Improvement of Filter Estimates Based on Data from Unmanned Underwater Vehicle with Machine Learning Insansiz Sualti Aracindan Alinan Verilere Baǧli Filtre Kestirimlerinin Makine Öǧrenmesi ile Iyileştirilmesi

2020-10-15
Erol, Berna
Cantekin, Recep Fatih
Kartal, Seda Karadeniz
Hacioglu, Rifat
Gormus, Kurtulus Sedar
Kutoglu, Senol Hakan
Leblebicioğlu, Mehmet Kemal
© 2020 IEEE.In this study, the mathematical model of unmanned underwater vehicle is obtained in 6 DOF. The navigation sensor data are generated from mathematical model response. Extended Kalman filer and Unscented Kalman filter is applied to estimate noisy sensor data. For the EKF, nonlinear model is linearized around the equilibrium points. For the UKF, nonlinear system model is used. The estimation performance of EKF and UKF are compared. Estimation has been improved by applying support vector machine algorithm, which is machine learning, for unscented Kalman filter estimates. All this study is modeled in MATLAB/Simulink and PYTHON environment.

Suggestions

Experimental test of the acoustic-based navigation and system identification of an unmanned underwater survey vehicle (SAGA)
KARTAL, SEDA; Leblebicioğlu, Mehmet Kemal; Ege, Emre (2018-05-01)
In this study, a nonlinear mathematical model for an unmanned underwater survey vehicle (SAGA) is obtained. The structure of the mathematical model of the vehicle comes from a Newton-Euler formulation. The three-dimensional motion is realized by a suitable combination of right, left and vertical thrusters. The navigation problem is solved by a combination of the inertial navigation system and acoustic-based measurements, which are integrated to obtain more accurate vehicle navigation data. In addition, a ma...
Vision-based Navigation and System Identification of Underwater Survey Vehicle
Kartal, Seda Karadeniz; Leblebicioğlu, Mehmet Kemal; Ege, Emre (2015-05-19)
In this study, a nonlinear mathematical model for an unmanned underwater survey vehicle is obtained. The inertial navigation system and vision-based measurement systems are modelled. The magnetic compass, depth sensor and pitot tube are used in order to support vehicle's attitude, velocity and depth information. The state errors are estimated with error state estimation algorithm from the noisy measurement data. The navigational data of the vehicle can be obtained accurately using the extended Kalman filter...
Experimental Study on the Effects of Winglets on the Performance of Two Interacting Horizontal Axis Model Wind Turbines
Ostovan, Y.; Uzol, Oğuz (2016-10-07)
The focus of this experimental study is to investigate the effects of winglets on the performance of two interacting similar horizontal axis model wind turbines. For this purpose, a downwind winglet is designed and manufactured to be attached to the blade tips of the upstream turbine. A set of wing extensions with the same length as the winglets is also produced to be compared to the winglets. Power and thrust coefficients of both turbines are measured with winglets as well as with wing extensions attached ...
Experimental test of vision-based navigation and system identification of an unmanned underwater survey vehicle (SAGA) for the yaw motion
Kartal, Seda Karadeniz; Leblebicioğlu, Mehmet Kemal; Ege, Emre (SAGE Publications, 2019-05-01)
In this study, a nonlinear mathematical model for an unmanned underwater survey vehicle (SAGA) is obtained. The structure of the mathematical model of the vehicle comes from a Newton-Euler formulation. The yaw motion is realized by a suitable combination of right and left thrusters. The navigation problem is solved by using the inertial navigation system and vision-based measurements together. These are integrated to more accurately obtain navigation data for the vehicle. In addition, the magnetic compass i...
Aerodynamic investigation of a model scale helicopter rotor in ground effect
Şahbaz, Mehmet; Kurtuluş, Dilek Funda; Sezer Uzol, Nilay; Department of Aerospace Engineering (2017)
In this thesis, ground effect issue which is a vital topic for helicopter industry is investigated with model scale.For this purpose, a test setup with model helicopter rotor is established. With setup, ground effect is investigated with inclined ground and ground without inclination. Thrust and torque values are obtained for different rotation speeds with constant collective pitch. Comparison with literature is done and reliability of the test setup is proven. Inclined Ground Effect and Ground Effect with ...
Citation Formats
B. Erol et al., “Improvement of Filter Estimates Based on Data from Unmanned Underwater Vehicle with Machine Learning Insansiz Sualti Aracindan Alinan Verilere Baǧli Filtre Kestirimlerinin Makine Öǧrenmesi ile Iyileştirilmesi,” 2020, Accessed: 00, 2021. [Online]. Available: https://hdl.handle.net/11511/70218.