Military Review

Tank-robot: opportunities and prospects

Recently, the possibility of creating crewless tanks (BET), or, as they are commonly called, tanks-robots. This problem is subject to progress in aviation When creating unmanned aerial vehicles (UAVs), many are interested, but often the focus is on issues that are not relevant to the essence of the problem and without taking into account the capabilities of technical means.

Before discussing the possibility of creating BET, it is necessary to decide what purpose is set for this object, what tasks are assigned to it, methods for solving the set tasks, and technical means ensuring their solution.

The purpose of the development of BET is obvious: to remove a person from a tank to save his life in the process of using this combat vehicle. At the same time, BET must ensure the solution of all tasks assigned to a linear tank without losing the quality of their performance. By definition, a tank is an armored tracked vehicle with powerful weapons designed to be used as the main striking force of the ground forces, which determines the tasks it performs.

In addition to the tank, a number of special purpose vehicles are being developed at its base: reconnaissance, demining, repair and evacuation vehicles and a number of other special vehicles. These objects belong to another class of armored vehicles and require separate consideration.

The tank can be used for its intended purpose in various conditions of combat use, such as: in an offensive on rough terrain or in urban areas, reconnaissance, defense, on the march. In this case, not in all cases the use of BET will be justified, for example, on the march and in defense it is impractical, the use of crew tanks is more efficient here.

When performing the assigned tasks, the tank must move on the ground, search for targets and destroy them using the capabilities of the crew and technical means. Remove the crew from the tank and ensure its control in two ways - to make the tank autonomous or remotely control it.

According to the theory of automatic control, BET can be completely autonomous as an automatic control system (tank-robot) or remotely controlled by the operator as an automated control system (robotic tank). These are two completely different classes of machines. There may also be a hybrid control scheme when the object operates autonomously and, if necessary, the operator can take over control.

Comparison of the BET development with the development of the UAV is incorrect, since the conditions for the use of the UAV in the airspace are much "softer" than the working conditions of the BET on rough terrain, in a rapidly changing environment, among natural disturbances that prevent solving diverse tasks.

At the beginning of the 90's I had to discuss the problems of creating the information and control system of the tank with the developers of the control systems of the space shuttle Buran and the Energia launch vehicle, which they tried to connect to this work. According to their estimates, the solution of tasks by a control system on a tank is not inferior in complexity to rocket and space technology, and in some matters is more difficult.

Tank robot

The BEP in this version must move independently on the ground, overcome obstacles (including suddenly appearing), search for a target, choose the most dangerous, determine the type weaponsaim and shoot.

Only a highly intelligent control system can perform all these operations without human intervention. There is no artificial intelligence, and even more so an “artificial neural network,” as some craftsmen suggest, and is not foreseen in the near future. All this while from a series of fiction.

The maximum that can be realistically realized is the movement of an object under a rigid program on a previously reconnoitered area in order to reconnoiter and identify enemy fire weapons. If necessary, with the possibility of interception control operator. To achieve more at this stage is not yet possible. In this design, the robot tank can not solve the tasks assigned to the line tank.

Robotic tank

The BEP in this version should solve all the tasks assigned to the crew of the tank by issuing commands from remote operators. To do this, the facility must be provided with the possibility of obtaining information and remote control:

- electronic video surveillance system driver;
- nodes and mechanisms that ensure mobility;
- electronic devices for monitoring and searching targets (body, heat, radar);
- automatic loader;
- A system of aiming and firing;
- navigation system.

The BEP must have a crypto-resistant and interference-free channel for transmitting commands from remote operators and a noise-proof video channel for transmitting images from the video surveillance system of the driver’s mechanic and from surveillance devices and target searching.

All these systems on the BEP should be linked to a digital information management system. One may ask: on which of the existing tanks does all this exist? There are no such tanks today, potentially only a tank designed by the Armata, which contains some basic systems, is ready for this.

In this regard, the statement of the UVZ director on the creation of a robotic tank based on the T-72B3 does not hold water, on this tank there is practically nothing to implement such a concept. This is an unsubstantiated statement by the director, not the chief designer, who does not represent what set of issues must be solved for such a tank.

Technical means

The problem of creating BET is not in the tank itself; conceptually and in terms of its layout, it can be unchanged, but in the absence and complexity of creating the necessary systems to ensure its high-quality control. The most problematic of these are the video surveillance system for driving and orienteering, the transmission of control commands and the navigation system.

Video surveillance system

Existing television systems do not provide for the creation of a circular three-dimensional picture of the terrain, they only allow to get a flat picture, which is not enough for orientation in the terrain. This problem has not been solved on any of the objects of armored vehicles.

Closest to her decision on the Israeli tank "Merkava". In the “Iron Vision” system developed for this tank, which receives signals from many video cameras placed around the tank’s perimeter, a three-dimensional picture is created via a computer and displayed on the helmet display of the operator. Without such a surveillance system, the BET concept cannot be implemented.

Control channel

This element of the control system is the most problematic and vulnerable on the part of the enemy. The equipment for transmitting digital information via radio communication channels existing today in the ground forces is not sufficiently effective and cannot guarantee the transfer of control commands in the conditions of opposition from the enemy.

The use of hardware ZAS can provide the necessary cryptographic resistance and eliminate the possibility of the enemy intercepting the control of BEP. It is possible to increase the noise immunity of the channel with the help of special data transmission equipment, but the enemy still has the ability to effectively suppress the communication channel when used on a limited area of ​​the EW system, which is now observed in UAV control systems. To solve this problem, it is necessary to consider the possibility of creating information transmission channels on other physical principles that exclude their suppression.

Navigation system

This element should include two components: global GLONASS / GPS and inertial navigation systems. The global system allows to determine the coordinates of the BEP and, according to certain algorithms, the coordinates of the target, but the determination of the spatial position for low-speed and fixed objects is very problematic. This requires an inertial navigation system using a gyro platform at the facility. The combination of these navigation systems will allow you to accurately determine the coordinates of the BET, its position in space and the direction of motion necessary to ensure the firing. It should be borne in mind that the adversary is able to suppress the global navigation system in some areas.

Base for BET

BET can be developed specially crewless and it will not provide for full crew accommodation or the possibility of retrofitting a linear tank with the necessary systems. The development of a special BT makes it possible to reduce the amount of reserved space and the mass of the tank by eliminating crew members. With this concept, a new unit of armored vehicles appears, the organization of its production and operation, as well as transportation to the place of use is necessary.

More promising is the concept of using the base tank as a base, in which the remote control capability is already incorporated in its standard systems. The tank can be equipped with the necessary systems at the factory or already in the army as needed and used as BET. In addition, it will be difficult for an adversary to determine the place and time of BET use, since linear and crewless tanks will hardly differ in appearance. This concept was laid in the developed tank "Boxer"; it can be laid in the tank "Armata". This concept allows any line tank to be made unmanned.

The BET control machine must also be built on the base of the linear tank chassis, and the operators' workplaces must be equipped with instruments and systems for the jobs of the crew members of the linear tank.

An important question is the number of BET control operators. Operators must perform all the functions of crew members in BET movement control, target search, firing and unit management, that is, there must be three people. It is possible to reduce the number of operators to two people, in this case, the command should be provided by the crew commander, and the quality may be lost when combining the functions of searching for targets and firing at one operator.

As a result, it can be said that BET can be created, only it will not be a tank robot. To do this, there is still no necessary technical means. So far it can be a robotic tank with a remote control that performs the tasks of a linear tank at the command of remote operators.
Photos used:

Subscribe to our Telegram channel, regularly additional information about the special operation in Ukraine, a large amount of information, videos, something that does not fall on the site:

Dear reader, to leave comments on the publication, you must sign in.
  1. Larum
    Larum 22 October 2018 10: 24
    AI slowly appears, and how it will be - so that was where to put it.
    tvk what's next to the terminators
    1. Saxon
      Saxon 22 October 2018 12: 11
      It does not appear ... AI elements have long been used in weapons. Not so wide yet, but ...
      In the USSR, work began in the early 70s. The first products with AI elements were adopted in our country in the early 80s ...
      Were it not for the collapse of the Union, probably the tanks would have been without crews :). The concept of a tower without a crew, in part, has plans for creating, ultimately, a robotic tank.
      1. Larum
        Larum 22 October 2018 13: 56
        This is clear. AI - as a complete computer control.
        Yus already has unmanned warships
        1. 17085
          17085 22 October 2018 16: 13
          Quote: Larum
          This is clear. AI - as a complete computer control.
          Yus already has unmanned warships

          Yeah, with a black curly automatic loader
          1. Larum
            Larum 22 October 2018 16: 52
            And google? can they manage without blackies?
            1. 17085
              17085 22 October 2018 17: 31
              I’ll say without Google that wanting doesn’t mean having googling, show the link, but for now, don’t tell.
              1. Brylevsky
                Brylevsky 22 October 2018 18: 01
                Here you are: we are talking about ships, right? And if you are talking about AZ in a tank, then the Americans studied it, and refused to use it.
                1. 17085
                  17085 23 October 2018 00: 16
                  145 tons, not a ship. anything but a ship.
  2. NEXUS
    NEXUS 22 October 2018 12: 07
    And URANIUM-9 isn’t it a forerunner to such robots? The only question is the autonomy of such a system and the ability to make decisions yourself, which requires AI.
    1. Vita vko
      Vita vko 22 October 2018 15: 37
      Quote: NEXUS
      Is URAN-9 really not the forerunner of such robots?

      But why mention it to the author of the article, Yuri Apukhtin. The purpose of his article is formulated in one sentence.
      The statement of the director of UVZ on the creation of a robotic tank based on the T-72B3 does not hold water, there is practically nothing on this tank to implement such a concept.
      Well, for the "stupid" the author repeats it again.
      This is an unfounded statement by the director.

      And then comes a whole series of arguments completely divorced from reality.
      Although on the same VO there were worthy publications about the lack of Uranium-9 detected in Syria and the very statement of the UVZ management about the need to use a more secure, more powerful and fire-resistant platform. Hence, a completely logical version of the unmanned modernization of the T-72 arose, which in any case will be several times cheaper and more effective than the invention of a "bicycle" for new army robots.
    2. voyaka uh
      voyaka uh 24 October 2018 23: 46
      Uranium-9 is the right idea. True, I would start with small cheap wedges with a machine gun. To develop tactics of use in battles without fear of losing. The effect can be huge.
  3. Peter Tsar
    Peter Tsar 22 October 2018 12: 38
    The future belongs to robotic systems. IMHO, a complete rejection of man on the battlefield is coming. To prepare a soldier and make him fight and gain experience, and besides, it costs a lot of money to not die. But to cook up a "terminator" and train an operator sitting many kilometers from the battlefield, these are prices of a different order. Naturally in mass production. After the failure of one machine, the operator simply switches control to another unit. The person is alive, experience and skills remain.
    1. Dersturm
      Dersturm 22 October 2018 13: 52
      it has not yet flashed a nuclear bomb on the operator))
      1. Peter Tsar
        Peter Tsar 22 October 2018 17: 44
        So far, the kernel will multiply all by zero, yes. But not far off is possible the invention of some kind of prodigy compared to which nuclear weapons are babbling. Or they will learn how to stuff robotics on new physical. principles. Immune to Amy.
    2. gunnerminer
      gunnerminer 22 October 2018 15: 32
      -Imho is coming a complete rejection of -

      At the same time, they do not reduce the VUNC, do not cancel the call. The loyal T-55, T-64 and fighters with a bottle of KS, at best with RPG-7, fall off. .
      1. Peter Tsar
        Peter Tsar 22 October 2018 17: 37
        Well, he is not coming tomorrow or in a year. Powerful plug with technologies, and not only with us. We are waiting for a technological breakthrough equal to the internal combustion engine. In the meantime, will have the old fashion, bayonet and butt.
      2. Peter Tsar
        Peter Tsar 22 October 2018 17: 41
        It is also necessary to consider that there are high-tech enemies, and the war with them is corresponding. And there are bearded men and other rebels. And what to crush with their ballistic missiles and electronic warfare systems. Well, you know.
  4. Brylevsky
    Brylevsky 22 October 2018 14: 02
    Why do we fixate on the fact that it is inside the tank that there must be a "brain" that can "think"? Why is it impossible to make the "brain" outside the tank, and the tank has only weapons of destruction, plus position sensors in space, coupled with the means of displaying the environment and a data transmission and reception terminal? Thus, the "brain" can be placed on remote servers - the performance of many computers is higher than one ... and technologies do not stand still, bid data, for example ...
    1. tank-master
      tank-master 22 October 2018 15: 50
      Read the article carefully, the weakest point is the communication channels for land equipment.
      1. Brylevsky
        Brylevsky 22 October 2018 16: 50
        Thank you for reading. In my opinion, the weakest point is artificial intelligence. I also read it: "
        To solve this problem, it is necessary to consider the possibility of creating channels for transmitting information on other physical principles that exclude their suppression.
        Can't IR laser be used to transmit control commands? Or ultraviolet? I understand - aerosol interference can be delivered, such as "Curtains" ... but the source of interference can be destroyed? Well, or a completely "simple" scenario is possible. Imagine something like this algorithm embedded in a tank computer: "Follow the operation area according to the route. Upon arrival at the position, start searching for targets. All detected targets without a signal from the radio frequency identifier must be destroyed." Of course, a lot of questions arise: how will the tank's computer itself plot a route on the battlefield, in a continuously changing combat situation; by what criterion to determine whether the target is destroyed or a repeated attack is required; these are, of course, not all questions ... but. Do you need an intellectual on the battlefield, or a soldier? How much intelligence does a kitchen cockroach or housefly have? Never mind, they live ... If you drop or send a platoon of such "non-intelligent" tanks to the area of ​​the proposed operation, they will mow down everything living there. Option - all living things without an RFID tag on themselves. Of course, it will not be a warrior, but a butcher, but so what? The main thing is the result. And it is possible to create such a robot now.
  5. gunnerminer
    gunnerminer 22 October 2018 15: 29
    Another chimera.
    1. Brylevsky
      Brylevsky 22 October 2018 17: 10
      Come on, come on. We can't stop progress. As soon as the robot becomes cheaper than an infantryman, completely different wars will begin, this is obvious. It is generally accepted that AI is still "too weak". However, he already beats a person not only in chess, but also in Go. Have you heard anything about quantum computers? They are still too expensive and too big, but this is only for now. Once I studied in a Soviet school, there we, among other things, were taught to count ... on abacus (in stores there were such, large, wooden ...) and slide rule, and computers were then only in research institutes. And what? Not so many years have passed, and computers that once cost fabulous money are now in every family and even my generation cannot imagine life without them. I propose to discuss this topic again (about "another chimera") in 5 years.
  6. Anton Capucin
    Anton Capucin 22 October 2018 19: 02
    Ammunition can be done with EMP in order to disable such equipment. And not with the help of a nuclear base, when all the electronics burn out. Or to make kamikaze robots when they destroy themselves and destroy equipment around them. EW is fine, but very expensive and specialized. Still not a fact that will help. It’s also possible to use drones, it won’t be able to cope with a swarm, all the more expensive, but it’s completely possible to burn everything in the affected area. I am sure they are working on it, we have the technical base and a head on our shoulders. If we consider modern games, they just tell us about it, and the capture of enemy equipment, and ammunition. In this case, you can make sure that the person is not injured.
  7. albert
    albert 22 October 2018 19: 12
    The concept art of the tank developers stole from Red Alert? lol
  8. Horse, people and soul
    Horse, people and soul 22 October 2018 22: 35
    Only a highly intelligent control system can perform all these operations without human intervention. There is no artificial intelligence, and even more so an “artificial neural network,” as some craftsmen suggest, and is not foreseen in the near future. All this while from a series of fiction.

    In fact, this is not so. When talking about BET, then by inertia they think of a conventional tank that will be controlled remotely by the operator or very limited independently. But such a tank is blind, deaf and nerdy like our neighbors at the Baltic Sea.

    The future BET is a massive, cheap real robot without any optional human control, made in several versions for ground use (assault, reconnaissance, mine clearing machines, evacuation for equipment, sanitary evacuation for personnel ...), where its main will be network-centric capabilities.

    A flock of such robots, together with UAVs, see the battlefield wider, react faster, the task is redistributed faster, so faster that a person cannot process such a volume of information with such speed. Man only poses a task, changes it, or cancels execution. And the robots in the pack themselves determine step by step steps among themselves. They are looking for offensive paths, looking for possible cover, looking for enemy firing points and distributing them among themselves for suppression, etc.

    What one robot saw - then all robots know. They can use the weapons of other robots if it is impossible to use their weapons, a closed position, the ammunition has ended, etc.

    For such robots do not need a machining center. Blockchain All robots in the pack are AI.
  9. Horse, people and soul
    Horse, people and soul 22 October 2018 22: 47
    Friends, a huge number of complex technical problems requiring real-time response are already solved today without the use of AI.

    Each of the commentators here understands what AI is, somewhat in its own way.

  10. Photon
    Photon 23 October 2018 01: 06
    "Artificial intelligence and even more so" artificial neural network ", as suggested by some craftsmen, today does not exist and is not expected in the near future"
    "Existing television systems do not provide a circular three-dimensional picture of the area"

    It seems that the author spent several decades in suspended animation, he recently thawed. After suspended animation, the author has not yet managed to show anything, not even a simple camera that can stitch panoramas, track moving objects, highlight faces, eyes, and watch smiles. And about the use of neural networks, in particular Yandex, for him it is absolutely fantastic.
  11. M. Michelson
    M. Michelson 23 October 2018 06: 39
    Well, there is another option. Since cross-country driving is the most difficult, leave only the driver. (Although I somehow will not envy him ...)
  12. Leomobil
    Leomobil 25 October 2018 23: 09
    For people superficially immersed in this subject, a quite typical logic of reasoning. The author initially draws his conclusions on controversial grounds, which in the end lead him to the only possible option (far from unambiguous). Well, for example, consider Without a crew tank (BET) - without a crew tank, this means that the crew is there, only it is not in the tank, but where is it? Hiding in another tank that goes next? Or is there a bus with crews behind a tank regiment, each with a remote control on its chest? And this is the main task at the moment?
    Another statement "a tank must perform all the tasks of a conventional tank" - why is it all from BET? Reconnaissance is more convenient from the air, small arms on a separate platform, serious protection of only ammunition, aiming on commands from outside with a minimum of complex electronics in the tank, only actuators and a feedback channel on the state of units and systems. And this is only the first level of counter questions. If you go deeper, it turns out that the BET consists of no crew modules, and who will control them :)? The crew in Zhooo ... the bus rides from behind and controls the Tank :) (well, if not only visually). It will not be a tank, but a warehouse of sensors and wires, and the control channel will be loaded with the exchange of information on the current state like the MCC at the cosmodrome. And so on the list.
    That is, the initially incorrect assignment of tasks to the "Main Striking Force of the Ground Forces" leads the author away from the Main task to solving a large number of minor ones. It will not work out of a tank (even an Armata) to make a BET like a Katyusha - Yars (from a guards mortar - a ground complex of the Strategic Missile Forces).
    1. Vladimir 5
      Vladimir 5 17 December 2018 00: 10
      Leomobil. You correctly note that the author himself creates the framework for BET and then wanders in them. Modern armament and use is going along the path of integrating all means and forces into the battlefield system with network-centric control, further developing with horizontal "swarm" and other connections at a higher level on the basis of artificial intelligence .... BET was initially determined to play, more as an armored carrier of weapons and fire impact in the system of organizing a battle, and he does not need many of the previous qualities, other means are assigned to perform them in the general network of the management and information field ..... Criticism of Armata, in not understanding the possibility of an easy quick transformation into an unmanned option from the basic crew item T-14 ..., Other necessary components: for communications, control and other things, it is already the task of other departments from the military-industrial complex ... For the creation of autonomous samples with integration into a common system with network-centric control of robotic military equipment. needs coordinated work of many research institutes and enterprises, the Coordinator should be the Ministry of Defense of the Russian Federation with its own research institutes, as a customer, because there is no real competition between monopolists in the Russian Federation, which predetermines stagnation without conditions for a qualitative leap, only modernization and slow lagging development ...
  13. Metlik
    Metlik 17 December 2018 00: 32
    "Artificial intelligence and even more so" artificial neural network ", as suggested by some craftsmen, today is not and is not expected in the near future. All this is still from a series of fiction."
    Unfortunately, this is a reality. Only some will deny the obvious until the robot arrives at their home.
  14. antique
    antique 19 January 2019 21: 55
    I can hardly believe that Russia will have its own full-fledged AI. Microelectronics is in its infancy, there aren’t enough programmers, mathematics has gone somewhere. Who and on what basis will develop? Office of nanowires? The joystick? This seems to me the limit of our technological level. There are no anti-tank missiles, on the principle of firing and forgetting. Can not. Is Tukhachevsky reincarnated?
  15. Tech Priest
    Tech Priest 18 March 2021 16: 57
    Dear author of the article, your statement about the impossibility of creating a robotic tank based on the T-72 is extremely unfounded. It would be nice for you to read at least this little article: “A. V. Kalinin, V. P. Noskov, I. V. Rubtsov The system of auto-driving of the T-72B tank "