Drone War Is Coming

[stromboli]

http://www.wired.co.uk/news/archive/2016-01/08/drone-swarms-change-warfare

The summer of 2016 will mark a significant moment in the future of air warfare: swarms of US Navy robots will leave the laboratory and be tested in the field. In the past, small numbers of extremely expensive manned aircraft were the norm, but, in the next few years, large numbers of cheap, expendable drones will be deployed in real-world situations. The US Navy's Low-Cost UAV Swarming Technology (LOCUST) programme is fusing unmanned aircraft into a swarm, and will demonstrate 30 drones flying together somewhere over the ocean.

The challenge won't be just getting them in the air at the same time -- what makes a group into a swarm is its ability to co-ordinate.

"A swarm operates as a unit," says Stephen Crampton, CEO of Swarm Systems, an autonomous-systems startup based in Hertfordshire. "It has a mission that it has to carry out, and it is self-reconfiguring so that if one drone gets taken out, the others autonomously change their behaviour to complete the mission."

Any aircraft, manned or unmanned, can be brought down by a single missile, but a swarm can take multiple hits and keep going.

Being able to absorb this type of damage makes the drone swarm hard to stop, especially when defences are geared towards shooting down single planes. Designing a system in which drones are aware of each others' movements is a big breakthrough.

"It is the communication and/or awareness of each other that represents the major leap forward from remotely controlled and independent platforms to those that allow for collaborative behaviour," explains Lee Mastroianni, project manager of LOCUST. The programme has been working with the Georgia Tech Research Institute to develop a system in which individual drones will position themselves autonomously, flying in formation without being told explicitly where to go.

"Precise formation control of large numbers of UAVs while conducting various manoeuvres is a major objective," says Mastroianni. Managing the swarm requires a new approach to control: instead of remotely piloting a single drone, the operator manages the swarm. He describes how the operator's interface will handle "aggregation" and "disaggregation", his terms for drones joining or leaving the swarm. A single drone might detach to get a closer look at a target, and return or carry out an attack.

The main hardware for LOCUST is the Coyote drone, a metre-long unmanned aircraft produced by defence manufacturer Raytheon. Coyote was designed as an expendable reconnaissance asset with folding wings, so that it can be fired from the tubes used for dropping sonar buoys on anti-submarine aircraft. Once clear of its launch tube, the Coyote's wings flick out and it can fly for up to 90 minutes on battery power, beaming back video from 30km away.

But the hardware is less important than the sensors and software enabling the drones to act in a swarm. "I am largely UAV agnostic," Mastroianni says. "The key is a modular UAV that can easily acceptdifferent payloads depending on which missions are desired and can be produced cheaply enough that they are one-way."

Adaptability is important because different payloads are required for different types of mission: the drones may be equipped with video cameras or other sensors, jammers to interfere with enemy radar or they might carry explosive warheads for kamikaze-style attacks. This is why "low cost" is part of the name -- the drones must be cheap enough to be expendable. "Reaching the point where an entire swarm of UAVs costs less than a single missile is the general objective," Mastroianni says. The goal seems realistic enough when you consider that the US Navy's Harpoon anti-ship missiles cost around $1.2 million (£770,000) each.

Offensively, the swarm can overwhelm adversaries: anti-aircraft batteries simply don't have enough missiles to stop them. And although the Coyote drones may be too small to sink a ship on their own, they could potentially knock out radar, missile launchers or other key systems, leaving them vulnerable to other attacks. In defensive mode, a swarm can form a protective cordon against fleets of fast boats like those used by Iran's Revolutionary Guard. The swarm might carry out high-risk reconnaissance missions, collecting imagery or other data from targets too well-defended for a Predator drone or a manned aircraft to approach.

Mastroianni says the biggest challenges for the swarm are not technical, but more based on perception: safety policies treat unmanned aircraft as if they are manned, meaning that they are highly regulated. And there is the added complication that the drones are largely autonomous rather than being individually remotely piloted, so they are not under human control at all times, like current drones, which makes policymakers cautious.

"Establishing trust in autonomous UAV systems is not only the biggest challenge, but a major objective," Mastroianni says. Swarms at sea are a start, but the real impact will be when they engage in land warfare. Stephen Crampton, CEO of Swarm Systems, says the cluttered environment where drones have to avoid trees, buildings and power lines is far more difficult than open water. Autonomous sense-and-avoid for small drones is still in its early stages, but as processors get more powerful, it is becoming more reliable. Crampton says that other advances such as deep learning and neural networks also offer potential solutions and the technology is advancing rapidly.

Swarm Systems demonstrated a swarm of quadrotors at a Grand Challenge event for the Ministry of Defence in 2008. But the first step into the market is an individually-piloted drone called Nano -- Crampton compares it to "flying binoculars". "When you've got a squad that is pinned down, they need flying binoculars, which they can send over the hill or around the corner," he says

British soldiers already have Black Hornet, a palm-sized helicopter with cameras, but it is useless in strong winds; Crampton claims the Nano will be able to fly 99 per cent of the time. The company's background makes the Nano a first step towards tactical swarms of many drones working together. "You can see soldiers having a peripheral reconnaissance swarm," Crampton says.

Battery life is a big issue for small drones. But a swarm can have a "hive", a base station where individual drones return for recharging while the rest continue their mission. To the operator, unaware of charging going on in the background, the swarm's endurance is unlimited. This approach is relatively easy for fixed bases; Crampton says a mobile hive for soldiers on patrol is more challenging. A walking robot such as Boston Dynamics' BigDog, however, offers a good starting point.

Next year will be a breakthrough year for drone swarms -- once trust in swarms is established, development is likely to be rapid. Then Crampton's "peripheral reconnaissance swarm" will change the face of the battlefield -- especially when the drones are armed

.

Bought my grandson Noah a drone 2 years ago. He can fly that thing so accurately that he can park it over your head and follow you around with it. I've seen demos on the internet of drone swarms doing complicated maneuvers. They come in multiple sizes and I know they have been used tactically by the U.S. in Afghanistan.

A coordinated drone swarm could overwhelm any type of defense, if not with weaponry in sheer numbers. Flying mines directed at a group of men could kill at considerable distance with pinpoint accuracy. Nasty shit.

[Munch]

Did anyone else see the movie Toys with robin williams? Makes me think of the plot from that where the evil uncle wants to use children to play video games thinking its a toy when its really drones.

[Hydra009]

https://www.youtube.com/watch?v=bkNV2nWUVYw

[Shiranu]

The downside to entering a new era... the ways we find to kill each other come along for the ride.

[Valigarmander]

It's okay, we're safe. They'll only be used to kill brown people in poor countries.

[stromboli]

It's okay, we're safe. They'll only be used to kill brown people in poor countries.

The bad news is that because of commercial applications already forthcoming, drones are available to a wide body of consumers. It would be a lot easier to weaponize a commercial drone than just about any other readily available delivery system I can think of. Something like a couple of kilos of high explosive is well within the capability of a drone, the bigger ones have long range and can deliver with pinpoint accuracy. You could stand off a considerable distance and fly drones at the White House, for instance.