基於智能模糊控制器的機器人夾爪之實現

文化大學機構典藏 CCUR > 理工學院 > 工學院 > 機械工程系暨機械工程學系數位機電研究所 > 博碩士論文 > Item 987654321/30111

jsp.display-item.identifier=請使用永久網址來引用或連結此文件: https://irlib.pccu.edu.tw/handle/987654321/30111

题名:	基於智能模糊控制器的機器人夾爪之實現 Implementation of Robotic Gripper Based on Smart Fuzzy Controller
作者:	黃瑄傑 Huang, Syuan-Jie
贡献者:	機械工程學系數位機電碩士班
关键词:	夾爪視覺系統壓力模組伺服馬達智能模糊控制 Grasping gripper Vision system Pressure module Servo motor Smart fuzzy controller
日期:	2015-06
上传时间:	2015-08-05 10:06:21 (UTC+8)
摘要:	這幾年來，機器人從實驗室或是工廠等單純的環境漸漸地將邁向充滿變數無法掌握的複雜的環境，也就是我們的生活裡，為了克服環境的不確定性捨棄了複雜的機器人運動學及絕對座標的運算，本研究採取以感測器回授的環境參數交由設計好的模糊控制器產生相對的控制量給馬達使用，架構共分為三個步驟，分述如下：步驟一:以相機連接電腦進行影像擷取，利用LabVIEW的影像辨識功能對所擷取到的影像進行處理，取得待抓物在固定物距的情況下在影像x座標所占有的像素值，依據放大倍率得到待抓物的寬度。步驟二:先將馬達運轉至物體寬度，再透過模糊控制器在運行時壓力與馬達之間的變化，進而分析出待抓物為軟的物質或硬的物質(此時模糊控制器輸入為壓力感測器，輸出為馬達所位移∆X值)。步驟三:透過步驟二的結果，給予模糊控制器適當的壓力目標值並測試是否能供抓取，若不行則做修正，直到成功抓取為止。本研究針對在複雜環境下進行抓取提出了整個架構流程，與模糊理論及影像辨識的結合，且已順利的利用LabVIEW程式設計軟體完成整個系統的實現。 In recent years, the operating environments of robots have moved from simple ones toward complex environments in which the disturbances, load’s or system parameters’ variation are significant. To overcome the environmental uncertainties, an alternative grasping gripper is designed and implemented for robotic manipulator system in this study. To avoid the redundant computation of inverse kinematics, the relative coordinates are adopted in the proposed architecture. The proposed grasping action consists three steps as follows. Step 1: The camera is connected to a computer for image capture. The LabVIEW function of image recognition is used to process the captured image. When the grasped object is in the fixed position, LabVIEW gets the pixel values in the x coordinate of the image. Step 2: The angle of motor is turned to the width of the object first, then the fuzzy controller is executed to detect the soft and hard characteristic of the grasped object by analyzing the variation of the pressure and the PWM of motor. (The input linguistic variable of the antecedent part is the pressure value and the output linguistic variable of the consequent part is the ∆X value of the motor). Step 3: According to the detecting results of Step 2, the membership value of the fuzzy controller are tuned to the appropriate pressure target value. If it can’t grasp the object successfully, then goto Step 2 and try again. This study has proposed the whole grasping architecture for the robotic gripper, which combines the fuzzy theory and image recognition. The effectiveness is verified by some experimental results, and the proposed architectures are implemented in the home-made robotic grasping gripper in laboratory.
显示于类别:	[機械工程系暨機械工程學系數位機電研究所] 博碩士論文

文件中的档案:

档案	描述	大小	格式	浏览次数
index.html		0Kb	HTML	304	检视/开启

检视Licence

在CCUR中所有的数据项都受到原著作权保护.

数据加载中.....