Akademik Veri Yönetim Sistemi
Tezin Türü: Yüksek Lisans
Tezin Yürütüldüğü Kurum: Orta Doğu Teknik Üniversitesi, Mühendislik Fakültesi, Makina Mühendisliği Bölümü, Türkiye
Tezin Onay Tarihi: 2017
Öğrenci: LİSAN OZAN YAMAN
Danışman: KIVANÇ AZGIN
Özet:The Global Satellite Navigation System (GNSS) is widely adopted for common positioning system due to its precision, cost and effectiveness. Despite its advantages, GNSS receivers are susceptible to signal degradation both intentional cases such as jamming/spoofing and unintentional cases like signal blockage in urban environment due to tall buildings. On the other hand, dead reckoning navigation system such as Inertial Navigation System (INS) is immune to external interferences and it can supply continuous navigation solution. However, the immunity comes with a price of unbounded positioning error growth with time due to mainly the Inertial Measurement Unit (IMU) sensor errors which continuously integrated into INS mechanization process. In order to bound inertial navigation system stand-alone navigation precision below some threshold, commonly GNSS or any other navigation aiding systems can be integrated with INS. Moreover, the IMU sensor errors are the crucial source of INS performance degradation factors and extensive laboratory tests are held by IMU manufacturers to calibrate inertial sensors for various types of application where the precision is critical. Even IMU is calibrated in laboratory environment, inertial sensors commonly have residual error terms left from calibration process. In mass production inertial sensor facilities such as MEMS products, manufacturers may not calibrate via laboratory processes due to necessary extensive labor and cost. That is, many low cost inertial sensors especially belong to automotive grade IMU, born uncalibrated and suffer from error terms. Therefore, robust algorithms and procedures for calibrating inertial measurement units especially low cost-low grade group of sensors in the field without need of precision laboratory equipments are promising. In this thesis, the development of integrated navigation algorithm that can be used for multi purpose including inertial sensor field calibration algorithm is carried out. First of all, the fundamental aspects of inertial navigation system, and its integration with GNSS receiver is exploited. The idea of calibrating the inertial sensor without use of extensive laboratory equipment is blended with Extended Kalman Filter (EKF) based INS/GNSS integration filter. Furthermore, for land vehicle navigation purpose Zero Velocity Update (ZUPT) and Non-Holonomic motion Constraints (NHCs) also integrated in the developed algorithm. Single and multi-run simulation studies are carried out together with static and dynamic field tests to show the performance of the integration filter. The dynamic calibration procedure deduced by the simulation study is applied to various MEMS inertial measurement units. The full verification of modular integrated algorithm is studied via land vehicle dynamic tests with sub 100 $ IMU and GPS receiver combination.