A COMPARATIVE STUDY ON TWO METHODS OF DECOUPLING A SIX-AXIS ACCELEROMETER WITHOUT AND WITH A GYROSCOPE

Methods of decoupling a 6-axis accelerometer are attracting more and more attention because of their great significance for the accuracy, efficiency and stability of acceleration calculation. This paper introduces the basic decoupling methods without and with a gyroscope first. The decoupling accuracies of numerical accelerations are compared based on the definition of the comprehensive error, and the efficiencies are compared based on the computational time. In order to compare the stability of the two methods, the influence of step size is given, and the comprehensive errors are compared. Possible error sources that affect the stability of the decoupling methods are listed and divided into three categories, and the influence of every error source is analysed and presented. In the simulation, the gyroscope-based method is verified to achieve better accuracy, efficiency and stability than in the case of the method without a gyroscope. The experimental results agree well with the theoretical analysis.