Abstract
The projectile dynamics model is non-linear system. The specifications of projectile used are 12.7 x 99 mm caliber. In the projectile motion, to achieve the target can be estimated by using some estimation methods. In this study, the estimation of 12.7 x 99 mm caliber projectile motion is computed by using Unscented Kalman Filter (UKF) method. In this simulation, we estimate model of the system by measuring each variables position. The estimation results show the performance of UKF and KF in non-linear systems is different, based on their respective characteristics. The final result of this study shows that the error estimation of UKF is 81 % smaller in x-position, 85 % smaller in h-position and ν-position, 84 % smaller in γ-position. UKF performance is very good for non-linear system and the approach with KF algorithm has less optimal result.
Original language | English |
---|---|
Pages (from-to) | 184-193 |
Number of pages | 10 |
Journal | Procedia Computer Science |
Volume | 144 |
DOIs | |
Publication status | Published - 2018 |
Event | 3rd International Neural Network Society Conference on Big Data and Deep Learning, INNS BDDL 2018 - Sanur, Bali, Indonesia Duration: 17 Apr 2018 → 19 Apr 2018 |
Keywords
- Estimation
- Projectile Motion
- Unscented Klaman Filter (UKF)