Kalman Filter Analysis of an Animal Head-Motion Estimation System

Head motion is a parameter that scientists at NASA Ames Research Center are interested in observing during flight experiments. The purpose of determining head motion is to understand the physiological effects of the flight environment upon Rhesus monkeys. Previously, the engineers at Ames Research Center used angular rate sensors to develop head motion velocity HMV systems. Although advantages exist for using angular rate sensors to determine head motion, several disadvantages have prompted the engineers at Ames Research Center to investigate new methodology for designing HMV systems. One method employed to avoid the problems associated with using angular rate sensors uses an accelerometer configuration. However, accelerometers are noisy and contain both deterministic and stochastic errors. Hence, this thesis explores using the Kalman filter as a covariance analysis tool to minimize the accelerometer errors and develop an animal head-motion estimation system. Furthermore, the results of several experiments show that an accurate depiction of head motion is obtainable.