Slashdot Mirror
Making Joysticks Obsolete
Alien54 writes "NASA has been taking a beating about their in ability to get the really important things done. ABCNEWS has a story that reveals exactly what they have been working on: Computer control of things like joystick interfaces and keyboards by reading the muscles directly. No more joysticks or keyboards. Just gestures to control, say, an aircraft. A more advanced gaming interface indeed! There is also a short video showing the current state of the art." Imagine what this device could do for the por-- I mean, gaming industry.
Imagine it: your plane's pilot suddenly has to sneeze, or he develops a nervous tic, or he has a muscle spasm.
"Folk's, I apologize for the rough flying back there, the pilot had to scratch an itch..."
What fun!
About a month ago I watched a television show about people with disabilities. During the show they had this mind control based program attached to a person's head which allowed them to think of which way to turn things. The program was dropped by the USAF (or they said it was) but continues for paraplegics.
I don't recall the complete set up they had, but it seemed like electrical wiring attached to the backside of the persons head allowed them to think about moving objects in a specific direction, sent signals to their wheelchairs or in another case implanted motors in one crippled persons hand to commit the actions.
Anyways the USAF was testing it previously since they were stating that pilots had too many buttons, controls, gauges to monitor, so they were looking for alternative methods of having them manuever the aircraft when they were flying. If I'm not mistaken it (the technology) was highly used in NASA as well.
So aside from all this semi informative stuff. I would rather have NASA spend time fscking around to get things done properly then to keep having them waste money crashing, and losing aircraft.
2600 being run by Peter Pan
Want Root?
Here's the information to what I was speaking about for those interested in the above thread I posted.
http://www.brainfingers.com/technical.htm extreme thanks to Michael (don't know if he wants his last name posted) for pointing me in the right direction with the link.
Want Root?
To make the obligitory Douglas Adams quote for this thread, you need to sit annoyingly still to keep listening to the same station. If you controlled an airplane like this, you'd soon have the first case of an airliner crash caused by a flea.
Quattuor res in hoc mundo sanctae sunt: libri, liberi, libertas et liberalitas.
The interesting part of all this stuff is the material the controller is made of.
:>
- --
Some cut-off spandex from a jumpsuit, and coat button snaps sewn on backwards for the electrodes to connect to.
NASA Engineers with duct tape and a toolshed can do pretty much anything.
BTW, For info on the Neural Engineering project at NASA Ames who is working on this project, see The Neural Engineering group
-----------------------------------------------
Basic frustration: touch feedback is much faster than visual feedback. Visual is limited to maybe 2Hz; tactile is maybe 10x that. This is one reason gloves-and-goggles virtual reality systems are so painful to use.
True force feedback is very effective. Most of the stuff on the game market is treated as a subwoofer, not a real force feedback device, so you can't judge by that.
One of the neatest force feedback devices I've come across was from the now-defunct Kraft Telerobotics. They had a backhoe you could control with a force-feedback handgrip. The feedback was good enough that you could feel what you were digging, and could reach down into a trench full of muddy water and dig around pipes by touch. Unfortunately, they got into NASA contracts instead of heavy construction and were never heard from again.
And have they thought about how they're going to seal the controls to keep the vaseline from leaking into everything?
Seriously, I can just see the humans of the future as big blobs just laying in their Poetic Tech Chair twitching and jerking their way through life with attendant Aibos and Palm-creatures zipping around them. Actual contact won't be necessary as we can stop and smell the roses right on the information highway.
"Beware of he who would deny you access to information, for in his heart, he dreams himself your master."
I saw a segment on this on some news show -- it slips my mind what it was. Anyway, watching the guy demonstrate the technology, I have to say, it looked pretty uncomfortable. He moved his hand in ways that would surely cause carpal tunnel after prolonged use. It looked like he was bending his wrist to a degree that I don't think would be necessary while using a joystick.
Also, when it comes to using a joystick, people can readjust their hands on the controller if they become uncomfortable. As far as I know, you'd always have to move your hand in the same way to get this technology to work.
-- dR.fuZZo
Um, the problem is that you don't know before hand which problems will be solved by which research...
Sure we can specifically target things like AI for ocean probes, but what's from stopping someone just like you saying 'That's sooo expensive, why not just do AI for automatic vacuum cleaners instead?'
The difference being that for Hoover or Eureka, good enough is what sells, and AI research will stop when the vacuums figure out how to navigate, plug themselves into an outlet, and not get destroyed by running into the street. AI research for Mars, or even the ocean, will hopefully design for robustness, learning capability, flexibility, and reliability.
How do we know that traveling to Mars doesn't unlock some sort of cancer cure as a side effect of making humans more fit in space? How do we know we won't cure baldness or the common cold by designing anti-radiation treatments? How do we know that we won't find the secret to room temperature superconducting electronics in the design of a lightweight yet sturdy heat transfer support structure (since there is no convection or radiative heat loss, you have to use other means)?
There is applied research, where you know the problem and try to find an answer, and there is pure research, where you don't know the problem, and thus finding an answer is as much about finding the problem as anything else. Going into space offers a lot of pure research problems; materials sciences, medicine, genetics, biology, physics, mechanical engineering, electronics, etc. How do we know how they all fit? If we did, we would already know how to solve problems and questions; we don't, so the public has to be satisfied with the fact that such lofty, grand, and otherwise pointless expeditions do have many positive secondary effects.
Traveling to Mars may unlock the secrets of buckyball transistors, or new lubricants that use buckyballs, or whatever. We don't know, yet.
Geek dating!
GPL Deconstructed
political_news.c: warning: comparison is always true due to limited range of data type
As someone who works with a related technology (Brain Actuated Technology), I wanted to address some of the concerns I've seen a few people raise. Before I get too deeply into the discussion, I'll refer you to a previous reponse I made to the "Surfing The Net With Brainwaves?" article. If you are curious to see what I have said already about the subject, check it out.
I'm a software engineer that works with a device called a Cyberlink that allows you to control the mouse cursor (and other peripherals) using a combination of Brain/Body signals (EEG, EMG & EOG).
From the electrode arrangement in the NASA picture I saw, it looks like they using EMG (electromyograph) signals to detect discreet electrical impulses for specific muscles. There is a lot of electrical energy involved in actuating the muscles in your body, the hard part is figuring which muscle signals of the multitudes are the ones you care about.
One of the most difficult aspects of these types of technologies is resolving a "rest state". Energy is expended even when you are trying to hold a bodypart, like your arm, still. If the movement of a joystick/mouse/wheel was mapped directly electrical activity in your arm, wrists and hands you would have to worry about keeping them stationary to begin with so you wouldn't generate interference (and cause the device to move left when you wish it to stay where it is). Electrically, we are very noisy...
If you aren't concentrating, it may cause control to "drift". For example, if I move the mouse around with my hand and I let go, the mouse cursor usually will remain exactly where you left it on the screen. But with devices that use raw biosignals, it is hard to "turn-off" electrical signals present in your body so that the device interpreting them will stop reacting to them. When I use the brain actuated mouse interface with the Cyberlink, its kinda hard to keep the mouse from NOT moving. Sure, I can move it up/down/left/right when I want to, but when I don't want it to move, it is hard to prevent it from "drifting" slightly in some direction. These are aspects of the technology I am working on fixing. As such, I am intimately familiar with most of the difficulties involved.
In the case of picking up impulses meant for your hands and arm, it becomes even harder because we use our hands all the time for other tasks.
But the good news is that these things can be tuned. There are ways around these limitations and work is being done to resolve these issues. I suffer from tendonitis in my wrists, so *I* at least have a very vested interest in making this technology work. :) This technology is tremendously useful for people with physical disabilities and there is work being done to make the technology more appealing to able-bodied persons as well. Your concerns are duly noted. (by me, at least).
If you have additional questions or concerns for someone who is familiar with this technology, feel free to drop me an email and I will do what I can to answer your questions. :)
Mmm, brain-controlled railguns... ;P
PS: An interesting side-note is that it seems to take around 100 milliseconds (1/10 of a second) for a signal to be sent from your brain to get to your index finger and trigger movement *click*. By using a reflex tester (http://www.reflexgame.com/) the quickest I can seem to consistantly react to the screen changing color and click 'stop' seems to be 0.33 seconds. While using the Cyberlink (with electrodes on my forehead) I am able to consistantly react in only 0.22 seconds. So, my reaction time is about 1/10 of a second faster if I don't have to wait for the signal to travel all the way down to my finger. I'm already pretty dangerous in games like Unreal and Quake, but now I can fire that much faster. It is, however, a very odd feeling. You fire faster than you are expecting to. Its surprising. I keep thinking "Whoa, I fired already?" Eventually, we'll all be able to be LPBs of a different variety. ;P
Cheers,
Michael
About 3 years ago, I had the privelege of trying out a neurological interface system that was being developed by a private company in massechusetts. The system was being demonstrated through a two dimesional skiing game. The user put his finger in a sort of sensor tube, with electrodes along its length, and the skier changed direction according to their muscular twitches. The actual orientation of the controller finger didn't matter, but after a few minutes, you found that by thinking "left, left," really hard, your body would actually make the skier move. The tech was pretty cool, and, simialar to this NASA program, a way to interface the human mind more closely to a machine.
Unfortunately for this program, the ski program differs greatly from flying a combat, or even civilian aircraft. Since so many muscle movements are involuntary (what if the pilot sneezes?) The type of close interface provided by a neuromuscular interface is not ideal for an immediate reactive function. In the Apache gunship, a combat helicopter used extensively by the army, in-helmet sensors track the pilot's eye movements for hands free targeting. However, because the weapons don't automatically deploy, the gunner/pilot must consciously fire, and there is a buffer between the eye interface, dangerous consequences. A direct link flight system as examined in the article lacks a requirement for conscious authentication, and in my opinion, therefore makes the aircraft far too volatile, and prone to accidental adjustment. I know I sent the skier crashing into the trees more than once!
My Karma is so good, I'm the Dalai Lama...or something.
Me: "How many times do I have to tell you... put the beer down."