本文之主題為研究無人載具如何在平面上追蹤一移動之目標物,並透過硬體實驗,完成無人載具對目標物之追蹤控制。我們從最基本的機構設計著手,逐次增加馬達、感測器、控制器等裝置,希望所完成的無人載具能在避開環境中障礙物的條件下,即時追蹤移動的目標。 無人載具在追蹤目標物時的狀況可分為二種: 一種是在無障礙物的平面上,另一種則是有障礙物的存在。在無障礙物的平面上,我們用簡單的導引率實現;在有障礙物的平面上,結合路徑規劃演算法,可尋找出一條避開障礙物之最佳追蹤路徑。完成最佳追蹤路徑的設計後,我們即可利用全球定位系統與電子羅盤來獲得載具的位置與姿態,經由模糊控制器,達到無人載具對目標物追蹤控制之目的。由實驗的結果顯示,我們所提出的方法確實可達到載具對目標物的追蹤控制。
The main theme of this thesis is to design an autopilot system for an unmanned vehicle to track a moving target on a flat plane. A tracking controller is developed along with hardware experiment. First, we design the hardward of unmanned vehicle, and add gradually device like motor, sensor, controller and so on. The goal is to have the unmanned vehicle being able to track a moving target and simultaneously avoid obstacle in the environment in real time. There are two scenarios when an unmanned vehicle tracks a moving target:one is moving without obstacle, the other is with obstacle. On a flat plane without obstacle, we can use a simple guidance law to track a moving target. If obstacles exist, we need to use path-planning algorithm to design an optimal path from the vehicle to the target which bypass obstacles. After finishing the design of the optimal path, GPS receivers and electronic compass are adopted to obtain the position and the attitude of the vehicle, respectively. A fuzzy controller is then designed to accomplish the goal of tracking. From the experimental results, the proposed methodology indeed can effectively control an unmanned vehicle to track a moving target.