本文提出運用一混合式邏輯動態系統為模型架構，演繹設計實現能自主行進 於玉米田間之控制載具。本文以可程式控制器作為載具之控制核心，透過類比轉 數位模組接收感測器類比訊號，經數位編解碼後，由内建比較器作多段輸出，提 高對植株環境變異偵測之解析度。控制載具之實際功能測試是在玉米田間進行，
能夠模擬一般喷灑或運送作業機器在作物株間行進運作。本文提出之技術方法可 應用推演於其他型式自主行進載具其有限狀態機的演繹設計及實現。

In this paper, based on the modeling scheme of a mixed logic dynamic system, a direction-speed controller is designed and implemented by using a Moore finite-state machine to build up an automated guided vehicle (AGV) in corn field. Four photo-sensors, two ultrasonic sensors and an electro-gyro are outfitted in an AGV to sense the environment variation while the AGV goes through row spaces in a corn field.
In accordance with geometric positions and by setting threshold values from multi-sensors configured on a vehicle, a logic approach detecting and identifying the relative space variations from sensing states with leaves interference was constructed to deduce and simulate sequential state of logic values (0 and 1) to the controller. By integrating an analog to digital converter (ADC) and the built-in comparator in a programmable logic controller (PLC), analog signals from multi-sensors were digitized and coded into multi-bit input signals to improve the resolution to sensing plant state variations. The PLC was designed as a control core unit of an AGV to implement a binary dynamic processing and discrete event system. Furthermore, the function test to this controlled vehicle was implemented in the real corn field, and the general procedures like spray and harvest carried out by the same vehicle were also simulated among plants. The results indicate that the performance in the above methodologies work well and this example can also serve as a case study to show how to design a hybrid control system for the other similar applications.