發(fā)布時(shí)間：2018-03-05 02:36

  本文選題：環(huán)境感知　切入點(diǎn)：激光雷達(dá)數(shù)據(jù)融合　出處：《南京理工大學(xué)》2017年碩士論文　論文類(lèi)型：學(xué)位論文

【摘要】：無(wú)人車(chē)在軍事與民用方面具有廣泛的應(yīng)用前景。隨著物聯(lián)網(wǎng)、人工智能、計(jì)算機(jī)科學(xué)等相關(guān)技術(shù)的發(fā)展,無(wú)人車(chē)的外在環(huán)境也日臻完善。環(huán)境感知作為無(wú)人車(chē)系統(tǒng)的重要組成部分,對(duì)整個(gè)車(chē)起著至關(guān)重要的作用。本文針對(duì)無(wú)人車(chē)環(huán)境感知中的兩個(gè)重點(diǎn)與難點(diǎn)問(wèn)題進(jìn)行研究,設(shè)計(jì)了基于雙激光雷達(dá)的環(huán)境感知處理架構(gòu),并基于該架構(gòu)研究與實(shí)現(xiàn)了結(jié)構(gòu)化環(huán)境下的低矮道邊檢測(cè)與非結(jié)構(gòu)化環(huán)境下的負(fù)障礙檢測(cè)兩個(gè)環(huán)境感知的難點(diǎn)問(wèn)題。本文的主要研究成果與創(chuàng)新點(diǎn)如下:1、針對(duì)以往單個(gè)單線、多線激光雷達(dá)點(diǎn)云密度小,對(duì)于特殊場(chǎng)景感知能力差的特點(diǎn),研究設(shè)計(jì)了基于雙多線激光雷達(dá)對(duì)稱(chēng)式安裝的環(huán)境感知與信息融合處理架構(gòu),并對(duì)該架構(gòu)下的點(diǎn)云密度進(jìn)行定量分析。實(shí)驗(yàn)表明,該架構(gòu)大大提高了無(wú)人車(chē)車(chē)體前方觀測(cè)區(qū)的點(diǎn)云密度,減小了車(chē)體周身盲區(qū),可以解決無(wú)人車(chē)難度較大的環(huán)境感知問(wèn)題。2、根據(jù)雙激光雷達(dá)水平安裝的雷達(dá)掃描特點(diǎn),分析了雷達(dá)點(diǎn)在障礙區(qū)的分布特性,提出一種新的結(jié)構(gòu)化環(huán)境下的低矮道邊的檢測(cè)算法。算法使用了基于梯度一致性的點(diǎn)云分割方法,該方法可對(duì)雷達(dá)點(diǎn)進(jìn)行快速分割,高效的將雷達(dá)點(diǎn)分割為地面點(diǎn)與障礙物點(diǎn)。然后利用路面點(diǎn)與柵格地圖提取出候選道邊點(diǎn),最后分別使用最小二乘與改進(jìn)的RANSAC算法進(jìn)行道邊提取。實(shí)驗(yàn)結(jié)果顯示,點(diǎn)云分割算法具有良好的分割效果,改進(jìn)后的RANSAC算法具有較高的實(shí)時(shí)性,能夠滿足無(wú)人車(chē)的需求。3、針對(duì)非結(jié)構(gòu)化環(huán)境下的負(fù)障礙檢測(cè)問(wèn)題提出一種新的感知方法,該方法不依賴(lài)于地面平整度,通過(guò)局部點(diǎn)云分布特征進(jìn)行檢測(cè)。首先,將雷達(dá)點(diǎn)云映射到多尺度柵格,統(tǒng)計(jì)各柵格的點(diǎn)云密度與相對(duì)高度等特征并做標(biāo)記;然后,從點(diǎn)云數(shù)據(jù)中抽取負(fù)障礙幾何特征,將柵格的統(tǒng)計(jì)特征與負(fù)障礙的幾何特征進(jìn)行多特征關(guān)聯(lián)找到關(guān)鍵特征點(diǎn)對(duì);最后,將特征點(diǎn)對(duì)聚類(lèi),劃分負(fù)障礙。方法已成功運(yùn)行在無(wú)人車(chē)上,實(shí)驗(yàn)表明,該方法具有較高的實(shí)時(shí)性和可靠性,在非結(jié)構(gòu)化環(huán)境下具有良好的檢測(cè)效果。上述研究成果均已成功使用在"行健一號(hào)"無(wú)人車(chē)上,該車(chē)多次參加"中國(guó)智能車(chē)未來(lái)挑戰(zhàn)賽",并在比賽中取得優(yōu)異的成績(jī)。
[Abstract]:With the development of Internet of things, artificial intelligence, computer science and other related technologies, As an important part of the unmanned vehicle system, environmental awareness plays an important role in the whole vehicle. The environment sensing processing architecture based on dual lidar is designed. Based on this architecture, the paper studies and realizes two difficult problems of environmental perception, which are low lane edge detection in structured environment and negative obstacle detection in unstructured environment. The main research results and innovations in this paper are as follows: 1, aiming at single line in the past. Because of the low point cloud density of multi-line lidar and the poor perceptual ability of special scene, the environment sensing and information fusion processing architecture based on symmetrical installation of dual-line lidar is studied and designed. The experimental results show that the point cloud density of the observation area in front of the vehicle body is greatly increased, and the blind area around the body is reduced. It can solve the problem of environment perception, which is difficult for unmanned vehicle. According to the radar scanning characteristics installed horizontally with double lidar, the distribution characteristics of radar points in obstacle area are analyzed. In this paper, a new algorithm for detection of low edge in structured environment is proposed. The algorithm uses a point cloud segmentation method based on gradient consistency, which can segment radar points quickly. The radar points are divided into ground points and obstacle points efficiently. Then the candidate edge points are extracted from the road surface points and raster maps. Finally, the least square algorithm and the improved RANSAC algorithm are used to extract the edge points. The experimental results show that, The point cloud segmentation algorithm has a good segmentation effect, and the improved RANSAC algorithm has a high real-time performance, which can meet the requirements of unmanned vehicles. A new perception method is proposed for the negative obstacle detection problem in unstructured environment. This method does not depend on the smoothness of the ground, and detects the distribution characteristics of the local point cloud. Firstly, the radar point cloud is mapped to the multi-scale grid, and the point cloud density and the relative height of each grid are counted and marked. The geometric features of negative obstacle are extracted from point cloud data, and the statistical features of grid and geometric features of negative obstacle are correlated to find the key feature points. Finally, the feature points are clustered. The method has been successfully run on an unmanned vehicle. Experiments show that the method has high real-time and reliability. The above research results have been successfully used in the "Xingjian 1" unmanned vehicle, which has participated in the "China Smart vehicle Future Challenge" many times, and has achieved excellent results in the competition.
【學(xué)位授予單位】：南京理工大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類(lèi)號(hào)】：U463.6;TN958.98

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 肖克來(lái)提;鄧志東;;基于激光雷達(dá)的多屬性柵格地圖的創(chuàng)建[J];信息與電腦(理論版);2016年08期

2 朱學(xué)葵;高美娟;李尚年;;一種智能汽車(chē)的實(shí)時(shí)道路邊緣檢測(cè)算法[J];北京聯(lián)合大學(xué)學(xué)報(bào)(自然科學(xué)版);2015年04期

3 趙振宇;周銳;池沛;;基于模糊推理Petri網(wǎng)的無(wú)人機(jī)智能決策方法[J];兵工自動(dòng)化;2015年03期

4 馮學(xué)強(qiáng);張良旭;劉志宗;;無(wú)人駕駛汽車(chē)的發(fā)展綜述[J];山東工業(yè)技術(shù);2015年05期

5 張賢啟;余有晟;劉俊才;;無(wú)人駕駛汽車(chē)的發(fā)展及可行性[J];山東工業(yè)技術(shù);2015年04期

6 馬飛躍;王曉年;;無(wú)人駕駛汽車(chē)環(huán)境感知與導(dǎo)航定位技術(shù)應(yīng)用綜述[J];汽車(chē)電器;2015年02期

7 段建民;鄭凱華;周俊靜;;多層激光雷達(dá)在無(wú)人駕駛車(chē)中的環(huán)境感知[J];北京工業(yè)大學(xué)學(xué)報(bào);2014年12期

8 郭欣;;自動(dòng)車(chē)輛識(shí)別與裝車(chē)系統(tǒng)研究[J];工礦自動(dòng)化;2014年08期

9 任佳;杜文才;白勇;;基于貝葉斯網(wǎng)絡(luò)自適應(yīng)推理的無(wú)人機(jī)任務(wù)決策[J];系統(tǒng)工程理論與實(shí)踐;2013年10期

10 戴斌;聶一鳴;孫振平;安向京;;地面無(wú)人駕駛技術(shù)現(xiàn)狀及應(yīng)用[J];汽車(chē)與安全;2012年03期

，

本文編號(hào)：1568412