本文選題：Arduino　切入點：Enhanced　出處：《西安科技大學》2013年碩士論文　論文類型：學位論文

【摘要】：隨著工業(yè)自動化的高速發(fā)展，傳統(tǒng)的人工入庫已不能滿足倉儲自動化發(fā)展需求。本文的設計是以深圳市中科鷗鵬智能科技有限公司提供的基于Arduino Enhanced Board控制板的車載機械臂智能小車為實驗平臺，針對傳統(tǒng)的人工入庫存在的問題，設計了基于智能小車的模擬入庫系統(tǒng)，基于智能小車的模擬入庫系統(tǒng)實現(xiàn)的功能是小車巡線行走，到達取貨點后超聲波遠距離測量物體位置，由機械臂抓取物體，然后進行顏色識別并放置到與物體顏色對應的位置完成入庫，最后小車回到初始位置。 論文是基于智能小車的模擬入庫系統(tǒng)的軟件設計，同時配合硬件的電路，完成了整體系統(tǒng)的制作，并對軟、硬件進行了聯(lián)機調試，為深圳市中科鷗鵬智能科技有限公司開發(fā)了第一代可行走的機械臂小車。硬件模塊包括Arduino Enhanced Board控制板、巡線模塊、超聲波模塊、顏色識別模塊、機械臂模塊；軟件模塊包括巡線模塊、超聲波測量模塊、顏色識別模塊以及機械臂控制模塊。 硬件方面首先對各個模塊所用到的器件進行選型、分析與比較，針對不同的器件設計相應的驅動電路，然后對各個硬件模塊進行設計，并對各個硬件模塊的功能進行測試，最后將所有硬件模塊整合在一起進行調試。 軟件方面主要是針對各個硬件模塊編寫對應的代碼，其中重點是顏色識別模塊和機械臂控制模塊軟件設計。顏色識別模塊是針對傳統(tǒng)的閾值法識別精度較低、魯棒性較弱的問題以及基于神經網絡的最小色差法計算量大、難以應用的問題，提出了基于TCS230傳感器的模糊識別顏色算法。算法實現(xiàn)過程是首先根據少量的標準樣本確定出模糊關系，將TCS230傳感器實時測得的RGB數據模糊化，根據離線求出的模糊關系解出模糊輸出，通過解模糊確定出識別的物體顏色。模糊識別算法不僅識別精度較高，計算量較少，而且魯棒性較強，能夠應用到很多自動化系統(tǒng)。機械臂模塊是采用傳統(tǒng)的D-H法離線對機械臂進行建模，求出變換矩陣，通過對矩陣齊次變換得到逆解方程，求解方程后將結果寫入程序以控制機械臂。D-H法相對于遺傳算法等智能算法計算量較少，而且易于實現(xiàn)，適合應用在本文設計的智能小車系統(tǒng)中。 智能小車模擬入庫的100次實驗結果證明，設計的智能小車模擬入庫系統(tǒng)對物體顏色識別率高，搬運精確。同時系統(tǒng)的安全性較高，，而且響應也較快，為倉儲的入庫系統(tǒng)的發(fā)展提供了參考。
[Abstract]:With the rapid development of industrial automation, The traditional manual storage can no longer meet the requirement of warehouse automation development. The design of this paper is based on the Arduino Enhanced Board control board, which is provided by Shenzhen Zhongke Gupeng Intelligent Technology Co., Ltd. In view of the problems existing in the traditional manual storage system, an analog storage system based on intelligent trolley is designed. The function of the simulation storage system based on intelligent trolley is to inspect the track and measure the position of the object at a long distance by ultrasonic wave. The object is grabbed by the robot arm, then color recognition is carried out and placed in a position corresponding to the color of the object. Finally, the vehicle returns to the initial position. The paper is based on the intelligent car simulation system software design, and with the hardware circuit, completed the production of the whole system, and the software and hardware on-line debugging. For Shenzhen Zhongke Gupeng Intelligent Science and Technology Co., Ltd, the first generation walking robot is developed. The hardware module includes Arduino Enhanced Board control board, line patrol module, ultrasonic module, color identification module, robot arm module; The software module includes inspection module, ultrasonic measurement module, color identification module and manipulator control module. In the hardware aspect, the devices used in each module are selected, analyzed and compared, the corresponding driving circuit is designed for different devices, and then each hardware module is designed, and the function of each hardware module is tested. Finally, all the hardware modules are integrated together for debugging. In the software aspect, the corresponding codes are mainly written for each hardware module, in which the emphasis is on the software design of the color recognition module and the robot arm control module, and the color recognition module is aimed at the low accuracy of the traditional threshold method. The problem of weak robustness and the problem that the minimum chromatic difference method based on neural network is difficult to be applied, A fuzzy color recognition algorithm based on TCS230 sensor is proposed. Firstly, the fuzzy relation is determined according to a small number of standard samples, and the RGB data measured in real time by TCS230 sensor is fuzzied. The fuzzy output is solved according to the fuzzy relation obtained from the off-line, and the object color is determined by the fuzzy solution. The fuzzy recognition algorithm not only has higher recognition accuracy, less computation, but also has stronger robustness. It can be applied to many automation systems. The traditional D-H method is used to model the manipulator off-line, and the transformation matrix is obtained, and the inverse solution equation is obtained by homogeneous transformation of the matrix. After solving the equation, the results are written into the program to control the manipulator. D-H method has less computation than the intelligent algorithm such as genetic algorithm, and it is easy to realize. It is suitable for application in the intelligent car system designed in this paper. The results of 100 experiments show that the designed intelligent vehicle simulation storage system has high recognition rate of object color and accurate handling. At the same time, the security of the system is high and the response is fast. It provides a reference for the development of warehousing system.
【學位授予單位】：西安科技大學
【學位級別】：碩士
【學位授予年份】：2013
【分類號】：TH692.3

【參考文獻】

相關期刊論文 前10條

1 盧川英;于浩成;孫敬輝;孟中;;基于TCS230傳感器的顏色檢測系統(tǒng)[J];吉林大學學報(信息科學版);2008年06期

2 唐圣學,劉波峰,徐東峰;基于模糊神經網絡的顏色識別方法[J];傳感器技術;2003年11期

3 李致金;吳文娟;;基于MSP430微控制器的帶USB通信的顏色識別系統(tǒng)[J];電子技術應用;2009年08期

4 趙光藝;趙光興;欒曉霄;;基于顏色實現(xiàn)溫度測量的方法及實驗研究[J];儀器儀表用戶;2010年03期

5 張英華;李明海;趙傳峰;李麗;楊小弟;;基于顏色傳感器的水質硬度在線自動測試儀的研制[J];分析儀器;2010年05期

6 閆賀;吳懷宇;彭晟遠;鄭秀娟;;四自由度模塊化機械臂運動學建模[J];信息技術;2011年10期

7 黃大昌;史寶庫;劉冰;馬汝東;;基于顏色傳感器的酚酞堿度自動測量裝置[J];黑龍江電力;2012年01期

8 廖鋒;;機械臂逆運動學模型控制新方法[J];計算機工程與應用;2012年12期

9 李愛勤;張緒坤;徐剛;徐建國;顧震;;顏色檢測系統(tǒng)及其在果蔬干燥中的應用[J];江西科學;2009年01期

10 王安敏;楊起;;一種結合SVM的顏色識別系統(tǒng)研究[J];機械與電子;2012年04期

本文編號：1575976