移動機(jī)器人導(dǎo)航關(guān)鍵技術(shù)的研究與實(shí)現(xiàn)
發(fā)布時間：2018-11-24 11:16
【摘要】：移動機(jī)器人在幫助社會產(chǎn)生巨大效益的前提之一就是它能夠自主操作,不受到人為參與。其核心技術(shù)之一就是自主導(dǎo)航。自主導(dǎo)航是移動機(jī)器人在預(yù)先已知或者未知地圖環(huán)境的情況下,從一個地點(diǎn)自主導(dǎo)航到另外一個地點(diǎn)來完成人們給機(jī)器人預(yù)先安排的任務(wù)。機(jī)器人從一個地點(diǎn)行走到另外一個地點(diǎn)需要考慮的問題有三個,包括機(jī)器人所在地圖環(huán)境中的位置、機(jī)器人所到達(dá)的目的地、機(jī)器人如何到達(dá)指定地點(diǎn)。以上三個問題實(shí)際上就是導(dǎo)航系統(tǒng)三大內(nèi)容:地圖創(chuàng)建,機(jī)器人定位,路徑導(dǎo)航。本論文是以水利水電行業(yè)電力智能巡檢機(jī)器人項(xiàng)目的研發(fā)為基礎(chǔ),其重點(diǎn)研究的是機(jī)器人移動導(dǎo)航技術(shù)的實(shí)現(xiàn)。其中包括對地圖環(huán)境的建模,主要采用拓?fù)涞貓D抽象表示機(jī)器人導(dǎo)航所在的環(huán)境信息;采用ORB-SLAM和超寬帶結(jié)合定位方法對機(jī)器人定位;提出一種基于貪心策略處理包含中間節(jié)點(diǎn)集的最短路徑算法;提出了一種簡單但高效的局部路徑避障算法;以及處理機(jī)器人行走過程中所遇到的偏離航向、速度、方向等問題。本文所完成的工作以及貢獻(xiàn)詳細(xì)說明如下:(一)利用PHP+MYSQL+IIS開發(fā)一套web上位機(jī)系統(tǒng)。主要用于工作人員查看機(jī)器人狀態(tài)、機(jī)器人位置;編寫代碼實(shí)現(xiàn)了Dijkstra最短路徑算法,即給定目的地后,系統(tǒng)能夠生成一條全局最短導(dǎo)航路徑;編寫代碼實(shí)現(xiàn)了基于貪心策略來處理中間節(jié)點(diǎn)集的最短路徑算法;可以通過可視化界面控制機(jī)器人前后左右運(yùn)動;利用了多線程和非阻塞通信機(jī)制實(shí)現(xiàn)機(jī)器人與上位機(jī)系統(tǒng)通信、傳輸全局導(dǎo)航路徑數(shù)據(jù)、機(jī)器人所在位置數(shù)據(jù)、控制機(jī)器人前后左右運(yùn)動數(shù)據(jù)。(二)利用C/C++語言設(shè)計了一套移動機(jī)器人導(dǎo)航主體軟件系統(tǒng)。機(jī)器人主體采用了兩臺mini電腦,一臺控制機(jī)器人運(yùn)動,進(jìn)行局部路徑規(guī)劃的避障算法,控制機(jī)器人運(yùn)動速度和方向,以及運(yùn)行超寬帶定位算法。一臺處理ORB-SLAM算法和Kinect攝像頭數(shù)據(jù),并將所得到的數(shù)據(jù)通過以太網(wǎng)的形式傳遞給另一臺電腦。(三)本文結(jié)合了ORB-SLAM定位算法和超寬帶定位算法各自的優(yōu)點(diǎn),設(shè)計出在地圖環(huán)境下不同位置采用不同的算法定位方法。機(jī)器人在直線行走時采用ORB-SLAM定位算法,而機(jī)器人在轉(zhuǎn)彎時則采用超寬帶定位算法。這樣極大地提高了機(jī)器人的定位精度,并且所受到的外界干擾也有明顯改善;同時本文還針對機(jī)器人在全局路徑導(dǎo)航過程中,所出現(xiàn)的中間節(jié)點(diǎn)集情況下,提出一種基于貪心策略的最短路徑算法。并且通過反復(fù)實(shí)驗(yàn),證實(shí)了本算法有很強(qiáng)的實(shí)用性;最后本文在局部路徑導(dǎo)航中針對水電巡檢機(jī)器人特定的地圖環(huán)境提出了一種簡單但高效快速的避障算法,能夠讓機(jī)器人迅速地避開障礙物。
[Abstract]:One of the prerequisites for mobile robots to produce great benefits in helping society is that they can operate independently without human participation. One of its core technologies is autonomous navigation. Autonomous navigation is a mobile robot that navigates from one location to another to complete the tasks that people prearranged for the robot under the condition that the mobile robot knows or does not know the map environment in advance. There are three problems that the robot has to consider from one place to another, including the location of the robot in the map environment, the destination reached by the robot, and how the robot can reach the designated place. The above three problems are actually three main contents of navigation system: map creation, robot positioning and path navigation. This paper is based on the research and development of power intelligent inspection robot project in water conservancy and hydropower industry, which focuses on the realization of robot mobile navigation technology. It includes modeling the map environment, mainly using topological map to express the environment information of robot navigation, using ORB-SLAM and ultra-wideband localization method to locate the robot. Based on greedy strategy, a shortest path algorithm including intermediate node set is proposed, a simple but efficient local path obstacle avoidance algorithm is proposed, and some problems such as deviation from course, velocity and direction are dealt with in the course of walking. The work and contribution of this paper are described in detail as follows: (1) A web host computer system is developed by using PHP MYSQL IIS. It is mainly used to check the status and position of the robot, write code to realize the Dijkstra shortest path algorithm, that is, after a given destination, the system can generate a global shortest navigation path. The shortest path algorithm based on greedy strategy is developed to deal with the intermediate node set, and the robot movement can be controlled by visual interface. The multithreading and non-blocking communication mechanisms are used to realize the communication between the robot and the host computer, to transmit the global navigation path data, the position data of the robot, and to control the moving data before and after the robot. (2) A mobile robot navigation software system is designed by C / C language. Two mini computers are used in the main body of the robot, one is used to control the motion of the robot, the obstacle avoidance algorithm is used for local path planning, the speed and direction of the robot are controlled, and the ultra-wideband positioning algorithm is run. One handles ORB-SLAM algorithm and Kinect camera data and transfers the resulting data to another computer in the form of Ethernet. (3) combining the advantages of ORB-SLAM localization algorithm and UWB location algorithm, this paper designs different location algorithms in the map environment. The robot uses ORB-SLAM algorithm when walking in a straight line, while the robot uses ultra-wideband localization algorithm when turning. In this way, the positioning accuracy of the robot is greatly improved, and the external interference is obviously improved. At the same time, this paper proposes a shortest path algorithm based on greedy strategy for the intermediate node set in the global path navigation. And through repeated experiments, it is proved that this algorithm has strong practicability. Finally, this paper presents a simple but efficient obstacle avoidance algorithm for the specific map environment of hydropower patrol robot in local path navigation, which enables the robot to avoid obstacles quickly.
【學(xué)位授予單位】：電子科技大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TP242
本文編號：2353484
[Abstract]:One of the prerequisites for mobile robots to produce great benefits in helping society is that they can operate independently without human participation. One of its core technologies is autonomous navigation. Autonomous navigation is a mobile robot that navigates from one location to another to complete the tasks that people prearranged for the robot under the condition that the mobile robot knows or does not know the map environment in advance. There are three problems that the robot has to consider from one place to another, including the location of the robot in the map environment, the destination reached by the robot, and how the robot can reach the designated place. The above three problems are actually three main contents of navigation system: map creation, robot positioning and path navigation. This paper is based on the research and development of power intelligent inspection robot project in water conservancy and hydropower industry, which focuses on the realization of robot mobile navigation technology. It includes modeling the map environment, mainly using topological map to express the environment information of robot navigation, using ORB-SLAM and ultra-wideband localization method to locate the robot. Based on greedy strategy, a shortest path algorithm including intermediate node set is proposed, a simple but efficient local path obstacle avoidance algorithm is proposed, and some problems such as deviation from course, velocity and direction are dealt with in the course of walking. The work and contribution of this paper are described in detail as follows: (1) A web host computer system is developed by using PHP MYSQL IIS. It is mainly used to check the status and position of the robot, write code to realize the Dijkstra shortest path algorithm, that is, after a given destination, the system can generate a global shortest navigation path. The shortest path algorithm based on greedy strategy is developed to deal with the intermediate node set, and the robot movement can be controlled by visual interface. The multithreading and non-blocking communication mechanisms are used to realize the communication between the robot and the host computer, to transmit the global navigation path data, the position data of the robot, and to control the moving data before and after the robot. (2) A mobile robot navigation software system is designed by C / C language. Two mini computers are used in the main body of the robot, one is used to control the motion of the robot, the obstacle avoidance algorithm is used for local path planning, the speed and direction of the robot are controlled, and the ultra-wideband positioning algorithm is run. One handles ORB-SLAM algorithm and Kinect camera data and transfers the resulting data to another computer in the form of Ethernet. (3) combining the advantages of ORB-SLAM localization algorithm and UWB location algorithm, this paper designs different location algorithms in the map environment. The robot uses ORB-SLAM algorithm when walking in a straight line, while the robot uses ultra-wideband localization algorithm when turning. In this way, the positioning accuracy of the robot is greatly improved, and the external interference is obviously improved. At the same time, this paper proposes a shortest path algorithm based on greedy strategy for the intermediate node set in the global path navigation. And through repeated experiments, it is proved that this algorithm has strong practicability. Finally, this paper presents a simple but efficient obstacle avoidance algorithm for the specific map environment of hydropower patrol robot in local path navigation, which enables the robot to avoid obstacles quickly.
【學(xué)位授予單位】：電子科技大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TP242
【參考文獻(xiàn)】
相關(guān)期刊論文 前8條
1 王東署;段誼海;王佳;;未知環(huán)境中移動機(jī)器人的環(huán)境探索與地圖構(gòu)建[J];鄭州大學(xué)學(xué)報(理學(xué)版);2014年03期
2 趙世偉;馬少池;吳書明;林洋;;紅外線與超聲波組合定位技術(shù)研究[J];價值工程;2014年14期
3 ;中國工業(yè)機(jī)器人領(lǐng)跑智能制造裝備產(chǎn)業(yè)升級[J];可編程控制器與工廠自動化;2014年03期
4 肖竹;王勇超;田斌;于全;易克初;;超寬帶定位研究與應(yīng)用:回顧和展望[J];電子學(xué)報;2011年01期
5 朱大奇;顏明重;;移動機(jī)器人路徑規(guī)劃技術(shù)綜述[J];控制與決策;2010年07期
6 李季;孫秀霞;;基于改進(jìn)A-Star算法的無人機(jī)航跡規(guī)劃算法研究[J];兵工學(xué)報;2008年07期
7 張殿東;無線射頻識別(RFID)技術(shù)[J];電信技術(shù);2005年02期
8 孫維君;歐式空間中向量的叉積及其應(yīng)用[J];山東理工大學(xué)學(xué)報(自然科學(xué)版);2004年01期
相關(guān)碩士學(xué)位論文 前2條
1 楊飛;基于三維激光雷達(dá)的運(yùn)動目標(biāo)實(shí)時檢測與跟蹤[D];浙江大學(xué);2012年
2 于克;室內(nèi)自主移動機(jī)器人的定位研究[D];燕山大學(xué);2009年
，本文編號：2353484
本文鏈接：http://sikaile.net/kejilunwen/zidonghuakongzhilunwen/2353484.html
最近更新
教材專著
