Pride Jazzy Select power chair stripped down for conversion to R/C Electric mower (YardBot). I plan to add a carry tray mounted above the batteries/blade motor as well as an attachable pull-behind wagon for hauling more yard trash/garden supplies. The battery box will be removed to mount the mower deck in it's place. I'm trying to stay with the original controller and tap into the joystick control with an R/C receiver.
I will try to keep this updated as I've seen quite a few people wanting a cheap way of adding R/C control to a power chair. My main batteries were low so I can't get an accurate read on the joystick output but it looks like it uses a reference voltage for 'zero' (no movement) then swings + or - for directionional info. I believe it operates on 5v so I the reference point would be half but I'm reading 4.2v and swinging from 3.8v to 4.6v (referenced to ground, not the center tap). Will update when I have exact values plus the pinout for interfacing.
The R/C I pulled from a toy truck has 2 channels that swing above and below 0v by an amount equal to the input voltage (in this case 7.5v). I'll need to convert that to coincide with the voltages expected by the power chair controller.
The Jazzy Select power chair has a motor controller mounted on the base with a 3-wire cable (red, white, black) running up to the armrest control pad. The control pad has a power button, horn button, speed control dial and joystick. The 3 wires are +V, Gnd and Data. I'm not finding much info on the Internet and service manuals don't seem to be available so right now I'm guessing. The joystick pad has 24v going in but only sends 5v to the joystick. I'll have to check if the speed control dial has any affect on the joystick input voltage. From what I gather so far it all seems to work according to a reference voltage which might be programmed into the motor controller. Soon as I nail down the pinout and specific voltages I'll build a circuit to convert the output from the R/C receiver to the required levels the control pad expects to see from the joystick.
After that the mechanical engineering starts where the battery box is removed and a mower deck inserted between the motor/controller section and the swivel wheels. The 2 12v batteries will mount on top of the deck with the blade motor. Plans will include a 'come-here' feature to free the user from having to drive it to their location. Maybe a 'follow-me' function too.
9/25/13
YardBot has a name, B.R.Ü.T. - Basic Roaming Utility Transport (couldn't think of a good 'E' word) and now it's too late to go work on the project (plus I can't find my small breadboard) so I'll give a progress later this week.
Hauls stuff around yard, not autonomous.
Control method: R/C controlled but avoids obstacles on it's own
I plan to make a robot platform based on a electric wheelchair, an arduino (I just bought it so I need to learn first) and a Spektum R/C controller, so you are ‘collected’ so I can watch your progress as I learn too…
That’s why I put this up. I didn’t find a whole lot of info even though it’s getting popular, especially for lawnmower builds. For starters it will just be an RC electric mower and ‘yard mule’ but if I don’t get too sidetracked with other projects I’ll add more features. Any info I uncover will be posted here.
I believe I have the same model … or nearly the same model - got it several years ago, when I was less electronic savey
Anyway, if the controller is like mine (mine has long ago died) - it had “many” safety features - one was a reference voltage where it was very terse. I don’t remember all the details. The parts I do remember :
no manuals on the electronics (this isn’t an open source wheelchair after all)
lots of irritating safety features - which are probably great for someone actually using this as a wheelchair
the MOSFETs used really were impressive - for the amount of current they controlled - (I fried these too
If you don’t know what your doing, you can easily fry stuff - I’m better, now I occasionally brown my electronics, but I don’t burn or fry it
If you build your own controller - the inductance of the large wheels and motors can play havoc with your H-bridge (don’t go here)
Good luck - btw I’m about to re-open my wheelchair into a new project, Not bothering to attempt to build my own H-bridge again, but thinking of ganging a coupld sabertooth controllers together.
I did notice a lack of available technical info on these power chairs, no service manuals offered online. If the built-in safety features become a problem there’s an option to re-program the controller but I suspect the software and info will be hard to acquire. I’m going to try working with the controller first.
The videos I’ve seen of the remote controlled Jazzy base shows them zipping around using a replacement controller but that’s not how I expect a lawnmower to operate. Those safety features (anti-tipping, slow ramp up to full speed, etc) could be a plus on a RC mower or even a large robot (my other project, not yet listed).
I wanted to ‘talk’ directly to the motor controller but that looks more complex than what it’s worth so I’m looking at interfacing the RC receiver with the joystick control board. Seems to me it should be a simple circuit to convert the ± voltage output of the receiver into values similar to the joystick output. However, since the answer currenty eludes me plus I’ll eventually add a microcontroller for obstacle avoidence or other features, I now plan on doing the conversion with an Atmel328.
While I’m waiting on the components for that I’m moving along on my bigbot project which I’ll post soon. That’s based on the other chair I got in the deal, a Drive Medical Cirrus Plus.
Was going to comment, just saw that your project is almost at it’s one year anniversary… Curious what became of it…
I am working on my first robot, which i plan to post some information on tonight when I get home. It will be a little crawler, maybe a remote monitoring and navigation type bot.
My ultimate goal is to make use of a wheelchair base like you have done. I was lucky to get one that my brother came across which appears to only have a problem with the batteries.
Yep. hard to find one cheap with good batteries. What make/model do you have? Invacare, Pride Jazzy (this is a Select), Drive Medical (BigBot Project base)?
This project stalled so I switched over to my BigBot project, https://www.robotshop.com/letsmakerobots/node/36684, but just recently started back at it. I’m at a temporary stopping point with the BigBot and working on the body’s frame, arms (and how to attach) and head. I need to get the body frame done to attach the motion sensors.
Meanwhile I’m back to work on this one and hopefully I’ll have an update to this page when I get past the current hurdle. Actually I should be working on it now instead of goofing around on LMR (again).
I started with a test platform, https://www.robotshop.com/letsmakerobots/node/32968, so I could get the basic code right and get familiar with the Arduino, software and sensors etc. so I would recommend a smaller project to start, then transfer what you learned to the wheelchair.
Note that the dremal box is just temporaily holding the electronics to keep them dry (it was sprinkling rain out that day but my younger son wanted to try it)
I now have aluminum skin on it, but I have a long ways to go.
That first video looks like what I have… I love how maneuverable that thing is… Curious… did you tap into the joystick controls or are you controlling the motors directly?
I could not find an easy way to tap into the controller that allowed a quick and snappy response. The builtin controller is designed to allow control, but it seems to have some built in throttling. For example… when using the build in controller you can go backwards and quickly snap forward and the machine will just stop and go forward. With direct control (not the built in controller) you can go backwards and snap the throttle forward and the thing will practically do a wheelie.
That’s why I’ve been trying to tap into the stock controller, I want those safety features on a lawnmower. So far it’s been beyond difficult, at least for the one I have, and I’m about to pack it in and go back to work on my BigBot Project. I’m hitting too many obstacles and don’t have the skill to do what I was wanting.
