This is the start of the Future Vision Robot. This electric wheelchair was left at UAT with no batteries and a non working motor controller. The first task that we had to do was dismantle the wheelchair and get to the core of the control system. We spent some time taking apart the joystick controller, this is the round PCB in the pictures below, we spent some time with an oscilloscope and with LabView and a DAQ (Data Acquisition) Device trying to under stand the serial commands that were being sent by the controller. After a few hours we realized that the joystick was also broken and there is no information online for how to communicate with the motor controller.
The next test for us was to test the motors… 3 power supplies and some fancy daisy chaining, and a melted wire later… we found out that both of the motors do work, so it was not a total waste of time.
I have since gotten a new speed controller for it and hopefully will be having another post on here soon with the next step of this project. Any small additions I will update in my blog and once i get to the next step I will all the updates in a new project post.
Great project!
Many have tried to ‘reverse engineer’ wheelchair protocol controller and have not completely succeeded. There is such a stringent safety requirement when it comes to wheelchair control and protocol. Most controllers use the industry approved Multiple Master – Multiple Slave (M3S) protocol (Google M3S). It uses a Controller Area Network (CAN bus) as the network layer which carries the M3S packet which contains control and safety data. Additionally, a ‘Safety’ command is sent at the same time the M3S data command is transmitted to ensure that the operator did in fact produce a movement command via joystick. This is to prevent a rogue/ runaway wheel chair. Most joysticks however are analog and can be emulated with a digital potentiometer IC using SPI or I2C to command the motors. I am working some bugs out to accomplish this with the four wheel chairs I have.
You might want to consider using a SaborTooth 2X50 to handle the larger wheelchair motors. I know it’s more money ($250), but able to handle A LOT!
Keep up the outstanding work!
Thanks,
When i received the wheel chair the communication and/or motor controller were not working, so the time that i did spend trying to reverse engineer the system was also to see if it might work. And about the motor controller, at the time that i bought 2X25 they did not have the 2X50, so i am hoping that i will be able to get this to work this week.