PDA

View Full Version : has anyone ever tried to build a mech



wind27382
02-14-2008, 03:13 PM
just a little curious I know that with today technology it's impossible. but I wanted to know if anyone has ever tried to build a mecha, or a gundam, or landmate. etc. because if you can pull it off i want to be the first to get the video.

MYKL
02-14-2008, 03:21 PM
There a quite a few people futzing around with building one. You saw the one that the guy in Alaska built right?

I don't see it as impossable with what we have avialable today. I believe it is just a matter of the right talents and drives meeting with the freedom to create.

You have to give the world a reason to support anything you create.

asbrandsson
02-14-2008, 07:27 PM
Hello,

I think that that hits the nail on the head. If you really want something to take off it has to have a justification for people to want it. The guy in Alaska created something that is kind of a neat toy, like the robot car that eats other cars. It is cool to look at, but has not particular purpose or real use.

Asbrandsson

MYKL
02-14-2008, 07:51 PM
Don't get me wrong though, think there are some real uses for an ambulatory machine.
Firefighting. Rescue. Forestry. Entertainment. Construction.

It would of course instantly be assimilated by whatever government it was built under the protection of...

Alex
02-14-2008, 08:57 PM
Has anyone every seen this mech:


http://www.youtube.com/watch?v=FtovadY4lYk

I still find it a little hard to swallow though...

DresnerRobotics
02-14-2008, 09:51 PM
^ Ive seen that before in other videos. "Weapons" were completely fake before, maybe they work now... but it doesn't actually walk, it has wheels built into it's feet and it just sort of shuffles.

MYKL
02-15-2008, 11:09 AM
Yeah, its sk4t3r b07... Heh.

But it is closer than we were before to actualizing an full sized mech.

Our little humaniod bi-peds are barely past the shuffle thing themselves. Put my RN-1 on a soft carpet and he becomes a different bot. Put him in the grass and he's done. The wheeled and multilegged platforms are still vastly superior. Sure I can program my little bot to walk up stairs. But my Radio controlled stadium truck can drive right up those stairs and make a four point landing when it launches off the top.

I think we need to rethink the construction of the skeletal system of our little bots. I am thinking about getting some foam carving it into the shape of a Landmate type miniature mecha and then layering this with fiberglass likethe animatronics/puppet guys do. You wait for the sculpt to cure then you cut it open at appropriate junctures and fish out the foam. You are then left with a light weight yet very resilient exoskeleton within which to mount the electrics and actuators. IMHO the electrics, batteries and most of the servomotors can be located with the main trunk and upper limbs. Like you and I.

This place and the company found here stirs the mind...

^_^

Matt
02-15-2008, 11:30 AM
I'm totally not posting this to be inflammatory, so no offense is meant to anyone who likes Mechs. I just thought this article was funny and also makes many good points. So I'm just throwing it in here to add to the discussion. Go ahead and agree, disagree, whatever...

Sci-fi rant: Why giant mecha robots are stupid by Jay Garmon (http://blogs.techrepublic.com.com/geekend/?p=1148)

Dave
02-15-2008, 11:54 AM
I see I'm a little late to this thread, but if you're interested in Mechs, you should check out Neogentronyx (http://www.neogentronyx.com/). It's just one dude working on it, and he ran out of money so the project is stalled. Anyhow, it may not work, but it looks bad ass.

http://www.trossenrobotics.com/images/forums/tn_img_0218.jpg

It also looks extra brutal since it's kind of rusty and covered in snow. It still looks like it could come to life and start killing at any moment. On a side note; if you don't want to be killed by marauding robots, you should join the resistance (http://forums.trossenrobotics.com/showthread.php?t=1421).

MYKL
02-15-2008, 12:49 PM
Thats the one I was talking about in Alaska! There are videos of it shuddering about...

It was a valiant attempt. It would make a fine steampunk decoration.

There are so many lightweight yet structurally sound alternatives to this kind of thing. You just shouldn't expect it to be easy to make a forklift walk. Have you seen the Big Dogs? I think they are called mules too.

http://images.google.com/imgres?imgurl=http://www.defensetech.org/archives/images/2xWalker.jpg&imgrefurl=http://www.defensetech.org/archives/cat_drones.html&h=289&w=259&sz=29&hl=en&start=11&tbnid=19x1H6NAZsBmQM:&tbnh=115&tbnw=103&prev=/images%3Fq%3Djohn%2Bdeere%2Bharvester%2B6%2Blegs%2 6gbv%3D2%26hl%3Den%26safe%3Doff%26sa%3DG

John Deere/Plustech has a four legged forestry machine that is cool as heck.

http://images.google.com/imgres?imgurl=http://forevergeek.com/images/walk99big.jpg&imgrefurl=http://forevergeek.com/gadgets/the_walking_forest_machine.php&h=303&w=400&sz=33&hl=en&start=1&tbnid=iGsPqoJO1iHSrM:&tbnh=94&tbnw=124&prev=/images%3Fq%3DPlustech%2BWalking%2BMachine%26gbv%3D 2%26hl%3Den%26safe%3Doff%26sa%3DG

I read the Anti-Mech post Matt posted a link too. Sadly the mans arguement is logically sound. But, Before the combusion engine a powered wagon was chastised as being a waste of space and energy because you could ride a horse or even a bicycle with more efficiency. The face of this world was changed by artests who value innovation. The fun factor of making a unique way to do something mundane is undeniably what makes the human mind the most impressive example of design. Get on a freeway ad look at the vehicles and drivers. A person drives what makes him/her feel good in most cases even if it is not the most efficient trasport they could use. Teen aged persons driving overpowered SUVs with reduculous sound systems... They could be on a scooter with a hybrid motor...

I am not a proponant of the whole accidental, survival of the fittest sadness that really is a downgrading slight to a wonderfully designed plane of existance. I am also not a religious zealot. I just, as an artest and an engineer of sorts cannot ignore the fact that we as creators have yet to construct anything that is as impressive as even the most base lifeform.

So why limit yourself to Logical, Ergonomic, Efficient, Reasonable ETC... creations?

Make your frickin' Armoured suit and trot it out into the street and touch the lives of your fellow man. Thats what artests have been doing since the begining. There is no efficient reason to paint a picture of a mountain or a beautiful woman. Go and see it yourself. There is no logical reason to create a tool that produces music and to sing along with it.

There is only our instinctual desire to create and please one another.

Thats reason enough for me to keep making new things. Like giant mecha that are stupid...

^_^

asbrandsson
02-15-2008, 07:23 PM
I'm totally not posting this to be inflammatory, so no offense is meant to anyone who likes Mechs. I just thought this article was funny and also makes many good points. So I'm just throwing it in here to add to the discussion. Go ahead and agree, disagree, whatever...

Sci-fi rant: Why giant mecha robots are stupid by Jay Garmon (http://blogs.techrepublic.com.com/geekend/?p=1148)

Hello,

I think that this article is way off base and shows that alot of people are not aware that Robotics has come a long way in the last while. Some strides are made by the day. Bipedal Robotics is moving at a great pace. Even hobby robots can climb stairs now. I think that real change will come with wider use of pico and mini pc motherboards as well as quad core and up processors. A robot like Asimo uses 10 pcs to operate, which I am sure use single core processors. I am sure if it used 5 quad core processors it would be able to work faster and stronger.

Asbrandsson

Alex
02-16-2008, 01:26 AM
Well said MYKL!

I have to admit, I was more on the side of humanoids being an impractical way of creating a robot, I am and always will be a HUGE fan of humanoids, but still I thought they were a bit impractical. However, after reading your post you really gave me a different perspective on the whole discussion. I never was an artist myself, so it really helps me to think outside of the box when I listen to an artist's perspective on a matter.

thanks:)

jdolecki
02-16-2008, 09:29 AM
With todays technology you could build one. Companies like Catapiller, Case, Komatsu have the resources and technology.

But why would you want to? The energy expended to run the thing would surpass the value of any job it was doing.

Are you going to build one to walk to work? Fight wars? Farm? Until a Power source that capable of providing cheap, unlimited supply is devloped I think they will just remain a novelty.

wind27382
02-16-2008, 03:28 PM
I was just wondering. i know that right now it's real impractical to construct a mech. but I meant more towards a landmate. their so much smaller so I feel that they would take a lot less resources. Not to mention they have far more capabilities. such as only being ten feet tall so they will be able to used within city limits without causing mass panic.

MYKL
02-18-2008, 10:43 AM
I believe the first paractial suits will be like landmates. There are already power assisted suit in development for handicapped persons and for loading applications. The next step is a lightweight powersuit with man-machine haptics that allow the pilot to move directly through the machine. The 'armour' could be simply for protection from heat/cold/atmoshpere/debris. They don't need to be the plated juggernaughts that some of the animes depict. Powersources will evolve with demand. Global warming is causeing a flurry of alternative energy source research. Build a usful machine and someone will see profit/benefit in spending time/resources upon developing a powersource to make it work.

Ryan.Marlow
02-22-2008, 10:57 PM
Don't get me wrong though, think there are some real uses for an ambulatory machine.
Firefighting. Rescue. Forestry. Entertainment. Construction.

It would of course instantly be assimilated by whatever government it was built under the protection of...

Here's another project you all may not be familiar with: Mechanized Propulsion Systems, Inc. (found online at www.MechaPS.com (http://www.MechaPS.com)). We (I have actually been a part of the Team now for three years) are the world's first company dedicated specifically to developing commercially viable bipedal heavy vehicles, or mechs.

What we're doing is very much grounded in both real technology, on the production side, and in real market needs, on the sales side. We are not, as the "Giant Robots are Stupid" columnist might expect, building 100 ft. tall, invincible flying things for a cool half-billion dollars, nor are we making interesting lawn ornaments that move but serve little purpose. We have been building a prototype vehicle using existing, off-the-shelf technologies, and designing it to serve not just one function but a multitude of functions (in fact, task adaptability will be a major selling point). The above quote about " Firefighting, Rescue, Forestry, Entertainment, Construction" could not be more on point - those are all among the fields we have already done preliminary market research in, and with very promising results, I might add.

The MPS Mark I (prototype) is about 90% complete for parts, and 30-40% on software, though you sadly won't find any pictures of it on our site (A. IP protection, B. it's hideous, not the best PR). Attached image is a concept render of what we expect the third version of the machine to look like - very similar in structure, just a lot easier on the eyes. What we're doing usually stirs up a lot of skepticism, but that's expected, so if you have questions I'd be more than happy to answer them.

-Ryan Marlow
Public Relations and HR
MPS, Inc.

P.S. As for the government assimilating our project, at this point that would be nice (funding is a wonderful thing), though we expect that our AI software will catch their eye long before the bipedal vehicle concept has had time to catch on.

asbrandsson
02-23-2008, 03:28 PM
Hello,

It is interesting to see the mech that you are working on. I think that often times the government is like the bank they do not understand development of non-traditional industries.

There must be a lot of off the shelf parts, like hydraulics and commercial vehicle cabs. I could see how development of these products is much more expensive then hobby humanoid robots which can be developed on a shoe string buget.

Asbrandsson

asbrandsson
02-23-2008, 03:34 PM
The energy expended to run the thing would surpass the value of any job it was doing.

Hello,

That is not really true, an Hydraulic system does not really use more or less power then any other system. I do think though that it would be hard to develop a mech that can run a hundred kms an hour like a more traditional automobile.

Asbrandsson

Ryan.Marlow
02-24-2008, 01:04 AM
Hello,

That is not really true, an Hydraulic system does not really use more or less power then any other system. I do think though that it would be hard to develop a mech that can run a hundred kms an hour like a more traditional automobile.

Asbrandsson

Well, the hydraulics are a bit of an issue in that you actually have massive power loss when you convert engine rotation into hydraulic power, i.e. at the point of the hydraulic pump. While it may not seem like it, this is a relatively well-understood and expected thing for any machine using hydraulics, so the engines are bigger to compensate. Enough hydraulic power to move a mech that quickly makes it very heavy, and you get a vicious cycle wherein you need a heavier engine to push it forward, but have just made your load heavier, so you need more power, so you need a heavier engine, etc.... One way around this - get the leg power direct drive from the engine via a simple scissor mechanism. I won't lie - I say this because it's what we're doing - but I've sat in on extensive discussions of each of the possible power sources to move a moderately sized mech and their implications, and just about the only way to do it with sufficient power/weight characteristics is direct drive from an internal combustion engine. If you do that, you technically can get it to run at over 100 mph, but you still don't want to - there are tons (figuratively and literally) of structural safety issues, plus the issue of terrain vision; if you were running at 100 mph, think about how hard it would be to see the ground in front of you to step safely, especially if you were not on a prepared smooth surface like a road.

Exciting tidbit: the fictional technology of "myomer fibers", that is, metal that contracts like human muscle, already exists in the form of nitinol and carbon nanotube wires treated for linear contraction. Future generations of our machines will likely attempt to resolve the complex electrical and thermal interference issues in using nitinol (or CN's, if they go down to less than $10,000/lb. like they are now) so that we can give the machines the same fluid articulation of human muscle. Bonus: these fibers can exert dozens of times as much force per weight as human muscle, and require miniscule amounts of electricity, making this potential future mech technology highly energy efficient.

asbrandsson
02-24-2008, 09:08 PM
Hello,

Is it that that pump needs to be bigger to generate preasure or that the motor needs to be bigger to make the pump generate more preasure?

I think that alot of computer assisted piloting is needed to make a venture like this possible. Although it might just be the kind of thing that needs some experience to get used to. Driving a skid steer in not all that complicated once you get going at it, but it seems impossible if you just jump in the seat and try hit some levers or buttons.

Asbrandsson

Ryan.Marlow
02-24-2008, 10:33 PM
Hello,

Is it that that pump needs to be bigger to generate preasure or that the motor needs to be bigger to make the pump generate more preasure?

I think that alot of computer assisted piloting is needed to make a venture like this possible. Although it might just be the kind of thing that needs some experience to get used to. Driving a skid steer in not all that complicated once you get going at it, but it seems impossible if you just jump in the seat and try hit some levers or buttons.

Asbrandsson

Ah, correct on all three counts.

I'm pretty sure both the engine and the pump have to be bigger - the pump to generate more pressure, the engine to supply the needed power to drive the larger pump.

Lots of computing power: You're not joking. We have already specified a product roughly the size of a normal computer tower that contains upwards of 80 industrial-grade, low-power, low-heat chips with roughly 1 Ghz each, and have seven potential bays in the cockpit to put these systems into. The AI discussion gets pretty complicated - it's based on the human prefrontal cortex and includes things with fun names like "Reflexive Central Executive", "Hidden Agency Manager", "Active and Passive Agents", "Sensory-Ego-Sphere", and my personal favorite, "multi-level multidimensional hierarchical neural network". The vehicle is in fact so heavily "computer assisted" that we joke that it is "piloted by a computer that takes suggestions from a pilot", which is really more truth than joke, especially in things like safety situations in which the machine can override the pilot from inadvertently doing something that can harm a human being.

Some getting used to: yes, it will. In fact, I think that will be a big selling point - once people have piloted the machine, they will begin to understand some of the more abstract benefits it confers: the ability to think about and solve problems in 3D space with the same intuition you use for your own body, but on the scale of a large vehicle. Although the AI itself is also designed to learn about itself, its environment, and its pilot. So it won't just be the pilot adapting to the machine, both will adapt to each other. And that data is stored on a portable Pilot Data Pattern so they can jump it into a new machine and have it immediately understand their control tolerances.

Alex
02-25-2008, 09:47 AM
though we expect that our AI software will catch their eye long before the bipedal vehicle concept has had time to catch on.What sort of AI software are you working on? Do tell! That is of course if you're allowed to:)

I don't want this thread to get hijacked though, so if you can talk about the AI software, post in the Software & Programming (http://forums.trossenrobotics.com/forumdisplay.php?f=21) forum.

EDIT: Yet again, I fail to read a complete thread before I comment:( Still though Ryan, I (and I'm sure many others here) would love to hear more about your AI work in the S&P forums I mentioned above.

MYKL
02-25-2008, 10:08 AM
Mr. Marlow,

Have you built proof of concept models of the leg assembly? I check in to your site once a month or so looking for updates and what I've found sounds encouraging but all that I've really seen is a video of your powerplant and a lot of parts. There are the CG bits showing your construct lifting two semi trucks... Fun...

Have you built any working models of the leg linkages you've designed to see if the thing will walk with the constraints of real life acting upon it? It must be great to be able to focus with a group of like minded individuals on something this unique and that will make the segway look like a skateboard as far as public appeal.

I personally agree with the idea that light weight is the way to start. Once you have the basic construction, a working model, down then you can refocus on making it bigger and better. Otherwise you may end up with something like what you see in the back yard of a well intentioned welder in Alaska.

I hope you are successful. I'd love to work on something like what you are doing. As would the rest of the people that post in here.

wind27382
02-25-2008, 03:00 PM
food for thought.


http://www.imagebulk.com/images/wind27382/medium84528400376b2f302jb6.jpg

Ithink it's just a prop but it would be great if it worked.

tom_chang79
02-26-2008, 09:06 PM
I have a bunch of 1/100th scale gundam and a few 1/144th scale gundam models. This 1/1 scale model rocks!

Hmm, how do we know that the Japanese government isn't already building one in a hush-hush project for their defense? :wink:

Ryan.Marlow
02-27-2008, 12:07 AM
Mr. Marlow,

Have you built proof of concept models of the leg assembly? Have you built any working models of the leg linkages you've designed to see if the thing will walk with the constraints of real life acting upon it?

...There are the CG bits showing your construct lifting two semi trucks... Fun...

I hope you are successful. I'd love to work on something like what you are doing. As would the rest of the people that post in here.


MYKL,

The Mark I prototype is the proof of concept. We have built miniature mock ups of the system to test its basic mechanics, we have had experienced engineers develop, examine and approve our full leg and foot designs, and a past Team member actually built an entire proprietary set of physics modeling software (he now works, of course, for Microsoft) to test our designs, and found them to be sound. All that remains is to build the true, 1:1, real machine, and test and tune it, which is what the Mark I is about.

The mech lifting trucks video: Okay, I have to chuckle at that. I think I told them not to put that on there... this is what happens when we plan to spin off the fan site from the business site, and then keep the fan material parked on the business site until we finish the fan site development. Still, it is rather amusing.

To anyone who might be interested in getting involved, I would be more than happy to get in touch with you and talk to you more in depth about it (just PM me, or email me through the form on the Contact portion of our site at www.MechaPS.com/contact (http://www.MechaPS.com/contact)). As a company, we are just growing out of our early developmental phase into one of much more serious business, so although we cannot provide a full-time salary, we do compensate in terms of stock options. The time commitment is also very flexible, both in terms of how much it is (set by negotiation between you and a supervisor, not by an arbitrary quota) and when it is (can accommodate generally any outside commitments, be they academic or professional, so long as the negotiated amount of time and work is met). For anyone who would love to see this succeed, the best way to see that is from the inside - to contribute to the company and be part of - and have stock in - that achievement, and the ones future beyond it.

MYKL
02-27-2008, 10:46 AM
Mr. Marlow,

I hope that I did not convey any disrespect for you and/or your company. I think what you are all contributing to is the dream of many people. All it takes is for someone to break ground, show initiative and put themselves on the line for what they believe in. I'm sure that since you've drawn a moderate croud of people to work with you, you have a credible amount of data and R&D to move in the direction you have. I see that you've built a frame for the torso. And that you've gathered a plethora of parts, off the shelf and custom to put into your project. I will continue to follow you and your creation in the hopes of seeing a new set of names join the Wright brothers.

Be well sir and 'make a leg'.

^_^

SN96
02-27-2008, 05:47 PM
In response to Matt's article link, I can only say: trying to solve difficult problems brings new technology that can be used for other more practical applications. And last I checked, ASIMO can walk up stairs and move fast. :veryhappy: