Destroying Russian tanks is just the beginning for US AI drone autopilot.

FPV video shows a Ukrainian kamikaze drone closing in on a Russian armored vehicle. The video feed cuts out, the Russian radio breaks into jamming and the pilot loses control. This would normally end the mission, but video from another drone shows the AI-powered FPV guiding itself straight to the target and taking direct aim. It is one of the first strikes using a new smart autopilot supplied by American company Auterion that prevents jamming and has a higher hit rate than human pilots.

The Skynod S Autopilot was unveiled last week but is already on the frontline in Ukraine. The ability to lock on from long range makes jamming useless. I've seen several combat videos that I can't share for security reasons, and caveats apply, but as far as can be gauged, the system seems effective. As Auterion CEO Dr. Lorenz Meyer told me, automated targeting with FPVs is just the beginning.

Auterion's App Store allows developers to write their own apps for Skynode S as well as download them as needed. Meyer says terminal guidance apps for attack drones are currently the most popular.

“Terminal guidance is like MS Office on a laptop,” says Meyer. “This is the minimum requirement that everyone expects.”

Open Source Drone Intelligence

Autorion has a long lineage in the drone world, having led the open source movement for more than a decade. In 2008 Meier, then at ETH Zurich, set out to build an autonomous drone. The result was Pixhawk 4, known globally simply as PX4, an open-source flight controller that provides drone developers with a flexible set of tools. There are now 10,000 PX4 developers worldwide, including some who use it for robots and other machines.

The Pentagon clearly believes in Auterion's open-source software as they put it at the heart of their BlueSUAS initiative for small US-made military drones. It was implemented as an alternative to Chinese-supplied drones that come standard with existing security risks and are banned by the US military.

Auterion previously developed the Skynode X, a powerful autopilot and mission computer that provides drone autonomy with machine vision, object recognition, visual navigation and a variety of third-party apps. The newly released Skynode S offers similar capabilities but at a much lower price.

“We're offering a 'Ukraine-ed' price point in Ukraine and have licensed the software for free to the Ukrainian government,” Merritt says. dollar.”

This puts the Skynode S in the same class as thermal imaging in terms of significantly increasing the cost of a $500 FPV, but offers much more capability. And costs can decrease as production increases.

“Skynode S … will be produced in thousands, which is unprecedented,” says Meyer.

It's no surprise to learn that Oterian is also involved with the Pentagon's Replicator program to produce large numbers of low-cost AI-powered drones.

100% fatal (so far)

The guidance system provides optical lock-on: the operator identifies the target and flags it for autopilot while the drone is out of jamming range. It can then proceed through the 'jamming bubble' even if the operator loses contact. According to French Army Chief of Staff General Pierre Schull, 75% of drones are currently downed by jamming.

Meyer says his AI terminal guidance system used in Ukraine has so far scored a 100% hit rate, compared to 20%-40% for FPVs controlled by a human. He doesn't expect to maintain an unbroken winning streak but believes he can achieve 90% in the long run.

This means the two biggest causes of misses—pilot error and jamming—can be eliminated.

This means that with a guidance system like the Auterion S, four times as many drones can pass, and then have a higher hit rate thanks to smart guidance. It's a sobering thought that an already high FPV hit rate can be dramatically increased with such a simple modification.

And while the current version guides the drone to the target, future developments will likely feature target selection, where the AI ​​chooses the most dangerous location. For example they could circle to hit the rear of a turret on Russian tanks causing a devastating explosion, or the ammunition stack on a self-propelled gun.

Open source software makes it easy to develop new applications. In June, Auterion's Dronecode nonprofit hosted a 48-hour hackathon for NATO in Krakow, challenging teams of coders to develop a system to visually identify targets and a flight path to intercept them. prepare They used an early version of Skynode S with AuterionOS, similar to the product just released.

“It was a world first,” says Meyer. “Developers were able to focus solely on developing the app.”

But technology is capable of much more than that.

Boosting Bombers and Aiding Interceptors

GPS and other satellite navigation is heavily jammed in both Ukraine and Russia, even affecting some military-grade systems and dropping US-supplied Excalibur guided artillery rounds and JDAM guided bombs, as well as drones. The usual solution is expensive twin navigation modules, but these become inaccurate at long ranges and are expensive.

AI systems are also capable of visual navigation, using the drone's camera to find its way by matching objects on the ground, just like early pilots, using images from previous reconnaissance flights. Some Ukrainian drones already use AI software called Eagle Eyes to do this.

Meyer says the Skynode S could add an extra twist: terminal guidance. Once the drone reaches the target area, the object recognition app can visually acquire the target and hit it with precision. In theory, this means that even a strike from several hundred miles away could find a specific piece of machinery in an oil refinery, strike the center of a radar dish or pass through a specific window. Or just going home to a tank parked far behind the front line as seen by satellite.

The setup has already been tested on the Russian Lancet and the Ukrainian equivalent of the Shahid drone.

“Testing proved it could hit targets accurately,” says Meyer. “The system will hit the frontline in the coming weeks.”

The Skynode S can also help in air-to-air combat. Meyer expects that autonomous 'dogfight' apps will soon be available that will enable an FPV drone to plot an intercept course, avoid and destroy an adversary without any operator input. will outrun the attacker, detonating when he gets within killing distance.

Ukrainian FPV drones recently began intercepting Russian spy drones at high altitude. Skynode S can significantly increase the number and success rate of these engagements, while also ensuring that Ukrainian drones can outrun Russian interceptors.

An AI pilot can help with other tasks as well. For example, some skilled FPV pilots are able to perform dive bombing attacks, which have the precision of a kamikaze attack but without sacrificing the drone. The main limitation seems to be the lack of experienced pilots. But a diving app may soon be just a download away.

Automated drone mining, or using jam-proof drones to drop tire-shredding caltrops on specific roads behind enemy lines, or just resupplying friendly forces can be very easy.

The big picture

Looking at the bigger picture, Meyer says Skynode S is already integrated with Ukrainian battlefield management software such as Delta and Krupiva. These command and control systems gather information from drones, satellites and other sources and combine it into a single picture visible to commanders on the ground. He demonstrates this with a video of an AI-powered spy drone that spots objects and returns data.

Targets are always verified by a human operator, but the AI ​​system has already done most of the work of finding objects of interest from the video feed. Since each drone no longer needs an operator, it will provide more detailed 24/7 surveillance and real-time monitoring of the battlefield. And because it only needs to communicate when it finds something, AI-powered intelligence needs far less bandwidth to send far more useful information.

AI also offers the possibility of adding capabilities such as change detection, automatically comparing a scene to how it was last time to spot new mines, trenches, vehicle tracks or hidden positions. It looked like

Putting AI at the edge can dramatically speed up the process of gathering, processing and disseminating battlefield intelligence. Thousands of smart, autonomous surveillance drones may eventually become more important to fighting the war than terminal guidance.

The name is Skynode, not Skynet.

Skynode S also supports drone swarm control: one operator can control multiple drones. So an operator might have several reconnaissance drones and a stack of FPVs ready to launch, using the scouts to vector the FPVs to find and engage targets.

This makes it sound like the entire kill chain could be automated. It is important here that there is always a human on loop monitoring operations, even if they no longer need to be in the loop.

Meyer emphasized that Skynod S is not about creating autonomous killing machines. He compares the level of automation to the way guided missiles are currently used. The FPV operator is doing exactly the same job as the Javelin missile operator by locking onto a target: both are 'fire-and-from' weapons that first land on targets set by a human.

“I'd have serious concerns about anything more autonomous,” Meyer says. “But there's a lot that can be done without allowing drones to choose targets autonomously.”

It will take some work to fully explore the new capabilities offered by Skynode S. But with software developers, low-cost hardware, and a ready supply of drones to fit in—and a battle to win—things are likely. to be fast.