Which design feature should a remote combat robot have?

The future is now. This is the place for mecha and science.
Post Reply
False Prophet
Posts: 955
Joined: Thu Dec 31, 2015 7:40 am

Which design feature should a remote combat robot have?

So, I have been making up the setting for my original fiction, and it needs remote-controlled combat robot that would completely replaced human soldiers. At first, I was tempted to use a Terminator-esque machine, but then I realized that directly copy the human shape is not good enough, because in many ways our body shape is unsuited for the intense environment (for example, standing on two foot do lower our ability to jump, or the spine has to take the weight of the whole body in an unstable manner, for example.) The only saving grace is that because we basically model the world to fit us, a robot that mimics our shape would have an easier time interacting with the environment.

So, I want to ask you guys: How should a future combat robot look like, assuming that all the problems of hardware (powe source, muscle, etc.) have been solved?
User avatar
Dark Duel
Posts: 4833
Joined: Mon Aug 21, 2006 6:39 pm
Location: A blue City in a red State

Re: Which design feature should a remote combat robot have?

First, I would suggest a quadruped or hexapod design - 4-6 limbs - like say, the Gells-Ghe or GiTS's Tachikoma and other assorted multiped tanks.
You'd have greater stability on the ground than a biped, and theoretically you could also design it with a lower profile than you might otherwise get with a Terminator-style humanoid. Alternately, ditch the limbs altogether and make it a hovering unit instead.

Second, hands are overrated. It strikes me as an unnecessarily complex mechanism compared to a design that incorporates the necessary weapons or what have you directly into the body or limbs, as the case may be - look at the GaZuOOT, for just one example, or some of the Guntank variants.
If need be, you can design it in such a way that either more than one interchangeable weapon are built into the same limb (think a simplified, fixed version of the GN Sword, for instance), or else make it so that the weapons are interchangeable and mounted onto a fixed hardpoint. If you absolutely have to have it be able to pick up something, a simple claw works just as well as a five-fingered hand, and is a lot less mechanically complex.
Finally, putting the cockpit in the most heavily armored section of the main body makes sense.

Just a few thoughts.
// ART THREAD // NOT ACCEPTING REQUESTS

"You can learn all the math in the 'verse, but take a boat in the air you don't love, she'll shake you off just as sure as the turn of the worlds. Love keeps her in the air when she ought to fall down. Tells you she's hurting before she keens. Makes her a home."
User avatar
Areku
Posts: 1216
Joined: Mon Feb 11, 2008 12:00 am

Re: Which design feature should a remote combat robot have?

It depends on what type of remote control structure you're interested in. Are these robots semi-autonomous (mostly able to function without remote control but needing human input for weapons use and periodically for certain tasks), direct dependent control (human continuously in the loop, robot can barely move without remote control) or somewhere in between? Or is the robot capable of a full range of actions on its own, with humans basically exercising veto control ("No, robot, that's actually a civilian, don't shoot")? Also, how many robots does a single human control? Is it one robot per person, or does one human oversee a team of robots or does a team of humans oversee an entire network of robots? If it's a network of robotics, I'd recommend studying up on swarm robotics, though that would probably be getting away from what one normally associates with "remote control".

Also, how does this paradigm factor into your narrative? Is the story structured around the morality and controversy of such a system? There are a multitude of angles you could take this, the first and most obvious being the instinctive repulsion to the idea of a robot killing a human, but other issues include responsibility and accountability (current ethical struggles include identifying who's to blame for an error when you have someone on the ground calling in a strike, approval for the strike getting sorted through regional or theater command, policy-makers dictating rules of engagement and often getting directly involved in what used to be small-scale decisions, and a soldier in a trailer in California digitally "pulling the trigger", with modern multi-national collaborations further complicating the debate; Eye in the Sky provides some excellent perspective into this without getting preachy), the effect a robotic force has on the other side's morale (some are psychologically devastated by the idea of risking their life to fight something that can be easily replaced by a factory, others use the idea to demonize the side using robots/drones and galvanize their efforts), the debate over whether or not lowering one side's human cost actually makes them more inclined to start wars and distribute violence, and what effects this can have on the humans controlling the robots (many drone operators struggle with psychological whiplash from digitally pulling the trigger by day and going home to their families every night). Or are the remote controlled robots simply a means to deliver an action-packed sci-fi setting?

If they're being used to replace human soldiers, is there an emphasis on their use on a "main" battlefield (with tanks, planes and artillery), or are they being used more for urban combat and occupation and policing (if urban, is the opposition guerilla in nature or more like a traditional military force)? How much do you favor stylistic design over realistic depictions?

Depending on the answers to those questions, the "best" robot can take a huge range of forms.
User avatar
Gelgoog Jager
Posts: 1640
Joined: Mon Jun 05, 2006 9:09 pm

Re: Which design feature should a remote combat robot have?

I would suggest something alongside the mechas from Kill Command (AKA "Identify"), which are indeed AI-controlled units meant to replace regular human infantry:

http://www.starburstmagazine.com/images ... mand-4.jpg
http://www.starburstmagazine.com/images ... mand-5.jpg
http://www.starburstmagazine.com/movie- ... and-images
False Prophet
Posts: 955
Joined: Thu Dec 31, 2015 7:40 am

Re: Which design feature should a remote combat robot have?

Areku wrote:It depends on what type of remote control structure you're interested in. Are these robots semi-autonomous (mostly able to function without remote control but needing human input for weapons use and periodically for certain tasks), direct dependent control (human continuously in the loop, robot can barely move without remote control) or somewhere in between? Or is the robot capable of a full range of actions on its own, with humans basically exercising veto control ("No, robot, that's actually a civilian, don't shoot")? Also, how many robots does a single human control? Is it one robot per person, or does one human oversee a team of robots or does a team of humans oversee an entire network of robots? If it's a network of robotics, I'd recommend studying up on swarm robotics, though that would probably be getting away from what one normally associates with "remote control".

Also, how does this paradigm factor into your narrative? Is the story structured around the morality and controversy of such a system? There are a multitude of angles you could take this, the first and most obvious being the instinctive repulsion to the idea of a robot killing a human, but other issues include responsibility and accountability (current ethical struggles include identifying who's to blame for an error when you have someone on the ground calling in a strike, approval for the strike getting sorted through regional or theater command, policy-makers dictating rules of engagement and often getting directly involved in what used to be small-scale decisions, and a soldier in a trailer in California digitally "pulling the trigger", with modern multi-national collaborations further complicating the debate; Eye in the Sky provides some excellent perspective into this without getting preachy), the effect a robotic force has on the other side's morale (some are psychologically devastated by the idea of risking their life to fight something that can be easily replaced by a factory, others use the idea to demonize the side using robots/drones and galvanize their efforts), the debate over whether or not lowering one side's human cost actually makes them more inclined to start wars and distribute violence, and what effects this can have on the humans controlling the robots (many drone operators struggle with psychological whiplash from digitally pulling the trigger by day and going home to their families every night). Or are the remote controlled robots simply a means to deliver an action-packed sci-fi setting?

If they're being used to replace human soldiers, is there an emphasis on their use on a "main" battlefield (with tanks, planes and artillery), or are they being used more for urban combat and occupation and policing (if urban, is the opposition guerilla in nature or more like a traditional military force)? How much do you favor stylistic design over realistic depictions?

Depending on the answers to those questions, the "best" robot can take a huge range of forms.
Wow, that is really helpful. Thanks!

In my view, these robots are normally fully controlled in an 1 master - 1 drone fashion. However, an operator can control more than one robots at an moment (last time I heard, the U.S Army is trying to reconfigurate the Predator to lessen the load and pave way for more multi-drone-control); or, in the case the command console (actually nothing more than an IFV that stays within 10km away from the battlefield), the robots can operates independently, however their priority command (unless reinstated before the battle begins) is to retreat.

And there robots are just the basic infantry. Jet fighters, MBT, etc. and all other attack vehicles have been replaced with remote-controlled drone.

For those who wondering what I am writing, it is basically like this: An intergalactic starship manned by a super A.I (Artificial Superintelligence, by today definition) travelled the universe looking for a new hospitable planet. After aeons, it reached its destination. However, the artificial human it created (It is the current consensus that it is impossible to transport live/cyro people to another planet, so better make clones from the DNA Map. See Michio Kaku's Sci-fi Science) could not survive the new environment, so the A.I mixed human genomes to those native animals of the new planet. The result is Newtype-esque human...

Nada nada nada...

What I want to tell is nothing as complex as you said, it is just this simple notion: Humanity is governed by codes that it can never resists - or, we will repeat the same mistakes over and over again even though we have gone forward so far from yesterday (think about a world completely obeys the spirit of Austrian Economic and Psychoanalysis. The setting is also heavily influenced by current events. too, with two countries led by Donald Trump and Putin's clones go to war to each other.)

It is the anathema of the latter Gundam series ("If we have a mean to understand each other, it will all be fine!"), if you think about it ("No, I can understand you, but why should you be anymore important to me or my Greater Good than when I don't understand you?")
User avatar
Gelgoog Jager
Posts: 1640
Joined: Mon Jun 05, 2006 9:09 pm

Re: Which design feature should a remote combat robot have?

Found another minor example form Gundam 00:

http://www.mahq.net/mecha/gundam/00/maj-v34ai.htm

This small MA is actually available in both manned and unmanned versions:

http://www.mahq.net/mecha/gundam/00/maj-v34.htm

And of course there's also the Automatons that A-Laws deploys from their Ahead MS:

http://gundam.wikia.com/wiki/Automatons
http://www.mahq.net/mecha/gundam/00/gnx-704t.htm
User avatar
MythSearcher
Posts: 1845
Joined: Sun Jan 20, 2013 4:36 pm

Re: Which design feature should a remote combat robot have?

From a more mechanical efficiency view, human legs are not that inefficient.
There have been studies about biped and more, and figured that biped is actually more energy efficient. Yes, it is unstable, but at the same time, that unstableness is what makes the efficiency of moving.

In fact, unstable is the way to go if you want to move fast anyway, piped or not, dynamic balancing is better than static balancing. In static balancing, which is the traditional way of thinking in designing robots about 20~30 years ago, you make a robot that tries to stablize itself at every step, and don't keep on going before it is stably balanced. In dynamic balancing, you are actually going through a series of falls, you don't balance yourself at every step, you just keep your fall in the direction of moving. BTW, in nature, humans and most animals use dynamic balance at least in rapid movement. The draw back of bipedal humans is that you cannot use static balancing when standing, you have to keep moving a bit to balance even when you don't really need to move, likely because you breathe and the CoM moves around a little and you need to take that into account. Current technology wise, a dynamic balanced robot do need to keep moving to stay put(there's also the reason of they are designed with pin sized legs and no way those are stable, likely a challenge more than a must), but in a more future robot, standing still can be as simple as locking down the legs at a balanced position, or extending a 3rd pod to achieve stablization.

At least two separate study I can bring up is one which let a Chimp walk on a treadmill, and measure the oxygen intake of it. The speed was kept unchanged, but the Chimp was encouraged to move on 2 or 4s. When moving on 4s, which is actually its more natural movement, it actually take in more oxygen compared to when just using its 2 legs. Another study actually captured Roaches running in slow-motion camera, and at its top speed, it turned from using 6 legs to just 2.
Bipeds are unstable and you need more brain power to balance, but it does have its own merits.

Another bio-mechanical related part, legs are more efficient when it is straight.
All those insect, spider legs? You have a bad design case.
In fact, most lizards and frogs' legs are not efficient as well, they introduced a torque when they have to first extend horizontally from their body before turning downwards. The longer the moment arm, the more strength you need to put in. It is not just a problem of moving, just balancing is bad enough, think of moving in 4s yourself, do you move easier with your arms straight down or with your arms extended sideways(like in the lowered push up position)?
Mammals usually have a pretty good design, some of the larger Dinosaurs as well. Their legs come straight down, there is little moment arm to multiply the force needed to support the body.

In this sense, the Tachikoma or others being suggested up there are actually not that energy efficient in the design of legs.

Anyhow, dynamic balancing will be a must for military robots, giving the possibility of losing a limb in combat, you can't really rely on the stablizing factor of more legs, even if you have a quatruped or hexapod, once you lose a limb, you need to change the way you move if you plan to return. Even a biped can jump back to base if a limb is loss in battle if you use dynamic balancing.(and yes, robots have been tested with this ability IRL)

Also, if your robots are replacing human soldiers, and they need to work in human living area, you might really want to have them designed with the size of humans. So straight up bipeds are a pretty good choice since other designs usually have a larger ground coverage area and will be limited in movement, thus cannot completely replace humans.
If you use wheels/tracks, which are more efficient in plains, you need to figure out how to move on stairs. Not impossible, there are wheel chairs that can do that, but not really that simple(by design they are pretty bad at doing so and introduces a lot of extra dead weight)
Hoverbots are not bad in terms of going through most of the terrain, but the problem usually comes at landing and taking off. It is pretty hard to get the equivalant of jumping or bursting in hover, thus the movement will be easier predict under fire.
Installing more than one movement system is not a good idea in efficiency, at lest one of those systems is dead weight when not in use.

And hands, if you are using giant robots, then no, you don't need hands but only turrets with weapons that have mounts and your base having heavy machinery to change them.
But if you are human sized and designed to replace human soldiers, you might really want your robots to have at least 1 hand with at least 3 fingers, and even if your robot is not a biped and not as tall as humans, its hand should at least be able to reach a height where humans can reach, for openning unlocked doors(i.e. turning the door knob) and pressing various buttons around, or even using a keyboard on a desk.

Eyes at human level will also help a lot, of course extra eyes at other levels will also help, but if you have a human controller, it'd be intuitive to control the movement of something at his/her own height. The world looks different at different heights, your sense of distance(and because of that, speed as well) actually changes.
False Prophet
Posts: 955
Joined: Thu Dec 31, 2015 7:40 am

Re: Which design feature should a remote combat robot have?

MythSearcher wrote:From a more mechanical efficiency view, human legs are not that inefficient.
There have been studies about biped and more, and figured that biped is actually more energy efficient. Yes, it is unstable, but at the same time, that unstableness is what makes the efficiency of moving.

In fact, unstable is the way to go if you want to move fast anyway, piped or not, dynamic balancing is better than static balancing. In static balancing, which is the traditional way of thinking in designing robots about 20~30 years ago, you make a robot that tries to stablize itself at every step, and don't keep on going before it is stably balanced. In dynamic balancing, you are actually going through a series of falls, you don't balance yourself at every step, you just keep your fall in the direction of moving. BTW, in nature, humans and most animals use dynamic balance at least in rapid movement. The draw back of bipedal humans is that you cannot use static balancing when standing, you have to keep moving a bit to balance even when you don't really need to move, likely because you breathe and the CoM moves around a little and you need to take that into account. Current technology wise, a dynamic balanced robot do need to keep moving to stay put(there's also the reason of they are designed with pin sized legs and no way those are stable, likely a challenge more than a must), but in a more future robot, standing still can be as simple as locking down the legs at a balanced position, or extending a 3rd pod to achieve stablization.

At least two separate study I can bring up is one which let a Chimp walk on a treadmill, and measure the oxygen intake of it. The speed was kept unchanged, but the Chimp was encouraged to move on 2 or 4s. When moving on 4s, which is actually its more natural movement, it actually take in more oxygen compared to when just using its 2 legs. Another study actually captured Roaches running in slow-motion camera, and at its top speed, it turned from using 6 legs to just 2.
Bipeds are unstable and you need more brain power to balance, but it does have its own merits.

Another bio-mechanical related part, legs are more efficient when it is straight.
All those insect, spider legs? You have a bad design case.
In fact, most lizards and frogs' legs are not efficient as well, they introduced a torque when they have to first extend horizontally from their body before turning downwards. The longer the moment arm, the more strength you need to put in. It is not just a problem of moving, just balancing is bad enough, think of moving in 4s yourself, do you move easier with your arms straight down or with your arms extended sideways(like in the lowered push up position)?
Mammals usually have a pretty good design, some of the larger Dinosaurs as well. Their legs come straight down, there is little moment arm to multiply the force needed to support the body.

In this sense, the Tachikoma or others being suggested up there are actually not that energy efficient in the design of legs.

Anyhow, dynamic balancing will be a must for military robots, giving the possibility of losing a limb in combat, you can't really rely on the stablizing factor of more legs, even if you have a quatruped or hexapod, once you lose a limb, you need to change the way you move if you plan to return. Even a biped can jump back to base if a limb is loss in battle if you use dynamic balancing.(and yes, robots have been tested with this ability IRL)

Also, if your robots are replacing human soldiers, and they need to work in human living area, you might really want to have them designed with the size of humans. So straight up bipeds are a pretty good choice since other designs usually have a larger ground coverage area and will be limited in movement, thus cannot completely replace humans.
If you use wheels/tracks, which are more efficient in plains, you need to figure out how to move on stairs. Not impossible, there are wheel chairs that can do that, but not really that simple(by design they are pretty bad at doing so and introduces a lot of extra dead weight)
Hoverbots are not bad in terms of going through most of the terrain, but the problem usually comes at landing and taking off. It is pretty hard to get the equivalant of jumping or bursting in hover, thus the movement will be easier predict under fire.
Installing more than one movement system is not a good idea in efficiency, at lest one of those systems is dead weight when not in use.

And hands, if you are using giant robots, then no, you don't need hands but only turrets with weapons that have mounts and your base having heavy machinery to change them.
But if you are human sized and designed to replace human soldiers, you might really want your robots to have at least 1 hand with at least 3 fingers, and even if your robot is not a biped and not as tall as humans, its hand should at least be able to reach a height where humans can reach, for openning unlocked doors(i.e. turning the door knob) and pressing various buttons around, or even using a keyboard on a desk.

Eyes at human level will also help a lot, of course extra eyes at other levels will also help, but if you have a human controller, it'd be intuitive to control the movement of something at his/her own height. The world looks different at different heights, your sense of distance(and because of that, speed as well) actually changes.
You brought up some really interesting idea, however, I want to ask these questions:

- It may be true that in moving normally, the straight legs are more efficient. However, what about jumping? Can the straight legs store as much as energy as V-legs, and would the straight legs handle the downward force of the landing better?

Call me strange, by I do believe that in my scenario of remote-controlled drones, operators would more likely to utilize jumping a lot more than in normal life. The reason is because in my setting, over 50% of the battles fought between two sides happened on open plains.

- Is roller-skating a good idea? I'm thinking of something in the line of Nightmare Frame.
User avatar
MythSearcher
Posts: 1845
Joined: Sun Jan 20, 2013 4:36 pm

Re: Which design feature should a remote combat robot have?

False Prophet wrote:
You brought up some really interesting idea, however, I want to ask these questions:

- It may be true that in moving normally, the straight legs are more efficient. However, what about jumping? Can the straight legs store as much as energy as V-legs, and would the straight legs handle the downward force of the landing better?

Call me strange, by I do believe that in my scenario of remote-controlled drones, operators would more likely to utilize jumping a lot more than in normal life. The reason is because in my setting, over 50% of the battles fought between two sides happened on open plains.

- Is roller-skating a good idea? I'm thinking of something in the line of Nightmare Frame.
The storage of energy depends on the distance you can change from going from the squat position to the straight position, thus not really related to whether the legs are straight or not when walking.
Insect type legs where it makes a lambda shape(or inverse V) might make the distance longer depending on design, but it is also extremely inefficient and likely waste a lot of energy in simple straightening for jump.(i.e. you might not be able to utilize most of the distance)
For the landing, the same applies, you want longer distance for spreading the impact, but you also want a shorter moment arm to optimize energy useage. Straight legs can squat down anyway.

I'd discourage jumping a lot. While in a jump, you cannot change your direction, thus your motion is extremely easy to predict and thus becoming sitting ducks against enemy fire.
Even if you have some kind of thrust or aerodynamic method to change direction in mid air, the initial acceleration will be much less than desirable.
I'd put my money on inproving stealth systems and stay low on the ground. In fact, open plains in real life aren't really that flat, keeping low profile actually make you pretty hard to detect, especially we are talking about robots here, they don't feel itches and they don't need to breathe. Staying put and not moving is their expertise. Inching forward in disguise would also be pretty simple.
If you watch documentary, you'd found out that nature has quite a lot of stealthy hunters in open plains.

Roller skates is a system with a lot of dead weight. When you skate, the legs are pretty useless, while when you walk the wheels are useless. You don't want duplicated systems on a military machine if you can avoid it. Yes, its cool, but at the same time the cost of installing all the system will sky rocket and the maintenance will be a headache. The supply of parts might also become a problem and cause you a lot of trouble(like machines gets damage but you don't have the parts to fix it) Hiding all the roller system inside the legs is not a fun engineering problem to solve. Human uses skates because we simple cannot remove our legs and install wheels, that is not the case in robots.

If your robots fight mainly in open plains, then you don't really need to care about legs at all.
Hovering system or at least tracks will be the way to go. Horizontal area is also not that important, you won't get stuck in between alleys, if its two stones, go around it or over it.
Change to legs when you get into a city, you don't need the robots to carry around the dead weight when fighting, carry it in your support vehicle and change the equipment per operation requirements.
False Prophet
Posts: 955
Joined: Thu Dec 31, 2015 7:40 am

Re: Which design feature should a remote combat robot have?

MythSearcher wrote:
False Prophet wrote:
You brought up some really interesting idea, however, I want to ask these questions:

- It may be true that in moving normally, the straight legs are more efficient. However, what about jumping? Can the straight legs store as much as energy as V-legs, and would the straight legs handle the downward force of the landing better?

Call me strange, by I do believe that in my scenario of remote-controlled drones, operators would more likely to utilize jumping a lot more than in normal life. The reason is because in my setting, over 50% of the battles fought between two sides happened on open plains.

- Is roller-skating a good idea? I'm thinking of something in the line of Nightmare Frame.
The storage of energy depends on the distance you can change from going from the squat position to the straight position, thus not really related to whether the legs are straight or not when walking.
Insect type legs where it makes a lambda shape(or inverse V) might make the distance longer depending on design, but it is also extremely inefficient and likely waste a lot of energy in simple straightening for jump.(i.e. you might not be able to utilize most of the distance)
For the landing, the same applies, you want longer distance for spreading the impact, but you also want a shorter moment arm to optimize energy useage. Straight legs can squat down anyway.

I'd discourage jumping a lot. While in a jump, you cannot change your direction, thus your motion is extremely easy to predict and thus becoming sitting ducks against enemy fire.
Even if you have some kind of thrust or aerodynamic method to change direction in mid air, the initial acceleration will be much less than desirable.
I'd put my money on inproving stealth systems and stay low on the ground. In fact, open plains in real life aren't really that flat, keeping low profile actually make you pretty hard to detect, especially we are talking about robots here, they don't feel itches and they don't need to breathe. Staying put and not moving is their expertise. Inching forward in disguise would also be pretty simple.
If you watch documentary, you'd found out that nature has quite a lot of stealthy hunters in open plains.

Roller skates is a system with a lot of dead weight. When you skate, the legs are pretty useless, while when you walk the wheels are useless. You don't want duplicated systems on a military machine if you can avoid it. Yes, its cool, but at the same time the cost of installing all the system will sky rocket and the maintenance will be a headache. The supply of parts might also become a problem and cause you a lot of trouble(like machines gets damage but you don't have the parts to fix it) Hiding all the roller system inside the legs is not a fun engineering problem to solve. Human uses skates because we simple cannot remove our legs and install wheels, that is not the case in robots.

If your robots fight mainly in open plains, then you don't really need to care about legs at all.
Hovering system or at least tracks will be the way to go. Horizontal area is also not that important, you won't get stuck in between alleys, if its two stones, go around it or over it.
Change to legs when you get into a city, you don't need the robots to carry around the dead weight when fighting, carry it in your support vehicle and change the equipment per operation requirements.
Hmm... Then I should probably stick with a twin-track drone... But which size would be good? They would have to be big enough to carry somewhat a lot of ammunition and fuel, since the battles can drag for quite a while, but small enough to be easily replaced/fixed. Something like the Renault FT tank?

But then, in that case, are they a bit too... easy to be defeated by mine, TOW missiles, etc. existing right at this moment in our world? Even if armor technology is better in the future, smaller drones do have to carry a thinner armor.

Also matter is the cross country ability. For example, a drone of that size may not be able to cross a 4 - 6 m ditch.
User avatar
MythSearcher
Posts: 1845
Joined: Sun Jan 20, 2013 4:36 pm

Re: Which design feature should a remote combat robot have?

False Prophet wrote:
Hmm... Then I should probably stick with a twin-track drone... But which size would be good? They would have to be big enough to carry somewhat a lot of ammunition and fuel, since the battles can drag for quite a while, but small enough to be easily replaced/fixed. Something like the Renault FT tank?

But then, in that case, are they a bit too... easy to be defeated by mine, TOW missiles, etc. existing right at this moment in our world? Even if armor technology is better in the future, smaller drones do have to carry a thinner armor.

Also matter is the cross country ability. For example, a drone of that size may not be able to cross a 4 - 6 m ditch.
I'd say keep it to around human weight but a bit smaller(denser).
You don't really need that much ammo since you can just keep them resupplied.
The more you have on them, the heavier and less efficient they are.
On the other hand, if you keep them small and light, they will be cheap and you can swamp your enemy with a lot of them, and if you arrange them in groups to resupply around the clock, you in effect have enough ammo on the field.
If it gets cheap enough, you don't really need to care if they get destroyed by TOW, its the enemy that is spending expensive equipment to destroy your cheap weapons instead of your expensive tanks.
Sadly, that's what human soldiers are, cheap weapons. Compared to armoured vehicles or any vehicles, at least. You really cannot replace the human soldiers with a much more expensive weapon, so small and destructable is better than larger and harder to destroy.
You don't need the robots to be hard to destroy, they just need to be similar to humans in their destructability. Since they don't really feel pain, they will already be better soldiers(at least with a longer combatable state under same conditions) And if they are smaller and presumably faster, they will be harder to hit.

I will personally stick with a Z shape design, the lower horizontal part are the tracks, engine and most of the mass, the slanted part can move up and down thus essentially you can move in a really low profile.
The upper part will be the main weapon, either you use laser weapons to keep the Centre of Mass lower, or you have to supply the weapon with an ammo chain running through the slanted part.
Come to think of it, works kinda like YMT-05? The middle portion will be greater in proportion though.

Most of the time they fight at the lowered position like small tanks, but in terrain with more up and downs, they can extend upward to about 1~1.5m to shoot. This will be less stable and probably lower the effective range by around 50~60%. But a weapon at 30~40cm is really not that useful.

If you want to get through ditches, I'd say find a way to go down, those are good for stealth.
You can also form bridges by extending forward, you need around a dozen of them just to form the bridge over a 4m gap.(most will be on one side acting as a ballast while 3~4 of them stacking on each other to form the actual crossing) So you might need around 20~25 in total to bring everyone to the other side(forming the other side's ballast once through.)
False Prophet
Posts: 955
Joined: Thu Dec 31, 2015 7:40 am

Re: Which design feature should a remote combat robot have?

MythSearcher wrote: I'd say keep it to around human weight but a bit smaller(denser).
You don't really need that much ammo since you can just keep them resupplied.
The more you have on them, the heavier and less efficient they are.
On the other hand, if you keep them small and light, they will be cheap and you can swamp your enemy with a lot of them, and if you arrange them in groups to resupply around the clock, you in effect have enough ammo on the field.
If it gets cheap enough, you don't really need to care if they get destroyed by TOW, its the enemy that is spending expensive equipment to destroy your cheap weapons instead of your expensive tanks.
Sadly, that's what human soldiers are, cheap weapons. Compared to armoured vehicles or any vehicles, at least. You really cannot replace the human soldiers with a much more expensive weapon, so small and destructable is better than larger and harder to destroy.
You don't need the robots to be hard to destroy, they just need to be similar to humans in their destructability. Since they don't really feel pain, they will already be better soldiers(at least with a longer combatable state under same conditions) And if they are smaller and presumably faster, they will be harder to hit.

I will personally stick with a Z shape design, the lower horizontal part are the tracks, engine and most of the mass, the slanted part can move up and down thus essentially you can move in a really low profile.
The upper part will be the main weapon, either you use laser weapons to keep the Centre of Mass lower, or you have to supply the weapon with an ammo chain running through the slanted part.
Come to think of it, works kinda like YMT-05? The middle portion will be greater in proportion though.

Most of the time they fight at the lowered position like small tanks, but in terrain with more up and downs, they can extend upward to about 1~1.5m to shoot. This will be less stable and probably lower the effective range by around 50~60%. But a weapon at 30~40cm is really not that useful.

If you want to get through ditches, I'd say find a way to go down, those are good for stealth.
You can also form bridges by extending forward, you need around a dozen of them just to form the bridge over a 4m gap.(most will be on one side acting as a ballast while 3~4 of them stacking on each other to form the actual crossing) So you might need around 20~25 in total to bring everyone to the other side(forming the other side's ballast once through.)
Ammunition in my setting is the caseless telescoped cartridge type with electric ignition, which makes the total weight of the gun and the ammo carried less than normal gun and metallic cartridges. However, as armor technology has advanced greatly (more on that in a latter post), and most human-manned vehicles have P-Field (telekinesis shield), I'd like the common caliber used by these drones to be around 8.5 - 9 mm, with power excels the .338 Norma Magnum (including improved armor-piercing capability), and most of the drones are armored enough to withstand threats up to 10 mm caliber bullet traveling at 1000 m/s (imagine the .416 Remington Magnum on steroid).

This is actually a shout-out to the current state of the ammunition vs. amour battle - Spartan Ammor System's Level III body amour with AR650 plate can actually stop 7.62 x 51mm at a very close range. And beside, thicker amour (on Exoskeleton suit - again, more on that latter) let me to have a reason for bladed weapons.

About the ditches, I could have an AVLB build on the same chassis, but that does take some time.
User avatar
MythSearcher
Posts: 1845
Joined: Sun Jan 20, 2013 4:36 pm

Re: Which design feature should a remote combat robot have?

False Prophet wrote: About the ditches, I could have an AVLB build on the same chassis, but that does take some time.
Ah, my design is actually a lesser version of that.
You have a bridge about 2 times the length of the robot(not counting the ballast main body part.)
You can stack them up like a pyramid to extend the length.
False Prophet
Posts: 955
Joined: Thu Dec 31, 2015 7:40 am

Re: Which design feature should a remote combat robot have?

So, here is the weapon configurations I intend for the drones:
- Mod. 0: Two co-axial 9.6mm machine guns; two four-tube grenade launchers.
- Mod. 1: One 60mm mortar, one 9.6mm machine gun
- Mod. 2: Two co-axial 7.62mm miniguns, two four-tube grenade launchers.
- Mod. 3: Four Drone-mounted Air-defense System (DMADS) launch tubes, each carries either one CLOS or infrared missile.
- Mod. 4: Two TOW missile launchers.
- Mod. 5: A 50 mm smoothbore Anti Vehicle Gun (AVG), one 9.6mm machine gun, one four-tube grenade launcher.
As you can see, Mod. 0 is for drone-to-drone combat, Mod.1 for artillery support, Mod. 2 for anti-infantry, Mod. 3 for drone-to-air combat (which is actually against flying drones), Mod. 4 and 5 for anti-armoured vehicles.

So, are these the optimal weapon arrangements?
User avatar
MythSearcher
Posts: 1845
Joined: Sun Jan 20, 2013 4:36 pm

Re: Which design feature should a remote combat robot have?

False Prophet wrote:So, here is the weapon configurations I intend for the drones:
- Mod. 0: Two co-axial 9.6mm machine guns; two four-tube grenade launchers.
- Mod. 1: One 60mm mortar, one 9.6mm machine gun
- Mod. 2: Two co-axial 7.62mm miniguns, two four-tube grenade launchers.
- Mod. 3: Four Drone-mounted Air-defense System (DMADS) launch tubes, each carries either one CLOS or infrared missile.
- Mod. 4: Two TOW missile launchers.
- Mod. 5: A 50 mm smoothbore Anti Vehicle Gun (AVG), one 9.6mm machine gun, one four-tube grenade launcher.
As you can see, Mod. 0 is for drone-to-drone combat, Mod.1 for artillery support, Mod. 2 for anti-infantry, Mod. 3 for drone-to-air combat (which is actually against flying drones), Mod. 4 and 5 for anti-armoured vehicles.

So, are these the optimal weapon arrangements?
Wondering if it can carry and fire 50mm AVG, why can't it carry more TOWs.(at least the missiles)
The recoil is pretty strong and obviously it can take that much force so likely weight as well.
False Prophet
Posts: 955
Joined: Thu Dec 31, 2015 7:40 am

Re: Which design feature should a remote combat robot have?

MythSearcher wrote:
False Prophet wrote:So, here is the weapon configurations I intend for the drones:
- Mod. 0: Two co-axial 9.6mm machine guns; two four-tube grenade launchers.
- Mod. 1: One 60mm mortar, one 9.6mm machine gun
- Mod. 2: Two co-axial 7.62mm miniguns, two four-tube grenade launchers.
- Mod. 3: Four Drone-mounted Air-defense System (DMADS) launch tubes, each carries either one CLOS or infrared missile.
- Mod. 4: Two TOW missile launchers.
- Mod. 5: A 50 mm smoothbore Anti Vehicle Gun (AVG), one 9.6mm machine gun, one four-tube grenade launcher.
As you can see, Mod. 0 is for drone-to-drone combat, Mod.1 for artillery support, Mod. 2 for anti-infantry, Mod. 3 for drone-to-air combat (which is actually against flying drones), Mod. 4 and 5 for anti-armoured vehicles.

So, are these the optimal weapon arrangements?
Wondering if it can carry and fire 50mm AVG, why can't it carry more TOWs.(at least the missiles)
The recoil is pretty strong and obviously it can take that much force so likely weight as well.
50mm AVG?

The reason Mod. 4 cannot carry more TOW is because the added detection system. Since even drones in my fiction have superior stealth ability on many wavelengths (which is something aircraft designers have been trying to do for years) and other forms of optical/audio detectors, countering them have become quite hard for the ground drones.
False Prophet
Posts: 955
Joined: Thu Dec 31, 2015 7:40 am

Re: Which design feature should a remote combat robot have?

So, I've pretty much codified how should a land-use combat drone look and works like, let's move to air-use.

I suppose that fixed-wing drones will be the main fighting force (taking the job of fighter, bomber, and attack aircraft), while rotor-use drones will work in reconnaissance, recovery, and other non-combat roles.

Basically not so different from what we're doing.

But the question I'd like to ask is: To what extend is the modularity of air-use combat drones? I've seen some police-use fixed-wing surveillance drones that the size and weight of the unit can be changed. But what about more than that?
User avatar
MythSearcher
Posts: 1845
Joined: Sun Jan 20, 2013 4:36 pm

Re: Which design feature should a remote combat robot have?

False Prophet wrote:So, I've pretty much codified how should a land-use combat drone look and works like, let's move to air-use.

I suppose that fixed-wing drones will be the main fighting force (taking the job of fighter, bomber, and attack aircraft), while rotor-use drones will work in reconnaissance, recovery, and other non-combat roles.

Basically not so different from what we're doing.

But the question I'd like to ask is: To what extend is the modularity of air-use combat drones? I've seen some police-use fixed-wing surveillance drones that the size and weight of the unit can be changed. But what about more than that?
I'd rather have a hovering/wing-in-surface-effect craft rather than land-use drones if air bourne ones are common.

I think I am talking on Mt Stupid here, or at most in the valley of despair, thus I am quite reluctant to reply, anyway, fixed-wing drones will have pretty different looks according to designed speed and size, the square-cube law makes it so that the smaller your craft is, the less scaled wing area it requires to obtain enough lift for its scaled down weight.(notice the weight scales down in cubic and the area scales down in square)

If you want slower drones, a front-swept is good, if you want faster ones, go for delta or back swept.
modularity is only good from less efficient units, in the police case, usually the power difference is huge(police having a huge advantage) but in military combats, I'd be really skeptical if your enemy it as weak as civilian mugs, at least that doesn't seem like a good plot line.
User avatar
Areku
Posts: 1216
Joined: Mon Feb 11, 2008 12:00 am

Re: Which design feature should a remote combat robot have?

MythSearcher wrote:I'd rather have a hovering/wing-in-surface-effect craft rather than land-use drones if air bourne ones are common.

I think I am talking on Mt Stupid here, or at most in the valley of despair, thus I am quite reluctant to reply, anyway, fixed-wing drones will have pretty different looks according to designed speed and size, the square-cube law makes it so that the smaller your craft is, the less scaled wing area it requires to obtain enough lift for its scaled down weight.(notice the weight scales down in cubic and the area scales down in square)

If you want slower drones, a front-swept is good, if you want faster ones, go for delta or back swept.
modularity is only good from less efficient units, in the police case, usually the power difference is huge(police having a huge advantage) but in military combats, I'd be really skeptical if your enemy it as weak as civilian mugs, at least that doesn't seem like a good plot line.
You're mostly right, but it's also worth noting that your payload ratio goes down as you make a smaller fixed-wing aircraft, as it requires a higher thrust-to-weight ratio to maintain any given speed as you get smaller. It's like a subtler, square-cube relationship between drag and thrust, with the smaller wing helping but not quite neutralizing the difference. So, you need a proportionally bigger engine to reach any given speed, driving down the proportion of your payload.

As for wing sweep, depending on just how slow you intend to go, unswept (straight) wings may be your best bet, as they are the most structurally efficient wing and provide excellent control and lift characteristics at low speeds. For example, the A-10 employs unswept wings because they provide the best handling and durability at the low speeds and altitudes that comprise its most important environments, and the characteristics of the wing also help it to maintain control even when it fires the GAU-8 (which recoils with slightly more force than one of the two engines). For drones, an unswept wing will also give you the longest loiter time at low speeds. Just don't expect to be breaking the sound barrier with an unswept wing, at least not with anything approaching efficiency.

Speaking of sweep, the Switchblade is a fun concept to explore.
False Prophet
Posts: 955
Joined: Thu Dec 31, 2015 7:40 am

Re: Which design feature should a remote combat robot have?

Hmm... So I should also taken into consideration the size, weight, and handling of the machine.

Ok. Since I intend for my fiction to have relatively few urban combat scenarios, so bomber is relative few and far between. For precision strike, delta-wing flying drones would be enough (though, which kind of delta-wing is ideal for UAV that is basically half the size of a MiG-21?)

Also, is it necessary to have inner bomb bay in UAV? I want these to have stealth capability, but also want to keep them small and simple to operate.
Post Reply