Home automation for people with disabilities


#1

A little personal confession: for a long time I’ve wondered why I do home automation projects. I’ve been doing it for so long that it’s almost a habit. Mostly the reason has been “because I can”, and I like solving interesting problems. There are definite benefits, sure, but when I meet a random non-technical person and tell them about the projects I build, they often wonder why I do it.

My perspective on that has changed a lot since late 2017, when I started working with a couple of friends who use electric wheelchairs for mobility. This is a huge topic with many nuances that I won’t go into right now, but I’ve just posted the video of a conference talk I gave last week about my recent assistive technology projects. Whatever your motivation for doing home automation projects, I hope you find this interesting and I’d love to hear feedback:

www.superhouse.tv/video-of-my-conference-talk-open-source-superhumans/


#2

Jon, I love that you have a passion for helping others with your vast knowledge and skill. It is actually really inspiring to me personally and I truthfully am striving to learn electronics to do the same. I am no where near your level of skill but hope to be so one day. I think that assistive technology is a horribly money dominated market and your work starts the conversations that are needed to try and make the impossible possible for so many with disabilities. I applaud you for your efforts and as always, I look forward to new information telling the world what you are doing. I am rooting for ya!


#3

Hi @jon I’ve watched a bunch of your Superhouse videos for some time. I was particularly interested when you mentioned you’d been working on some disability-related projects. I acquired a disability through a spinal cord injury about 40 years ago and have used a wheelchair ever since. I’ve generally been reasonably able to do things. As I’ve gotten older and experienced some extra medical issues along with some other disability-related problems, some of those things have become more difficult.

As an electrician in a former life and always having an interest in tech which I’m sure was started by my father who was also an electrician, it always seemed to me that technology could help a lot though it has been very expensive. Medical and disability-related equipment can be ridiculously expensive in fact as the companies making this stuff know they basically have the market at their mercy. Many modern off the shelf smarthome products are making that easier and more affordable but they don’t offer all the answers of course.

I’ve started in a small way myself using Amazon Echo and Dots and a bunch of commercially available devices made by Itead and a few other brands. I’ve also started looking into homebrew solutions like the FauxMo stuff to solve a few other problems. Waiting for some Ubiquity networking equipment and some blinds to be installed.

I’m particularly interested in making a remote control unit for my powerchair so that I can move it without being in it. This would be really helpful at times. My current chair has been under warranty so hacking it wasn’t an option. I had looked into doing it with an attachment that could simply operate the joystick on the chair using proportional servo control. I have no way to work out what protocols and signals might be sent by it’s control system but think an inline joystick emulator would be far superior.

I am excited to see someone with your knowledge and ability adding to the pool of people working on these sorts of devices. Really looking forward to seeing what the community come up with and hope to contribute in my own small way.


#4

Thanks Guru and Peter :slight_smile:

The price of assistive (and medical) technology is a really tricky topic. The prices are ridiculous, but on the other hand I can understand the sequence of events that has resulted in those prices. The people and companies who develop assistive tech aren’t necessarily trying to be unreasonable with their pricing, they (mostly) are doing their best to produce devices that make lives better, but internal forces such as the need to maximise profit combined with external forces such as the regulatory environment and the structure of medical services paid by insurance companies results in prices that look insane.

There is also the issue of economy of scale. Some of the most advanced technology in the world today is contained in our phones, which are miracles of engineering. If you look at everything a phone can do, it should cost way more than we pay for them. But because the market for phones is billions of devices, one-off engineering costs are amortised across a huge number of units.

While I was in New Zealand 2 weeks ago I heard an interesting story that illustrates this point, and ties in to the thing about insurance companies and regulation. I’ll keep this anonymous, because the person who told me about this works in the industry helping people with disabilities get set up with technology, and I don’t want to burn them. They said that a client had a need for a specific function that could be provided by either a $200 mobile phone with a piece of software on it, or by a $10,000 assistive technology device that was created specifically for the same purpose. The $200 phone would totally solve the problem for them, and of course would then provide a wide range of other features as well. The $10,000 device would only solve that specific problem.

But the $10,000 device is classified as a medical device, so it can be funded using medical insurance. The $200 phone is a consumer device, so their medical insurance refused to cover it. The person could buy the $200 phone themselves, buy they’re in financial difficulty and can’t afford it out of their own pocket.

The result? The insurance company pays for the $10,000 device instead of the $200 phone!

To the person who needs the device, this solves the problem. But the extra funds paid by the insurance company have to come from somewhere, and that “somewhere” is the fees that ultimately come out of everyone’s pockets.

So it’s a systemic problem, with many facets. I want to side-step the system as much as possible to efficiently solve problems and make people’s lives better, but the single biggest barrier is regulation around supply of assistive technology. Once again I understand why the regulation exists, but in many cases it simply prevents people from receiving technology that could make a big impact in their lives, or results in it costing $10,000 when it could cost $200.

Aaagggghhh, this has turned into a rant! I’ll stop myself now.

Regarding chair control, that’s something that Nick, Chris, and I have spent a lot of time working on and talking about future possibilities. We’ve achieved chair control using an Arduino to send fake joystick input to the Q-Logic controller on a chair, so we have the ability to remote-drive a chair but we’re not actively doing that right now. So far we haven’t added sensors or anything else for anti-collision, but that’s on the list and I think it would be important. Many people don’t realise that an electric wheelchair is a 200+ kg robot with very powerful motors, and it’s quite capable of ripping plaster off the wall or smashing a table if it’s not driven accurately. We want to add collision avoidance, SLAM (local mapping), and waypoint navigation, so it’s possible to tell the chair “go to the kitchen” or “park yourself in the corner” and have it just do it.

I could talk about this stuff for hours, and this post is too long already!


#5

Oh, Peter, I forgot to mention that I have a pile of Ubiquiti boxes sitting right near me, because I’m about to do an episode about the right way to set up WiFi for home automation. I’ve ranted against WiFi on various videos, but that’s only because I constantly have people telling me that I should stop using wires and use WiFi exclusively. WiFi is a critical technology for home automation and I use it a lot, so I figured that I should do a guide about how to set it up properly. That will be after another video I’m doing right now about how to install and configure Tasmota.


#6

@jon. Thanks for the reply. I do get where they are coming from. We are a limited market compared with potential smartphone buyers. While the NDIS often won’t fund hardware like a phone or tablet, it will fund disability-specific software for them. I still think there is a certain amount of corporate greed involved when a few dollars worth of parts sells for a couple of hundred dollars. A perfect case is the pulse/spO2 sensor used on hospital obs machines.

It’s interesting you should mention that their chairs use Q-logic. Mine has a Q-logic 2 controller and they now have Q-logic 3. I don’t know what the differences are. I think the Q-logic 3 now has the ability to be setup using bluetooth from the dealer’s Windows software rather just their proprietary cable. I know mine can control a phone or tablet via Bluetooth but I don’t really need that.

I hadn’t thought as far ahead as having the chair go somewhere autonomously on command. I had only considered controlling it in line of sight with a short-range small hand controller so wasn’t going to use any anti-collision sensors.

I had thought of just buying another Q-logic 2 joystick controller and hacking about with that and adding the mods directly to it, but even 2nd-hand they are expensive.


#7

This is certainly an interesting topic. I think that besides the insurance scam, economies of scale plays a huge factor along with compliance testing. You’re getting high-quality (hopefully) components which includes multiple injection molded parts when the production runs are in the hundreds or thousands instead of millions. Then the actual requirements on the engineering of the components is very expensive versus generic, non-compliance type products like novelty toys.
Have a look at this video from EEVBlog just talking about the power supply for a medical device.

Seems like it should be simple, right? A Mains to DC power supply - four diodes, a transformer, voltage regulator and a few passives should do it, right?

So the good news is also the bad news here - the medical grade devices are [generally] orders of magnitude better - higher quality components [hopefully!], better design, better; more stringent testing. So in one respect you’re using a sledge hammer to pound a nail, but in another respect you’re using a sledge hammer guaranteed to never smash your thumb and never miss the nail or bend it off to the side.

You could actually argue that some of the components are that expensive to AVOID insurance costs with being sued for faulty products.

And sadly, at the manufacturing company I work for, I quote out custom projects. I’ve had to tell people that it would actually be /more/ expensive to remove a standard option on a machine than to just get it and not use it. Because once we add in the engineering hours and anything custom to ‘cleanly’ remove an option and add filler plates, it is actually more expensive.

I literally had this scenario happen last week - a customer wanted to order cables ahead of time that come stock on the machine. They could order all the individual components (plug backshell, 14- conductor cable, and pins) for a somewhat reasonable price of less than $100. BUT if they wanted us to assemble the cables so they are pre-terminated, then it would be over $1,100 /each/ with a hefty one-time engineering charge just to create a new part number.

So that is the engineering world we live in…


#8

I’ve had to tell people that it would actually be /more/ expensive to remove a standard option on a machine than to just get it and not use it. Because once we add in the engineering hours and anything custom to ‘cleanly’ remove an option and add filler plates, it is actually more expensive.

Absolutely, this is a great point. I’ve had Freetronics customers ask me to supply boards without headers installed, which they assume means it will be cheaper. But no, what it means is that I have to communicate to the factory that when they do the next production run, they have to pull a certain number of the boards out of the normal process at a specific point, and skip some assembly steps. Then those boards have to be kept separate in packaging and inventory. And somehow they have to be tested, but now they don’t have headers, so the normal test jigs don’t work, so there has to be a modified test procedure for them. Oh, and if there’s a warranty claim against those boards, we need spare boards without the headers on hand for replacements… The problems just go on and on.

Doing a special run with a part deliberately omitted doesn’t make a board cheaper, it makes it many times more expensive to produce.