Ataletsel Ölçüm Birimi Hatalarının Belirlenmesi ve Düzeltilmesi

Abstract

Inertial navigation system determines position, velocity and attitude by using navigation equations and inertial measurement unit outputs. Accumulation of sensor errors propagated by integration becomes very crucial in long-term navigation applications. Inertial measurement unit errors are divided into two main groups; deterministic errors and stochastic errors. The purpose of this thesis is to improve the performance of an IMU by developing error compensation algorithms. 100 position static tests and 120 rate dynamic tests are performed to determine the deterministic errors. These identified errors are used in deterministic error compensation model. Implementation of the proposed theory results in 99.2% improvement in accelerometer errors and 99.3% improvement in gyroscope errors. A new hybrid wavelet transform-Kalman filter method for the determination stochastic errors is suggested.