【摘要】：隨著國(guó)防工業(yè)的不斷發(fā)展以及科學(xué)技術(shù)水平的不斷提高,大型航天產(chǎn)品的智能裝配技術(shù)得以發(fā)展,移動(dòng)視覺(jué)測(cè)量機(jī)器人在智能裝配的過(guò)程中起到了關(guān)鍵性的作用。視覺(jué)測(cè)量和路徑規(guī)劃是移動(dòng)視覺(jué)測(cè)量機(jī)器人任務(wù)規(guī)劃系統(tǒng)中的兩個(gè)重要的研究方向。對(duì)于待裝配的被測(cè)物,我們需要確定機(jī)器人的一個(gè)最優(yōu)觀測(cè)位置,使其在該位置進(jìn)行測(cè)量產(chǎn)生的誤差最小,讓機(jī)器人能夠更好地指導(dǎo)裝配過(guò)程。本文以單目視覺(jué)為測(cè)量方案,提出了一種機(jī)器人最優(yōu)觀測(cè)位置確定的方法。路徑規(guī)劃的目的一般是確定機(jī)器人在工作環(huán)境中移動(dòng)的最短路徑,本文對(duì)A*算法進(jìn)行改進(jìn),提出了基于改進(jìn)A*算法的路徑規(guī)劃方法�；跂鸥穹ǹ紤]環(huán)境中標(biāo)志點(diǎn)、被測(cè)物和觀測(cè)位置等因素提出了三種不同的地圖構(gòu)建方法。本文的主要研究工作如下:第一,對(duì)移動(dòng)視覺(jué)測(cè)量機(jī)器人觀測(cè)位置規(guī)劃問(wèn)題進(jìn)行研究。建立單目視覺(jué)測(cè)量模型。根據(jù)單目視覺(jué)成像原理確定測(cè)量方案,以此為基礎(chǔ)根據(jù)幾何關(guān)系和坐標(biāo)轉(zhuǎn)換分別針對(duì)二維和三維情況建立視覺(jué)測(cè)量模型。針對(duì)不同的工況建立最優(yōu)觀測(cè)位置測(cè)量模型,確定最優(yōu)觀測(cè)位置。第二,對(duì)移動(dòng)視覺(jué)測(cè)量機(jī)器人的路徑規(guī)劃方法進(jìn)行研究。闡述A*算法的實(shí)現(xiàn)過(guò)程,基于A*算法在柵格地圖中進(jìn)行路徑規(guī)劃方法的仿真。針對(duì)該方法會(huì)出現(xiàn)的碰撞問(wèn)題,改進(jìn)路徑規(guī)劃方法,使其能夠避免碰撞問(wèn)題。第三,對(duì)移動(dòng)視覺(jué)測(cè)量機(jī)器人的地圖構(gòu)建方法進(jìn)行研究。與傳統(tǒng)地圖構(gòu)建不同,本文的地圖構(gòu)建引入標(biāo)志點(diǎn)和被測(cè)物等因素,并以此為基礎(chǔ),分別針對(duì)單獨(dú)考慮標(biāo)志點(diǎn)因素,綜合考慮標(biāo)志點(diǎn)和被測(cè)物因素,綜合考慮標(biāo)志點(diǎn)、被測(cè)物和觀測(cè)位置因素三種不同的地圖構(gòu)建方法,利用基于改進(jìn)A*算法的路徑規(guī)劃方法進(jìn)行仿真,并分析對(duì)比不同方法的優(yōu)缺點(diǎn)和應(yīng)用場(chǎng)合。第四,通過(guò)實(shí)驗(yàn)驗(yàn)證最優(yōu)觀察位置測(cè)量模型和路徑規(guī)劃方法,并且同時(shí)驗(yàn)證移動(dòng)測(cè)量的優(yōu)勢(shì)。針對(duì)不同的實(shí)驗(yàn)搭建相應(yīng)的實(shí)驗(yàn)環(huán)境,通過(guò)機(jī)器人自身CCD相機(jī)和天頂全局相機(jī)對(duì)被測(cè)物進(jìn)行測(cè)量,分析測(cè)量誤差,根據(jù)得出的結(jié)論去驗(yàn)證本文提出的理論。
[Abstract]:With the development of national defense industry and the improvement of science and technology, intelligent assembly technology of large-scale aerospace products has been developed. Mobile vision measuring robot plays a key role in the intelligent assembly process. Vision measurement and path planning are two important research directions in task planning system of mobile vision measuring robot. For the object to be assembled, we need to determine an optimal observation position of the robot, so that the measurement error of the robot can be minimized, so that the robot can better guide the assembly process. In this paper, a method for determining the optimal observation position of a robot is proposed, which is based on monocular vision. The purpose of path planning is to determine the shortest path of robot moving in the working environment. In this paper, we improve the A * algorithm and propose a path planning method based on the improved A * algorithm. Based on the grid method, three different map construction methods are proposed based on the factors such as the mark points, the measured objects and the observation position in the environment. The main work of this paper is as follows: first, the position planning of mobile vision measuring robot is studied. A monocular visual measurement model was established. According to the principle of monocular vision imaging, the measurement scheme is determined, and based on the geometric relationship and coordinate transformation, the visual measurement models are established for two-dimensional and three-dimensional situations, respectively. An optimal observation position measurement model is established for different working conditions to determine the optimal observation position. Secondly, the path planning method of mobile vision measuring robot is studied. The realization process of A * algorithm is described, and the simulation of path planning method based on A * algorithm in grid map is carried out. To solve the collision problem, the path planning method is improved to avoid the collision problem. Thirdly, the map construction method of mobile vision measuring robot is studied. Different from the traditional map construction, the map construction in this paper introduces the factors such as mark point and object under test, and on this basis, separately considering the factor of mark point, synthetically considering the factor of mark point and object being tested, synthetically considering the mark point, considering separately the factor of mark point, synthetically considering the factor of mark point. There are three different map construction methods: the measured object and the observed position factor. The path planning method based on the improved A * algorithm is used to simulate, and the advantages and disadvantages of the different methods are analyzed and compared. Fourth, the optimal observation position measurement model and path planning method are verified by experiments, and the advantages of mobile measurement are also verified. According to the different experimental environment, the measured object is measured by the robot's own CCD camera and the zenith global camera, the measurement error is analyzed, and the theory proposed in this paper is verified according to the conclusion.
【學(xué)位授予單位】：哈爾濱工業(yè)大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類(lèi)號(hào)】：TP391.41;TP242

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 劉建華;楊建國(guó);劉華平;耿鵬;高蒙;;基于勢(shì)場(chǎng)蟻群算法的移動(dòng)機(jī)器人全局路徑規(guī)劃方法[J];農(nóng)業(yè)機(jī)械學(xué)報(bào);2015年09期

2 袁立;;現(xiàn)代飛機(jī)數(shù)字化柔性裝配生產(chǎn)線[J];航空科學(xué)技術(shù);2011年05期

3 趙娟平;高憲文;劉金剛;符秀輝;;移動(dòng)機(jī)器人路徑規(guī)劃的參數(shù)模糊自適應(yīng)窗口蟻群優(yōu)化算法[J];控制與決策;2011年07期

4 張辰貝西;黃志球;;自動(dòng)導(dǎo)航車(chē)(AGV)發(fā)展綜述[J];中國(guó)制造業(yè)信息化;2010年01期

5 齊勇;魏志強(qiáng);殷波;費(fèi)云瑞;于忠達(dá);莊曉東;;增強(qiáng)蟻群算法的機(jī)器人最優(yōu)路徑規(guī)劃[J];哈爾濱工業(yè)大學(xué)學(xué)報(bào);2009年03期

6 李德仁;;移動(dòng)測(cè)量技術(shù)及其應(yīng)用[J];地理空間信息;2006年04期

7 朱慶保;;復(fù)雜環(huán)境下的機(jī)器人路徑規(guī)劃螞蟻算法[J];自動(dòng)化學(xué)報(bào);2006年04期

8 郭英;盧秀山;馮尊德;;車(chē)載測(cè)量系統(tǒng)中兩傳感器相對(duì)位置關(guān)系的確定[J];工程勘察;2005年02期

9 楊淑瑩,董潔,曹作良;基于視覺(jué)導(dǎo)航系統(tǒng)的標(biāo)志識(shí)別方法[J];天津理工學(xué)院學(xué)報(bào);2002年03期

10 何大闊,王福利;基于可進(jìn)化性的快速遺傳算法[J];東北大學(xué)學(xué)報(bào);2002年07期

相關(guān)碩士學(xué)位論文 前2條

1 高超;導(dǎo)彈柔性裝配系統(tǒng)關(guān)鍵技術(shù)研究[D];哈爾濱工業(yè)大學(xué);2014年

2 吳強(qiáng);車(chē)載導(dǎo)航數(shù)據(jù)獲取與處理技術(shù)研究[D];西安電子科技大學(xué);2009年

，

本文編號(hào)：2404722