MWELLS@falcon.aamrl.wpafb.af.mil (Maxwell Wells) (11/29/90)
Notes on using accelerometers to measure head position. There are two types of accelerometer which may be useful for measuring head movement: electro-mechanical and electro-resistive. Electro-mechanical accelerometers work by supporting a mass on top of a piece of piezo electric material. Accelerations produce changes in force on the pe material and hence changes in the piezo electric voltage. Electro-resistive accelerometers work by suspending a mass on the end of a beam, upon which is adhered a strain gauge. Acceleration causes bending of the beam and changes in the resistance of the strain gauge. For translational acceleration (x,y,z axes) either type of accelerometer will work. Rotational acceleration can be transduced by using two translational accelerometers mounted so that their sensitive axes are orthogonal to the axis about which you wish to measure (eg accelerometers with their sensitive axes in the z axis (up and down) can be used to measure rotational acceleration about the y axis (pitch)). Initial consideration may lead you to think that a minimum of 6 translational accelerometers could be used to transduce 6 axes of motion (x,y,z and roll, pitch, yaw). But you would be wrong. For reasons too complicated to go into here (ok, I admit I don't understand it), you actually need 9 accelerometers. Piezo-resistive accelerometers are cheaper, smaller and more tolerant of abuse than piezo-electric accelerometers. However, piezo-electric accelerometers can measure down to d.c., which is useful for calibration against the 1g force of earth's gravity. Rotational accelerometers exist, but they are fragile, expensive and hard to calibrate. Having transduced acceleration, double integration will provide displacement, but, as has been noted on the net, with drift. Correction of drift could be done by some of the methods suggested by other contributors (photographic, optical etc) or by using a polhemus system. This latter course is something we have been considering as a method of increasing the update rate of head position ie double integrated head acceleration for speedy transduction of displacement, and constant polhemus updates for accuracy. I am not aware of any solid state (optical) gyroscopes that are small and light enough to be mounted on the head. There are some of the spinning wheel variety, which are fairly small (but much larger than accelerometers), but the wearer would have to put up with the inconvenience of 3 spinning, humming masses on his/her head. Also, gyroscopes do not completely eliminate the drift problem. They transduce velocity, which must be integrated to displacement. There are a number of papers published which describe the use of accelerometers for measuring head motion. I worked in the area a few years ago on a project for reducing the effects of vibration on performance with helmet-mounted displays. There are people who are currently using them for measuring head motion various vehicles. For more information, contact me, or write to: Human Factors Research Unit, Institute of Sound and Vibration Research, The University, Southampton, England SO9 5NH, Tel 011 44 703 595000 ext 2277 Max Wells