ICNPAA 2010: Mathematical Problems in
Engineering, Aerospace and Sciences

This paper addresses the use of artificial neural network (ANN) into Global Positioning System (GPS) aided low cost inertial navigation systems based on micro-electromechanical sensors (MEMS). The GPS technology dominates, nowadays, the positioning and navigation (POS/NAV) general market, and alternative POS/NAV systems are only needed because GPS does not work in all environments, or can not provide reliable solutions, under certain circumstances, during some time interval. There are different solutions to fulfill information during GPS blockage and integrated inertial sensors systems with GPS are frequently used with stochastic parameter estimation techniques. However, low cost inertial sensors have the disadvantage of accumulating continuous errors in great extension, leading to poor system performance. In this context a multi-layer feed-forward ANN is applied, to provide better NAV/POS solutions, during the lack of information in GPS outages portion of time. This work presents: an approach to model the input-output ANN signals based on a set of constrained land vehicle navigation equations; an adaptive ANN training Kalman filtering methodology and preliminary numerical simulation results based on urban vehicular positioning application data, acquired from low cost Crossbow CD400-200 inertial measurement unit and an Astech Z12 GPS receiver.