Performance Enhancement of Low Cost Non-GPS Aided INS for Unmanned Applications

Abbas, M. M.; Kamel, A. M.; El-Halwagy, Y. Z.; Albordany, R. A.

doi:10.21608/asat.2013.22287

Performance Enhancement of Low Cost Non-GPS Aided INS for Unmanned Applications

Document Type : Original Article

Authors

Egyptian Armed Forces, Egypt.

10.21608/asat.2013.22287

Abstract

Low cost Autonomous Navigation Systems (ANS) used in Unmanned Arial Vehicles (UAVs) usually depend on Global Navigation Satellite Systems (GNSS) as a primary mean of navigation. For military applications, it is not accepted to use GNSS as a sole mean for navigation as it is susceptible to jamming and signal interference. For that reason, integrated navigation systems that compromise GNSS and other sensors such as Inertial Navigation System (INS) are used. INS as a standalone navigation system suffers from a huge number of errors which makes it impossible to use the low cost and grade ones for long term navigation unless aided by other means such as magnetometers. This paper describes the INS, its basic principles, and types. The test results of low cost Non-GPS aided INS for unmanned applications that use commercial off-the-shelf Micro-Electro- Mechanical System (MEMS) accelerometers, gyroscopes, and magnetometers are introduced. An inertial measurement unit (IMU), different types of gyroscopes, accelerometers, and magnetometer theory of operation are illustrated. Real time attitude and drift computations are performed using MEMS IMU and magnetometer. A test result comparison between IMU and magnetometer is executed for attitude measurements. Magnetometer calibration is achieved for accurate attitude measurements. A Kalman filter is also used to smooth the output data for
IMU and magnetometers. Test results showed that attitude measurements using integrated gyroscopes outputs suffer from high drift rate in roll, pitch, and yaw. Using accelerometers aiding through a complementary filter, roll and pitch drift are compensated. Using magnetometers aiding, yaw (heading) is also compensated. Test results show that the sensors
used in the combo MEMS IMU board under test suffer from high drift in yaw which reaches about 11 degrees after 5 minutes. Using a well calibrated magnetometer aiding, the error in yaw is nearly compensated.

Keywords