居家型移動機器人之軟硬體設計與控制

Abstract

本論文主要為設計與研製一居家移動機器人之軟、硬體架構，希望本移動機器人能在動態且複雜的居家環境中，成為能自我引導與全方位移動的自主性移動機器人。在複雜且動態的居家境中，可藉著全方位移動機器人敏捷的行動能力，完成居家照護的工作。本研究主要完成整合與設計下列各部份軟硬體架構，包括：全方向輪的運動方式、DC連續旋轉馬達的控制、場域可程式化邏輯閘陣列(FPGA)、影像訊號之距離量測系統與五軸機器手臂的控制。 另外，本論文利用下列兩個實驗來驗證所研製的居家移動機器人的效能與可行性。第一個實驗為利用模糊控制器控制居家照護機器人的移動，以便執行抓取目標物的工作，進而完成使用者所指定將某一目標物移至另一地點的任務。第二個實驗是先利用以DNA為基礎的適應性學習法則模擬居家移動機器人完成自動導航任務，之後將其演化後的參數使用於所研製的居家移動機器人上並實際測試居家移動機器人的自動導航能力。

The objective of this thesis is to design and control a home mobile robot which is capable of the omni-directional and autonomous mobility in dynamic and complex environments. In narrow home environments, the mobile robot can perform nursing work by the ability of omni-directional mobility. This study includes the DC motor control of three omni-directional wheels, Field Programmable Gate Array (FPGA), the image-based distance measuring system and the control of a five-axis robotic arm. In addition, two experiments are given to demonstrate the applicability and feasibility of the proposed home mobile robot. The first is that a fuzzy controller is used to control the mobility of the home mobile robot such that the robot can move an assigned object from one location to another. The second is that first, through simulation, the adaptive DNA-based learning approach is used in the home mobile robot navigation, and then using a simple experiment verifies the ability of the robot navigation.