Robustness Assessment of Intelligent Information Fusion Techniques in Navigation Systems Encountering Satellite Outage

Most integrated strapdown inertial navigation systems (SDINS) and Global Positioning Systems (GPS) have been implemented using the conventional method of Kalman filtering with its drawbacks related to the requirement for an accurate predefined model for the whole system and existence of at least four satellites in the vehicle´s sight of view. In this article three well-known intelligent approaches in navigation literature will be assessed on a critical index called satellite signal outage, and robustness to this circumstance and computational burden of the algorithms will be contrasted. All networks are trained during the availability of GPS signal to map the error between the GPS and the SDINS. Then they will be employed to predict the error of the SDINS position components during GPS signal blockage. These approaches include RBF and extended neural networks as proposed neural solutions and neuro-fuzzy network as a powerful and applicable intelligent tool in engineering applications.