DESIGN AND IMPLEMENTATION OF NAVIGATION SYSTEM USING EXTENDED KALMAN FILTER ON HYBRID UNDERWATER GLIDER

• Main Author: Nugroho Setiawan, Handi
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/51334
• Institution: Institut Teknologi Bandung

Indonesian seas have abundant resources. In order to utilize marine resources properly, technology can be used to study the forms of life in the ocean. One way to study organisms in the ocean is by doing bathymetry (underwater mapping). Bathymetry can be used by scientists to study organisms that live on the ocean floor. Bathymetry maps can also be used to study coral reef flocks to aid monitoring and conservation of coral reefs. Autonomous Underwater Vehicle (AUV) plays a huge role in underwater mapping missions. AUV is an autonomous vehicle that operates under water. Using sensors and actuators, AUV can move around and carry out bathymetry missions. To be able to carry out bathymetry missions properly, an accurate navigation system is needed to ensure the quality of the data from the underwater mapping mission. Navigation system estimates position, attitude, and body velocity of a vehicle using several sensors. The main goal of this final project is the designing and implementing navigation system on ITB-HUG. ITB-HUG is an subclass of AUV called Hybrid Underwater Glider (HUG) which uses two modes, gliding mode and propulsion mode. Navigation system, which will be developed on ITB-HUG, uses Inertial Navigation System (INS) as the main sensor with measurement sensors in the form of Doppler Velocity Log (DVL), Global Positioning System (GPS), and depth sensor. Data integration from the main sensor and the supporting sensors will be carried out with the closed-loop error-state Extended Kalman Filter (EKF). This design of navigation vi system is implemented using Robot Operating System (ROS) framework. The navigation system is verified using Hardware-in-the-loop Simulation (HILS), where the kinematics and dynamics of ITB-HUG are simulated on a computer and the navigation system is embedded on a Single board Computer (SBC). From the verification results, it is found that the navigation system can meet the requirements on maximum position error for bathymetric missions, which are 20 m + 10% depth for horizontal position and 1 m for vertical position.