Battle robots in modern and future wars: myths and reality

One of the directions of development of the armed forces of the countries of the world is associated with robotics, automation of warfare. The main task is to reduce risks for military personnel. To reduce these risks on the battlefield, instead of people, it is supposed to send robots to the battle. In this regard, the battle with the participation of "terminators" no longer seems to be something incredible, because in different countries, including Russia, their versions of combat robots have recently been presented.

The prototype of combat robots can be considered exoskeletons - devices that reduce the overall load on the fighter. Further development - a full transition to robotization, when the machine must ensure the conduct of hostilities on its own.

It is quite likely that in a few years, in one or another theater of military operations, it will be possible to see the use of real "terminators" - combat robots capable of not just firing from a certain species weapons, but also analyze the situation, make a decision, exchange data with the command center.

At the same time, it must be said that the hypothetical use of combat robots in modern wars and the wars of the future has already managed to acquire many myths.

The Day channel has released an article discussing the development of robotic systems in the armies of the world on the example of the United States and Russia. The discussion is being held with the participation of military expert Alexander Artamonov:

Comments

[Grazdanin]

For the use of ground drones in combined arms combat, the time has not come. Technology has not yet grown. Only for special tasks mine clearance, counter-terrorism operations, emergency.

[ycuce234-san]

Yes, in general, they have grown, but this is not particularly necessary, so all this is being implemented slowly.
The main question in competing countries on the planet is how to make not just an autonomous robot, but a very reliable robot that can be trusted with strategic weapons and thereby raise the strategic missile forces of these countries to a new level.
There is such an old, but not at all outdated book - "Space weapons: the security dilemma", which concerns the "Star Wars" programs of the 80s and when reading it it is clearly visible that then, all space weapons, such as X-ray nuclear lasers, could do it back then, in the 80s, literally with a slight movement of the hand.
They couldn’t do one thing - to make both complex and reliable, and therefore safe, automation, which can be trusted with such weapons placed in space for a shorter period, say, 10-50 years, until the end of the warranty period of the automatic space station ... Such an automaton had no room for error during this entire period.
And now they do not know how to make systems as reliable as necessary, otherwise it would be possible to read in the news about missile launches and the placement of such automation in orbits. Therefore, the "security dilemma" was included in the title of the book.

[Grazdanin]

Such automation as you write is very complicated and expensive. The United States will build a station in the orbit of the Moon, stations or small "outposts" in orbit of the Earth, and will launch a large number of X37 orbiting aircraft. Thus, they will be able to place any weapons and equipment on aircraft, they will be mobile, they will be able to refuel in orbit, fly to Earth for maintenance or change weapons / equipment. Thus, they will provide 99,99% missile defense and the deployment of strike weapons that will reach their target anywhere on Earth in a matter of minutes.

[ycuce234-san]

Such automation as you write is very complicated and expensive.

This circumstance was also noted as secondary, and they did not forget about the maintenance and rearmament. But this does not change the main thing - such robots should be extremely reliable and it does not depend in any way on their specific implementation - whether the robot flies in space in an orbital plane or crawls along the lunar regolith with a mobile ground rocket complex on the route or exists in a different form.
And they must be reliable because from orbit the flight time for an orbital warhead is no longer whole minutes, but tens of seconds, a maximum of a minute, and people no longer keep up with such a speed of reaction of machines and they will have to fully automate the conduct of a nuclear war - with all the consequences ...
Therefore, modern shuttle robots can even fly into orbit - no one will entrust them with weapons until an appropriate scientific apparatus is created to control the complexity and performance of complex technical systems. And the shuttles and stations themselves are the same technologies of the 70-80s, which stopped in development due to the barrier of complexity and reliability.
And he also interferes with sending people to Mars, among other things - there is a noticeable risk that they will not reach because of a stale gasket or a computer choked with zero.

[gregor6549]

In connection with your comments, I recall the turbulent era of the introduction of robotics in enterprises in the last decades of the last century. Then robotic systems began to be shoved where necessary and where not. At the same time, the developers of these complexes, because of their blindness, tried in every possible way to imitate a person with his arms and legs, the coordinated management of which was more than difficult. Then it turned out that it was easier to create a machine park with robots built into the machines, rather than trying to pair with robots those machines that were previously controlled by a person. Well, all these complexes should be cost-effective i.e. with mass or large-scale production and where increasing labor productivity by attracting additional human resources is impossible for a number of reasons.
In an army with robots, a similar picture is observed. The word robot has become fashionable, under this word they are trying to knock out a sickly money, but there is little sense from "all sorts of shelesyakov". And it's not about technology at all. They are just enough. (As they would say in Odessa: "You want technologies - so I have them"
The problem is that while the Customer does not have a clear understanding of where, why and how robots should be used and what should be the efficiency / cost indicator for specific types of robots. I do not even mention the information protection of robots from enemy hackers. After all, it can be like Popandopulo's "There are a couple of machine guns, but they shoot at their own people"

[ycuce234-san]

There are plenty of "adequate" applications for robotics and automation in the military field.
Here aerial drones survived in competition with aviation and small artillery, now the issue with small drones remains.
Recognition and television surveillance systems, for example, in trenches, in guarding all kinds of warehouses or access to controlling armored vehicles, serious aircraft or artillery, and even turning on the ignition only in the face of the carrier or automatically changing the password in the truck. Control of soldiers (including automatic analysis of them by a psychiatric robot), delivery of medicines from automatic devices prescribed by doctors, delivery and inventory of ammunition and ammunition, control at checkpoints, etc. - computer cameras do not sleep from fatigue at the post and do not hide and do not relax even in the most fierce storm.

[Klingon]

Quote: Grazdanin

For the use of ground drones in combined arms combat, the time has not come. Technology has not yet grown. Only for special tasks mine clearance, counter-terrorism operations, emergency.

I agree, because for now these are just radio-controlled tanks with a small range, controlled by a joystick. Here is a wounded soldier on a self-propelled platform from the battlefield to take out, or to bring ammunition, for this algorithms should be enough

[Grazdanin]

Quote: Klingon

take out a self-propelled platform from the battlefield, or bring up ammunition, for this the algorithms should be enough

I don't think it will get stuck anywhere. There are too many parameters to analyze. Only under constant control.

[Thunderbringer]

It can still shoot from a certain position in the desired sector. For everything that moves there.
that is, to advance, take this position, complete this task for the right time, then return to where they are told.
Recognition works for itself, although it does not distinguish between civilians and military personnel with the required accuracy. Yes, this is not necessary. There are places in the war where civilians have nothing to do.

[Klingon]

Quote: Grazdanin

Quote: Klingon

take out a self-propelled platform from the battlefield, or bring up ammunition, for this the algorithms should be enough

I don't think it will get stuck anywhere. There are too many parameters to analyze. Only under constant control.

if the operator gets stuck, it will correct

[Grazdanin]

Here :) there are a lot of problems, first you need to teach how to ride, after that the problems are a carriage and a small cart. You cannot even use it purely on radio control, the communication range in open areas is several kilometers, in the city it is several hundred meters. To increase, you need repeaters, a lot of repeaters.
UAVs in the air and at sea are a different story, in 5 years they will become indispensable in these environments.

[Klingon]

Well, in general, yes, you're right. The robots have not yet matured to a real battle. And we still don't even have unmanned drums as such, neither the unmanned aircraft themselves nor the weapons for them

[Grazdanin]

Quote: Klingon

We don't even have drums without pilots.

In the USSR, they did not have time to do it, but no. 2 key problems: no efficient light engine, no microelectronics. For a UAV, these are the main components. Therefore, they make the C70, the size of which is the same as that of the Su57, the avionics are similar to it. As a demonstrator of technology norms, tests are necessary, but not suitable for a production car.

[Thunderbringer]

Why is it not good?
Very much the other way around. Why make a separate plane at all, if there is already a ready-made, flown structure?

[Grazdanin]

Flying wing and the normal layout of the same design? There is only avionics and a common engine can be. This S70 is more expensive than the Su57; the flying wing has a lot of restrictions. As a long-range reconnaissance bomber can be. But it will not be massive. In the United States, they are in no hurry to accept UAVs with a flying wing. Promising developments are going according to the normal scheme.

[Thunderbringer]

That's what I say. Maybe you shouldn't make a fence?

[Free wind]

I'm sorry, but ground-based drones have been around for a long time. Israel has patrol cars, fully autonomous, fires at the operator's command, used for about 8 years .. USA
for about 20 years they have been using a combat strike robot, "fire ant", can be in an anti-tank version, armed with missiles, and in anti-personnel, machine-gun armament. These are robots, not radio-controlled cars. The ant is used from ambushes, fire opens at the command of the operator. But I think it has long been replaced by more advanced models. Unmanned cars have been used for a long time, Teslas have only wound more than a billion kilometers, for all the time there is one accident, not fatal. Mamzelka on a bike, shot down.

[Grazdanin]

Tesla did not make full autopilot, they promise to finish it soon. It works only on public roads with visible markings in the absence of snow, rain, dense fog. Off-road autopilot is out of the question.
All military developments are of an experimental nature, after real tests they are subject to great criticism from the military. In the next 10-15 years, there will be no mass land drones.

[Thunderbringer]

It is really very expensive and so far unreliable.
Orientation technology based on lidars (as in the same Tesla) is not suitable for combat, and generally adverse weather conditions.
And recognition and orientation based on cameras still requires huge computing resources.

[Shuttle]

Just a whole new genre in VO - an article about the announcement of video transmission. As in the paper program on television during the USSR.

[nikvic46]

As soon as nuclear weapons appeared, it became clear that the war would not be waged for victory, but for destruction. Already now, aerial drones can cause significant damage to the enemy. The "unfinished" robot will not understand the military or civilian. It will be designed to destroy all living things. He will not kill his own. The codes on the form will work. A terrible thing.

[Grazdanin]

Uniform at the parade. In combat conditions, everyone has the same dirty camouflage something. People point blank can not figure out who is who. In Ukraine, there was a case when from 2-3 meters, talking with each other, they could not understand that they were facing an enemy.

[Thunderbringer]

The modern outfit provides a friend or foe responder. But this is also true for the future, like fully autonomous robots.

[Grazdanin]

Yea Yea. What would you kill for a couple of kilometers?

[Nikolaevich I]

Battle robots in modern and future wars: myths and reality... Well, I'm talking about myths! Remember that myth from Syria (?), When, according to it, "robotic platforms" attacked fortified positions of militants in Syria and split them! And they were ruled by the "hand of Moscow"! Yes ... yes ... (!) Straight from Moscow and ruled We will not rassusolate: whose hand is this .... and leg ....! But there was a lot of "excitement"! And the temperature was measured by the thermometers at the "speakers" over 40 ... wherever they were inserted! Actually ... it was noisy and fun! And only paaatoooom (!) Began to think: how is this possible and is it possible .... somehow it calmed down ... but had fun in the beginning great!

[Aleks2000]

But in fact, no matter how we make robots, no matter how we do, we still get tanks.
Only smaller.

All attempts to do something else failed (according to the media). Robot Fedya roam out in space. Robot Andryusha turned out to be a disguised person .....

[Grazdanin]

This is called IBD - imitation of violent activity. The army needs massive drone and reconnaissance UAVs. We need it yesterday. Our industry cannot provide them, the Ministry of Defense is covering up their helplessness.

[Thunderbringer]

Quote: Alex2000

But in fact, no matter how we make robots, no matter how we do, we still get tanks.
Only smaller.

Because tanks are still the most effective weapon on the battlefield.
No anthropomorphic mechanisms will achieve their reliability and effectiveness in the next 50 years due to extreme complexity, as a result of unreliability and high cost.
And also a power source, which is also not there. Only in films about the terminator.

[Aleks2000]

Tse is true.
What we can, we can.

And the dogs in Yusa are already on sale. They also have mini-tanks, but there seem to be fewer varieties ... There are even a couple of ships.

Well, the UAV business is slowly going on for everyone ...

[ANB]

Pattern recognition systems have advanced tremendously. And actually the algorithms for automatic guidance of weapons and opening fire are not very complex.
So it is even surprising that while robots are not yet used in hostilities.

[Grazdanin]

One plane can distinguish from another; there is no soldier from someone else's and will not be able to soon.

Quote: Grazdanin

Uniform at the parade. In combat conditions, everyone has the same dirty camouflage something. People point blank can not figure out who is who. In Ukraine, there was a case when from 2-3 meters, talking with each other, they could not understand that they were facing an enemy.

[ANB]

... and not soon

But the spicy vegetable knows him.
Face recognition by cameras has already been launched. In SQL. ru 6 years ago the tasks of counting the number of people in a room were considered trivial. I don't work with video, so I didn't even really understand the conversation. Voice and word recognition is already working pretty well.
Nobody knows what they write in closed laboratories.

[Grazdanin]

Quote: ANB

Face recognition by cameras has already been launched

I know that I myself have personally implemented these systems in work and developed a similar one. Therefore, I perfectly understand what they can and what they cannot. Here is a photo of a soldier, whose is he?

[Gato]

I know that I personally introduced these systems into my work and developed such

Likewise. The recognition probability clearly leaves much to be desired, even if there is no desire to counteract the system with all sorts of tricks. And the enemy will have such a desire - and a strong one.

[Gato]

Here is a photo of the soldier whose he is?

Judging by the St. George ribbon on the dugout and the method of lacing the ankle boots - ours.

[Grazdanin]

most likely, if the truth is not recently captured dugout is a photo accidentally found, so I myself do not know))

[ANB]

In my opinion, the creation of a security robot with the task of destroying all those who do not have the mark "their" is already technologically quite possible.

[Thunderbringer]

Quote: ANB

Pattern recognition systems have advanced tremendously. And actually the algorithms for automatic guidance of weapons and opening fire are not very complex.
So it is even surprising that while robots are not yet used in hostilities.

It's simple.
In combat conditions, it should be a compact, extremely reliable and unpretentious device. A classic green iron box with connectors that can withstand being hit by a tank and lying in a swamp.
Take an interest in what these recognition systems look like now. This is an unimaginable amount of boards, cables, and the like, mounted on aluminum nozzles. All this periodically fails, and requires the sensitive attention of specialists (for example, absolutely reliable control units such as electric motors still do not exist, they still periodically fail).
It remains either to use it permanently or on stand-alone reliable media. That is, he carefully flew off, returned, and fell into the sensitive hands of specialists.

[Gato]

Even the super-reliable missile warning systems have repeatedly failed and false positives. Giving a decision to open fire at the mercy of the ACS is still dumb. And everyone watched The Terminator and The Matrix.

[Thunderbringer]

A person fails even more often, by the way. It's just that people have developed a collective system for implementing decisions.
Like in an anthill. Observe. They drag, let's say something in the anthill crowd. At the same time, not all of them do it in the right direction, 10% (conventionally) try to drag it in the other direction. But the remaining 90% still hit the target.
Just like our opposition. They pull in the other direction, but as a result, the rest reach the goal anyway.

[Gato]

Pulled the other way

That's why it is the opposition. In an amicable way, the opposition should play the role of car brakes on a mountain road, while the authorities stupidly press the gas and turn the steering wheel in all directions. But this is not about us.
As for the collective decision-making system, this has long been implemented in the so-called. expert and analytical information systems.

[Thunderbringer]

In this case, it is about a battlefield with a large number of participants. They have no opportunity to use "expert analytical information systems".
Only the general order of the commanders and their own head.

[sen]

In some areas, the use of robots is already needed today. This is a fight in tunnels and the use of a "jihadmobile" - a robot. Between December 2015 and November 2016, the Islamic State group carried out 923 suicide attacks, about 70% of terrorist attacks were carried out using vehicles - terrorists exploded together with cars or trucks. 84% of IS attacks (776) were directed against military targets.

[Klingon]

are they representatives of the clumsy and pathos-propaganda TV channel Zvyazda, are they all minus? with a stubborn and always screaming leader who does not know how to distinguish tanks?

[Grazdanin]

The main thing is not to write about EW is bad, it's sacred!

[Thunderbringer]

The main thing is to write anything on Ukrainian sites.

[Indifferent]

I am surprised at the sofa specialists. In the mind - children! One electromagnetic pulse over the battlefield (either nuclear or simple) and all your robots will turn into pieces of iron. Robots can only fight with the Papuans. A lot of money will turn into trash. Smart people know about it, and if they are engaged in development, then to cut the loot!

[Grazdanin]

How many times has amy been applied? What is its radius of action? Is there protection from him?

[Procopius Nesterov]

Few people believed in tanks either. Airplanes generally seemed impossible, and in just 100 years what kind of evolution, from a barely flying ash-tree to jet fighters.
It can be anything of course, and we won’t see armada of robots, but if there is a chance that such a development option is possible, then there is no way to lag behind in this matter.