Please help Wheelchair for disabled help

Thread Starter

Vinod Antony

Joined Aug 8, 2015
2
Hello
Our wheelchair will incorporate an EEG sensor for control using brain (we plan to buy the neuro sky brain kit), MEMs accelerometre for gesture control, a HM2007 module for speech recognition and ultrasonic sensors for obstacle detection as this is a wheelchair also meant for the blind. My doubt is how to integrate all this into a single board.
Will it be possible to use just a single processor for all this?
If yes which one should we go for?
Can all the control be done through raspberry pi or arduino ?

Please help
 

blocco a spirale

Joined Jun 18, 2008
1,546
This is a very big question and you have provided very little information for what is a very complex project.

This is not how such things are designed.

I assume that you have already developed the $100 brain interface to the point where it provides the required outputs with sufficient resolution and reliability to control a wheelchair; Because, if not, why are you even asking whether or not it will fit on a single board or what type of processor to use when these are not the biggest challenges facing you?
 
Last edited:

GopherT

Joined Nov 23, 2012
8,009
Hello
My doubt is how to integrate all this into a single board.
My doubt is whether you have the skills to complete this project.

Could you please let us know what level of skills you have and what other projects you have completed to this point. Basically, I am hoping that you have more than bearing able to find an ECG module and knowledge that Arduino and Raspberry Pi exist.
 

Thread Starter

Vinod Antony

Joined Aug 8, 2015
2
Hello,
Thanks for the replies.
We are final year electronics and communication engineering students and this is our first main project we are undertaking (I hope this answers the "level of skills" part). I do have experience with Arduino (line follower robots).
First off, this is not an original project. We searched online for several wheelchair based electronic projects and came up with this one. We wanted to take up a challenging project and one which is as complex as this one.

We also have not started any work on it but will start soon once we get a complete design.

very little information for what is a very complex project
Sir, I would like to know what information you need in order to help.

Thank you
 

blocco a spirale

Joined Jun 18, 2008
1,546
Hello,

We also have not started any work on it but will start soon once we get a complete design.

Sir, I would like to know what information you need in order to help.

Thank you
Have you investigated the viability of your idea? I would hope that as final year engineering students this would have been the first thing you did.

So, to return to the obvious question; have you developed the EEG brain interface to the point where it can provide the outputs required to control a wheelchair or have you any basis to believe that this is even possible? There is no point proceeding with anything else until this stage is complete (although if you can develop the EEG interface to provide useful output this would have sufficient merit on its own)
 
Last edited:

AnalogKid

Joined Aug 1, 2013
10,987
I agree with blocco. Stories about things like this show up on 60 Minutes from time to time, and they sound great. But they take years to develop the software algorithms. Compared to the EEG part, everything else in your project is senior science fair level. Not trivial, but *relatively* easy.

And I strongly recommend against a single processor handling all functions. Better to have a separate uC for each of the subsystems, pick a communications phy layer like I2C, develop a messaging protocol so they can talk among themselves, and have an independent uC as master controller that does the user interface and integrates the information from all of the sensor subsystems into control signals for the chair, displays, etc. Separating the functions will drastically reduce development time, and make the whole project more manageable, testable, repairable, and upgradeable.

ak
 
Last edited:

crutschow

Joined Mar 14, 2008
34,285
I agree with AK that having each part of the system use a separate processor will significantly simplify the design process. It makes it much easier to break down the design for individual design groups to work on parts of the project.

You would also need some safety and redundancy circuits for a real system. But you can leave those out for a proof-of-concept design.
 
Top