System of Tracking and Positioning Automatically on Targets

• Main Authors: Jun-Ting Hsiao, 蕭浚庭
• Other Authors: Shih-Wei Kau
• Format: Others
• Language: zh-TW
• Published: 101
• Online Access: http://ndltd.ncl.edu.tw/handle/6z9s83

碩士 === 國立高雄應用科技大學 === 電機工程系 === 100 === The radar automatic tracking system and aerospace automatic positioning and navigation in the military originated the servo control system for industrial manufacture. After several decades of evolution and development in the control algorithms, the technology of servo control has gradually had more popular and diversified industrial applications in different fields, such as the British anti-exploding automatic multi-joint robot, space laser guided automatic focus control, pilotless drone bombing system, and network monitoring systems with image tracking functions. Thus, we use servo remote trajectory tracking and control system as the instructional tool for simulating the radar automatic tracking system. This thesis primarily uses the laboratory servo remote positioning system to simulate the servo control of the radar automatic tracking system. There are three research methods. The first is to search for the optimal positioning system and to establish the mathematical model, in order to find out suitable control principles or the optimal controllers based on system pros and cons, hoping that it can conform to actual radar tracking and positioning systems. The second is to establish remote network monitoring system to learn Ethernet master-slave remote network controls and the network structure; due to object obstacles, wireless network controls are used in some cases. In traditional proportional–integral–derivative controllers, the parameters are difficult to adjust and thus are difficult to pass on, so Radial Basis Function Neural Network Algorithm is used in this study. Self-learning and adjustment, superior selection, and continuous adjustment of PID controller parameters are used to gain the optimal control parameters in hopes of fast response and precise positioning of trajectory tracking. Finally, the experiment process uses irregular motion trajectories to make the servo positioning system achieve synchronous remote controls, in order to simulate radar automatic tracking of targets. In addition, the errors from the experiment could also be reduced to a minimum by the Neural Network PID controller, in order to enhance the positioning accuracy of the motors. Keywords; remote network control system, PID controller, Radial Basis Function Neural Network