For the recent release of X-Men Apocalypse, YouTuber [Allen Pan] from “Sufficiently Advanced” decided to make a rather ambitious project — a working Cerebro Helmet. Wait what?
When worn, it allows you to steer the person wearing it to the left or right using a series of impulses to the brain. It’s actually a well-researched technique called Galvanic Vestibular Stimulation — basically if you zap a nerve in your ear it messes with your balance. If you’re trying to walk in a straight line and the electrodes trigger, it causes a sudden imbalance and you tend to lean in that direction — thus changing the direction you are walking. Kind of like remote control cockroaches — but… with people…
It’s not quite how Cerebro works in the movies, but it is one hell of a cool project. [Allen] said he had difficulty finding people willing to put the electrodes on. He should bring it by our main office sometime!
I wonder if Galvanic vestibular stimulation is going to see usage in the world of VR. The Wikipedia article says that it’s being investigated for pilot training among other things.
Huh, didn’t think of that, I wonder if there are any other nerves that you can just kinda poke like that to mess with your balance. Could potentially solve some of the balance problems people have with VR.
It’s definitely popped up here and there as an immersion aid for VR, but from what I’ve seen it’s just too awkward for general commercial use. Plus, I was controlling Kieran on the boardwalk for about an hour, and his skin under the electrodes got pretty red and irritated. Obviously I didn’t have the fanciest set-up, but with how long people can binge video games I’d imagine that’d become a problem no matter how nice the electrodes were.
That is a very nice and honest response. I would tend to think that at least a few folks would probably get queasy as well but cool build and proof of concept. I had not heard of it before. Thanks :)
Maybe better conductivity between electrodes and skin would distribute the current on bigger area -> less irritation.
Have you tried using electrode gel (like the kind used for ECG/EKG electrodes)? Perhaps that would lessen the irritation.
An idea just occured to me. I wonder how hard it would be to hack together an endless hallway simulator for the Oculus Rift or HTC vive and pair it with your helmet such that you can walk down the hallway, seemingly endlessly, while your actualy just walking around and around in a big circle in the real world…
Nevermind, I just noticed that both of my thoughts have been answered already further down the discussion.
Samsung is well into developing a commercial GVS for VR under the name Entrim 4D. I’m personally working on adapting my perl and parallel port controlled GVS to esp2866 so I can control it within Unity using the arduino interfaces.
Are there any details on the electrodes? I tried actual AgCl gel electrodes, and those weren’t very conductive and stung quite a bit. The wet sponge method those tDCS guys use worked the best for me, but I imagine that wouldn’t fly for a commercial product.
This is exciting preliminary work, someone else already mentioned that Samsung is working on commercializing it. However, I may be mistaken but currently we only know how to do left and right, not forwards and backwards or up and down. The ability to simulate the sudden feeling of forward acceleration/deceleration would be critical for a proper VR implementation.
> The ability to simulate the sudden feeling of forward acceleration/deceleration would be critical for a proper VR implementation.
Just swap endolymph with ferrofluid and stimulate it using electromagnets.
Even the simplest GVS setup will have a little bit of pitch, roll, and yaw happening too. And response to the same stimuli will change depending on head orientation. Luckily with a VR HMD the head is tracked.
The summation of vectors from each part of the vestibular system with typical bilateral/bipolar GVS is mostly just left or right. But with arbitrary bilateral/monopolar (the ground is down towards the neck instead of the other electrode) stimulation you can summations that end up giving perceptions of forward or backward accceleration.
I think it’d be feasible for a company like Samsung to figure out how to account for these things with some kind per-user calibration.
ref: “Non-linear vector summation of left and right vestibular signals for human balance” doi: 10.1113/jphysiol.2009.181768
ref: “Probing the human vestibular system with galvanic stimulation” doi:10.1152/japplphysiol.00008.2004 – this one has some clear visualizations
It has been tried for redirected walking technique (you think you are going straight in VR while walking in circles in reality), but it just isn’t very practical.
There are also potential health and ethical issues – zapping someone’s brain is somewhat acceptable in a lab under very controlled conditions, but that’s not something you can put in a retail product where you have no idea about who and how will use it.
Imagine the lawsuit barrage should it cause an injury by having you walk into something, trigger an epileptic attack or something similar. And that’s just during “normal”, as-intended use. You can’t just cover your ass by sticking a disclaimer on it making it the user’s problem instead – courts will hand you your backside on a silver plate should it come to it, because a lay user can’t reasonably expect something like that to happen.
Fake ?
I tend to think it is not a fake, due to how realistically the “victims” swerved and nearly lost their balance. It looked right, to me. But it would be easy (except for the losing of balance part) to fake by having the helmet beep in the ear in which direction they are supposed to turn….
Real! But I knew this build would be a hard sell on the internet, so I tried my best to include the footage of the remote and having the poor Youtubers blindfolded. I suppose I could have included close ups during the cup challenge to show that nothing sneaky was added to the hardware that would let them know what direction to go in, but no matter what I wasn’t going to convince everyone.
It’s pretty easy to do this to mice and bugs enough volts in the right place at the right frequency and you are in control. I doubt it’s much harder with humans. From a technical stand point. You drill into a person’s brain implant some electrodes and suddenly it’s your fault they piss themselves ever time they hear stairway to heaven. People will sue you over anything these days.
Closed source, it seems. :(
Not much too it, just a DC voltage across your skull. Just tried (not recommended) a 15V (research says ~6V minimum, but a constant current, 1mA, is recommended) power supply connected to some wet q-tips. The world spins nicely in the direction of the current.
I’ll have a more detailed video up later. Not a full tutorial, but enough that if you know what you’re doing you’d be able to replicate it, and if you don’t then you shouldn’t be putting current through your brain anyhow. I used the RC car more for the spectacle, in reality it’d be much easier to set up with Arduinos and radio communication modules. The RC car board controls a mosfet H-bridge, and the source is 12v with a simple 2 transistor current source set to 3ma. Heads are really resistive though, so I never measured more than 2.5ma going through anyone’s skull.
this meshes nicely with two seemingly unrelated expierences ive had,
expierence 1 (very recent)
hold a neon bulb wire (or high-efficiency LEDs wired in reverse-parallel so the OCV doesnt damage [max Vr=5v]) and touch the other wire to the wet dirt while standing directly under 500kV powerlines, it will glow dimly… you are the collection plate and dirt is other plate of air-core capacitor… i would assume it will not work (as good or) at all when barefoot :P
expierence 2
walk under said lines and you occaionally loose balance in the direction of nearest phase line… while sober and feeling normal.
I wonder what happens when you stimulate both nerves? Do you fall down or split in half?
Technically you are affecting both nerves, depending on what side is the anode and which side is the cathode, the vestibular system is quieted on one side and stimulated on the other, causing your brain to interpret tilt. I suppose you could have a third electrode somewhere else so that both mastoid process electrodes could be at the same voltage, but I doubt it would have much effect on balance.
Oh, I see what your saying. This seems like it could have a lot more applications than what you’ve done’ but I can’t think of them now, keep up the good work.
Holy moley, this is an amazing and exciting application of a phenomenon I never even knew about!
Thanks! Now you know!
Old but still funny GVS installation: http://www.junji.org/saveyourself/
Haha yeah, ran across that when I was researching GVS. You’d think there’d be more stuff out there given how novel and old the technique is!
This helmet + Cardboard VR headset + smartphone and collision avoidance algorithm = autonomous walking humans !!!
Eat this Google autonomous driving car!
You can read HaD while this thing take you for a walk !
If the drivers, bikers, pedestrians would have all those kind of helmets, the traffic computes could direct them more efficiently on a non collision course. This reminds me of something.
All hail Plankton!
So, we have the technology to recreate movie “Gamer”? Is that right?
Yup. Now fetch Gerard Butler’s phone number.
I’m thinking of blind people. Add some bat-like sensors and get machine for walking without a blindman’s stick!
Now do it on a bike. :-)
Well, I guess they’d technically be wearing a helmet at least…
is it possible to get forward and reverse movement? left and right is limited. need to be able to make person move forward by remote!
Hey I was wondering if you could post an in depth diy tutorial on this so people that aren’t suffently advanced(see what I did?) can understand and build this? I know many people would appreciate that.