Most folks here seem to think that sneezing or other involuntary actions would cause catastrophic failure or crashes. Not the case (at least not in the case of NASA's 757 simulator (some earlier slashdot post.)
The neuroelectrical signals generated by your arm muscles (motor units) are rather unique per movement. (These signals are called MUAPS - motor unit activation potentials) I.e. you flex your fingers in a fist --> some signal, flex your biceps --> some different signal. The reason is that the resultant signal is an amalgamation of the firing of several different muscle groups (motor units) at different times, and with differing geospatial orientation (from the reference point of the electrodes).
So when you're moving your arm around, the electrodes are picking up a steady stream of these MUAP signals, which may overlap. Neural nets and some quick per-user-basis learning tests can easily separate out the important MUAPS from the non important ones. And also, the involuntary muscle contractions generated by a sneeze (or whatever) would be easy for the control system to pick out and ignore.
Well now that they have webserver on a chip (http://kyllikki.fluff.org/hardware/wwwpic2/), how about allowing student groups to send up a few of these on the next mission to mars? How about slashdot.mars (ok, more than a few chips).
Slashdot Mirror
Most folks here seem to think that sneezing or other involuntary actions would cause catastrophic failure or crashes. Not the case (at least not in the case of NASA's 757 simulator (some earlier slashdot post.)
The neuroelectrical signals generated by your arm muscles (motor units) are rather unique per movement. (These signals are called MUAPS - motor unit activation potentials) I.e. you flex your fingers in a fist --> some signal, flex your biceps --> some different signal. The reason is that the resultant signal is an amalgamation of the firing of several different muscle groups (motor units) at different times, and with differing geospatial orientation (from the reference point of the electrodes).
So when you're moving your arm around, the electrodes are picking up a steady stream of these MUAP signals, which may overlap. Neural nets and some quick per-user-basis learning tests can easily separate out the important MUAPS from the non important ones. And also, the involuntary muscle contractions generated by a sneeze (or whatever) would be easy for the control system to pick out and ignore.
Well now that they have webserver on a chip (http://kyllikki.fluff.org/hardware/wwwpic2/), how about allowing student groups to send up a few of these on the next mission to mars? How about slashdot.mars (ok, more than a few chips).