I know this is not a regular and daily question, but i want to know how can we create an ECG cap(according to my research this is the simplest and cheapest mechanism) to get at least 4 or 5 different signals so that i could use it to hook it up to a robot!!!
please enlighten me on how difficult it would be and how much the cap is going to cost.
kiriti:
I know this is not a regular and daily question, but i want to know how can we create an ECG cap(according to my research this is the simplest and cheapest mechanism) to get at least 4 or 5 different signals so that i could use it to hook it up to a robot!!!please enlighten me on how difficult it would be and how much the cap is going to cost.
Well, you're in luck. There's a game called MindFlex, and it contains an EEG head cap. There is also a library called Brain that allows you to use the output of the game (possibly requiring a little hacking.. not sure). Anyway, drive up to the top of this page, and enter mindflex in the search box. There are a lot of postings and other info on here about it.
The game can be had, retail, from Amazon for about $100, or from eBay for a llot less. While you're on Amazon, check out MindFlex Duel (link on same page as MindFlex Game), which has two headsets, and is about $50.
thank you for the reply and i looked into Mindflex but it seems to just measure the intensity of the thought that is it can give only high or low output, but i want to make an EEG cap which might be able to decode atleast 5 - 10 different thoughts so that it will be feasible for our robot, Is it true or can the mindflex can be used to decode different thoughts apart from the high and low intensity.
You should research the projects in this site, the Playground stuff, the Brain library, and reviews on the MindFlex.
There is no technology currently available (or at lest inexpensively), that will distinguish between 5 or 10 'thoughts'. You can train yourself to get certain waves to dominate your patterns, and once you do, you can use these to command something, but it isn't very selective, and it is time consuming, both in learning and in actually commanding actions.
so i have researched and have found only ON/OFF control with the arduino using the mindflex
but i wanted something like this...
where it can differentiate between left and right fist and both fists closing simultaneously, so is there a way to do it to make a homemade headset under 150$ which is sure to work and can fulfill my requirement??
or how about using 2 mindflex headsets? one on left and other on right side of the head so that high activity on left side will be detected on right half of brain and vice versa
kiriti:
but i wanted something like this...
Mind-controlled quadcopter flies using imaginary fists
where it can differentiate between left and right fist and both fists closing simultaneously, so is there a way to do it to make a homemade headset under 150$ which is sure to work and can fulfill my requirement??
Well, there you go. If you can find more details on that, and if what you find is within your skillsets, I'd say it would be worthwhile. The internet is your friend. You'll have to do the research, though, unless you can find someone else to hel with that.
Virtually impossible imho even with a medical eeg it would be viryually impossible to consiously effect an output.
Mri machines can detect better, ie thinking moove leg lights up a bit of brain, why not have a footswitch then.
Best bet would probably be to detect eyee movement.
Quote from link
"This is something of a first because it required precise mapping of the brain. "We were the first to use both functional MRI (Magnetic resonance imaging) and EEG imaging to map where in the brain neurons are activated when you imagine movements," LaFleur says. "So now we know where the signals will come from."
Therin lies the key.
A 48 channel eeg would likely break you budget IF you could acess the mapping data.
Also have to shave your head, or did last time i saw this done.
So i am still working on mindflex to make a simple robot move in 4 directions, i have successfully cut and merged the 2 headsets into one and will be using 2 different arduinos to decode each of the headset and then combine them to get my desired output
i think this would work as i have tried it on the game and i was alone able to move the ball in the direction i wanted with a bit of lag and a wrong output 1 in 10 times....
but the problem is the headsets when connected to arduino show the meditation and attention values 0, all the others are showing up except for these 2...
what must be wrong?
