Saab TERNAV, an Algorithm for Real Time Terrain Navigation and Results From Flight Trials Using a Laser Altimeter

F. Neregård, N.K. Johansson, N.P. Johansson, K. Lundberg, B. Gabrielsson

Abstract:	GNSS normally gives very accurate positions and is vital in many civil and military systems. However, even military GPS can be jammed quite easily and the interest for other more robust positioning systems has increased in recent years. Terrain navigation is one alternative to GNSS that is resistant to jamming and independent of external infrastructure. This paper presents a terrain navigation system developed at Saab Bofors Dynamics. The company has over 20 years experience in terrain navigation and has e.g. developed real time applications for the Swedish Air force including both the Viggen and Gripen aircrafts. The core algorithm has evolved over the years and is now based on Baysian statistical theory and the filters are implemented as discretised probability density functions in three dimensions. To obtain high integrity a number of parallel filters are used. The latest implementation has been done within a navigation demonstrator program where the algorithms have been adapted to a new radar altimeter and a laser altimeter. The system has proved to work well in a real-time environment and navigation performance for the recent flight trials is presented. The derivation of a non linear state space model for the terrain navigation problem will be described. This is followed by discussions of the modeling of system and measurement errors. The digital terrain elevation database contains ground elevation, but the altimeter will also detect trees in a forest and buildings in urban areas. This will result in measurement errors, and it is very important to model these errors in a reliable way. Most errors are taken care of in the measurement noise model, but it is inconvenient to model the largest errors as they appear quite seldom and under short periods. A large error can cause an erroneous position estimate. The idea behind parallel filters is to divide the measured altitude in time and feed different filters with different sequences.
Published in:	Proceedings of the 19th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2006) September 26 - 29, 2006 Fort Worth Convention Center Fort Worth, TX
Pages:	1136 - 1145
Cite this article:	Neregård, F., Johansson, N.K., Johansson, N.P., Lundberg, K., Gabrielsson, B., "Saab TERNAV, an Algorithm for Real Time Terrain Navigation and Results From Flight Trials Using a Laser Altimeter," Proceedings of the 19th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2006), Fort Worth, TX, September 2006, pp. 1136-1145.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In