Anyone ever hear of Battletech?

[Seret]

Well, bipedal / quadrupedal robots will become really useful the day they will be able to run efficiently on various terrains

I strongly suspect on most terrain they'd be outperformed on fuel efficiency by wheels. Range is a very important factor for the military. Walking is inherently inefficient because you're having to accelerate the whole mass of the vehicle vertically, even if you can get your leg mass down low enough to make moving the legs relatively cheap.

But yes, being aware of where the ground is is pretty important to a walking robot, along with the relative positions and accelerations of its legs, chassis, etc. There's a lot to process and react to at very high speed, which is why we're only just starting to develop reasonably capable ones now. Don't forget that all these Youtube videos of walking robots are advertising a commercial product. They don't show you the bits where the robot slips over, lands on it's robotic butt and can't get up again.

I don't think there would be any point in bipedal locomotion for military use (exoskeletons notwithstanding). Four or more legs would give better stability, lower profile, lower ground pressure, more weight carrying capacity and more redundancy. Even wheeled AFVs tend to have six or eight wheels.

[Red Iron Crown]

Have you guys looked around the internet yet? There are some interesting articles out there

That was a good read, thanks for the link. Interestingly, it cites wheeled vehicles as faster and tracked ones as better over rough terrain than walkers; the main advantage of walkers seems to be climbing over truly treacherous terrain.

Big Dog's robots are awesome and a bit scary. I feel like we're seeing early Terminator prototypes.

I can not let you get away with that one Bipedal motion is more efficient than wheeled propulsion, the latter just uses a lot more power to overcome that. Walking costs barely more energy than merely standing as it is basically a pendulum motion that uses its own energy to move forward. In short, you just control falling forward. Driving, on the other hand, means continuous friction between the wheel and the surface, friction within the wheel (deformation) and loss of energy due to the suspension. You also tend to touch the ground with more wheels at the same time, increasing the cost.

Legged vehicles can't coast. A wheeled vehicle can roll down a hill at virtually zero energy cost, similarly it can coast for quite a distance on level ground without expending much energy. Legged vehicles have to exert almost the same amount of energy to go down a hill or carry on at the same speed as they do to accelerate. They even have to exert energy at rest to maintain balance.

[Red Iron Crown]

Anyone interested in man-sized exosuit style mechs would do well to check out John Ringo's Posleen series. Alien tech enables power suits where the pilot is suspended in a biological gel that handles sustenance and waste removal as well as medical functions, enabling effectively unlimited endurance for the pilot. They use bipedal locomotion for ground operations, with powerful jump jets for quick bounds to make up for the speed penalty of legs. IIRC, they can also fly but the efficiency is low as they have now wings and are rather massive.

[KASASpace]

Okay, let me explain the reasoning for quicker reaction time:

The cockpit is fighter-aircraft like, with a similar HUD because the cockpit is mounted on the torso, which is a turret and can thus turn in many directions.

This HUD gives a great deal of information, and chances are tanks will not have such a HUD because only the gun barrel moves up and down, not the turret itself.

[KASASpace]

I strongly suspect on most terrain they'd be outperformed on fuel efficiency by wheels. Range is a very important factor for the military. Walking is inherently inefficient because you're having to accelerate the whole mass of the vehicle vertically, even if you can get your leg mass down low enough to make moving the legs relatively cheap.

If range were so important, then why did the Luftwaffe have 5oo km on the Bf-109?

Besides that, it would actually use a kind of battery being charged up for sprinting.

Now, range is actually better because the motor doesn't need to be as big as 60+ ton tank's. It merely needs enough power to use pumps and small pistons for the hydraulics of the vehicle.

[Camacha]

I strongly suspect on most terrain they'd be outperformed on fuel efficiency by wheels. Range is a very important factor for the military. Walking is inherently inefficient because you're having to accelerate the whole mass of the vehicle vertically, even if you can get your leg mass down low enough to make moving the legs relatively cheap.

That is pretty much what my earlier post was about. On firm, flat, paved roads wheels have an advantage. When the terrain becomes rougher, they start losing terrain as they have shorter suspension travel, meaning more movement of the body/chassis itself. The shocks also absorb a lot of energy, and wheel deformation and the sum of the multitude of wheels also come into play. This quickly adds up to a vehicle that bogs down (and expends a lot of energy) in places where legs do not. Long story short: wheels are great when you have proper surfaces, but start losing out in more demanding circumstances.

I am not sure where you get the idea that you would need to travel vertically when walking or running. Sure, there is possibly some bobbing, but the point is that you can keep your upper body quite steady while moving the legs beneath. Moving the legs is cheap, as its a pendulum motion. The return barely costs anything (and is friction free).

Not to be a hardass, but I think I mentioned all of these things before

But yes, being aware of where the ground is is pretty important to a walking robot, along with the relative positions and accelerations of its legs, chassis, etc. There's a lot to process and react to at very high speed, which is why we're only just starting to develop reasonably capable ones now. Don't forget that all these Youtube videos of walking robots are advertising a commercial product. They don't show you the bits where the robot slips over, lands on it's robotic butt and can't get up again.

I don't think there would be any point in bipedal locomotion for military use (exoskeletons notwithstanding). Four or more legs would give better stability, lower profile, lower ground pressure, more weight carrying capacity and more redundancy. Even wheeled AFVs tend to have six or eight wheels.

The calculative power required is not a problem - that is just a matter of time. Do not forget we already have quick robot arms that catch thrown rotating objects without a hitch. This is a bit more complex, but when you look at other developments you see that once you get the idea, things are quickly solved.

Finally, there is one good reason why bipeds might enter the army: the entire infrastructure is designed to deal with them. Instead of redesigning everything, you simply make the most of what you have. Choices are often made for practical reasons instead of being the ideal option.

Edited March 23, 2014 by Camacha

[Camacha]

Legged vehicles can't coast. A wheeled vehicle can roll down a hill at virtually zero energy cost, similarly it can coast for quite a distance on level ground without expending much energy. Legged vehicles have to exert almost the same amount of energy to go down a hill or carry on at the same speed as they do to accelerate. They even have to exert energy at rest to maintain balance.

Again, you are assuming easy sailing. Put a wheeled vehicle in sand or rocky terrain and the rolling quickly stops. Walking only costs a little more energy than standing in humans, so I would not be surprised if the same could be said for mechanic walkers. Idling is probably not a problem, as you can design some sitting/locked mode - a bit like how some birds lock their wings in flight.

[KASASpace]

Hmm, a tank usually will not roll well as the treads are designed to have as much traction as possible..........

Of course you might actually be able to conserver energy in the legged vehicle by using forward momentum to pick up the rear leg.

[RuBisCO]

Why not 4 legs or 6 legs or 8 legs, then you can have legs blown off completely and still be able to move around.

[Sillychris]

Why not 4 legs or 6 legs or 8 legs, then you can have legs blown off completely and still be able to move around.

The legs should have bayonets, too. Maybe even lightsaber bayonets or a similar can opener attachment.

[Seret]

That is pretty much what my earlier post was about. On firm, flat, paved roads wheels have an advantage. When the terrain becomes rougher, they start losing terrain as they have shorter suspension travel, meaning more movement of the body/chassis itself. The shocks also absorb a lot of energy, and wheel deformation and the sum of the multitude of wheels also come into play. This quickly adds up to a vehicle that bogs down (and expends a lot of energy) in places where legs do not. Long story short: wheels are great when you have proper surfaces, but start losing out in more demanding circumstances.

Sure, I don't disagree with any of this. The issue is whether on balance legs offer a net advantage. It's obvious they're an improvement over some types of terrain, but they're also sub-optimal over many others. To me it really seems like you'd only want legs if you were going to be spending a significant amount of your time in the roughest of rough terrain. For most vehicles that's not the case. I'm pretty sure legs will be remaining a specialist option, with wheels and tracks remaining the default.

I am not sure where you get the idea that you would need to travel vertically when walking or running.

You do, unless you can somehow lengthen your legs during the stride! Consider the motion through space of a hip joint, it will rise to a maximum as the leg under it reaches the vertical position, with a minimum height when the leg is at furthest extension to front or rear. On a biped that's mitigated by twisting the pelvis during the stride, but it's just one of a ton of little extraneous movements that sap energy. Watch over the top of a body of people as they walk if you don't believe me. The heads do bob up and down, they sway side to side, etc.

Not to be a hardass, but I think I mentioned all of these things before

You can be hardass all you like, as long as you're keeping it civil, realistic and interesting. Which you are.

The calculative power required is not a problem

Not in terms of processing power, no. It's a control problem, it's about finding the right algorithms. It looks like good progress is being made, in the last few years we've moved from robots that can walk to ones that can run.

Finally, there is one good reason why bipeds might enter the army: the entire infrastructure is designed to deal with them. Instead of redesigning everything, you simply make the most of what you have. Choices are often made for practical reasons instead of being the ideal option.

Are you talking about walking vehicles, or exoskeletons/human scale robots? That's definitely a consideration for the latter, although it's more of an issue for manipulators than locomotion. Robots with treads can get around our infrastructure fine as long as they can go up stairs, but being able to operate door handles, valves, etc designed for hands is a definite advantage for a robot. DARPA's recent rescue robot competition is a good example of why being anthropomorphic might sometimes be a good thing.

If you're talking about vehicles, all the infrastructure is designed for wheels, so that's definitely a vote for wheels. Just look at the fraction of our urban spaces we devote to roads, and military transports are designed for roll-on roll-off.

Walking only costs a little more energy than standing in humans

Most of the sources I'm looking at put standing at about 1.2 MET, while walking is more like 2.5. So walking takes about twice the energy of standing if you include the base metabolic load. Or to put it another way, standing takes about 20% extra energy above the baseline, while walking takes 150% more. So in humans the process of walking costs about seven times more than standing,

Edited March 23, 2014 by Seret

[KASASpace]

Why not 4 legs or 6 legs or 8 legs, then you can have legs blown off completely and still be able to move around.

With that many legs you would have one heck of a time maneuvering and even worse maintenance issues.

[Sillychris]

With that many legs you would have one heck of a time maneuvering and even worse maintenance issues.

I both agree and disagree with you, as both statements are partially true and partially false.

In regard to maintenance, 8 legs would be a more complex mechanical system than 2 legs. However, with 8 legs the reliability of any individual leg is not as important. Thus the mechanical maintenance would be more complex but not as important. In regard to stabilization maintenance, 8 legs is a piece of cake to control whereas bipedal motion is actually a very challenging problem. The computer systems and programming required to command more legs than 2 is ironically simpler than strictly 2.

In regard to maneuvering, I think it really is situation specific. In tighter quarters, 2 legs is better. On loose or sticky/sucky ground, more legs is better. I'm not sure if 2 or 8 legs would be better on steep terrain, though. I tend towards more legs on steep terrain since when I go rock climbing I have to use all 4 limbs and hence I become a 4 legged vehicle. Of course, if your mech's gun arms had hand like objects, this would eliminate that problem.

I think we need a mech modeled after a scorpion. Whole lotta legs, two light guns in the front and a big badass rear mounted gun. Nice and low like a tank, too.

[magnemoe]

With that many legs you would have one heck of a time maneuvering and even worse maintenance issues.

4 or more legs has multiple benefits: first you don't need so advanced legs and an advanced stability system. Look at the difference on the legs of four legged and two legged robots, the two legged ones need advanced feet, the four legged ones uses just pads, in short you only need an knee joint and a simple forward/ back hip joint who can be rotated a bit for turning.

You have an low center of mass: harder to hit, less damage if it falls, the possibility to add wheels on the body for road movement.

For military use I hardly see any use of an two legged robot. An two legged robot has benefits in thigh areas designed for humans nowhere else. The military solution for checking out an house would be an miniature robot.

[Mecha Pants]

You have an low center of mass: harder to hit, less damage if it falls, the possibility to add wheels on the body for road movement.

Another option would be have the legs end in wheels, and thus have the best of both worlds. Just lock the wheel down when you're using it as a foot, and use the legs as articulated suspension when you're driving along.

[Seret]

Another option would be have the legs end in wheels, and thus have the best of both worlds.

That would add a lot of complexity and mass to the leg, reducing efficiency. Running a driveshaft down the leg would be impractical so realistically you'd be talking electric hub motors. They're heavy things to swing around at the end of a leg (not to mention the high unsprung mass reducing suspension performance when using the wheels). You want to keep the mass of the legs as low as possible. Magnamoe's suggestion of putting wheels on the hull is a bit more practical if you want a hybrid wheel/leg vehicle.

[Sillychris]

Running a driveshaft down the leg would be impractical so realistically you'd be talking electric hub motors. .

I think driveshaft technology could be seriously improved by the time mechs become feasible. Perhaps more lightweight methods of transferring power could be developed... unless they already exist?

With a little bit of airtight tubing, you can spin a lightweight air motor with extremely high power/weight while storing the powerplant wherever you like. This is the principle of air-hand tools which are preferred for high power applications where the size and weight of an electric motor would be prohibitive. So the technology for lightweight drive wheels on the feet already exists today and it will only improve with time.

Of course, the mech could always have free wheels on the feet equipped strictly with brakes and derive it's propulsion from thrusters of some type near the center of mass. To avoid reaction mass, a super-efficient electrically powered future propeller could be utilized

[Awaras]

Of course, the mech could always have free wheels on the feet equipped strictly with brakes and derive it's propulsion from thrusters of some type near the center of mass. To avoid reaction mass, a super-efficient electrically powered future propeller could be utilized

Or it could just propell itself with leg motions, like a person on rollerblades...

[Sillychris]

Or it could just propell itself with leg motions, like a person on rollerblades...

I nominate Awaras for best mental image award.

I eagerly await the day of rollerblading mechs! They could have ski/skate attachments for colder regions, too.

[KASASpace]

4 or more legs has multiple benefits: first you don't need so advanced legs and an advanced stability system. Look at the difference on the legs of four legged and two legged robots, the two legged ones need advanced feet, the four legged ones uses just pads, in short you only need an knee joint and a simple forward/ back hip joint who can be rotated a bit for turning.

You have an low center of mass: harder to hit, less damage if it falls, the possibility to add wheels on the body for road movement.

For military use I hardly see any use of an two legged robot. An two legged robot has benefits in thigh areas designed for humans nowhere else. The military solution for checking out an house would be an miniature robot.

Actually, we have more than the technology required to build the legs, even if bipedal, because it isn't complicated, but it is advanced.

Ever hear of Isogrids? They're used in rockets, and so they have a good structural quality compared to a normal skeleton.

Ever hear of HyperCarbon? No? Then I suggest looking it up. It is in it's effective infancy, but it is lighter, stronger, but more expensive, than titanium.

Now, for the smoking gun..........

BuckyPaper. A carbon nanotube sheet that currently shows promise as armor for multiple applications.

For what I call "first generation" mechs hydraulics and RH Armor will be used, as well as titanium isogrids. (c. 2025-2045)

however, second generation mechs will use hydraulics and perhaps a fiber of sorts that acts much like a tendon, and titanium isogrids. (c. 2035-2055)

Third generation will have tendon-fibers and bucky paper armor. Titanium structure. (c. 2045-2065)

Fourth generation will have a hypercarbon structure. (c. 2055-2075)

Fifth generation, Z-pinch fusion reactor. (c. 2065-2085)

Of course that's just me.

Now, I think both argument sides in the discussion of tanks vs. mech are wrong, as we shouldn't be comparing the two, but rather discussing where a mech would fit in an army.

[DaveofDefeat]

I could see a mech as a useful maneuverable weapons platform for counter insurgent operations. Perhaps with a highly modular design that would allow you to equip different weapons packages to suit a specific mission.

[Sillychris]

Now, I think both argument sides in the discussion of tanks vs. mech are wrong, as we shouldn't be comparing the two, but rather discussing where a mech would fit in an army.

A mech would be best suited fighting giant monsters, I think. Like Kraken and giant fire breathing worms. They'd also be great for morale and propaganda.

[Seret]

I think driveshaft technology could be seriously improved by the time mechs become feasible.

They probably will, improving current driveshafts is a materials issue. A non-metal with better torsional stiffness per unit weight that was as practical as a metal would find its way into driveshafts quickly.

With a little bit of airtight tubing, you can spin a lightweight air motor with extremely high power/weight while storing the powerplant wherever you like. This is the principle of air-hand tools which are preferred for high power applications where the size and weight of an electric motor would be prohibitive. So the technology for lightweight drive wheels on the feet already exists today and it will only improve with time.

Hey, I like pneumatics, pretty much everything I've worked on in my career has had pneumatic bits in it somewhere. However, they don't really scale up straight from windy tools to vehicle propulsion. Pneumatics aren't particularly efficient or lightweight (once you start chucking in compressors, reservoirs, surge tanks, etc) and fairly bulky, which is a no-no for AFVs. They're good for high-impulse intermittent use, but aren't suited to propulsion.

Bottom line though, even if you had motors and power transmission that weighed nothing, having a bloody great heavy wheel and tyre at the end of your leg would be sub-optimal. Best idea is the one above to simply put the wheels on the hull instead of the end of the leg. Much simpler, completely doable without invoking future tech, and you'd be able to fold the legs when not in use to get a vehicle that was similar size to a normal wheeled one (so narrower and lower).

[KOCOUR]

As hard as i try, i just cannot find any reason for existence of a Mech.

What would the benefits be?

Pros:

It might carry huge weapons.

Well one soldier with SOFLAM and longrange radio supported by longrange artillery, and one CAS plane or military satelite can do Hugeweapons strikes much better (he is small and not so easy to spott) and cheaper.

have overviev of battlefield because its tall.

Planes and satelites are better in that.

and looks awesome.

looking awesome is not really tactical advantage. Actually its better if you are not seen at all.

Cons:

Big easy target

Too complicated.

Either really fragile or too heavy for anything.

I did play mechwarrior and i enjoyed it then, i tried Mechwarior online, it was fun too, I have read few Battletech books, And Robot Jox was one of my favorite films as a kid, but iam also a millitary freak. I play Arma alot, i used to play Airsoft when i still had time for it, iam still activelly involved in historical combat reananctment, and iam training with real pistols and Assault rifle twice a year. And in light of what i know about combat and tactics, mechs IRL doesnt make any sense to me.

What make sense, is advanced comunication, aoutonomous drones, longrange missiles, and skilled, properly equipped forward observers.

Also Powerarmor like in Fallout or in StarshipTroopers (the book) or Combat space suit like the ones in Alastair Reynold´s Revelation Space might be usefull, If they would be possible.

[KASASpace]

Has anyone heard of a Tyrannosaur?

It weighed more than 6 tons, no matter long, short, metric, or imperial.

It could run, too.

Now, let's say you could control one on the battlefield.

And then you put a bunch of missiles and guns on it.

*No advantages* as SOME people would say.....

Even though it will strike fear through anyone who faced it?

Psychological warfare.

In response to KOCOUR:

Pros:

1.) Battlefield maneuverability increase

2.) Heavy payloads over rough and hard to traverse terrain

3.) Heavy infantry support

4.) Can see more of what is in front of you

5.) Can turn on a dollar coin (dimes are too small )

6.) Can actually crouch for cover

7.) Imitates almost all possible human motions

8.) Can't be knocked down as easily due to calculations which constantly check the mech's attitude (orientation) and a Gyro

Cons:

1.) Bigger target

2.) Not really big guns, as too much kickback for mech to handle

3.) not much in the way of speed

The mech has one axes that it is a large target, whereas a tank has two axes of the type.

Not to mention the mech can sustain multiple hits that would knock it over.

On to design, I think some type of inverted legs would assist in running, perhaps something along the lines of a Tyrannosaur's leg shape?