An Inertial Navigation Data Fusion System employing an Artificial Neural Network as the Data Integrator

Michael Forrest, Tim Spracklen and Nick Ryan

Abstract:	This paper presents a novel application employing Artificial Neural Networks (ANN) for the fusion of data from the various sensors in an Inertial Navigation System (INS). The application uses a Self Organizing Map (SOM) to process the incoming data and provide an improved estimate of the system position. Currently, the most common method for the fusion of data sources is through the use of Kalman filters. Although Kalman filters represent one of the best solutions currently available, they lack certain necessary properties that are intrinsic to ANN's. Presented within this paper are simulations to establish the feasibility of such a system in concept. Initial experimental results are also presented.
Published in:	Proceedings of the 2000 National Technical Meeting of The Institute of Navigation January 26 - 28, 2000 Pacific Hotel Disneyland Anaheim, CA
Pages:	153 - 158
Cite this article:	Forrest, Michael, Spracklen, Tim, Ryan, Nick, "An Inertial Navigation Data Fusion System employing an Artificial Neural Network as the Data Integrator," Proceedings of the 2000 National Technical Meeting of The Institute of Navigation, Anaheim, CA, January 2000, pp. 153-158.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In