I'm making a game about robots and I need some sort of hand-wavey explanation for why they can move somewhat quickly in terms of motor control.
My limited knowledge of robotics would make me think that pretty much all power in kinematics would have to be derived from the servos in the joints. Is there an alternate theory on robotic kinematics I can latch onto and "futurize" or is there some sort of fictional tech I can apply to explain why the servos are so quick to respond and can lift such heavy weight?
The idea is that they are controlled by a single human that controls the mechs with some sort of neural interface. They go through intense physical therapy to develop motor controls for these mechs to the point that an experienced pilot will feel like the mech is a part of their physical body. I was thinking I could support this by saying that the humans develop neuronal structures in the motor cortexes as well as the cerebellum to help their control.
So how do I support this on the mech side? I've come across muscle wire which contracts when a charge is rung through it so maybe I could go with that tech and scale it up?
Also the mechs aren't super huge. They can fit a human inside them. They're only like 50ft tall.
Jonathan Johnson
If its a futuristic game just use the nuclear fusion meme to justify the power needed to make them move that fast. Also say graphite makes it strong enough to handle itself.
Lucas Edwards
Yeah I'm not concerned about power so much as the control and actuation.
Hunter Ross
Nobody in a vidya will question that.
But throw in psionics if it worries you. Just the word, psionics, all you need.
James Wood
Carbon Nano Tube Muscles. Input from laser scanners.
Ayden Barnes
>> artificial muscle No. Just no.
Electric motors already have fucking insane power to weight ratios. But this power is in the form of high rpm low torque. This is bad for robots, even the fastest robot ain't gonna spin its limbs around at 10,000 rpm.
What we really need are high torque to weight ratio motors. The simplest way to get this is just a method of efficiently transforming high rpm to high torque in a compact manner.
A new electrostatic motor has been proposed that should have a high torque to weight ratio.
Molecular nanotechnology can supposedly allow the construction of electrostatic motors with petawatt per cubic meter power densities.
So you can handwave things and go lelz room temperature superconductor. Superconductors allow you to make motors with stronger magnets, which means more torque.(superconductors are only used as more powerful magnets, not coils)
Physical therapy is the right way to put it. No matter what, the mech will have different dynamics than the human body. Thus the operator would have to relearn how to walk so to speak.
If you are plugging into to the motor cortex or whatever, the mech need not have all the same muscles as the human body.
Honestly, we don't have a very good idea of what the cerebellum does.
Grayson Russell
That neural interface could be used to control a series of complex gyrostabilizers. For example, the unit will naturally want to maintain balance via hard-coded instructions; an autopilot of sorts will permit movement but none of that funky anime mecha maneuver bullshit. The neural link overrides that and tells the system "yes, I want you to throw yourself in that direction". Would have a steep learning curve but the potential for extremely fine motor control.
Yes, I stole this from Battletech.
Aaron Collins
Mechs only make any sense in null gravity, so make the mech fights happen in space or really low gravity.
Jaxson Taylor
>>> artificial muscle >No. Just no.
biological muscles are the most efficient. But since a version those of those that works in robots don't exist we need to create an alternative. How about servo motors with little hamsters and hamper wheels in them. Each joint has it's own hamster. It'll be the most efficient design ever!
robots tend to fail in 3 ways. • actual cost, as in money • power supply • cooling
Even if you have infinte money, getting a power supply to fit inside a mecha is difficult. That's why neon genesis evagelion were powered by cables. But then even if they're powered by an external source, with regular use servos will become damaged from heat. 3rd problem can be ignored for video game, unless you want to implement heat management as a game mechanic. 1st can also be explained as simple progression of technology. As items become more evolved and popular the costs tend to go down. 2nd problem can be solved by clicking on /x/ and asking for alternative energy or over unity energy.
Elijah Bell
I was just going to go with the le compact nuclear reactor core meme for the power source. That or "quantum batteries" which are apparently a thing and can allegedly store a fuckton of power and information.
The cost is high but you play as a paramilitary mercenary group that recieves funding from multiple governments (or whoever has deep enough pockets to request a job). I think that is enough to cover it in-universe.
I was going to ignore cooling more or less except for energy weapons.
>> artificial muscle >No. Just no.
Can you explain why not, exactly?
I appreciate the electrostatic motor explanation. I'm also using ball and socket joints too so I'm wondering how I could explain those. Maybe a magnet system also explained away with room temperature superconductors?
I'm no neuroscientist but I'm pretty happy with my explanation. I'll take your advice and avoid specifically mentioning parts of the brain though because those do seem like pitfalls. Thanks for that.
Noted.
Samuel Adams
also for energy weapons I'm going to make "plasma" weapons fire superheated slugs since that's the most convincing version I've heard (thanks, Deus Ex).
I also want to do laser weaponry but I don't understand how they're damaging or what sort of effect it'd have on an armored robot. Is it type of burning? I imagine there is no stopping power.
Jonathan Cruz
Photons have momentum, so If you want a laser to have stopping power, it can. Gonna need a lot of photons though... Google solar sails for a demo.
Alexander Hughes
fuck, I was about to make this exact same thread. I was wanting to make a fighty robot game myself. I thought I was being cool and original.
Any more info on your game, OP?
Lucas White
I don't think laser weaponry is going to be particularly effective against heavy armor, like what you'd probably expect from a 50 foot combat mecha. My opinion's based on projectrho, but I think it's sound.
About a fifth of a way down the page, ctrl+f 'holes of increasing aspect ratio'.
In short, the ejected detritus from each blast'll make a cone that makes it harder to drill through.
Of course, artistic license and gameplay though.
Brayden Watson
Fuck it user. There's too many problems with the physical logistics. Just use magic, like in Guilty Gear.
>In the year 2934 humans discovered magic. >Over the next several decades they figured out how to combine it with technology to create ridiculous superweapons and also penis enlargement supplements that really work. >It was an age of wonder. >And then the robots came!
Caleb Robinson
Instead of making the mechs big, make the humans small. Like this entire war is being fought on a table top by shrunken down soldiers and figurine sized mechs.
It doesn't actually solve any of your problems but you've contorted the problem in a way that will satisfy most pleb gamers. For real, /v/ is legit the dumbest board on Veeky Forums after /pol/.
Eli Fisher
>Honestly, we don't have a very good idea of what the cerebellum does. pretty much the opposite of correct
Ayden Barnes
Thanks for these. I guess I'll just have to follow the "rule of cool" and throw science to the wind when it comes to lasers.
Not much. I'm loosely a part of /agdg/ but I don't go to the threads. I you want an indie mech fix there is Project Nimbus on Steam and then there is another user making a game called Project M through /agdg/ and you can find a demo on the demo day pages (this is NOT my game).
I don't want to go full magic but there's always Clarke's Law.
I'm not really a fan of this idea. Plus it's easy to have everything to scale since I can just work directly with meters in my development pipeline.
Christopher Walker
You could still make lasers useful if they don't have good anti-armor properties. Just because they can't penetrate a tank or a mecha doesn't mean they don't have applications in shooting down missiles, aircraft, or lighter targets (E.G. soldiers, civilian targets).
Daniel Torres
Lots of small thruster holes on arms, legs, torso, etc. Aesthetically pleasing, open ended design platform, grounded in reality.
Camden Flores
Nanomachines
Jeremiah Collins
>>biological muscles are the most efficient. Lol no. Biological muscle is 30% efficient at converting chemical energy to mechanical energy. Electric motors have achieved efficiencies of >robots tend to fail because of cooling industrial robot arms are for the most part, passively cooled. Cooling isn't much of a concern, except for the CPU
>>Can you explain why not, exactly? Geometry. What we want out is rotational motion and torque, muscles produce linear motion. It is better to produce rotary motion out. Pic related is a schematic of an artificial muscle system. As the muscle pulls closed , the muscle pulls off perpendicular. This means muscle power is wasted because it doesn't go to torque.
>>I'm also using ball and socket joints too Spherical electric motors exist. I have even seen proposals for robot arms with spherical electric motors.
Ok so on superconductors, superconductors can only act as really powerful static magnets. We can't really switch superconductors on and off, because that's not how superconductivity works. This is a pain in the ass, because we can't just take a regular motor and add superconductors and expect everything to work better. Having fuck powerful static magnets helps a lot though.
But, your ball joints can be sufficiently hidden away that no one will be able to tell if they use weird as spherical motors. Asimo has ball joints, but it doesn't use spherical motors to do so.
we sort of think it does something involving internal state estimation, feed forward models, skill learning, or some combination thereof. It has 3/4 of the neurons in the brain yet only takes up 10% of the space in the brain. We have yet to determine why.
I don't think.you realize how powerful electric motors are. You won't have a problem moving fast, just make them out of carbon nano tubes or some magic shit. Also, just have a strong AI manage the interface between human and machine, better yet, just have the player be a strong AI.
Adrian Rodriguez
>I don't want to go full magic but there's always Clarke's Law. When you find yourself nitpicking at these details you've maybe got rework your focus and figure out what sort of story you want to tell.
