移動機器人之階層模糊邏輯控制

Abstract

ㄧ般而言機器人在未知的環境中移動，通常需要使用大量感測器來感測周遭環境變化，以達成自主導航和完成工作任務之目的。若使用模糊邏輯系統來設計控制器，由於大量感測器訊號當作輸入，導致模糊規則數會變得非常龐大。因此，本論文使用階層模糊邏輯控制（Hierarchical Fuzzy Logic Controllers，HFLC）方法設計一移動機器人控制器，完成移動機器人在未知環境中可以自主移動且夾取目標物，階層模糊邏輯控制器將行為分成六種子動作行為模式以便減少模糊規則數。本論文之機器人之硬體架構包括：全方向輪控制、DC馬達控制、雷射測距儀量測、紅外線距離感測器、視覺系統、RF通訊模組與五軸機器手臂之控制。最後，本文透過沿牆面自動導航、抓取目標物、靜態避障、動態避障等實驗來驗證機器人執行各種動作行為和測試移動機器人整體的效能。

For automatically navigating in an uncertain environment, a mobile robot generally requires the huge number of sensors. When the fuzzy logic system is used to design the control of the mobile robot system, the design of the fuzzy rules becomes very complicated because of the large number of sensors. Therefore, this thesis uses a method of Hierarchical Fuzzy Logic Controllers (HFLC) for the mobile robot so that the design of the fuzzy rules becomes simple. To achieve auto navigation and mission, the method divides the behavior of the mobile robot into six mode of the behavior. To verify the effectiveness and applicability of the proposed method, we design and implement a mobile robot, and the main structure of the mobile robot includes the DC motor control of three omni-directional wheels, laser sensor, radio transceiver module, distance measuring sensor, a vision system and the control of two five-axis robotic arms. Also, we perform some experiments, including auto navigation, target object searching, static obstacle avoidance, dynamic obstacle avoidance and so forth.