本文關(guān)鍵詞： 拆除機(jī)器人 SLAM 多傳感器信息融合定位 機(jī)器人操作系ROS　出處：《安徽工業(yè)大學(xué)》2017年碩士論文　論文類型：學(xué)位論文

【摘要】：隨著城市的改造及工業(yè)的發(fā)展,拆除機(jī)器人的使用日益廣泛。目前拆除機(jī)器人不具備自主移動(dòng)能力,以人工遙控操作方式為主,一定程度上限制了作業(yè)效率與精度。拆除機(jī)器人的自主移動(dòng),需要借助傳感器信息同時(shí)進(jìn)行空間自定位與環(huán)境感知,這個(gè)過(guò)程被稱為即時(shí)定位與地圖構(gòu)建(SLAM)。該技術(shù)涉及傳感器信息處理以及多個(gè)數(shù)學(xué)模型,是拆除機(jī)器人自主移動(dòng)的前提。本文首先從SLAM的一般性問(wèn)題出發(fā),闡述SLAM中定位與建圖的關(guān)系;建立SLAM的概率學(xué)模型,以擴(kuò)展卡爾曼濾波和粒子濾波算法為基礎(chǔ),對(duì)SLAM算法的具體實(shí)現(xiàn)進(jìn)行探討;在MATLAB中對(duì)兩種SLAM算法進(jìn)行仿真對(duì)比實(shí)驗(yàn),以性能較好的算法作為拆除機(jī)器即時(shí)定位與地圖構(gòu)建的理論基礎(chǔ)。其次,針對(duì)拆除機(jī)器人SLAM過(guò)程中的定位問(wèn)題展開(kāi)分具體析:探討拆除機(jī)器人定位過(guò)程中存在的問(wèn)題與難點(diǎn),對(duì)拆除機(jī)器人SLAM過(guò)程中涉及的模型及兩種定位方式進(jìn)行了分析與建模,并在MATLAB中針對(duì)拆除機(jī)器人掃描匹配定位進(jìn)行了算法驗(yàn)證;在此基礎(chǔ)上,基于多傳感器信息融合技術(shù),融合里程計(jì)定位與掃描匹配定位信息,解決履帶滑移帶來(lái)的定位問(wèn)題;以上述理論為基礎(chǔ),通過(guò)融合定位算法對(duì)Fast SLAM算法進(jìn)行改進(jìn),作為拆除機(jī)器人即時(shí)定位與地圖構(gòu)建算法。最后,以上述改進(jìn)的SLAM算法為基礎(chǔ),基于開(kāi)源機(jī)器人操作系統(tǒng)ROS的開(kāi)源包進(jìn)行編碼改進(jìn),在機(jī)器人仿真器Gazebo中建立仿真環(huán)境,對(duì)比里程計(jì)模型定位與融合算法定位的建圖效果。在此基礎(chǔ)上,基于拆除機(jī)器人進(jìn)行硬件平臺(tái)和軟件平臺(tái)的搭建,在實(shí)驗(yàn)室對(duì)改進(jìn)后的算法進(jìn)行實(shí)驗(yàn),構(gòu)建了精度較高的實(shí)驗(yàn)室地圖,完成了對(duì)拆除機(jī)器人即時(shí)定位與地圖構(gòu)建算法研究。
[Abstract]:With the improvement of city and the development of industry, the demolition robot is used more and more widely. At present, the demolition robot does not have the ability to move independently, and it is mainly operated by manual remote control. To a certain extent, the efficiency and precision of the operation are limited. In order to move the robot independently, it is necessary to use the sensor information to simultaneously carry out the space self-localization and the environment perception. This process is called instant location and map building. The technology involves sensor information processing and multiple mathematical models, which is the premise of removing robot's autonomous movement. This paper starts with the general problem of SLAM. This paper expounds the relationship between location and map building in SLAM, establishes the probabilistic model of SLAM, discusses the realization of SLAM algorithm based on extended Kalman filter and particle filter algorithm, and makes a simulation and comparison experiment on two SLAM algorithms in MATLAB. The algorithm with good performance is used as the theoretical basis of the real time location and map construction of the demolition machine. Secondly, the localization problems in the SLAM process of the demolition robot are analyzed in detail: the problems and difficulties in the localization process of the demolition robot are discussed. This paper analyzes and models the models and two kinds of localization methods involved in the process of removing robot SLAM, and verifies the algorithm of scanning matching localization in MATLAB, and based on this, based on multi-sensor information fusion technology, The location problem caused by crawler slip is solved by combining the location information of mileometer and scanning, and based on the above theory, the Fast SLAM algorithm is improved by the fusion localization algorithm. Finally, based on the improved SLAM algorithm, the coding of the open source package based on the open source robot operating system ROS is improved, and the simulation environment is established in the robot simulator Gazebo. On the basis of this, the hardware platform and software platform are built based on the demolition robot, and the improved algorithm is experimented in the laboratory. The laboratory map with high precision is constructed, and the algorithms of real time localization and map construction of the dismantled robot are studied.
【學(xué)位授予單位】：安徽工業(yè)大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP242

【參考文獻(xiàn)】

相關(guān)期刊論文 前8條

1 芮強(qiáng);王紅巖;王欽龍;萬(wàn)麗;蓋江濤;周廣明;;履帶車輛轉(zhuǎn)向性能參數(shù)分析與試驗(yàn)研究[J];機(jī)械工程學(xué)報(bào);2015年12期

2 王法勝;魯明羽;趙清杰;袁澤劍;;粒子濾波算法[J];計(jì)算機(jī)學(xué)報(bào);2014年08期

3 朱磊;樊繼壯;趙杰;吳曉光;;未知環(huán)境下的移動(dòng)機(jī)器人SLAM方法[J];華中科技大學(xué)學(xué)報(bào)(自然科學(xué)版);2011年07期

4 李世飛;王平;沈振康;;迭代最近點(diǎn)算法研究進(jìn)展[J];信號(hào)處理;2009年10期

5 石杏喜;趙春霞;郭劍輝;;基于PF/CUKF/EKF的移動(dòng)機(jī)器人SLAM框架算法[J];電子學(xué)報(bào);2009年08期

6 賈銀亮;張煥春;經(jīng)亞枝;;Bresenham直線生成算法的改進(jìn)[J];中國(guó)圖象圖形學(xué)報(bào);2008年01期

7 鞠純純;何波;劉保龍;王永清;;基于粒子濾波器的SLAM的仿真研究[J];系統(tǒng)仿真學(xué)報(bào);2007年16期

8 羅榮華,洪炳昒;移動(dòng)機(jī)器人同時(shí)定位與地圖創(chuàng)建研究進(jìn)展[J];機(jī)器人;2004年02期

相關(guān)碩士學(xué)位論文 前2條

1 王帆;基于卡爾曼濾波和粒子濾波的移動(dòng)機(jī)器人同時(shí)定位與地圖創(chuàng)建研究[D];西安工程大學(xué);2012年

2 王培勛;基于子地圖連接的機(jī)器人同時(shí)定位與地圖構(gòu)建研究[D];中國(guó)海洋大學(xué);2010年

，

本文編號(hào)：1501758