SIMULATION OF THE PROPORTIONAL INTEGRAL DERIVATIVE OF IMAGE STABILIZER WITH 3-DOF

• Main Author: Fessenden, Jireh
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/43339
• Institution: Institut Teknologi Bandung

Utilization of UAV with the help of AI to detect threat cannot substitute the task of water police not because the AI not work well, but the extreme condition at sea shore that prevent camera to take footage that AI system can analyze. Therefore, 3 DOF mechanical image stabilization is needed to compensate the disturbance that comes from the extreme sea shore environment. Simulation is done to depict the performance of the 3 DOF mechanical image stabilization to minimalize the impact of the disturbance of wind in the form of sine wave. 3 DOF mechanical image stabilization is an inertially stabilized platform (ISP). ISP system have multi base motion system. To simplify modeling the motion system, chosen one base as reference base as common ground to observe the multi base motion system. Therefore homogeneous transformation matrices are needed to transform the parameter needed from one base to the reference base. The motion of the system depends on the orientation of the end-effector or camera therefore inverse kinematic system is used to govern the motion of the system. The movement dynamics of the system govern by the Lagrange equation. To simplify the computation, robotic manipulator; one of the form of Lagrange, equation is used to govern the movement dynamics of the system. The system is control by PID cascade control system to stabilize the dynamic of the system and to minimalize the disturbance. The disturbance have two component, there are amplitude (rad) and frequency (Hz) of the disturbance. The simulation shows that the amplitude of the disturbance is take more effect on the stabilization system that the frequency of the system. Optimally the system can stabilize the disturbance with maximum parameter at amplitude of 25o and frequency of 0.6 Hz that produce maximum error 8.6o. For the disturbance more than those parameters, it will unreliable and likely to blow away the drone. The addition of camera make the moment of inertia of the base where camera attach is not zero therefore the image stabilization is not only focus on minimalize the disturbance effect. That condition burdens the system.