GPS/INS Integration for Land Vehicle Navigation Using Source Difference Method

D. Bhatt, P. Aggarwal, P. Bhattacharya, V. Devabhaktuni

Abstract:	Land vehicles navigation systems mainly rely on the positioning information coming from Global Positioning System (GPS). However, GPS receivers require direct line-of-sight to four or more satellites, which is not always feasible. In order to provide continuous navigation solution, GPS information is usually combined with Inertial Navigation System (INS) data. Generally Bayesian filtering techniques like Extended Kalman Filter (EKF) or Unscented Kalman Filter (UKF) are employed for integrating GPS and INS units. However, major problem lies in defining the stochastic error models along with accurate a priori information for these filters. Alternatively to overcome these limitations of these filters, Artificial Neural Network (ANN) have been widely implemented for low-cost INS units. In case of Micro Electro Mechanical System (MEMS) INS units, the input/output relationship to be learnt by ANNs is highly complex. Hence, the ANN model has residual or training error associated with it, which results in degraded navigation performance for these MEMS sensors. This study exploits the idea of incorporating efficient knowledge based neural network technique, i.e., Source Difference Method (SDM), as an intelligent scheme for INS/GPS integration utilizing Position and Velocity Update Architecture (PVUA). The performance of the proposed method is examined by evaluating the solution accuracy under GPS outage conditions. The obtained results validate the efficacy of the SDM approach over the regular ANN model for several GPS outages.
Published in:	Proceedings of the 24th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2011) September 20 - 23, 2011 Oregon Convention Center, Portland, Oregon Portland, OR
Pages:	1811 - 1817
Cite this article:	Bhatt, D., Aggarwal, P., Bhattacharya, P., Devabhaktuni, V., "GPS/INS Integration for Land Vehicle Navigation Using Source Difference Method," Proceedings of the 24th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2011), Portland, OR, September 2011, pp. 1811-1817.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In