TL;DR: Seven hard reasons why analog cable control was and remains the way to go for any critical use / application. Background is that bicycle industry has a clear relationship with electronic wireless brake control: they don’t use it. For great reasons. How does that spell out from a prosthetic arm user view? I use the transcript of a great bicycle part video and comment about aspects that relate to prosthetic arm control. 

Video transcript and commentary, “would wireless bicycle brakes actually work?”

(C) own art / AI

[LINK TO VIDEO]

Text Video Content (C) copyright GCN Tech Show – Dec 16, 2021 GCN Tech Show This week on the GCN Tech Show: Alex and Ollie go through what we know about wireless braking and ponder the mechanics involved. Plus, new Canyon Grails, bearings that don’t need grease, Bianchi brings production home, the year in numbers by Strava and all your favourite GCN Tech Show fixtures!

should we go ahead and maintain points yes which is this week about wireless brakes and when we’re going to see them on road bikes we’ve seen wireless gears on road bikes and they have a number of advantages they can be lighter easier to set up because you haven’t got to route the cables through the frame and stuff like that yeah and wireless brakes in theory would have these advantages too not having to root hydraulic lines through your fork and your frame would well make maintenance far easier building your bike much easier and just reduce complexity of the system it could also save you a lot of weight because our cyclists love lightweight stuff we do but it’s not without its challenges for example and this is probably the biggest one what if it stops working well if it stops working like when you’re on a descent you have no breaks so a big challenge is how do you build in fail safes into such a system so that if the signal does cut out or the battery goes you can still stop well how would the brake lever communicate with the brake caliper i mean how would the caliper even be actuated

Fact 1: not a person in the world has an idea about what to exactly do in order to RELIABLY bring a failed myoelectric prosthetic arm control back at once where users and industry and academia apparently are stalling at control errors far from useful [link, link] – and bicycle industry cannot bring a failed electronic brake system back as well – where industry and users alike seem to totally chill about that.

because currently in hydraulic systems you’ve got a piston in the lever which is then pushing hydraulic fluid to the pistons in the caliper making them retract and open up so it does beg the question how would the lever communicate to make that caliper actuate yeah you have to have some kind of mechanism in the caliper as well yeah we’ll find out more about the challenges involved of wireless brakes and if it’s something we’ll ever see on bikes i actually spoke to paul cantor from sram who’s one of their lead engineers oh yeah does he have anything good he did actually it was very interesting chat we had and well the first thing he pointed out was it is something that sram has looked at but as an academic curiosity or as like a thought experiment it’s not something that they’re like at all working on and it’s not something they’re anywhere near to actually making a reality but it is something they thought about they’re not coming next week then okay no um and he did the first thing he said was while it’s great to get rid of cables and reduce that complexity from the system you take one thing away but then you bring in a load of other problems that you have to solve so the first sort of issue he highlighted was the size of the caliper would get really big he reckoned it’s not going to be good for aero is it no but you know you’ve then got to by not actuating the caliper with fluid like we’re i’m gonna leave it like we mentioned you you then have to have some kind of mechanism in the caliper itself like a little motor or something there are ways you could do it but his understanding was that with the current levels of tech it’s gonna make it big

Fact 2: myoelectric arms add redundant motor and battery bulk and weight and not a person in the world has actually explicitly stated that these are superfluous and useless as long as there are existing bones and muscle; on bicycles, the same is undisputed as it constitutes the reality but: funny how a bike mechanic terms two simple reliable brake cables a thing of ‘complexity’.

and then he said that reliability was a problem because if you take the current wireless tech on etap it’s capable of doing millions and millions of shifts without like a failure a failed shift or a loss of connection and it is incredibly reliable and works really well but they said for a breaking situation even like one in one million shifts not going through they said that would be unacceptable for them yes because obviously the ramifications of when you need to break and can’t break are huge it kind of makes me wonder of what’s possible with today’s tech and what are some of the like complications that they’re going to have to try and overcome well he’s

Fact 3: real life failure rates must not exceed a far, far lower rate, than what current academic research plays with and finds cool and bicycle users and industry seems to acknowledge that and be cool with it, millions and millions of shifts and brakes and ‘incredibly’ reliable,  but that is really where it’s at in terms of failure rate acceptability. 

well he he said you know in theory it would be possible to build something with today’s tech it just probably would be quite big because of the caliper size but you could you know feasibly make that wireless communication more reliable and and potentially get there in that respect and then the caliper itself could be actuated by small little motors and you could develop something similar things to what are in derailleurs i guess that kind of small little motor yeah yeah but then you’d obviously need to perhaps build in some fail safes too in in case it did fail or the battery went or whatever actually talking to fail so second we’re thinking of other industries or applications where they have a fail safety option and the thing that comes to mind is lorries so i don’t know if you’re familiar with lorry braking systems are you i’m not no okay so they work on a completely different principle to a bike and a car so on our bikes when the brakes are at rest the pistons have retracted so the brakes are off and the wheel spins really freely whereas in a lorry when the brakes are at rest and there’s the systems all switched off the brakes are effectively locked on so that means should there be a problem with any of the system and it was to cut out fail or whatever it would just apply the brakes so that is a like really good fail safe because if it all goes wrong brakes just apply yeah i mean you could there are potential failsafes you could build in like that but then it’s that thing of like how does that fail safe apply because does your brakes just lock on as soon as like imagine riding your bike as well i’ve just got this thing in my head of going down a big descent pulling the brakes imagine they didn’t work i mean that would terrify you in the first instance and then you’re going are they just going to apply like lock on are you going to go fly over the handle your brakes fail then it’s just like yeah just anchors on or do they just gradually apply and you’re sort of touch and go whether you’re going to slow down enough to make the corner well this this this actually then raised another thing so after saying that it was possible with the tech in theory to kind of build something

Fact 4: electronic fail safe mechanisms may suck big time, be super bulky, and bike specialists recognise and respect that.

there’s another issue that i hadn’t thought of at all which is the sort of ergonomic feedback you get from a brake lever so when you pull the brakes you feel the modulation through the lever you feel the amount of pressure that you’re putting on and you get feedback in turn from the brake caliper because they’re physically connected to each other whereas with an electronic system you’re not going to have that ergonomic not connected because they’re not connected so you’re going to have to build in some kind of way of artificially recreating that modulation that feedback into the lever which again is increasing complexity so it’s like you’ve removed one thing but you’ve created a whole thing i was just thinking you saying that made me sort of think that’s not going to be easy it’s not going to be cheap and it’s going to make the lever system far bulkier than what it would

Fact 5: cable control offers analog scalable comfortable light weight elegant permanently available reliable feedback / sensory feedback control that does not add bulk, and bike mechanics have clearly understood that. 

and that’s a time when everyone’s trying to make the levers as small as they can well the advantage is you do remove the hydraulic piston from the lever which in the reservoir which is quite a big sort of thing that’s in there but you you yes it’s it’s going to potentially increase complexity and the the thing is with what i would liken it to for people who maybe don’t understand what i’m trying to explain here is it’s a bit like if you play a driving game on a on a console and i’m always just press i know you are you just press a button on your controller to accelerate or to break and that’s just like it’s like an on off switch isn’t it it’s like it’s just on yeah i mean there’s a bit of more you can’t change how much you’re applying but it’s completely different to driving in a car and having a pedal that you can push your foot down on where you’ve got so much control and modulation and that’s what it’s like with the brake lever in your hand that’s a that’s like a really key like issue this isn’t it because imagine losing all of the ability of having feedback from the brakes you’re slamming them all the time you’re gonna oh sorry i didn’t realize i pulled a lot of my wheels every time yeah that terrifies me so for wireless shifting having that just basically you press a button effectively and it’s on off shift or no shift for that application perfect much more simple better suited but you are going to need to engineer this mechanical sort of human feedback loop into a brake lever which i think is fascinating it is fascinating

Fact 6: everyone is trying to make control systems effective, cheap and light weight, not expensive, cumbersome and heavy / bulky, and bicycle builders have recognized that.

so i guess in conclusion wireless braking is kind of a bit science fiction at the moment yeah (…) 

Fact 7: just because it is science fiction does not mean the idea sits on sound principles of engineering, as recognised by bicycle specialists.

Summary of cable control vs. myoelectric / remote control facts

As we can understand also from many prosthetic arm experiences, an electronic bicycle brake system is not one that is anywhere near bike rider acceptance.

The same is true for myoelectric arms: most notably, such systems seem to offer a staggering degree of lacking control and hardware reliability, most hand designs lack grip angle comfort, and they complicate life by adding unnecessary bulk and mentally insane cost. That, in essence, summarizes the myoelectric experience precisely.

The points from above, again:

• Fact 1: not a person in the world has an idea about what to exactly do in order to RELIABLY bring a failed myoelectric prosthetic arm control back at once where users and industry and academia apparently are baffled – and bicycle industry cannot bring a failed electronic brake system back as well – where industry and users alike seem to totally chill about that. Myoelectric failure cannot be prevented, that failure and the pure junk allure of that technology are here to stay. 
• Fact 2: myoelectric arms add redundant motor and battery bulk and weight and not a person in the world has actually explicitly stated that these are superfluous and useless as long as there are existing bones and muscle; on bicycles, the same is undisputed as it constitutes the reality. Any myoelectric failure prevention system as required add on to the already superfluous bulk will add more dead weight to the prosthetic arm. 
• Fact 3: real life failure rates must not exceed a far, far lower rate, than what current academic research plays with and finds cool and bicycle users and industry seems to acknowledge that and be cool with it. Acceptable real life failure rates are insanely lower than what academia “plays with” since over 40 years. 
• Fact 4: electronic fail safe mechanisms may suck big time and are big and large and bulky and bike specialists recognise and respect that. The negative aspects of myoelectric failure may have to get more stage presence. 
• Fact 5: cable control offers analog scalable comfortable light weight elegant permanently available reliable feedback / sensory feedback control that does not add bulk and bike specialists have understood that. Cable control wins our hearts each and every time – which does not automatically mean any uneducated tinkerer is able to understand and build these well, that, no. 
• Fact 6: everyone is trying to make control systems effective, cheap and light weight, not expensive, cumbersome and heavy / bulky, and bicycle builders have recognized that. Fully reliable cheap and robust lightweight controls are absolutely in – not out. 
• Fact 7: just because it is science fiction does not mean the idea sits on sound principles of engineering, as recognised by bicycle specialists. It may be just bad engineering to blindly run after Science Fiction ideas – and imitating these is not the same as providing a robust framework. 

What academic researchers tell us in 2023

(C) own art / AI

The above are hard facts and the contrast image used here is to exhibit the far reaching social failure of academic research and prosthetic industry.

• Both generally tend to treat all arm amputees as people they do not need to listen to. That is why the difference to bicycle industry however is so striking.
• The academic research uses “prosthetic arm” as a trigger word to get funding. Funding authorites are not well informed as to what is going on but before they ask arm amputees, they ask these researchers, so the funding authorities also seem to treat arm amputees generally as people they do not need to listen to. Last 15 years, academic research that produced tangible results that improve prosthetic arm design for real work: zero. They are, however, entitled to their opinions. Suit yourselves.
• Most industrial developers build what the customers want, and the customers aren’t arm amputees. Never were, never will be. They build prostheses for the prosthetic technicians – with few notable exceptions. So to most industrial developers, arm amputees are also people they do not need to listen to. They are however also entitled to their opinions, it is absolutely legal to think that. Suit yourselves.
• Prosthetic technicians need devices that give up / stall before they break so no one can be blamed with not covering costly repairs; myoelectric technology does that: it offers promise but no actual help. They also need devices that cost an insane fortune simply because they live off some 30% off handling the hardware alone. That much is rational. Deplorable but rational. So what we as arm amputees are to them is not even relevant – as long as they can sell myoelectric prostheses. Body powered prostheses are too functional and comfortable when built well, but regardless how well they’re built, there’s always a next failure point – so building bad body powered prostheses is what is going on at the moment. See the recent failure of a typical Fillauer product.

These shed a light into the belief systems of what we have to assume are authoritative voices of myoelectric prosthetic arm researchers:

• “[In CYBATHLON 2016 and 2020,] the prosthetic arms “race” was won by
  athletes wearing body-powered prostheses with one degree of activation only, a sad but instructive tuition for all of us [1]”. How is that sad, how can that even be sad when the authors have no own products that were ever geared towards near-perfect reliability?
• “The user’s intention can be obtained to control the artificial hand by using different interfaces, such as electromyography, electroneurography, and electroencephalography. This and other sensory information can be exploited by different learning mechanisms that can help the user adapt to changes in sensory inputs or outputs, such as reinforcement learning, motor adaptation, and internal models.”[2]  We did not know cables were even a thing. Like, at all. Right.

Note the difference in expression towards failure-prone technology between academic researchers of prosthetic arms and bicycle tech aficionados – the technical problems are the same.

@article{castellini2020upper,
  title={Upper Limb Active Prosthetic Systems—Overview},
  author={Castellini, Claudio},
  journal={Wearable Robotics},
  pages={365--376},
  year={2020},
  publisher={Elsevier}
}

@misc{gentile2023perspective,
  title={A Perspective on Prosthetic Hands Control: From the Brain to the Hand. Prosthesis 2023, 5, 1184--1205},
  author={Gentile, C and Gruppioni, E},
  year={2023}
}