Quantization and Colored Noises Error Modeling for Inertial Sensors for GPS/INS Integration

In this paper, modeling approaches for quantization and colored noises have been proposed. To accommodate the quantization noise, a modified inertial navigation system (INS) error dynamics is developed in this paper, and the quantization noise is incorporated into the modified INS error dynamics as augmenting driving noise. The three kinds of colored noises are modeled by using an equivalent differential equation driven by a unit white noise, and a technique is developed in this paper to augment the Kalman Filter of GPS/INS integration using this equivalent differential equation. Experimental test results show that the proposed stochastic error modeling approaches for quantization and colored noises significantly improves the accuracies of the estimated inertial drifts and the navigation solutions.