DESIGN AND SIMULATION OF NAVIGATION SYSTEM FOR HYBRID AUTONOMOUS UNDERWATER GLIDER

• Main Author: Ath Thahirah Al Azhima, Silmi
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/42120
• Institution: Institut Teknologi Bandung

The navigation system was developed to produce an estimate of position, speed and attitude of underwater vehicle. There are two sensors that used, IMU as the main sensor and DVL as the correction sensor. In the process of calculating the position there is an integral that can increase noise of estimation. The estimation position will have a large error. The method is needed to minimize the noise. The Extended Kalman filter method was chosen as a sensor fusion to overcome the problems. The kinematic model chosen for EKF properties, so the system can stand independently and doesn't depend on the dynamic model of underwater vehicle. Software in the loop simulation and hardware in the loop simulation are performed to analyze the estimation results will be compared with the output from the dynamic model of underwater vehicle. From both simulations was got a small error in each of the estimation result. The difference betwen speed and position error is 10?1, this is due to the integral process after the speed error estimation.