A New Online-Identification Algorithm for Odometer's Scale Factor

Abstract

A identification in a nonlinear stochastic system - strapdown inertial navigation system is addressed. A new online-identification algorithm of odometer's scale factor is proposed. From zero velocity update (ZUPT), the velocity errors in inertial navigation system propagate in the Schuler loops, and based the slow-varying property of velocity errors, the algorithm identifies the odometer's scale factor using the changes of velocity during a short time. A tracking differentiator and a fadeout memory LMS algorithm is adopted in the identification to make the algorithm more practicable. The algorithm can track the odometer's scale factor varying with time, and therefore be suitable for navigation case in which a vehicle runs a long time and roads vary a lot. Simulation indicates that this new algorithm can identify the odometer's scale factor fast with small errors, so as to improve the accuracy of navigation.