本研究之主旨為發展一具有戶外自動導航控制與障礙物偵測系統之自主無人載具,此系統包含主控站、載具和GPS參考站。主控站負責載具之控制與路徑規劃,並將GPS參考站衛星接收機接收到之資料透過無線網路傳送給移動站進行GPS差分定位,移動站則包含了第二組衛星接收機、馬達、雙眼相機、電子羅盤,負責將載具偵測到之環境數據回傳給主控站。 為了避免載具在路徑追蹤模式時偏離規劃路徑,我們採用模糊PID控制器控制載具之線速度與角速度,其中線速度項之回授由載具與參考點之距離誤差求得,角速度項之回授則由載具與參考點之視線角誤差求得,透過模糊理論解出PID控制之K_p、K_i和K_d,調整載具之線速度與角速度,配合電子羅盤提供載具之姿態,精確的掌握系統之動態性能,使無人載具順利地回到規劃之參考路徑上。 障礙物偵測系統採用雙眼相機,利用左右影像疊合產生平面視差之原理得到前方障礙物資訊。經由實驗結果認證,透過GPS差分定位,結合上述感測器偵測之障礙物資訊與路徑規劃,我們所設計之無人載具確實能回到參考路徑上,並順利抵達終點。
The purpose of this research is to develop an unmanned vehicle system with outdoor automatic navigation control and obstacle detection. The whole system consists of three parts : main station, moving vehicle, and GPS reference station. The moving vehicle includes a GPS receiver for vehicle positioning, a servo motor system for motion control, an electronic compass for vehicle’s attitude judgment, a stereo camera for obstacle detection and avoiding. The GPS reference station includes another GPS receiver for positioning data. The main station is to monitor and control the whole system. To prevent the vehicle deviating from the reference path, we use fuzzy PID controller to control the moving vehicle by setting the vehicle’s velocity and angular velocity. The feedback of velocity term is obtained from distance between the position of the vehicle and the reference point, and that of angular velocity term is obtained from the line-of-sight angle from the current position to the reference point. By using fuzzy theory, we can adjust the parameters of PID immediately. The vehicle can then be driven back to the reference path in the path-tracking mode. For obstacle avoiding, we use stereo camera to detect the obstacles in sight. Experimental results show that combining fuzzy PID controller, obstacle detecting system and GPS positioning, an unmanned vehicle can indeed follow the path to the destination without colliding the obstacles.