· I agree with SRLM that this sounds phony. The rules seem fairly vague, particularly when you compare them to the rules of contests offering prizes less than a hundredth this amount.· They also haven't generated much press (like none) for an event with such a large prize. My first guess would be that it's just a hoax. My second guess would be that it's a fishing expedition looking for autonomous robot projects. Given the military implications of a device that would meet their specifications, you might find yourself contacted by some fairly unpleasant people if you fielded an entry that appeared to have any· promise. It bothers me that they don't even have a picture of the "similarly matched combat robot" your entry is supposed to spar with.
· That being said, it is an interesting project. There are companies that market gas turbine engines for model aircraft and I've seen at least one turboshaft version. That powering a compressor for the pneumatics would probably provide the best power/weight ratio you could get without an enormous investment of time and/or money. The issue of dynamically balancing robots came up in another forum a while back and the book I mentioned earlier appears to be the only title on the subject, so you might try to find a copy of it. MIT Press.
· Do they make kevlar pneumatic cylinders? Kevlar is pretty much the optimum material for pressure vessels and I'd think you could achieve a great weight savings in the pneumatics with kevlar cylinders. If they aren't commercially available, you might consider making your own.
· If I remember correctly, the biped robot in the MIT project was tethered to an arm which rotated around a central column. If you used such a setup to do software development, you could conterweight the arm so you wouldn't have to worry about the power/weight ratio and just use steppers or servos·for actuators as a first step in developing the robot model.
· I just thought that as a third guess the prize money might be ligit, but offered by a madman who intends to mass-produce an army of autonomous robots to take over the world. Just saying.
We do need a sort of Gyro to keep us stabilized.
Spin around in a circle about 20 times so that the fluid in your inner ear is is spinning and you fall flat on your face because your "Gyro" is still spinning when the rest of you comes to a stop. Seems that the Vestibular system is linked to hearing, sight, and balance.
I am new to all of this, I’m happy just to get my wheeled robot to do things. ·
I don’t know how important it is to have a humanoid robot, but I do think walking robots will prove to be important. I personally don’t much care if the robots will have two or eight or twenty legs or any combination in between. But for the reasons that Pi Guy has mentioned regarding places a legged vs. a wheeled robot can go, I think wheeled robots will be relegated to flat (read office/warehouse) environments and legged robots for most everything else (including multistory homes).. ·
The question that has crossed my mind is: How many wheeled creatures do you know in the wild? Other than the rolling action of some sea urchins and single celled animals, I can’t think of any. I once heard of a spider that would roll down hill when it was a question of survival, but I can’t verify if that was true or not. The point being that legged creatures seen to outnumber wheeled/rolling creatures by a considerable margin. I would guess there is a good reason for that.
I don't really have the biology background to answer that, but I can try. Wheeled cretures are not found in the wild because a)how do you keep a biological wheel alive, fed with nutrients, and connected to the nervous system? and b) how do you propel the wheel? As a side note, there are wheeled animals (they use large coconut type wheels) in the fiction series of the Golden Compass. I found that particularly fascinating when I read it as a child.
I don't really have the biology background to answer that, but I can try. Wheeled cretures are not found in the wild because a)how do you keep a biological wheel alive, fed with nutrients, and connected to the nervous system? and b) how do you propel the wheel?
The rotifer might be pretty close - it came to mind immediately when you said this:
@Pharseid: The "similarly matched combat robot" is the winner of battlebots, I believe (you might want to double-check me, SLRM).
Also, I just wanted to say that using things like a jet engine and Kevlar cylinders sounds a little ambitious for my *future* project. If I do use Kevlar, however, it will probably be in places like the hands and feet.
@Ultra: are you just trying to torture me???
@Deadwhale: because animals don't have access to rubber sealants and metal pipes.
@Whit: Thanks for the link. It is definitely what I was talking about when I said "Pressure vs. Position" with pressure being the best form of control; however it still has some problems. I'm not sure if it is due to poor coding, a poor sensor selection, or some mechanical issues but it does not have the smooth motions I envision. In order for it to walk like a human it needs to allow it's foot to go back farther as it walks, as well as let it easily roll and swing back to the front. As you can tell, this is obviously not happening.
If I were you I'd do a web search for Kevlar cylinders, if they're standard products they don't add any complications to the project at all (other than perhaps expense). And the turboshaft engine is an existing product. But really, I trying to point out that you'll want to save every pound you can. Another alternative for acuators might be air muscles.
The only problems I'd be worried about are weight and size. But, before you get too excited about this humanoid, remember I still have another 2 years to go be fore I will attempt it. You're better off worrying about my FHR (five horse robot) than the future humanoid since the FHR is one of the robots I plan to make to 1) give me experience before I try such a task and 2) show others, such as potential sponsors, that I am a very capable roboticist. If these precursor projects don't work as planned, you might as well say goodbye to the pneumatic humanoid until I complete college.
If it is indeed Kevlar pistons you are referring to, then no, I do not believe they are a standard item. When I searched for them all I got were brake pads.
Whit, that robot definitely has some problems to fix, however I'm not going to explain them all right now because I have homework to finish.
In this thread I've mostly been pointing out what roboticists do wrong, so I decided to include a link to an interview of my favorite roboticist. You can read this interview here:
Comments
· That being said, it is an interesting project. There are companies that market gas turbine engines for model aircraft and I've seen at least one turboshaft version. That powering a compressor for the pneumatics would probably provide the best power/weight ratio you could get without an enormous investment of time and/or money. The issue of dynamically balancing robots came up in another forum a while back and the book I mentioned earlier appears to be the only title on the subject, so you might try to find a copy of it. MIT Press.
· Do they make kevlar pneumatic cylinders? Kevlar is pretty much the optimum material for pressure vessels and I'd think you could achieve a great weight savings in the pneumatics with kevlar cylinders. If they aren't commercially available, you might consider making your own.
· If I remember correctly, the biped robot in the MIT project was tethered to an arm which rotated around a central column. If you used such a setup to do software development, you could conterweight the arm so you wouldn't have to worry about the power/weight ratio and just use steppers or servos·for actuators as a first step in developing the robot model.
· I just thought that as a third guess the prize money might be ligit, but offered by a madman who intends to mass-produce an army of autonomous robots to take over the world. Just saying.
-phar
"On every humanoid I’ve seen, a gyro is used to give them balance. If humans don’t need gyros, than why should a humanoid?"
I think we forgot about the inner ear.
http://en.wikipedia.org/wiki/Vestibular_system
We do need a sort of Gyro to keep us stabilized.
Spin around in a circle about 20 times so that the fluid in your inner ear is is spinning and you fall flat on your face because your "Gyro" is still spinning when the rest of you comes to a stop. Seems that the Vestibular system is linked to hearing, sight, and balance.
·
I don’t know how important it is to have a humanoid robot, but I do think walking robots will prove to be important. I personally don’t much care if the robots will have two or eight or twenty legs or any combination in between. But for the reasons that Pi Guy has mentioned regarding places a legged vs. a wheeled robot can go, I think wheeled robots will be relegated to flat (read office/warehouse) environments and legged robots for most everything else (including multistory homes)..
·
The question that has crossed my mind is: How many wheeled creatures do you know in the wild? Other than the rolling action of some sea urchins and single celled animals, I can’t think of any. I once heard of a spider that would roll down hill when it was a question of survival, but I can’t verify if that was true or not. The point being that legged creatures seen to outnumber wheeled/rolling creatures by a considerable margin. I would guess there is a good reason for that.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Prune Free.
I don't really have the biology background to answer that, but I can try. Wheeled cretures are not found in the wild because a)how do you keep a biological wheel alive, fed with nutrients, and connected to the nervous system? and b) how do you propel the wheel? As a side note, there are wheeled animals (they use large coconut type wheels) in the fiction series of the Golden Compass. I found that particularly fascinating when I read it as a child.
I found this video and if you look under the "rubber socks", there seems to be an articulated foot. Pretty interesting video.
http://www.youtube.com/watch?v=_CQ5AKaEi3U
In this one, the robot, "Dexter" jumps.
http://www.youtube.com/watch?v=ZnTy_smY3sw
I know it is attached to a cable, but it is very impressive - and much different than the Asimo walk or run.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Whit+
"We keep moving forward, opening new doors, and doing new things, because we're curious and curiosity keeps leading us down new paths." - Walt Disney
Post Edited (Whit) : 3/30/2009 10:12:53 PM GMT
http://en.wikipedia.org/wiki/Rotifer
I don't know exactly how "wheel-like" that feeder thing is, but the rotifer may have "solved" the problems you're raising.
@Pharseid: The "similarly matched combat robot" is the winner of battlebots, I believe (you might want to double-check me, SLRM).
Also, I just wanted to say that using things like a jet engine and Kevlar cylinders sounds a little ambitious for my *future* project. If I do use Kevlar, however, it will probably be in places like the hands and feet.
@Ultra: are you just trying to torture me???
@Deadwhale: because animals don't have access to rubber sealants and metal pipes.
@Whit: Thanks for the link. It is definitely what I was talking about when I said "Pressure vs. Position" with pressure being the best form of control; however it still has some problems. I'm not sure if it is due to poor coding, a poor sensor selection, or some mechanical issues but it does not have the smooth motions I envision. In order for it to walk like a human it needs to allow it's foot to go back farther as it walks, as well as let it easily roll and swing back to the front. As you can tell, this is obviously not happening.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
PG
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
PG
-phar
http://www.cumulus-soaring.com/mh-cylinders.htm#Cylinders
The only problems I'd be worried about are weight and size. But, before you get too excited about this humanoid, remember I still have another 2 years to go be fore I will attempt it. You're better off worrying about my FHR (five horse robot) than the future humanoid since the FHR is one of the robots I plan to make to 1) give me experience before I try such a task and 2) show others, such as potential sponsors, that I am a very capable roboticist. If these precursor projects don't work as planned, you might as well say goodbye to the pneumatic humanoid until I complete college.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
PG
-phar
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
PG
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
PG
Check this out - http://www.youtube.com/watch?v=5R-cyCXrucA&feature=player_embedded
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Whit+
"We keep moving forward, opening new doors, and doing new things, because we're curious and curiosity keeps leading us down new paths." - Walt Disney
@Whit: Did you see the video of the MIT biped doing a flip? That's what I call dynamic balance.
-phar
In this thread I've mostly been pointing out what roboticists do wrong, so I decided to include a link to an interview of my favorite roboticist. You can read this interview here:
http://www.pbs.org/wgbh/nova/sciencenow/0304/03-bion-nf.html
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
PG