Randy, Hmmm...good questions. A preposition makes that question from impossible to highly improbable. Do you beleive a man walked on the moon?
To me, if someone was saying, "Hey, let's put a man on the moon", would sound completely nuts even with today's technology. But they did it how long ago? 34 years.
Those men and women involved in that project were true visionaires. They focused on whatever would work and kept positive. They did it despite all those against them (engineers, chemists, mathematicians, physists, and the astronauts are heros in my book and I don't give a damn who thinks otherwise :o). There are even nuts today who beleive it never happened because they can't conceive of it.
I'm no visionairy. I am trying to build a "hammer" and I need help to do that. The real question becomes, "Can you see someone building a hammer?".
Sharkey, The efficiencies suck....around 40% at best (about like a stepping motor but much lighter...but the power-to-weight is great...hot, but great). It needs work to minimize power. Also, for power...think utility. :o)
I have a friend trying to get hold of a used magnetic-bearing microturban (5kW). Now, I know this isn't the thing you want bringing you a cup of tea in the evening, but think "Utility Class". Yes, the images just blow you away (actually the new systems are quiet and not quite how the imagination portrays them ;o)
To make up for immediate power, looking to lithium ion batteries that can be charged off the generator. Working those final details out will take some time. Know any good sources?
I like the idea of the pushups, but I am going to just program the PIC to run it to failure as soon as time permits.
Martin, First, I appreciate your argument. I was there myself. I agree there should be something better...but until then. ;o) Also I love the "tape-sliding stepping-motor"...The sliding tape is very important to the mechanics...keeps things from breaking.
When the additional mechanics are added to move the kevlar ribbon from the magnetic gap, you increase weight as well as other complications. Since you have the coil separate, now you need a frame and a mechanism to squeeze the ribbon external to the coil. The mechanism will have a certain amount of flex which will increase the travel required and reduce the force that could be exerted.
The ribbon must be very, very thin because the magnetic field force falls off at a rate of a square of the distance. This also dictates that the range of motion must also be very small.
The step range is huge in magnetic terms for a stepper motor. Take a good stepper like a double stack NEMA 34 from PacSci. Each time a step is made, the rotor travels ~0.016 inches at the circumferance (actually, it is much bigger than that because there is an air gap between the magnet and lamenated ferrite). Now, the magnetic field must be strong enough to turn the rotor at that distance. In addition, the rotor contains many magnets, several coils, and the shaft/rotar. Plus, there needs to be a tremendous amount of material to guide the field lines to handle the multiple path they must follow.
Now, steppers are great (personally, very partial to API Motion which is now owned by PacSci). They can apply force to "absolute" position. This device doesn't need the frame or the large amount of iron (just enough to guide the specific fields). There is no rotor, shaft, multiple coils, etc. The device doesn't provide a "step" like a stepping motor (wouldn't it be great if it did).
Now, with reduced weight from the overhead of the stepper (and now even counting gears, shafts, and frame for them), and with much smaller operating ranges, the power-to-weight is greater.
I've said it a few times now. You need iron. I don't care how far you are moving. To paraphrase: Use as much iron as you'll need, not less.
Now, if you are content with very low efficiency numbers (I even doubt 40%) then, by all means, don't use nature's free source of magetic field strength, burn battery power to create it.
Re: that. Can you peak-and-hold the solenoids to reduce power consumption?
Okay, but what does a "utility class" robot (was this thing meant to be a biped?) _do_? If it's going to be burning 5kW, it'd better be doing something valuable.
Is it meant to just move stuff around? What can it do that a forklift with a PUMA bolted to the top won't do better? Or, for that matter, what can it do that you can't hire for minimum wage?
Sharkey, Yes, that is part of what the PIC is helping control. Part of that is an expiratory PWM.
It is to be a biped. And the 5kW is overkill (haven't found anything smaller :o).
This is just a stepping stone, nothing more. It is an "open source" baseline. What good is an Asimo? What good is a SDR-4X? Hopefully, at least George 1.0 can be of some use.
Remember, there is a reason the name has "1.0" in it. Maybe around "4.0" it will be more what we all want.
Martin, I agree, and to re-paraphrase: "Use as much iron as needed, and not any more". ;o)
Do you have a good source for larger neodymium magnets? Not opposed to them, just never found any that I could use. Also, they will need shielding but that is no different than a coil. I'll try anything.
Oh, right. 'Expiratory' isn't a term I'm familiar with.
What good are they? Bugger all. They're PR stunts and tax sinks for large R&D organizations.
Autonomous bipedal hominids are a waste of time, because the planet is already overrun with self-repairing, self-maintaining ones. What on earth is the point imitating them?
Prosthetics, though, that'd be a worthwhile field for your SynthMuscle developments.
Larger? What do you mean? Large neodymium magnets are downright dangerous. I do mean DANGEROUS, they could smash your finger like it isn't there. Be careful.
Search the www for "neodymium" there are a good number of reputable soruces.
What do you mean by "never found any that I could use"? What are your specifications?
Shielding? Explain? Now we are going to have muMetal shields in this thing?
Sharkey, For prosthetics, that is the first place they are being used. A friend of mine is quadraplegic. There is a version Limb Controller is built with a USB port (via a National 9603). It provides the PID interface.
A notebook will serve as the interface between the Arm and his mouse (specialized and mounted to his chair). He can change the programming through a VB program acting as the Central Controller.
There is a great need to a Utility Class droid. There will be bad uses, there will be great uses. It is just a tool, like a hammer.
Martin, I have literally thousands of neodymium magnets (for magnetic bearings on certain types of equipment). They are dangerous. What I need is one that is 0.6x0.6x0.3 inches big with the poles at center of the faces. Grade N42 or better.
To use magnets of that strength will require managing their field lines...imagine otherwise ;o)
Why do stereo matching? It's computationally very expensive. Most animals don't use it, they use the computationally much cheaper optic flow. And even animals which do use don't use it for high speed stuff.
-- Chris Malcolm snipped-for-privacy@infirmatics.ed.ac.uk +44 (0)131 651 3445 DoD #205 IPAB, Informatics, JCMB, King's Buildings, Edinburgh, EH9 3JZ, UK
Yeah, I'm not working on stereo vision now. Now I'm working on texture-based segmentation with a single camera. It seems more biologically plausible and should be computationally more efficient. Once I have a segmented image, even stereo should be easier to perform.
Stereo vision comes into its own when you have a random background of branches and leaves. It would be rather difficult to segment your random dot autostereograms.
Matching segments from two stereo images sort of works. The main problem I have is two segments in one image joining into a single segment in the other (or next) image or vice versa.
The other issue was segments that went beyond the left or right side of image. This made it tricky to calculate their centroids to determine depth.
For simple systems I think the practical solutions come from goal determined analysis. For example if your gaol is to detect and maybe even recognize people then movement is the easiest way to segment them from the background and determine their outline, particularly the head part which can act as a reference to the rest of the body.
As for texture based segmentation I find my images really don't show much texture unless the camera is very close to the surface or the textures are very large like patterns on wall paper.
One method that I found works is simple local thresholding the image into blob segments. I imagine these could be used as a framework for more intense analysis?
If you were interested I could send you an example? Maybe you could give some suggestions how to improve the code? If so let me know via the newsgroup. Any posts to this email address will go to spam land which I never download.
PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.