601Views10Replies

Author Options:

Hi! Answered

Hi!

Nobody has posted here and I'm curious if this project is active.

I will be working on regiggering something to be a sort of myoelectric prosthetic. I don't have one yet as I'm still recovering, however I might as well think about how I could help the situation.

I found a great hobbyist website, Spark Fun. From them I ordered an Arduino for $32.

The Arduino is a n atmel microcontroller with a usb interface, allowing programming the chip and communication with the running program.

For myoelectric control, voltages would be derived from skin voltages. I'm no physiologist, but if some muscle activity can be discerned from the (low voltage) input, it could be used.

The PWM outputs are ready-made to control common servos, so a motor-operated claw could be controlled.

With the arduino's ability for independent operation as well as interfacing to the computer to revise the microcontroller's software and reprogram, it should be fairly straightforward to achieve a custom controller for a robotic claw of sorts.

Now, where do I get the claw (at a reasonable price)? Or, maybe it'd be better to custom design one?

Stay tuned.

Comments

The forums are retiring in 2021 and are now closed for new topics and comments.
0
1rmedman
1rmedman

13 years ago

I only have 2 sensors controlling a pincer type movement; a harness in which I can move my shoulder to release and lock the elbow. I don't know of any prosthetic that moves the individual fingers. Placing the sensors would be easy, I would use the shoulder. A mouse sleeve with a remote type ir sensor could have been invented long ago for far less and would be far more useful than my 25G+ pincer. Electronic components are getting cheaper while the mechanical hand is getting more complicated and would be way too expensive, I agree. I had no idea hey made clickable tracker balls; I'll look for one. btw, I've been searching for someplace to buy wired sensors for a few hours and haven't made any headway. So if someone reading this knows, I would appreciate it.

0
1rmedman
1rmedman

13 years ago

Im a ae amputee with a myoelectric arm; its pretty useless. Im sure with a LOT of practice id be juggling eggs, but a year in and im still crushing cans. I think copying the design of the human hand is rather silly. It's hard to do and pointless. Id like a hand to be more useful than human. I would need it to have several digits that I could change the orientation of. A nice design might be 3 pairs of opposible 'fingers' that could move independently. Anyway im no engineer. I would think that a usb connection alone, allowing control of a pc mouse, would be far more useful to me than the claw and hand I have now. A simple sleeve with myoelectric sensors could accomplish that.

0
Kiteman
Kiteman

Reply 13 years ago

How much control can you exert over your arm? Is it a basic open/close like a pincer, or can you control fingers individually? You'd need to be able to consciously control at least two sets of muscles (I think, maybe three) to work a mouse, then you can combine left-right, up-down and clicks into one sleeve. I can see your point about copying the traditional hand, but most appliances are designed to be used by the standard human grip. Imagine how much more expensive it would be to design and produce things to match your chosen arm design. Have you thought of switching to a tracker ball? Gross arm control could replace the fine finger/wrist control, and the clicks could be made by pressing the whole tracker.

0
JoeHS
JoeHS

14 years ago

I use an OttoBock myoelectric hand, great when it works. So far, every 8 months I have a serious mechanical problem. I work in an office. I push a pencil and tap "lightly" on a keyboard all day. I hold a MS Electrical Engineering from Va-Tech, and a BS Mathematics. I just can't believe that a prosthetic myoelectric hand could not be built for less than a few grand. The electronics is the lesser problem. I don't deal with microcontrollers nor MES sensors. But I think that I could get over the learning curve for that quickly enough. The mechanics of the hand itself is another story. I'd like to a see a motorized hand design; solve that problem and I think a 'home made' prosthetic myoelectric hand would be possible. I am open for collaboration. regards, Joe.

0
mm1ssNuke
mm1ssNuke

Reply 13 years ago

I am still working on my LEGO device, I think with some glue and motors it might be the cheap and repairable machine you are looking for.

0
yerg2k
yerg2k

Reply 14 years ago

Hi Joe,
It seems to me the major challenge is matching a design to many people's requirements.

But at minimum a design should be rugged, and pack a big punch in a small package. The question becomes, how can the greatest amount of force be applied to objects of different sizes with least mechanical complexity and most efficient transfer of electricity into work. Least mechanically complex means less parts to break. It would be wise to plan the design to be able to take large stresses that you might expect now and then. I like to daydream about things, and one of them is "if I had a metal hand" that could transfer motion to my arm properly (to not hurt) and be so rugged that I could say, kung fu chop a board in half, or make a fist and be able to knock on a door and make some noise. </daydream>

I have the controlling-it part figured out. I haven't spent much time thinking of ideas to overcome this challenge in physics. Evidently others have, however I have no knowledge of how rugged plastic can be when used in a hand.

I was thinking, the aspects a design should be measured and chosen by simplicity, efficiency, and power throughout range of motion.

Something simple came to mind. I'll be looking into physics of solenoids. Let's collaborate.

Best Regards,
Derek

0
JoeHS
JoeHS

Reply 14 years ago

I agree, keep the design simple. This avoids generating maintenance problems. I plan to begin experimenting with PIC microcontrollers and try to learn more about MES (myoelectric signal) sensing. A rugged design is nice; being able to maintaing the prosthetic personally is the the real goal. Designing a socket to fit on the arm is a mechanical project not too terribly complicated, definately 'doable'. Reasonably good looking cosmetic gloves are in the accessible price range. Suspension sleeves cost me $45 each. The two huge hurdles: myoelectric sensor and controller, and the motor & mechanics for the hand itself. Building the electro-mechanical hand being the complication. Anyway, by next week I should start working with a PIC microcontroller. After a couple of weeks experimenting, I should be able to know more about the feasabililty. Google searches provide a lot of useful info. I am thinking that DC motor and gearing is the best way to go for the hand. I doubt if a solenoid provides the needed level of control etc. Existing motors and gearing is what is needed. Building that stuff is chasing rainbows; off the shelf items is the way to go. The only mechanical part that might need building is the mechanical moving "skeleton" of the hand. Keeping it simple. I like my otto bock hand. The thumb moves and the two fingers move together; simple and effective. When it works, it works well. I'd also like to know if there is anyone else out there lurking in cyberspace. regards, Joe.

0
PlastBox
PlastBox

13 years ago

From searchmobilecomputing.techtarget.com :

Myoelectric signals are detected by placing three electrodes on the skin. Two electrodes are positioned so there is a voltage between them when a myoelectric signal occurs. The third electrode is placed in a neutral area, and its output is used to cancel the noise that can otherwise interfere with the signals from the other two electrodes. The output voltage is processed using a device called a differential amplifier. The output of this amplifier has much higher voltage than the myoelectric signals themselves. This higher voltage, which produces significant current, can be used to control electromechanical or electronic devices.

Winner of Black Widow contest

It seems reading data from musclecontractions through the skin shouldn't be too hard. Those people built the electronics around a $1 microcontroller and it seems to take the space of about 2 arduino-cards (which is decent, considering it is built from scratch).

Please, feel free to contribute to my thread on this subject in my thread on arduino.cc

0
jfk
jfk

14 years ago

yerg -

Great to have you involved. This type of initiative is something that has been proposed by us (http://www.openprosthetics.org/myoelectric), and something that several volunteers have expressed interest in, but on which not much has been done.

A graduate student in engineering in South Africa has actually completed a thesis on this topic and has a cheap board design that we are awaiting authority to publish.

Proposed microcontrollers on which such a device could be based are an open-source design from MIT, the gumstix (http://gumstix.com/), the PIC or dsPIC, and others.

As far as the electromechanical hand or gripper is concerned, there is not much out there. A good idea might be to repurpose a toy or industrial gripper. Cost is an issue. Commercial ones are based on Maxxon brushless DC motors with integral transmissions (http://www.maxonmotor.com/).

This effort is currently lacking organization and published results. We'll see if we can't get something started soon.

Thanks for your interest and ideas.

Jon Kuniholm, Open Prosthetics Project