones per month and 400 of the FPV type attack drones, but this will depend on getting contracts from the military.
So far, the feedback from trial missions has been positive.
“It feels like a video game. I just set the waypoints, choose the targets, and watch it work,” says one user quoted by Andrii.
“It’s fascinating to watch from the coordination center. I just wish we could increase the warhead and push the range to 500 kilometres,” says another.
The important element of StratForce’s solution is the software. The drones, both motherships and FPV attackers, can be of any size, shape or configuration. If users want a large, jet-powered mothership like China’s SS-UAV, or bigger, fixed-wing attack drones, or even robot boats or tanks carrying drones, these could be accommodated.
The crucial first step, long-range delivery of multiple small autonomous attack drones, has now been taken. We do not yet know how effective this Version 1.0 is, or its limitations. But it is likely to give Russia some serious problems in the coming months.
