Powered armour combat

[[R_H]]

1. IFF hardcoded: If your insurgent manages to crack your robot wirelessly, a hardcoded IFF would prevent firing in the direction of the soldier.
Likewise, your soldier should have a hud with IFF, to verify all the robots near him/her are under control.

And this wouldn't be problematic against an opponent with some competance in electronic warfare because...?

2. Self Destruct: Nothing so dramatic as an explosion. Something more like thermite over the computer and drivetrains. Maybe a can of compressed gas, steam pressure, or and coiled springs pop the robot, rendering it useless to insurgents.

Steam? Springs?

If they'd be able to use it as shrapnel (for example), it would still be of potential use to them.

[Uncluttered]

And this wouldn't be problematic against an opponent with some competance in electronic warfare because...?

Because the opponent would have to physically tamper with the roms in the robot to turn this off. Or the insurgent would have to crack the IFF encryption, on the battlefield.

It's not foolproof, that doesn't mean it's a bad idea.
An insurgent sophisticated enough to break encryption in realtime, has better things to do anyway.
Also, Newsflash. Not all opponents are competent in electronic warfare.

Steam? Springs?

This is about making exploitation of the robot difficult. Not about making it impossible. That's a fools errand.
This isn't movieland. The idea of self destruct is to render the Robot useless as-a-ROBOT.
You don't need to destroy the whole thing, just key parts.
Destroying the electronics is straight forward.
However, this leaves a physical platform behind they can use as an IED, or for some sort of decoy.

You don't want dramatic explosions, you just need mechanical faults.
Motors overheat deliberately, tires overinflate, robot arm throws itself to pieces, battery shorts and bursts into flames, thermite melts the engine, reactive armor ignites, radar overheats, coolant mixes into the fuel, air pressure pops casings, a joint may pop off, and a metal ring expands, making it difficult to re-attach without special tools and methods.

If they'd be able to use it as shrapnel (for example), it would still be of potential use to them.

Are you kidding about the shrapnel? Why aren't you more worried the insurgent could use the scrap metal arms to bludgeon you to death?
They can get better quality shrapnel at the hardware store.

[[R_H]]

And this wouldn't be problematic against an opponent with some competance in electronic warfare because...?

Because the opponent would have to physically tamper with the roms in the robot to turn this off. Or the insurgent would have to crack the IFF encryption, on the battlefield.

It's not foolproof, that doesn't mean it's a bad idea.
An insurgent sophisticated enough to break encryption in realtime, has better things to do anyway.
Also, Newsflash. Not all opponents are competent in electronic warfare.

News flash, Iraqi insurgents were able to tap into UAV feeds.

Besides, the next 100 years aren't going to be spent fighting illiterate peasants, there might actually be opponents who are competent (in EW).

Steam? Springs?

This is about making exploitation of the robot difficult. Not about making it impossible. That's a fools errand.
This isn't movieland. The idea of self destruct is to render the Robot useless as-a-ROBOT.
You don't need to destroy the whole thing, just key parts.
Destroying the electronics is straight forward.
However, this leaves a physical platform behind they can use as an IED, or for some sort of decoy.

You don't want dramatic explosions, you just need mechanical faults.
Motors overheat deliberately, tires overinflate, robot arm throws itself to pieces, battery shorts and bursts into flames, thermite melts the engine, reactive armor ignites, radar overheats, coolant mixes into the fuel, air pressure pops casings, a joint may pop off, and a metal ring expands, making it difficult to re-attach without special tools and methods.

How about the time honored method of getting some one to blow it up?

All the ideas you proposed sound fairly movieland to me.

If they'd be able to use it as shrapnel (for example), it would still be of potential use to them.

Are you kidding about the shrapnel? Why aren't you more worried the insurgent could use the scrap metal arms to bludgeon you to death?
They can get better quality shrapnel at the hardware store.

2. Self Destruct: Nothing so dramatic as an explosion. Something more like thermite over the computer and drivetrains. Maybe a can of compressed gas, steam pressure, or and coiled springs pop the robot, rendering it useless to insurgents.

Really, hardware stores sell good quality shrapnel? I'll have to ask the next time I'm in one. You missed my point, which was, they'd still find some way of using it. Why bother with all those elaborate self-destruct mechanisms then?

[Purple]

Why bother?
Well because they can find much less use of it as shrapnel or melting it down for scrap than they could if they could take over it and get control over your brand new Killbot 2000.

Also, I propose a strong explosive charge rigged to detonate on a 2 - 5 minutes timer. Before the timer runs out the user has to input a signal to the robot to reset it. If not, the droid explodes violently and if at all practical with enough force to kill anything in about 200 - 300 meters wide area.

[Sarevok]

@Uncluttered

How do you propose "hard-coded" IFF works ? As in what target attributes would it use to determine whether it is friendly or not ?

[[R_H]]

Why bother?
Well because they can find much less use of it as shrapnel or melting it down for scrap than they could if they could take over it and get control over your brand new Killbot 2000.

Then don't use it in a manner which will get it hijacked, or use encryption? What a novel idea...

Also, I propose a strong explosive charge rigged to detonate on a 2 - 5 minutes timer. Before the timer runs out the user has to input a signal to the robot to reset it. If not, the droid explodes violently and if at all practical with enough force to kill anything in about 200 - 300 meters wide area.

So if they're able to take over the robot, what prevents them from fooling the timer?

Moving along, you now have a robot packed with explosives, basically a robotic suicide bomber, which will blow up if someone tampers with it. Great idea. DEAD CHILDREN EVERYWHERE.

[Simon_Jester]

@Uncluttered

How do you propose "hard-coded" IFF works ? As in what target attributes would it use to determine whether it is friendly or not ?

Best bet would be to use "Is this object wearing a radio beacon that pings in the designated code sequence? If so, it is not a target. Is this object wearing a uniform my recognition software tells me is friendly? If so, it is not a target. Does this object have something I can identify as a weapon? If not, it is not a target."

Or just remotely operate the thing and let the operator worry about it, like with a UAV.

[Purple]

Then don't use it in a manner which will get it hijacked, or use encryption? What a novel idea...

It is always good to have a last resort option for when all else fails.

So if they're able to take over the robot, what prevents them from fooling the timer?

This part was actualy more intended to make sure the robots don't just stand still and become as easy to pick up as an artillery peace with the crew dead.

Moving along, you now have a robot packed with explosives, basically a robotic suicide bomber, which will blow up if someone tampers with it. Great idea. DEAD CHILDREN EVERYWHERE.

So? In war people die.

It's not like any sane army would deploy killer robots into a war zone where they have to watch who they shoot. You need humans for that since they can make that judgment call, and more importantly since they can take the blame and not the government behind them.

And I don't see anything particularly bad with the logic that each army should rather have 1000 enemy children die than lose one of their own soldiers.

[Uncluttered]

And I don't see anything particularly bad with the logic that each army should rather have 1000 enemy children die than lose one of their own soldiers.

Wow purple. I never thought to put you into the sociopath camp.
I am truly dissapointed.
Say hello to Norade for me.

OK. Lets do this from a Kissinger point of view.
You'll piss everyone off when you kill children.
1000 enemy children dead = 10000000 new enemies.

It's bad math, even for a sociopath.

[Purple]

News flash. I am a sociopath. Perhaps I should have made it clear earlier on I guess but I figured everyone knew already.

[Uncluttered]

@Uncluttered

How do you propose "hard-coded" IFF works ? As in what target attributes would it use to determine whether it is friendly or not ?

Best bet would be to use "Is this object wearing a radio beacon that pings in the designated code sequence? If so, it is not a target. Is this object wearing a uniform my recognition software tells me is friendly? If so, it is not a target. Does this object have something I can identify as a weapon? If not, it is not a target."

Exactly.

Pretty much the same way IFF has worked since the dawn of time.
It's interesting that when I mention it, it suddenly becomes science fiction around here.

"Hard Coded", means that the features are built into the hardware (or software) in such a way that they cannot be modified. This isn't simple copy protection.

Haxxor skilz will not help here. It's like trying to hack your car into a plane.

Or just remotely operate the thing and let the operator worry about it, like with a UAV.

Mostly this, but IFF prevents an insurgent, from killing the operator and turning the big robot guns onto platoon after platoon.

Right now, UAVs are relatively rare, and this is highly unlikely while they are based in CONUS.

In the future, ground combat Robots will be controlled more locally because of politics, bandwidth, and lag time. High altitude UAVs don't have jerky movements, and lag is not a big problem, but eventually it will be.

Local control might be a bunch of guys in an air conditioned trailer on a base a few miles away, a VR helmeted soldier riding in a Humvee, or a guy with a monocle HUD crouching around the corner trying not to shit his pants.

The more operators you have, the more opportunities the insurgent has to exploit robotic control.

[Uncluttered]

News flash. I am a sociopath. Perhaps I should have made it clear earlier on I guess but I figured everyone knew already.

In topics about weapons systems, this isn't a problem.
(Like say: child rearing.)

But, this isn't DOOM. Real war is frequently counter-insurgency.
There are "rules" to describe the habits and motivations of a human population.

A relevant rule subset to this discussion.

The population's loyalty is dictated by the answer to these questions:
1. Which side is going to win?
2. Which side threatens the most?
3. Which offers the most protection?

If you aren't willing to sacrifice your soldiers to protect the civilians, you will LOSE the war.

[Stark]

News flash. I am a sociopath. Perhaps I should have made it clear earlier on I guess but I figured everyone knew already.

I guess you should be used to the idea that your ideas are worthless, then. Your ideas will be judged on their merit, or lack of, and nobody gives a fuck about your self diagnosed edginess.

[Purple]

I guess you should be used to the idea that your ideas are worthless, then. Your ideas will be judged on their merit, or lack of, and nobody gives a fuck about your self diagnosed edginess.

Do be so kind to spare me your high and myghit better than thau speeches.

If you aren't willing to sacrifice your soldiers to protect the civilians, you will LOSE the war.

But is it not worse for public opinion if your soldiers die than if enemy civilians do? Especialy if you can simply use propaganda to dismis enemy civilian deaths as propaganda and staged images. I mean it's not imposible to do.

Or you could just have small explosive charges laid out all around the machine, not strong enough to blow it up just to blow up joints or computer parts and other stuff that can't be repaired without sophisticated technology. But that is taking all the fun out of it.

[Stark]

Do be so kind to spare me your high and myghit better than thau speeches.

Spare US your totally absurd self diagnosis. Your personal views of your ideas don't actually affect their validity - whether or not you're a Hot Topic sociopath doesn't defend them; they're still stupid. You even KNOW they're stupid!

[Uncluttered]

News flash, Iraqi insurgents were able to tap into UAV feeds.

So, is this an argument against encryption?
Because those feeds were not encrypted.

Besides, the next 100 years aren't going to be spent fighting illiterate peasants, there might actually be opponents who are competent (in EW).

Even with quantum computers, breaking encryption in real time is going to be really really hard. By then, the robots will be smart enough that many of these suggestions and issues will be obsolete.

We might as well be having a conversation about the interception of carrier pigeons.

How about the time honored method of getting some one to blow it up?

I didn't say you couldn't still do that.
The problem is, sending a soldier out there to blow it up kind of defeats the purpose of using robot doesn't it.

All the ideas you proposed sound fairly movieland to me.

Try reverse engineering something sometime.
Most of what I'm describing, is just a methodical removal of failsafes.

Really, hardware stores sell good quality shrapnel? I'll have to ask the next time I'm in one.

Silly me. I thought hardware stores sold nails.
I guess I should be looking for breakfast cereals there instead.

You missed my point, which was, they'd still find some way of using it. Why bother with all those elaborate self-destruct mechanisms then?

They are welcome to use it as an ashtray, a doorstop, or a coffee table.
However, there should at the very least, be a minimum of development put into denying them using it.....wait for it..........AS A ROBOT !!

When combat robots become ubiquitous, it's not going to be easy for a soldier or more importantly, a civilian journalist, to tell the difference between a "loyal" platform, and a compromised lobotomized rolling bomb.

[Uncluttered]

I guess you should be used to the idea that your ideas are worthless, then. Your ideas will be judged on their merit, or lack of, and nobody gives a fuck about your self diagnosed edginess.

Nobody gives a fuck about your unpleasent internet tough guy persona.

But is it not worse for public opinion if your soldiers die than if enemy civilians do?

To an extent.
Sadly, it depends on media coverage. It's a good idea to do the right thing, and protect the children. You never know when someone is going to leak footage of your massacre to the world.
It's probably a good idea to design your robots on this doctrine.

Especially if you can simply use propaganda to dismiss enemy civilian deaths as propaganda and staged images. I mean it's not impossible to do.

If your propaganda is perfect, you won't need an army.
Eventually, the truth will come out.
More and more of the brown skinned, foreign talking, strange people that you unfortunately labeled "enemy civilians" pack camera phones, and are on the worldwide network known as the internet.
Blackwater learned this the hard way.

Or you could just have small explosive charges laid out all around the machine, not strong enough to blow it up just to blow up joints or computer parts and other stuff that can't be repaired without sophisticated technology. But that is taking all the fun out of it.

Oh absolutely. They can and will do this. I refrained from mentioning this, because around here, "Small explosive" means a gram of antimatter.

Besides, explosives make shipping and handling more complicated.

[Purple]

Oh absolutely. They can and will do this. I refrained from mentioning this, because around here, "Small explosive" means a gram of antimatter.

I meant something more along the lines of the equivalent of the filling of a small pistol bullet exploding inside a paper bag duck taped onto the main frame or something.

Also if you can weaponise antimatter I don't think you need killer robots.

[Uncluttered]

I meant something more along the lines of the equivalent of the filling of a small pistol bullet exploding inside a paper bag duck taped onto the main frame or something.

Yes. like a squib, or an explosive bolt.
I was worried that the phrase "small explosive" would unleash an avalanche of misunderstanding from a group that normally debates the merits and methods of blowing up planets.

Also if you can weaponize antimatter I don't think you need killer robots.

Yes you do. But now you can make BIGGER killer robots.

[[R_H]]

So, is this an argument against encryption?
Because those feeds were not encrypted.

No, it's pointing out that they're not a bunch of ignorant peasants, in addition to the bandwidth issue

Even with quantum computers, breaking encryption in real time is going to be really really hard. By then, the robots will be smart enough that many of these suggestions and issues will be obsolete.

We might as well be having a conversation about the interception of carrier pigeons.

Big assumption.

My point was, just because the GWOT isn't RAR PANZER WAR, doesn't mean that war between nation-states (or parties similarily well equipped/trained, example Hezbollah) is not a possiblity within a short timespan.

I didn't say you couldn't still do that.
The problem is, sending a soldier out there to blow it up kind of defeats the purpose of using robot doesn't it.

The use of more and more robots, however autonomous, will not eliminate the need for combined arms. The ATGM didn't eliminate MBTs from battlefields.

Try reverse engineering something sometime.
Most of what I'm describing, is just a methodical removal of failsafes.

Why bother with failsafes and preventing reverse engineering? If I'm not mistaken, very little military hardware has built in self-destruct devices. If you don't want to lose it, don't use it.

Silly me. I thought hardware stores sold nails.
I guess I should be looking for breakfast cereals there instead.

Remember to get the hypervelocity cereals, none of that subsonic shit.

They are welcome to use it as an ashtray, a doorstop, or a coffee table.
However, there should at the very least, be a minimum of development put into denying them using it.....wait for it..........AS A ROBOT !!

Because lots of military hardware has that...find some other way to blow it up.

When combat robots become ubiquitous, it's not going to be easy for a soldier or more importantly, a civilian journalist, to tell the difference between a "loyal" platform, and a compromised lobotomized rolling bomb.

Could argue that implementing a self-destruct mechanism implies that the 'bot is not to be trusted. Other, less drastic measures would be implementable which would reduce the possibility of a ROGUE ROBOT.

Or you could just have small explosive charges laid out all around the machine, not strong enough to blow it up just to blow up joints or computer parts and other stuff that can't be repaired without sophisticated technology. But that is taking all the fun out of it.

Oh absolutely. They can and will do this. I refrained from mentioning this, because around here, "Small explosive" means a gram of antimatter.

Besides, explosives make shipping and handling more complicated.

So why exactly are you harping on about needing a self-destruct then?

Or just remotely operate the thing and let the operator worry about it, like with a UAV.

Mostly this, but IFF prevents an insurgent, from killing the operator and turning the big robot guns onto platoon after platoon.

Right now, UAVs are relatively rare, and this is highly unlikely while they are based in CONUS.

In the future, ground combat Robots will be controlled more locally because of politics, bandwidth, and lag time. High altitude UAVs don't have jerky movements, and lag is not a big problem, but eventually it will be.

Local control might be a bunch of guys in an air conditioned trailer on a base a few miles away, a VR helmeted soldier riding in a Humvee, or a guy with a monocle HUD crouching around the corner trying not to shit his pants.

The more operators you have, the more opportunities the insurgent has to exploit robotic control.

I assume subverting control of the robot would be easiest when the operator is close to the enemy (ie around the corner, which makes him more vulnerable if he's not directly paying attention to his surroundings and/or visibly controlling the 'bot).

Best bet would be to use "Is this object wearing a radio beacon that pings in the designated code sequence? If so, it is not a target. Is this object wearing a uniform my recognition software tells me is friendly? If so, it is not a target. Does this object have something I can identify as a weapon? If not, it is not a target."

Or just remotely operate the thing and let the operator worry about it, like with a UAV.

Providing IFF-capability sounds very complicated and expensive, seeing how there's probably always going to be an operator in the loop. Might as well just go with RO then.

How would IFF be implemented (I'm guessing in a manner which reduces false negatives)?
Weapon(like object) present?
Uniform?
IFF?
Engage?

And I don't see anything particularly bad with the logic that each army should rather have 1000 enemy children die than lose one of their own soldiers.

Wow purple. I never thought to put you into the sociopath camp.
I am truly dissapointed.
Say hello to Norade for me.

OK. Lets do this from a Kissinger point of view.
You'll piss everyone off when you kill children.
1000 enemy children dead = 10000000 new enemies.

It's bad math, even for a sociopath.

I find myself agreeing with Uncluttered (with the spirit, perhaps not the content), perhaps for the first time in this thread

[Uncluttered]

No, it's pointing out that they're not a bunch of ignorant peasants, in addition to the bandwidth issue

Who argued that position?
Sounds like you are off to see the wizard with that one.

Big assumption.

My point was, just because the GWOT isn't RAR PANZER WAR, doesn't mean that war between nation-states (or parties similarily well equipped/trained, example Hezbollah) is not a possiblity within a short timespan.

Even if the U.S. decided to go to war with gotee wearing U.S. from a parralel universe, breaking encryption in real time is HARD.
Hezbollah wasn't actually doing this. IDF weren't using their radios properly.

I didn't say you couldn't still do that.
The problem is, sending a soldier out there to blow it up kind of defeats the purpose of using robot doesn't it.

The use of more and more robots, however autonomous, will not eliminate the need for combined arms. The ATGM didn't eliminate MBTs from battlefields.

What are you actually arguing against?
This is not an argument against combined arms, or an argument for replacing all human soldiers with robots. It's about augmenting a human force with robots.

If you have to send a soldier into a dangerous scenario to destruct a damaged robot, then your robots aren't augmenting very well in that instance.
If all your soldier has to do is walk over and blow it up, then he/she might as well just drag it to the repair shop.

Why bother with failsafes and preventing reverse engineering?

Your statement makes me hope you don't work for any military contractors.

Why bother with failsafes......

So your lithium battery doesn't overcharge and blow the fuck up, starting a fire and lighting off the ammo, killing a tent full of jarheads nearby it should have been helping guard.
So the motors or hydraulics don't stress the frame of the arm, twisting itself into a pretzel.
So the robots fuel pump shuts off when fuel pressure suddenly drops, preventing a leaking fuel line from starting a fire and burning to death the soldier using the damaged robot for cover.
So the robots machine gun doesn't swivel around and shoot itself in the radiator.
So the battery disconnects via a contactor when the DC motor controller is damaged by a bullet, preventing the robot from a sudden accelerating into your soldiers as they get ready to fire a morter.

......preventing reverse engineering?

So your patent isn't stolen by a bigger military contractor.
So your robot soldier isn't mass produced in a chinese factory by the thousands.
So someone doesn't find a zero day exploit you didn't know about.
So the information in your robots CPU isn't used for espionage.

If I'm not mistaken, very little military hardware has built in self-destruct devices. If you don't want to lose it, don't use it.

That's because very little military hardware has a droid brain on board.
You will find that things like drones, DO have some sort of self destruct on board. This is a silly thing to argue.

Remember to get the hypervelocity cereals, none of that subsonic shit.

I'll make sure it's fortified with iron.

They are welcome to use it as an ashtray, a doorstop, or a coffee table.
However, there should at the very least, be a minimum of development put into denying them using it.....wait for it..........AS A ROBOT !!

Because lots of military hardware has that...find some other way to blow it up.

Most military hardware has a procedure to render it "useless" to the enemy. I'm just talking about the hardware doing it to itself.

When combat robots become ubiquitous, it's not going to be easy for a soldier or more importantly, a civilian journalist, to tell the difference between a "loyal" platform, and a compromised lobotomized rolling bomb.

Could argue that implementing a self-destruct mechanism implies that the 'bot is not to be trusted.

I don't really care how the robot feels. If a war robot has feelings, it's a poor design.

Other, less drastic measures would be implementable which would reduce the possibility of a ROGUE ROBOT.

If I can't use failsafes and anti tamper devices, tell me, how am I going to stop a "ROGUE ROBOT"? Am I going to take away its cell phone and TV privileges?

So why exactly are you harping on about needing a self-destruct then?

Because it sucks when the "ignorant peasant" cracks open the robot, and using parts from an RC car, pilots it back to you with about two pounds of pure love onboard.

The more operators you have, the more opportunities the insurgent has to exploit robotic control.

I assume subverting control of the robot would be easiest when the operator is close to the enemy (ie around the corner, which makes him more vulnerable if he's not directly paying attention to his surroundings and/or visibly controlling the 'bot).

Yes. Exactly. Which is why it's likely to eventually happen, even with soldiers and robots guarding him.
This would be an argument FOR not against IFF.

Providing IFF-capability sounds very complicated and expensive, seeing how there's probably always going to be an operator in the loop. Might as well just go with RO then.

The soldiers were kicking ass, right until their internet went out....

How would IFF be implemented (I'm guessing in a manner which reduces false negatives)?

There are many ways to do it.
The simplest way, is for the IFF to deny the robot a certain arc of firing based on the location of the soldiers wearing the transponders.
This would be short range only, maybe even infrared, or it becomes an Intel exploit, and a missile lock.
Perhaps the soldier can also designated entire areas as off limits. A school, a house, etc.

I find myself agreeing with Uncluttered (with the spirit, perhaps not the content), perhaps for the first time in this thread

The object of the conversation is as much about getting rid of faulty personal conclusions, as much as it is about having a persuasive argument for the other persons.
If you won't disagree with me, who will?