Wounded Soldier Gets Robotic Hand Replacement

Dextrus V1.2 Robot Hand

[Neal Muzzy], a local member of the Cedar Valley Makers makerspace, just made news on Open Bionics for his robotic prosthetic hand called Dextrus v1.2 which he made for his friend, and wounded war veteran, [Taylor].

In just two months, [Neal] worked with his friend to make this robotic prosthetic with the goal of having it more functional and easier to use than [Taylor]’s current prosthetic. The very first prototype was made by using the open-source Dextrus design, to test fit, and control using EMG sensors. Once they determined it would work — onto customizing!

They call it Dextrus V1.2, and it works better for [Taylor] than the original — but that’s the whole point of the Open Hand project — starting with a base design, and making it better. If you’re not familiar with the Open Hand Project, it was originally crowd-funded on Indiegogo, and is now an organization to make robotic prosthetic hands more accessible to amputees. We wrote about it in Hacklet 41 – Prosthetic Projects.

http://www.youtube.com/watch?v=seQPeJvXet8

Now this robotic prosthetic might be rather limited, but like [Neal] says:

In its completed form, this arm may not end up being the go-to for daily usage compared to the professional-made prosthetics that Taylor has, but that’s fine because this project will have served as the test bed for trying out new features and creating a control program to work as easily and intuitively as possible.

The next time Taylor is having a professional-made prosthetic arm put together, he will be able to provide this 3D-printed arm as an example of every feature and program behavior that he will want the new arm to include.

But, in the meantime, [Neal] is already working on adding a 2-axis wrist and multiple grip modes to the prosthetic. Awesome.

[Thanks for the tip Nathan!]

47 thoughts on “Wounded Soldier Gets Robotic Hand Replacement

  1. Double amputee, that’s sad, if he’s a vet why the f**k doesn’t the government give him some money to get some commercial myoelectric prostheses? Cool hack, screw the government.

    1. He has commercial-made ones, but the frustration / reason behind this project is that they are not intuitive to use, lacking something as simple as an indicator light to show whether grip or wrist control is active. For that, plus adding an array of other features, I feel we can improve over his existing prostheses.

        1. It depends what exists, there’s a limited market for replacement limbs (but keep on sending the soldiers out, it’ll grow). It would be nice if the existing companies took customer feedback, although I’m sure they do a lot of work with their users already. Maybe they need to give Neal a job, or at least a consultancy.

          I don’t think the problem’s lack of money, I think it’s just that it’s a small market, and modern technology. New technology arrives every day that’d be really useful in replacement limbs. Maybe a bigger problem is lack of electronics expertise among the users. They don’t know what’s possible to make, or how the prostheses, or their software, work. Engineers don’t have daily experience using prostheses. Short of a few massive soldering mishaps, that’s probably gonna persist.

          The answer is stuff like this, 2 friends who each have half of the puzzle. So more communication is needed between design engineers, and their product’s users. It’s a business organisational problem. Since it’s a small market, and I imagine users are pleased with whatever they get, there’s not a lot of pressure to get things exactly right.

          One good thing is the proliferation of powerful low-power embedded processors, it means a lot more options might be possible in future. Related to that, perhaps the users could be given software that lets them configure the hand, and even “program” it with some sort of graphical language. A little button or option somewhere will bring it back into factory mode, if somehow you mess it up.

          1. @Greenaum
            It *is* lack of money, at least in the sense that the small market means it doesn’t represent enough profit, or enough economy of scale for it to be a priority for grant money, etc. I don’t know that there’s a good solution. It would be better with a UHC system keeping the prices down, but even then I fear money would go toward issues that affect more people. Don’t know of any way to really eliminate the limited bang for buck problem. Sad.

        2. Taylor did a TEDx Talk at UNI (University of Northern Iowa) recently, but I do not see the videos as published yet. That being said, he spoke about this project, and that the Prosthesis lab at Walter Reed had an open door policy, and any and all Ideas the guys had or wanted to try, they were more than willing to give a shot. Taylor is actually a quadruple amputee, with portions of all his limbs removed, which is not the norm, and presented special complications all around. This all started when he mentioned to neal that he was having issues using the prosthetic to point things out, and joked maybe they should put a laser pointer on his arm… neal decided to make it happen. Taylor was impressed and mentioned some other issues he had with his current prosthesis, such as strange requirements to switch modes, specifically needing to flex his muscle both ways at the same time (open and closing the hand) in order to switch modes, thus this project began. And thats not the end of it, some of the other additions on the drawing board include an LCD touchscreen, built in USB mouse, and higher power servos with current sensing to handle detection of when the object is sufficiently gripped. More to come in the future and we will be sure to update all you hackaday readers with our progress. Thanks!

          1. That’s really good! Maybe Taylor could start doing some outreach stuff in veteran’s associations, brainstorming sessions for what people want in their prosthetics. Bring a few of the lab guys along. It seems there’s a lot you could learn from each other, maybe make it a 6-monthly thing. Commercial manufacturers might want to get involved, or consult with the group. You could take a hand to bits and show how it actually works, empower the users.

            Of course it’s not just war veterans who need prostheses, but I suppose it’s a good way to find a few of them in one place, and I’m sure it’s something they talk to each other about.

    2. That would mean sorting out the stupidly big mess that is how the US deals with veterans.
      The fact he has got anything means, sadly, he is “lucky” to have got though the system that far.

      But then, its easier to spend billions on spying systems, aircraft carriers and thousands of unneeded tanks then it is to sort out paperwork.

  2. Just a note:

    prosthetic is an adjective.
    prosthesis / prostheses are nouns.

    You can say a prosthetic limb/hand/arm/device or a prosthesis but it is all too commonly (incorrectly) just called a prosthetic in many articles.

    1. “a prosthetic” is short for “a prosthetic limb”. It’s perfectly acceptable English. Can we give the grammar nitpicking a bit of a rest please? Nobody gives a shit. It really doesn’t help, and it’s starting to annoy. It’s already nearly driven Brian off the deep end.

      If I wanted grammar lessons, or even spelling, I’d go hang round some English language-based website. Where I’m sure they could kick your arse (noun) with both hands tied behind their backs. I don’t care. Engineers don’t have to be perfect at spelling. Hackers only have to be good at hacking. If you were their English teacher, you might have some valid reason for this bullshit.

      This applies to everyone, thanks.

      1. When did constructive critisism become offensive?
        The guy didn’t say it in a bad way, he just made a note about the proper use of words.
        Why is that annoying, because he pointed out an error?
        Maybe I’m old fashioned, but I was taught that when someone points out a mistake you’ve made, you thank them for bring it to your attention, so you can make improvements in the future.
        Learn from mistakes, rather than averting your eyes and pretending it didn’t happen.

        1. Thing is, I’m sure we’re all satisfied with our own ability with the English language. And if we weren’t, Hackaday isn’t the place we want to learn. It’s a “mistake” but language mistakes are irrelevant to the subject of Hackaday.

          I think the problem might partly be that engineer-types think with more precision than other people, which brings a tendency to pedantry, which is a virtue in engineering, electronics, programming, etc.

          But in other life it’s just rude. It seems like one-upping. Try constantly correcting people’s language in person, see how popular it makes you.

        2. @Royell
          Your mistake is in thinking the criticism is constructive. It’s a pedantic criticism in the truest sense: absolutely no meaning is lost, and the usage is common enough that it doesn’t make the passage awkward.

          You’ve also assumed that it was an accident or ignorance, and not a conscious choice to sacrifice correctness for flow or tone; that it’s not a regional use difference you’re not aware of; and even that it wasn’t a deliberate mistake to troll pedants. Hubris, like pedantry, is a common foible among engineery types.

        3. bringing it to your attention that you said bring instead of bringing.

          see, it’s annoying when some idiot corrects you for a typo, when everyone knew exactly what you meant anyway.

  3. This is such a rough initial concept – the final one will be sleek, flesh-tone, and won’t have electronics zip-tied to the outside of it. This one proved that we can make something that fits and is controllable, so now I can invest massive design time in this to come up with a full-featured version. Taylor’s existing prosthetic has myoelectric sensors, but to switch between grip control and wrist rotation control, both muscles must be held tense at the same time, with no indicator showing whether the mode switched or not. Something as simple as that indicator light would remove a frequent frustration, and that led me to think we can do better. Just wait, what is shown here is such a very crude first try – I won’t stop until we have something truly impressive, full-featured and greatly improved over the existing prosthetics he has today.

    1. Not always. Posted below is an article about a similar, if not identical situation. With free market, it usually comes down to, 1: Is there a customer base? 2: How big is it? 3: Can we get support from insurance/government?
      In the case of prosthetic limbs, there is a decently sized customer base. Currently growing, but not a truly steady growth rate. = bad investment. Also, when it comes to injuries like these, no two are truly the same. Spending millions on development of a refined, easy to use, and truly high quality prosthetic are wasted if there’s only say, 20-30 amputees with the type of injury that matches the design. Even if you have a prosthetic that meets the needs of a wide group, you still have to depend on investors to wait through the government/insurance vetting process. Which can take awhile.
      You really do need the government to step up and not only support existing companies, but offer contracts to companies/entrepreneurs who can design/manufacture near universally adaptable kit.
      http://www.technologyreview.com/featuredstory/531761/paralyzed-again/

    2. This isn’t a market thing at all, it’s a research lab consulting with it’s users.

      If you want an example of free markets, compare the price Americans pay for private healthcare, with the amount people pay in countries with national health services.

      1. We don’t have private health care in this country ever since the anti-trust laws of the 1930’s, which were the beginning of artificial constraints and favors to the politically connected, aka fascist economics. If you want to criticize how much we have to pay for health care in the US, look no further than our out of control government.

    1. Oh damn I feel so bad, I missed that he was missing both. Whatever I’ll be sure to work extra hard to get guys like this super forearms. These projects are amazing mostly because of the smile they put on the amputees face!

  4. One problem with most of these 3D printed hands is the arrangement of the joints doesn’t follow that of the human hand. Fingers aren’t all the same length and the metacarpophalangeal joints (MCP) where the fingers attach to the palm are not all in a straight line. The staggered and curved positions of the joints make it easier to grasp irregular objects.

    Hasn’t anyone designing one of these hands *looked at their own thumbs*? The motion of the trapeziometacarpal joint (TMC) (AKA first carpometacarpal) joint naturally folds the tip of the thumb to between the MCP joints of the middle and ring fingers. Hold up one of your hands and move the thumb in and out, without trying to move it towards either edge of the palm. A more natural TMC angle with a manual adjustment to change where the thumb tip folds into would make the prosthetic more compatible with grasping different objects. At least one of the very expensive hands has a TMC that can be manually adjusted in and out while the MCP is closer to a natural angle.

    For a myoelectric control, if the nerve signals controlling the TMC joint angles can be isolated, or retrained to other signals, it would be possible to make the thumb motion even more like a natural thumb. The thumb has the most complex joint motions in the hand. The TMC pivots a large amount in two axes and can rotate a bit. The MCP has a large range of bending motion and a slight side motion.

    Making any prosthetic hand that can duplicate the sideways movement of the MCP joints will require more advances in miniaturization and motor strength, as well as advancements in interfacing and signaling to control those motions. The index and pinky MCPs have a large sideways motion while the middle and ring can move sideways about half as far.

    With myoelectric hand and arm prosthetics there’s a ‘catch 22’ that goes along with how much of the natural arm is there, and whether its deformity or injury. The less natural arm there is, the more room there is for actuators and electronics, but the farther away from the end points of nerve signals for controlling the motions of the hand – if the nerves are even there, which they may not be with some congenital deformities.

    With 3D printing there are no constraints on the position and angles the finger joints can be, thus there’s no ‘need’ to make the designs so artificial and machine-like with straight line and right angle placement.

    1. I suppose that’s Step 2. Though the human brain can compensate for a lot, if it means existing prosthetic hand users have to use a different style of grip. It’s not too long ago that prosthetic hands were just motorised claws. And before that, unmotorised!

    2. The hand that I used in this build is the Dextrus V1.2 developed by Open Bionics. It has the fingers arranged in a staggered arrangement of knuckle positions, with a shorter little finger, and a different digit for the thumb. The thumb also has a servo articulating it between a flat and opposed position. The video of my arm controlling the hand shows the articulation better. When we tested it on Taylor, I was missing a few motor controllers that day and could only do the thumb and index finger.

      Moving forward, Open Bionics has some amazingly more lifelike designs that are printed in flexible filament, of which I have a skin colored spool in the mail right now. Their hands have always seemed to be more realistic, more naturally moving, than the others I have been able to find available for download.

      Really, our project will do more to develop the control scheme. I have recently had a successful test of an arm socket which allows Taylor to spin the wrist using his natural arm movement (not the myoelectric sensors), and this can combine to make an easy menu select. Twist all the way to one extreme, and a button is pressed. While that button is pressed, the various lights indicating grip modes become active, and the myo’s detecting a “close” or “open” muscle pulse will index it one position at a time, so you can twist the wrist, “close, close, close”, and then rotate back to normal now in a different mode (ie, going from full-grip to now controlling the wrist flex up/down). Quicker, more intuitive, and customized to work the best for Taylor’s preferences.

    1. If a metal hook was better, people would wear metal hooks, and the chaps here would be working on better hooks. All this prosthetic myoelectric nonsense would seem to be worth bothering.

        1. Oh and additionally, people have been using hooks for centuries and there are even today people who use hooks, and sure a large percentage because they can’t get anything else, but also a percentage who just find it the best available option even in the developed world, although I understand there are hooks that have a sort of grabber action attached to it, but still basically hooks.

          And yes we should and are developing better stuff and it’s available already for a certain price for decades, but I was talking about the design in the video of this article compared to a hook, which isn’t beating a hook so far. But since it’s the homemade version it takes time – as I said originally.

          1. I can only presume Taylor knows about hooks, and yet he chooses to use this robot thing. Between him and you, assuming you’re a quadlimbed person like me, I’d take his opinion.

          2. Proof of concept. Proof. Of. Concept. This is what amateurs came up with and put together in my garage. We’re not claiming that this one has any practical usage, as currently pictured. This one proves that it can fit, and that while worn, it can be controlled by Taylor. Those are the only things that this one was made to test.

            Now that we know it can fit and can be controlled, the focus will be on making the design have more features (grip modes, wrist control, etc) and then slimming things down and streamlining them to where it looks more like a natural arm in terms of size and color. Then we can start comparing it to others.

          3. Nevertheless, there are still plenty of people who use robotic hands, there are companies that sell them, as we know. So obviously some people prefer them to hooks. And I’d assume they’re the experts on usable hand replacements.

Leave a Reply to twdarkflameCancel reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.