Peformance comparison of Autonomous neural network based GPS/INS integration

Abstract

In positioning and navigation applications, Inertial navigation system (INS) and Global positioning system (GPS) technologies have been widely utilized. Each system has its own unique characteristics and limitations. Therefore, the integration of the two systems offers a number of advantages and overcomes each system inadequacies. The proposed schemes are implemented using the Autonomous neural networks (AUNN) — the cascade correlation network (CCN) and the Feedback cascade correlation network (FBCCN) that was able to construct the topology by itself autonomously on the fly and achieve prediction performance with less hidden neurons.