On the development of a conceptual intelligent navigator for a low cost DGPS/MEMS IMU integrated system

Integrated systems provide an enhanced navigation system that has superior performance in comparison with either a stand-alone Global Positioning System (GPS) or Inertial Navigation System (INS) as it can overcome each of their limitations. In addition to the quality of the sensors applied in such an integrated system, the core component of such a system is the multi-sensor data fusion or integration scheme. Currently, the Kalman filter approach has been widely recognized as the standard optimal estimation tool for current INS/GPS integration scheme, however, it does have limitations, which have been reported by several researchers. In addition, the most concerned impact caused by these limitations is the accumulation of positional error during GPS signal blockages. Therefore, this study exploits the idea of incorporating one of the most well known artificial intelligent approaches, Artificial Neural Networks (ANNs), as the core algorithm for developing an intelligent scheme for INS/GPS integration applications. The proposed scheme, a conceptual intelligent navigator, has the ability to generate, store, and accumulate navigation knowledge which is applied by the proposed scheme to fill in the gap during GPS signal outages. The proposed technique is expected to reduce the accumulated positional errors caused by conventional approaches (i.e. Kalman filter). An extended Kalman filter (EKF) that has 15 states and the proposed scheme have been tested using MEMS-based INS data, collected in land-vehicle field test, to evaluate the performance of both methods during several GPS signal outages.