I'm directing a small team of highly skilled, intelligent operatives to raid a top secret research facility. Unfortunately, our reconnaissance has revealed that not only is this secure facility protected by competent and well trained guards, they also seem to have an advanced illusion or holographic projection system. We're unsure exactly how this works but from observation we know it can:

  • Generate convincing moving 3d images of people on-the-fly

  • Disguise real objects or people by overlaying a 3d image on top of what's really there (ie a door can be made to appear as part of a wall)

  • Generate 3d-localized sounds, simple special effects, and voice-changes on-the-fly

  • It can't make things invisible to more than one concurrent viewpoint

As a savvy commander, I'm worried about the implications that this could have on our raid including but not limited to:

  • Enemy guards could disguise themselves as one of my operators on-the-fly and thus gain a surprise advantage

  • Guards could use the limited invisibility of the system to hide from a single person at a time

  • The holographic system or its controller could disguise one of my operatives as an enemy and cause us to accidentally shot them

  • Enemies could hide inside holograms and shoot or use concealed positions to flank us

My team consists of ~8 skilled operatives who are all equipped with cutting edge, but mundane technology. Suppressed rifles with subsonic rounds, advanced body armor, night vision goggles; the works. I don't have a lot of prep time so developing some new technology isn't really feasible but equipping all my operatives with ten-foot-poles or bags of ball bearings would be possible (I don't know how useful it would be though).

Question: What strategies or tactics can I employ to give my operation the highest chance of success and minimize commando casualties?

Some ideas I've already had:

  • Some sort of password system? All of my agents are capable of memorizing a 20-word passphrase before the operation and are good at math. Some kind of back-and-forth identify verification call-and-response could be used

  • Work in teams of at least two operatives to make invisibility illusions ineffective

  • Physically verify wall locations by probing with hands if walls are solid

  • $\begingroup$ In all that text about savvy ball bearings, what exactly is your question? $\endgroup$ – puppetsock Dec 17 '19 at 15:42
  • $\begingroup$ @puppetsockreinstateMonica Edited $\endgroup$ – Dragongeek Dec 17 '19 at 16:04
  • 1
    $\begingroup$ Worth checking how this "illusion system" works under specific conditions: smoke, fog, rain, etc. While human are used to such condition, it may reduce effectiveness of such nice technology... $\endgroup$ – Uriel Dec 17 '19 at 22:22
  • 2
    $\begingroup$ @Uriel Or coloured strobes: If you rapidly flashed red, green and blue light at a wall, would a portion that is really a disguised door be able to keep up, or would there be a "lag" where it didn't display the correct colour? $\endgroup$ – Chronocidal Dec 18 '19 at 9:17
  • 3
    $\begingroup$ Is there anything stopping them from constantly projecting a solid black box around your heads? $\endgroup$ – Omegastick Dec 19 '19 at 5:16

11 Answers 11



dogs and navy seals https://www.navysealmuseum.org/raven-nsm-multipurpose-canine

If the problem is the limitation of human senses, dogs are the answer. Dogs have gone to war since there were dogs, and commandos and dogs work well together. Illusions designed to trick humans will be obvious fakes to the dogs. They will smell right through disguises of any sort. The commandos will have to trust their dogs. I think they already do.

  • 20
    $\begingroup$ Doggos! Of course the answer is doggos! The goodest boys! Such an elegant and simple solution. +1 from me $\endgroup$ – rath Dec 18 '19 at 12:05
  • 10
    $\begingroup$ Is it just me or can any problem be solved with dogs? Bad at math? Dogs. Karen took the kids? More dogs! $\endgroup$ – Finn Dec 18 '19 at 13:46
  • 1
    $\begingroup$ I'm not sure an improved sense of smell would be sufficient. They're still animals and can be fooled via sight even if it's just to frighten or distract them. Their sense of smell (plus training) may not be enough to over-ride those behaviors even with training. $\endgroup$ – Ouroborus Dec 19 '19 at 7:55
  • 1
    $\begingroup$ @Ouroborus, there's a specific distinction in some breeds of dog, sight hounds and scent hounds. In this case you definitely want a scent hound. $\endgroup$ – Separatrix Dec 19 '19 at 11:57
  • 1
    $\begingroup$ @Ouroborus Humans get the vast majority of sensory input information visually. Dogs don't. They get large chunks from smell, hearing and vision. If different senses give conflicting information, humans will usually trust their vision. Dogs will be confused and decide depending on the specific situation. For example they tend to identify individuals (dog or human) primarily by smell. So if something looks like John but doesn't smell like him, it's probably not John. $\endgroup$ – quarague Dec 20 '19 at 14:13

Your password system won't work on a flanking enemy. They will kill you why you are asking the password.

IR visors will tell you where there are objects warmer or colder than the background. Those happen to be humans.

Make sure that your team wears a clearly identifiable mark visible in the IR, so that if they see sergeant John Doe without the mark they can kill him without thinking too much.

Also real doors and walls will have a different IR signature than air, so you won't be fooled into stepping into the void thinking it is a passage.

  • 7
    $\begingroup$ Ditto this, also if your operatives are spotted use smoke bombs or something similar. If no one can see your enemy loses the advantage of their cloaking system. $\endgroup$ – Dugan Dec 17 '19 at 15:48
  • 1
    $\begingroup$ "Make sure that your team wears a clearly identifiable mark visible in the IR, so that if they see sergeant John Doe without the mark they can kill him without thinking too much." but that takes time trying to identify each person they meet. John Doe could just be partially obscured, or it might be an enemy. From a distance you might not even be able to tell if it's supposed to look like John Doe or not, it's just a shape. $\endgroup$ – VLAZ Dec 18 '19 at 14:48
  • $\begingroup$ @VLAZ, it can be just a warm/cold ring around wrists and knees... something easy to spot $\endgroup$ – L.Dutch - Reinstate Monica Dec 18 '19 at 14:49
  • $\begingroup$ Additional useful tech - laser rangefinders can help pierce illusory scenery (and perhaps a handheld sonar "fish-finder" can be hacked to work in air to recognize hidden objects/enemies?) $\endgroup$ – G0BLiN Dec 18 '19 at 15:44
  • 2
    $\begingroup$ The IR idea depends on the electromagnetic spectrum the illusions can operate in. UV, lidar, and radar might be useless as well for the same reason. Sonar uses a completely different media, so might be a useful alternative. $\endgroup$ – Ouroborus Dec 19 '19 at 8:03

Inventing a new technology might be impossible, but I didn't see any restriction about purchasing existing technology. That being the case:

Laser Imaging / LIDAR

There currently exist handheld 3D laser-imaging devices which can be used continuously while in motion. If each member of the team (or even just a few) is equipped with such a device, they'd immediately be able to see through any static illusion. If the devices are networked, they could ostensibly combine their readings to form a more complete picture of surroundings.

A small RF transponder could provide reliable IFF between teammates. If the signals are encrypted, this would prevent the enemy from quickly spoofing them during the relatively brief period that the raid is occurring.

For a lower-tech solution:

Underslung Paintball Guns

Mount a relatively high-capacity paintball gun under their rifles, where a grenade launcher would typically be mounted on an M4. They can use this to test weather a surface is solid or illusory, without having to use separate gear, or move close enough to physically touch it.

This will also allow safe IFF checks, as an enemy disguised as a friend will need to have an illusion bigger than their actual body (the illusions cannot make them invisible, so they must "hide" within them, or they'll poke through this disguise); this means that a paintball striking a disguised soldier will pass through the illusion before impacting the person beneath.

For a quicker, but somewhat low-tech solution to IFF:

IR Pulses

Take apart readily-available remote controls to scavenge their infrared LEDs. These LEDs emit pulses that are invisible to the naked eye, but are readily apparent to IR sensors like those used on nightvision goggles. If the LEDs are mounted around each soldier's helmet, they can allow the team-members to quickly identify one another at a glance.

  • 2
    $\begingroup$ Regarding the paintball solution - OP states that the hologram generator is smart and fast enough to disguise people on-the-fly (e.g. disguise a friendly as an enemy - including localized voice change and motion matching), it seems reasonable that it will be able to incorporate illusory paint on the holograms and hide the real paint on surfaces that shouldn't have been painted were the holograms real... $\endgroup$ – G0BLiN Dec 18 '19 at 17:43
  • 1
    $\begingroup$ b.t.w. kudos on the 3D laser-imaging devices - that's an awesome idea! $\endgroup$ – G0BLiN Dec 18 '19 at 17:43
  • $\begingroup$ +1 for the IR pulses. I was going to post this as well. Illusions wouldn't be able to simulate the generation of an IR or other signal. The other thought I had was a sort of IR reflection patch so your IR transmitter could reflect off their gear but not the illusion's. $\endgroup$ – Tracy Cramer Dec 18 '19 at 18:14
  • 1
    $\begingroup$ @G0BLiN an easy way to deal with that is to use opaque paintballs which only you know the color of; the illusion can't know the color before the paintball hits the disguised enemy so either you will see it pass through or you will see it display the wrong color. $\endgroup$ – Skyler Dec 18 '19 at 18:29
  • $\begingroup$ @skyler - from a narrative perspective, I really like the paintball trick - it sounds clever, could be visually satisfying if used in any visual media (OP didn't say what they need this for) and would have worked perfectly with a magical illusion. On the other hand - if the holo-generator is good enough to overlay a human with a disguise (it can't possibly know what that human is going to do, say they'll move), covering up some paint should be trivial. $\endgroup$ – G0BLiN Dec 19 '19 at 7:15

Challenge and pass, brevity codes

Back in the Marines a security function taught as a basic requirement was the concept of a challenge and pass system. This is actually a wee bit more sophisticated than just "WHAT'S THE PASSWORD!" It started around WW2 where somebody might challenge an intruder by shouting "Lightning" expecting a friendly to give the response "Thunder." Problem is, an enemy waiting out of view in earshot can easily pick up what the challenge and pass is and copy it. It evolved further to be more disguised.

For example, on one training mission the challenge was beer and the response was pizza. But you don't just shout the words. You'd yell "WHO GOES THERE" then they'd say something like "A FRIENDLY!" Then you'd say something like "shit, good, I'm tired as hell and just want to get back home and have a beer" to which a friendly would have to say something to the tune of "I hear that man, I'm fuckin' hungry, I'd kill for some pizza right now."

Challenge/pass words were changed daily, by shift, or even mid shift and they were briefed to all involved personnel immediately upon changing. Also, there was always a duress phrase as well. If somebody was captured and forced under duress to divulge passphrases they would, ideally, inform the enemy of the real challenge, then give a false pass which would alert the guard that the approaching individual was an impostor. In the beer/pizza situation it was the word whiskey. So if somebody responded "Fuck that, whiskey is better" you'd know to shoot them because they were using information taken under duress from a captured comrade.

Brevity codes were just words designated to symbolize complex meanings or phrases while remaining short and concise. So somebody saying "vodka vodka vodka!" over the radio meant "Enemy breaching the wire, south side of compound, all quick response forces mobilize." Or something like that.

This concept can be improved upon

You may notice a trend here, in that the entire thing is disguised as jocular banter revolving around bar culture. There's a reason for that. Our operating environment was the Middle East, and Islamic culture prohibits alcohol. They do not possess a bar culture. This means, that even if a member of the Taliban gained the real password, there is almost a sort of cultural encryption involved. They lack the subtle nuances and cultural knowledge to be able to weave said password into a believable reply that doesn't arouse suspicion. Additionally, the entire system has a built in weaponization. If captured you have, however faint, a hope that you are not helpless. You can still protect your comrades and kill the enemy by possibly luring them into a trap by giving them a believable duress phrase.

This system is extremely crude and simple. But it is still effective. It plays off of cultural differences and the fact that language and culture are fluid constructs. Somebody not actively immersed in that culture has a very, very slim chance of grasping that subtlety and exploiting it without arousing suspicion and giving themselves away.

Ever watch Inglorious Basterds? There was a scene where the British saboteurs play everything perfect, but one of them orders two more drinks by holding up his pointer and middle finger in the classic UK/US hand gesture for two. In Germany the hand gesture for two is made with the thumb and pointer finger in a gesture that we native English speakers would assume meant "L for loser," in Germany it means the number two. Same thing, almost exactly. In that fictional scenario, simply forgetting/not knowing a seemingly unimportant detail like which fingers are used in common gestures got the entire crew killed.

Which brings me to my final point:

Intentionally Cultivated and updated Linguistics

The concept outlined above has been advanced upon to the point where military units utilize intentionally developed languages and cultures implemented and subconsciously imprinted by AI. Every single unit speaks its own subtly different dialect of the same language, complete with body language. To an outsider they all sound and look the same, but to an insider subtle cues and differences communicate volumes of data. Rank, unit, task, etc etc. All while sharing enough in common to intrinsically spot an outsider. The AI makes "random" changes which are wirelessly updated and seamlessly implemented to the point where a unit's soldiers do not even realise that their entire dialect and cultural body language cues have even changed. To them everything is as it always has been, and it's just blatantly obvious to spot an outsider a mile away. They stick out like a sore thumb!

But your crack squad of heroes have an inside scoop. They can not only imitate the enemy's appearance, but thanks to an experimental new AI based infiltration and espionage platform they can now decode the enemy's cultural/linguistic and dialectic countermeasures. They now can be subconsciously imprinted with the same inside lingual and body-language mannerisms as the enemy to fool the sentries and gain admittance to the facility. Not only that, but they can activly predict and update to match the changes which aren't quite as random as the enemy had hoped when they designed the system.

It's not perfect, they can only play the impostor for so long before being discovered, but it's juuuuuuust long enough to penetrate the enemy defenses deeply enough that if discovered the other counter measures and distractions are no longer effective or relevant.

  • 1
    $\begingroup$ "Challenge/pass words were changed daily, by shift, or even mid shift and they were briefed to all involved personel immiediatley upon changing." heck, with current tech, we can easily supply each operative (or probably sub-squad) with their own version of a "password authenticator" that gives a challenge and response (e.g., "lightning/thunder". The combo rotates, say, every 2 minutes and two sub-squads meeting will shout out the challenge/response they have in their authenticator. Basically, the idea is that the enemy will never be able to memorise a password combo or predict it. $\endgroup$ – VLAZ Dec 18 '19 at 14:45
  • $\begingroup$ @VLAZ - what if the enemy manages to get an authenticator? $\endgroup$ – G0BLiN Dec 18 '19 at 15:50
  • $\begingroup$ @G0BLiN good question, actually. I'm not sure. On one hand, if that happens, then your team failed. On the other, it seems like the thing that needs a contingency for, otherwise your team might fail even more. Quick solution might be link the operation to biometrics - if the device is "disturbed" (operative killed, or device removed from them), it stops working. But then it makes the device more complex and it increases the chance of a random failure. There is probably a better way lower-tech way to protect against theft but not sure. $\endgroup$ – VLAZ Dec 18 '19 at 16:41
  • 8
    $\begingroup$ I seem to recall that the term for "a sort of cultural encryption" is shibboleth. $\endgroup$ – Codes with Hammer Dec 18 '19 at 16:44
  • 2
    $\begingroup$ @BKlassen it will also significantly delay the authentication of two operatives meeting and trying to athenticate each other. It should be as fast as possible - few seconds delay fiddling with the device can get you killed or captured. $\endgroup$ – VLAZ Dec 19 '19 at 6:45

Assuming the projection system is good enough to figure out it should project a squirrel (using Willk's answer as a base for this one) there are fallbacks that can work in a pinch.

Strike operatives do not work alone.

Your 8-man section will subdivide in fireteams and pairs. These pairs know how to move together and cover each other. Let's say you're going down a straight corridor. You post 4 men on each side of the corridor. The ones at the front and the rear will move crouched to allow their teammates to shoot above their heads. You have 4 others left. If you need to turn a corner, the point-men can use mirrors or cameras to peek around corners, simultaneously, while the rest of the team is holding perimeter.

enter image description here

They know their formation.

Everyone knows where and how to stand. If someone is projected in your team's midst, your experienced operatives should keep it together. Maybe you have access to wearable IFF devices (Identify Friend or Foe). This would work automagically, using frequency hopping, encryption, the works. The enemy projection system cannot falsify that signal. Your commander's tablet will have status indicators displaying the vitals of each team member.

They know their equipment.

The men in the middle can use the IR / NV / Matter Density Indicators and alert the team to phantoms or threats. This is something that must be practised extensively before the op. You, as the commander, are responsible for setting the Rules of Engagement that would make your team effective with this equipment.

They create ideal fight conditions.

You won't go in alone. You command well-trained operatives so you're well funded. You want other teams to create a diversion. This will buy you some time as the enemy has to deal with your deception as well. The diversion teams should create a lot of noise and project a great deal of threat, but they will actually fall back once the going gets tough. Their job is done at that point.

They love their dogs.

Remember, dogs are still a part of this. They may get killed however, since Plot Armour is not always dog-sized.

Image Source



The flaw in the enemy's system is that it cannot conceal or disguise a person from more than one viewpoint at a time. So give each of your commandos a heads up display which floats a camera feed from a hovering drone transparently over part of their field of vision. Then use each commando's drone's existing selfie mode to keep it floating 10 feet behind, above and slightly off to the right of their position, facing forward.

Then add a support team a few miles from the combat zone, each flying two drones which are programmed to mirror each other's position from a slightly different angle. Have one or more support person for every commando connected by a private real time radio link. The support person's job is to watch the commando's flank and to fill the air with so many drones that the enemy can't use the selfie mode drones to locate the commando positions.

  • $\begingroup$ I don't understand the downvote, this seems like a useful suggestion. I recognize it from a popular computer game, it seems feasible $\endgroup$ – bytepusher Dec 18 '19 at 0:10
  • 3
    $\begingroup$ @bytepusher, thanks for the support. I've always wished that SE would require non-blank comments on down votes. I don't mind loosing the points, but I would prefer to know what the voter felt I did wrong. BTW... What computer game offers this? I thought I was mashing the available tech in a new way, but apparently I'm an accidental plagiarist. (not for the first time). $\endgroup$ – Henry Taylor Dec 18 '19 at 2:06
  • 2
    $\begingroup$ I think the "-1" may be because only "invisilibity" is concerned by the "one view point" contraint. Everything else is a 3D form that can be viewed from everywhere. $\endgroup$ – Echox Dec 18 '19 at 8:11
  • $\begingroup$ This HUDs could also communicate as an IFF system, which will let you know if the person in front of you is really an ally, or a disguised enemy. $\endgroup$ – Chronocidal Dec 18 '19 at 9:12
  • $\begingroup$ @HenryTaylor: I remember something similar from Battlefield 2142. With sufficient rank, one could get a drone to follow the soldier around. It would then automatically spot enemies, and they would be marked on the map / HUD. $\endgroup$ – bytepusher Dec 18 '19 at 19:52

TL:DR: see the last subsection

The door problem.


Public buildings are required to display evacuation plans. That tells you where doors are. Obfuscating those plans would be a clear violation of safety and OSHA wouldn't be amused. Additionally, somebody built that facility. And they didn't do it alone, off the top of their head. You might argue a top secret facility wouldn't have plans just laying around, but the point is there most certainly will be plans somewhere. It's up to your intelligence services to find them.

If you have a lot of time, you might try to infiltrate someone that will map out the complex and/or procure the plans. The plans may not actually be stored on site, but instead in some sort of national archive. An alternative to placing someone would be to flip an existing employee, with or without their knowledge. This runs the risk of creating a triple agent however, which is a bad thing for you. I would limit the use of interrogation, since you'll never be able to know if they're telling the truth until you go on site and verify by yourself.

As a general advice, do not engage an operation with incomplete intelligence. You can plan contingencies for things you know. It's things you don't know that you need to worry about. You can't expect success if you go in not knowing where the doors are.

Side note regarding doors

As a general rule, your security can't rely on the attacker not knowing about your security measures. So it's pointless to hide doors because you should assume attackers will already know where the doors are. The way you prevent physical access isn't by hiding doors, it's by requiring identification (e.g. ID badge), and making sure your doors (and walls, and windows) can withstanding some force.

If you decide no doors are present on floor plans, or there's no floor plans at all, a valid strategy to move through the complex then is to consider a wall is just a door you haven't blown open yet. There's no problem a sufficient amount of force can't solve after all.

The enemies-disguised-as-friendlies problem.

Strobe lights

Strobe lights are used IRL to identify friendlies. It's useful notably when you're way far away and people are just tiny smidges on a screen or a scope, and you need to shoot at the right people.

The holographic projectors will definitively be limited in the range of frequency they can display. As an example, if the projector works with visible light, it will be unable to reproduce an ultraviolet or infrared strobe light. Any other convenient frequency in the EM spectrum will do, or alternatively you could experiment with ultra/infrasounds. Obviously, emitting a signal means a knowledgeable enemy might be able to track it, so chose your signal wisely.

From there, the identification system is fairly simple: you have a small pack that emits your signal, you have a device that can read that signal (which could be the same device), and you have a way to tell your soldiers that information. And for that in particular, your soldiers presumably wear a helmet, which presumably has a visor. Since we're talking holographic illusion, I suppose adding augmented reality filters on a visor is a piece of cake. In the end, in practice, a friendly will pulse visibly, and an hologram will not.

Backup strategies

Your signal emitter may get borked, which means you will no longer be identified as friendly, which is a problem if rules of engagements are "shoot first, full stop". You don't need anything complicated, a simple password will suffice. WWII soldiers during the Invasion of Normandy used "Flash/Thunder" or a clicker to identify each other beyond visual range, and that will do the job just fine.

The enemies-disguised-as-flower-pots problem, which also encompasses the above two problems.


You can deprive a system from the power it needs to operate with a well placed bomb or two. If the facility is powered by a power plant, blow up the power plant or simply cut the cable. You'll also have to sabotage the backup generators in a similar way, because there will be backup generators.

Alternatively, an EMP might fry out enough electronics to make the system inoperable, at least in the sections where you need to be, so that you know you will not encounter an hologram.

Of course, sabotage like this because extremely complicated if the relevant facilities are hardened and/or underground. It will also bring attention to the fact that the facility is under attack. So not necessarily the most viable option depending on your scenario.

Holograms have a tell

All of the above are things you should consider for any operation anyways, but this is the real killer.

Equip your soldiers with a microcamera (which really should already be standard), a microcomputer, and a helmet with an augmented reality visor. We already know about one flaw in the system: it can't fool two points of view. Have your soldiers move two by two. Each soldier's kit will receive the video feeds from their teammates, and will be able to compare the two feeds. If it detect the same object in different angles, everything is fine. If one feed shows a flower pot and another shows a man, flag that as something to kill and show it to your troopers.

If you need a tell that a single operator can detect, then you could go for a flicker. Something imperceptible to the human eye but very visible to a camera, provided it isn't sync on the same frequency than the holographic projectors. This isn't really much of a stretch if you've ever tried to film a screen.

  • 1
    $\begingroup$ "Obfuscating those plans would be a clear violation of safety and OSHA wouldn't be amused." I would go out on a limb and say that an evil organisation does not care about OSHA violation. :P Just joking, it does seem plausible that there would be plans somewhere. I don't think it's a stretch to think it's possible to find them. Even if an "informer" (of some sort) maps the interior as you said. $\endgroup$ – VLAZ Dec 18 '19 at 14:39
  • $\begingroup$ @VLAZ I imagined just a regular top secret building where there needs to be plans, lines on the ground or signs so that Steve from accounting knows how to go to meeting room 12, but then it got me thinking what if there is an archetypal evil overlord that insists on no plans at all, and how to deal with that and I added a side note. So your input was useful. $\endgroup$ – AmiralPatate Dec 18 '19 at 15:15
  • $\begingroup$ There's no problem a sufficient amount of force can't solve after all." reminds me of "As the size of an explosion increases, the number of social situations it is incapable of solving approaches zero." :P $\endgroup$ – VLAZ Dec 18 '19 at 15:20
  • $\begingroup$ If the illusion system can fool exactly one person, it must be capable of fooling at least two, and no more than three simultaneous viewpoints... namely, the left and right eyeballs of a person (and maybe one backup). Otherwise, the illusion system is going to be imperfect. Accordingly, you don't even need cameras on pairs of soldiers to defeat it, you just need several cameras on each soldier's helmet and some way of spotting anomalies. If you have the support, have each soldier's live feed monitored by several off-site persons, each watching a different pair of cameras using VR glasses. $\endgroup$ – Matthew Dec 18 '19 at 18:17

For recognising your own troops vs others

We have the tech for this called Identify Friend or Foe (IFF). It's kind of misleading since it's identifying friendlies and thus anybody else is a "foe" but this is enough for your purposes. You just need to identify your own agents. You can even use stuff like NFC signals (for up close) or other radio signals that you can pick up with a receiver. If receiver picks up the signal you expect - you're clear.

In addition, you can add more measures

  • If possible, you can also keep constant monitoring of where each operative is. Give your team a radar to show positions or even Augmented Reality eyewear that has this as an overlay at the top. If they see a friendly target that isn't there on the radar, they can assume foe.
  • Similar effect can be achieved with constant communication. Everybody should be trained to issue periodic updates: "I'm at X, going to Y". Everybody else then can track this in their heads. If they mean somebody who isn't near where their last update said - it's a foe. It's taxing to keep track but not that much for a small enough team. If people go in pairs, then each pair only needs to track 3 other positions.

Recognising other (human) disguises of foreign troops

First of all, they can disguise as civilian or otherwise "not enemy" and still not a friendly (you already have a way of determining friendlies). Simple - anybody is to be assumed a foe. If possible, you can deploy less than lethal measures like stun guns, tear gas, and so on then subdue and neutralise the targets. Ziplock cuffs can be applied and the person just put aside to not be a problem.

This will minimise any civilian casualties if any are expected on the premise. It's also possible to simply shoot everybody and ask questions later. If no civilians are expected and hostiles can't be allowed to live, this is a viable option.

Recognising environmental disguises

This is not as easy but still doable. Say, there is a hologram over a door or a corridor making it look like there is a wall. Per the question, the hologram system can provide false visuals and sound but I doubt it can do those convincing enough for tech. You can use an ultrasonic distance measure device. They don't usually have a very many practical applications but would serve well in closer quarters - is that wall really there? You should get a reliable answer for that.

The hologram system might be able to issue false sound waves but it needs very serious computing power to match, calculate, and issue the false reverse sound wave response. So, while the holograms are sophisticated, I don't think they'll be able to fool sound distance measuring.

There are other types of distance measurements like lasers. This should be pretty similar to sound imaging but maybe very slightly easier to fool with a visual projection. I'm not an expert, though, it might even be harder to fool.

All in all, you can use distance measuring that doesn't rely on purely looking at an image to determine if there is something there or not. A combined technology that uses multiple scanning methods would be preferable but would need some more prep time, as it likely needs to be custom made.

However, detecting a false wall (or equivalent - false pile of stones is also an option) is half the battle. The smaller half. It is still a visual obstruction and even if your operatives know exactly what behind it. It might be a squad of enemies or an array of servers.

Assumption here - the holograms are not sophisticated enough to show a wall on one side but be clear on the other. That will allow enemies to just wait behind a "one way wall" and then shoot your operatives when they stand before them. But it seems pretty hard to do, so I assume that's not possible - enemies behind a false wall also see a wall but might be ready to open fire immediately when somebody emerges through it.

So, in this case recon and diversion tactics can be deployed - a drone can be send over in some cases to check what's behind an illusion, or a flashbang can be rolled through it. Overall, treat it as if it's a curtain and act accordingly - it obscures vision but it's not impossible to overcome. The squad could also scout around to find out what portion of space is unknown. It might be something with the size of a janitor closet or a basketball field. This can guide how to approach them.

It's also possible to...ignore the illusion. If there is a false wall hiding a small space, then you can just come back for it later. Drop a camera so mission control can keep an eye on it in case an enemy pops in from there but otherwise just move along.

  • $\begingroup$ "Assumption here - the holograms are not sophisticated enough to show a wall on one side but be clear on the other" Just display 2 walls, in close proximity (e.g. 3 inches apart, on the near and far side of a doorway) - one of these (invaders' side) displays a normal wall, the other (defenders' side) displays an image (possibly simplified shapes or coloured blobs) of the corridor/room through the doorway, and the enemy combatants' locations within. Not good enough to snipe, but enough to spray & pray or perhaps fire a shotgun. $\endgroup$ – Chronocidal Dec 18 '19 at 11:56
  • 1
    $\begingroup$ @Chronocidal yes, it's not impossible to poke holes in this "two side wall" (pardon the pun). You can also have cameras monitoring what's on the other side. Still, without some sort of restriction, it's very hard to write a proper answer. Especially, since the definition of the technology quite very wide. Perhaps the hologram tech can also casually fool radio waves, lasers, ultrasound and even thermal imaging. I've found going into "but what if it did <more stuff>" over and over again leads nowhere, so I'm sticking to mostly what's stated with some assumptions. $\endgroup$ – VLAZ Dec 18 '19 at 12:00

The easy way would be to cut the power supply as you go in. Apart from causing nine kinds of useful chaos, this will prevent their hologram machines from working. (From the description, the power requirements would be too much for it to be battery operated because of the wattage of light involved in daylight).


Existing answers seem to revolve around using IR goggles and other existing technology to circumvent the hologram technology. I'm going to try taking the approach that if the adversary can generate holograms in the visible spectrum they can and do also generate holograms in the IR spectrum. @Willk's answer about using working dogs is a good thought, but depending on the training of the dogs, the conflicting sensory input may render them ineffective, and will most certainly have a harmful effect on your operators, as even more so than the ordinary population they are trained to rely on their eyesight.

I will attempt to answer this while limiting the technological scope to the visible spectrum augmented by uninterrupted GPS and LTE-equivelant communication amongst the team. The team will have access to a single high flying surveillance drone, and no more than one quadcopter per team member. I will also add a range constraint on the hologram technology, it cannot project farther than 100 yards beyond the compound's fences.

It is also unclear what your objective is. Are you retrieving technology from your adversary or simply attempting to disrupt the adversary's existing operations? Is there a requirement for stealth or secrecy?

Given that your plot isn't to simply airstrike the compound, I assume your objective is to retrieve something. And given that the adversary could conceal cameras, and other monitoring devices, that stealth or secrecy is not a mission requirement.

Given that there are so many unknowns, and that you have some time, but not a lot of time, you should insert some operators ahead of time to perform surveillance, and perhaps even to "feint" at them, begin an attack to observe their response, depending on mission constraints.

A good start to the attack might be good-old-fashioned-chaos. Perhaps a prelude to the attack could be man-portable mortars or rockets. Perhaps this damages some of their projectors. At the very least, it helps your team identify holograms, as undamaged equipment after this is likely to be holographic, with the caveat that your opponent might be capable of projecting damaged equipment.

If the terrain allows, designating half or more of your team as marksmen would increase your team's awareness, while minimizing the number of operators in the "dangerzone". Given that induced friendly fire is a big threat, and the team's senses are going to be deminished by the hologram system, I would institute (and drill for a while) some rather rigid rules of engagement for them requiring double checking before actually shooting something. Using the uninterrupted GPS coordinate exchange, a shared map of team locations and other markers could be displayed on wrist mounted "phablets" for increased awareness, perhaps involving an overlayed camera feed from the surveillance drone, quadcopters, or even weapon-mounted.

The remaining operators would be minimally armed, limiting their numbers will also minimize the risk of friendly fire incidents. They will be the ones actually infiltrating the compound. Their primary weapon won't be actually shoot their adversary, instead to mark targets on the shared-map for the marksmen to hit. Laser marking could be used, but is also likely to be imitated by the adversary. These operators will also have direct quadcopter control to increase their awareness.

The biggest issue for them will be limited support from within buildings, perhaps limiting the usefulness of this plan. The biggest danger here will be gunmen hidden by holographic walls. Possible mitigations involve moving some of the marksmen mid-mission, clearing smaller buildings (or rooms) destructively with grenades, and as others have mentioned by marking rooms with paintballs or ink grenades, however paint splatters may be mimicked by the projections. Your team may be forced to clear a building room by room. In this case they will always want to enter a room simultaneusly from two perpendicular directions. The specifics will have to be tailored to the specifics of the terrain you encounter.

  • $\begingroup$ You've a fair point, given that real light-bending and distortion currently is limited to IR, so the real technology already is generating light-distortions in the EM-range that enhanced vision goggles are looking at. $\endgroup$ – Ruadhan Dec 18 '19 at 13:54

Going to come at this from the other direction, as the side in charge of the illusions and say that the soldiers have a very poor chance of success.


  • productive facility, not deliberate deathtrap / vault
  • illusions can not be distinguished from reality other than through physical interaction
  • illusions are omnidirectional (though flat, like painting voxels), no 1-way-mirrors
  • no corresponding hyper-intelligence, omniscience, infinite computing power

Unless we are entirely magic-based, the illusion system needs to have some way to see the area it is creating illusions in. Perhaps this is certain areas only (a glimmer of hope!). The bad news is that it probably only takes one area.

General tactics:

  • Blind the attackers with a white or black sphere around their heads

    • Already seems nearly insurmountable
    • Everything that follows assumes this somehow doesn't work
    • Likewise with walls/fog that stay directly in front of them
  • Illuminate their bodies or have overhead arrows point to them

  • Use projections as invincible commanders
    • Standing in the open at the corner and giving their current distance, etc.
  • Don't give time to calibrate or understand technique - overwhelm decisively once intrusion is understood
  • Waste ammunition
  • False floors, even slopes or staircases
  • Fill open spaces with a maze of false cover
  • Create the illusion of hundreds of defenders opening doors, windows, ceiling tiles, trapdoors standing up, etc. and firing simultaneously along with any number of real defenders

I think these tactics would fare very well against the current suggestions: dogs, passwords, teams. Depending on your flavor of illusion (you seem to be leaning towards "indistinguishable from magic") then lidar and infrared may be out as well. Paintballs and drones might be the closest, but obviously don't scale that well in a large and hostile facility.

Note that for any one way that the illusion is implemented (if very advanced but not magic) there will be useful ways to disrupt it. If you are going for that "scrappy and clever" feeling for your invaders then just pick an implementation (or 3, since they don't know which it is), and have them try different countermeasures on the fly.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.