Attitude determination algorithm using state estimation including lever arms between center of gravity and IMU
| Autor: | Ki-Yeol Seo, Sul Gee Park, Tae Hyun Fang, Sanghyun Park |
|---|---|
| Rok vydání: | 2016 |
| Předmět: |
020301 aerospace & aeronautics
0209 industrial biotechnology Lever Engineering business.product_category business.industry 02 engineering and technology Kalman filter Accelerometer Computer Science Applications Center of gravity Acceleration 020901 industrial engineering & automation 0203 mechanical engineering Control and Systems Engineering Inertial measurement unit Control theory Torque Observability business Algorithm Simulation |
| Zdroj: | International Journal of Control, Automation and Systems. 14:1511-1519 |
| ISSN: | 2005-4092 1598-6446 |
| DOI: | 10.1007/s12555-015-0251-4 |
| Popis: | In this paper, an enhanced attitude determination algorithm is proposed to decrease the estimation error by including an additive state variable for the lever arm. Attitude determination generally is carried out by measurements from an IMU (inertial measurement unit), which is typically located at the center of gravity of the vehicle. The IMU lever arm, which spans the distance between the IMU and the center of gravity, causes extra acceleration in the accelerometer and increases the error in attitude estimates. However, if the extra accelerations caused by the lever arm can be removed from the measurements of accelerometers, the increased attitude error caused by the IMU lever arm can be prevented. Because an IMU lever arm is fixed in a vehicle after installation, it can be considered as an additive element of the state vector in Kalman filter for attitude determination. The proposed algorithm is composed of a quaternion-based Kalman filter and includes an estimation of the IMU lever arm. In addition, in order to determine components of lever arm, the gross measure of modal observability is investigated for the system. An evaluation of the proposed algorithm is carried out by simulations with a noise model based on an actual IMU. Evaluations through simulations show that the proposed algorithm improves the performance with regard to errors. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink