Partner ex machina
The THeMIS crawler-type vehicle, affectionately called the Milrem pocket tank, no longer qualifies as breaking news. Now, however, the Estonian company wants to invest a million euros into a new system called Wingman that would allow the machine to move about the battlefield almost by itself.
It was Milrem's idea that got the most attention at the defense ministry's development grants competition last December where experts decided to pledge €200,000 of taxpayer money for the project.
“The reason this project won was mainly the fact that no other such guidance system currently exists for unmanned land vehicles,” said head of the defense investments department of the ministry Kusti Salm. “They have the ambition to make one.”
CEO of Milrem Kuldar Väärsi said that the Wingman, that currently exists only on paper, is a vehicle guidance system that uses sensors to paint itself a picture of the surrounding environment. A vehicle – for example Milrem's own THeMIS – can be ordered to follow a particular soldier or move from one place to another on its own.
“The goal is to turn unmanned land vehicles into battlefield partners for soldiers that can be guided simply and intuitively,” Väärsi explained. “That way the soldier could concentrate primarily on the mission, instead of guiding the machine.”
For that, the system would have to be able to make a number of decisions independently. For example, it would have to go around obstacles and plot an optimal route for itself. Images from the machine's cameras would be transmitted to the soldier in control via an augmented reality headset who would then be able to guide the vehicle using voice commands. More complex and responsible tasks, like opening fire, would always require a human decision.
The goal is not an easy one to achieve. While vehicles that can move autonomously in cities already exist, matters are much more complicated off road. “The most complicated part of the project is creating a deep neural networks-based artificial intelligence. It can be taught using machine learning algorithms that require huge amounts of data,” Väärsi explained.
The development goes beyond technology. Milrem will have to develop new tactics, training, and logistics for units using the system. Cooperation has been launched with the University of Tartu and the Defense Forces that have volunteered to help test the system.
Development of the Wingman system is estimated to cost as much as Milrem has already spent on THeMIS – more than €1 million. While the project deadline has been set for late 2019, Milrem wants to demo its first augmented reality guidance system in September.
“Over the next five years, we will definitely see autonomous land vehicles become a new battlefield capability,” Väärsi said. Estonia is not considering procuring autonomous vehicles at this time. THeMIS has been tested in the last two Spring Storm exercises.
If a camera drone buzzing overhead is just a nuisance at the beach, it constitutes a clear security threat concerning national defense objects. A startup based in Peetri village near Tallinn called Marduk Technologies is planning to build an affordable anti-drone system to blind or destroy the imaging systems of drones from a kilometer away.
The Ministry of Defense makes no secret that the idea of Marduk's engineers was the most exciting one at the competition. Drones have become a daily occurrence on national borders. Reports of intelligence drones hovering over nuclear power stations or military bases come almost every month. Drones have monitored Defense Forces trainings and are used for military activity in Ukraine.
Right now it is possible to scramble drones' GPS/GLONASS signals, throw nets, spray foam, or train hawks to catch them. The first laser and electromagnetic weapons have been developed, while they cost a fortune.
“Drones are sent up to collect something. If we can counter their purpose, the ability to record or transmit video, it is something the Defense Forces could use,” Kusti Salm said. The ministry supported Marduk Technologies with €70,000.
“The need for something like that came up at all manner of conferences in the past year: border control, military, civil defense. It is a new threat that has already become reality and is perceived to only get worse,” executive manager at Marduk Indrek Seppo said. He added that a team of Estonians spent months thinking of ways to solve the problem.
Their idea is to create a platform of robots that could accurately fix a laser beam on a drone and keep it there. Provided the signal is strong enough, ordinary photographic sensors wouldn't last long. Stronger lasers would ruin sensors for good.
The system would be able to target and hit all contemporary commercial drones and cameras from a kilometer away, before they even reach filming distance. “A laser beam scatters to an extent a kilometer from the source. If a camera's aperture is a few cm in diameter, a laser beam is at least ten cm wide a kilometer away,” Seppo explained.
Marduk Technologies plans to make the system available at an affordable price. That way it could interest a wide range of security agencies all over the world, as well as individuals.
Radars that can pick up drone activity are already on the market. The Estonians now have to develop a robot that can keep the beam fixed on the target. Once ready, the system will be able to protect entire areas from prying eyes in the sky, Seppo claims.
He said that Marduk's team is made up of a physics doctor, robotics engineers, and software experts with experience in building these kinds of systems from scratch. The platform to be called Marduk Shark should be operational by the middle of 2018 and will cost €200,000 to develop.
What if enemy communications could be picked up unnoticed over their heads? That capacity currently requires large unmanned aerial vehicles, like NATO's RQ-4 Global Hawk. However, such UAVs cost hundreds of millions of euros.
Viimsi-based Threod Systems wants to use new technology to make signal intelligence sensors so small they could be installed on smaller UAVs weighing up to 30 kilograms.
One such is Threod's own 2.2-meter Stream, meant for brigade-size units. “Aerial vehicles weighing up to 30 kg usually operate inside a range of 100 kilometers and at an altitude of 1-2 kilometers,” head of the company Villiko Nurmoja said.
Right now such vehicles can carry cameras and infrared sensors, in other words engage mostly in visual intelligence. If the machines could be equipped with different kinds of sensors, they could enter the world of signal intelligence or SIGINT. Because there is demand for the product, among others in the Estonian Defense Forces, Theored was given a grant of €66,815 at a recent defense ministry competition.
Nurmoja makes no secret that the entire subject is highly sensitive. Picking up enemy communications on the battlefield is currently possible mainly using ground equipment. The problem is that various objects, landscape, and signal noise disrupt detection close to the ground.
“There is less noise at higher altitudes. Further down you have all other radio communication, allied radio signals, electrical devices etc,” Nurmoja listed. Less noise makes it possible to pick up signals from further away and target enemy command centers, networks, and critical infrastructure without risking lives.
Another advantage of UAVs is the ability to pinpoint the location of enemy communications devices. “Triangulation is easy using aerial vehicles because they are on the move. They constantly measure distance between themselves and the signal source,” Nurmoja described. The location of the radio station would then be forwarded to the brigade commander to help them decide on a suitable course of action.
Even though the matter is hushed, similar efforts are being made by a few other sensor manufacturers in Europe and Russia. “´Weight is the biggest challenge as sensors could only weigh up to three kilograms. However, looking at recent innovations, I believe we can manage that,” Nurmoja suggested.
Threod plans to finish its device that would not be tied exclusively to the company's own drones and could be installed on other UAVs in 2018. “If someone manages to do this thing and have it weigh less than three kilograms, it will spark great interest,” Nurmoja is convinced. The first signal intelligence sensor prototype should take to the skies over Estonia for testing in October-November of this year.
One instead of eight
The Estonian saying “it's the size of a tank battery” remains not far from the truth even these days: military vehicles still require immense starting current – especially in Estonian weather.
“In cold weather, eight batteries is considered the absolute minimum. That amounts to a weight of 200 kilograms,” said project manager at Skeleton Technologies OÜ Egert Valmra. It can be no other way: temperatures below –18 degrees Celsius reduce the capacity of normal batteries threefold.
Skeleton is looking to develop the world's first supercapacitor-based military vehicles starting module that would stand in for eight standard batteries. Supercapacitors have other advantages over batteries: they are less sensitive to cold, will not catch fire or explode when hit.
The idea was born in April of last year when a military vehicle manufacturer approached the company. “The problem is created when large engines need to be started in cold temperatures in a situation where batteries are also used for other functions: for example when they are also responsible for heating and communications,” Valmra explained.
While the technology is aimed at tanks, the problem also plagues the Estonian Defense Forces: Estonia has over 80 SISU armored personnel carriers and will soon take delivery of more than 40 CV90 IFVs.
Large military vehicles do not use ordinary batteries but a special NATO 6T format instead. It includes stricter requirements for vibration, humidity etc. “We have not had to meet standards as strict as these before. Especially in situations where the vehicle might have taken a hit,” Valmra said.
Skeleton's plan is not to change the supercapacitor itself but to build a new shell around it. For example, it would be possible to use foam that turns vibrations into heat for isolation.
Because a NATO-format starting module could become a hot item on the world market, Skeleton merited a grant of €63,919 at a recent defense ministry competition. The first product to have undergone testing should be ready by the end of the year. The module is being developed in cooperation with the Defense Forces.
“As far as I know, it would be a world first. There have been 6T-format lithium ion batteries; however, their disadvantage is their high price,” Valmra said.
Skeleton Technologies' new starting module will probably cost more than the company's existing commercial module that currently goes for €1,390.
Valmra said that the NATO starting module would have an operational lifespan of 15 years or more than five battery generations. This means that by the time of the first maintenance, two or three years of use, the module will have saved the client the price of 12 batteries, or more than €3,000.