【摘要】：隨著人工智能技術(shù)的不斷發(fā)展與應(yīng)用,如今它已上升到國(guó)家戰(zhàn)略層面的高度。機(jī)器人作為人工智能技術(shù)的集成者,在實(shí)際用途中正在受到更多研究人員的關(guān)注。而移動(dòng)機(jī)器人的自主目標(biāo)識(shí)別與跟蹤應(yīng)用也是其智能化技術(shù)方面需要解決的一個(gè)核心問(wèn)題。如何將改進(jìn)的機(jī)器學(xué)習(xí)算法移植到移動(dòng)機(jī)器人上,并且使其處理跟蹤時(shí)光線變化、遮擋和復(fù)雜背景等問(wèn)題的方法更具有魯棒性,這是極富挑戰(zhàn)性的一項(xiàng)研究技術(shù)。該文在MT-R輪式移動(dòng)機(jī)器人平臺(tái)上,通過(guò)改進(jìn)的機(jī)器學(xué)習(xí)算法,實(shí)現(xiàn)了機(jī)器人的自主目標(biāo)跟蹤和對(duì)自身運(yùn)動(dòng)控制的功能,從而進(jìn)一步實(shí)現(xiàn)輪式移動(dòng)機(jī)器人的智能化應(yīng)用。本文的主要研究?jī)?nèi)容可以概括成:首先該文回顧了視覺(jué)目標(biāo)跟蹤的研究現(xiàn)狀,列舉出了不同視覺(jué)目標(biāo)跟蹤算法所采用的方法并分析了這些算法的缺點(diǎn)。著重說(shuō)明并分析了基于多實(shí)例學(xué)習(xí)的跟蹤方法和結(jié)合協(xié)同訓(xùn)練的算法。完成了對(duì)本文對(duì)采用方法理論基礎(chǔ)的鋪墊。同時(shí)也簡(jiǎn)要討論了移動(dòng)機(jī)器人在國(guó)內(nèi)外的研究發(fā)展?fàn)顩r。然后重點(diǎn)介紹移動(dòng)機(jī)器人內(nèi)部目標(biāo)跟蹤算法�；跈z測(cè)的目標(biāo)跟蹤算法通常依靠分類器來(lái)區(qū)分目標(biāo)和背景來(lái)達(dá)到跟蹤的目的,在分類器進(jìn)行學(xué)習(xí)的時(shí)候會(huì)對(duì)圖像分成樣本采樣和添加標(biāo)簽兩個(gè)單獨(dú)的步驟,但是這樣選擇的樣本是無(wú)目的性的,導(dǎo)致分類器的效果不穩(wěn)定。本文結(jié)合主動(dòng)學(xué)習(xí)的模型,提出一種新的樣本選擇的算法,基于多實(shí)例學(xué)習(xí)算法的框架,將主動(dòng)樣本選擇的策略加入到樣本采樣和標(biāo)簽分配之間,這樣可以選出有助于分類器學(xué)習(xí)的樣本,然后結(jié)合協(xié)同訓(xùn)練方法,防止由于誤差積累而導(dǎo)致的漂移,進(jìn)一步提高算法性能。通過(guò)在標(biāo)準(zhǔn)視頻序列上和其他六種算法進(jìn)行對(duì)比實(shí)驗(yàn),結(jié)果表明本文方法在目標(biāo)遮擋、光線變化等復(fù)雜條件下表現(xiàn)良好,具有一定的魯棒性。最后,在MT-R移動(dòng)機(jī)器人上運(yùn)行本文提出目標(biāo)跟蹤算法,結(jié)合機(jī)器人硬件驅(qū)動(dòng)策略,實(shí)現(xiàn)了移動(dòng)機(jī)器人的自主目標(biāo)跟蹤。并通過(guò)在不同的實(shí)際場(chǎng)景下進(jìn)行實(shí)驗(yàn)來(lái)驗(yàn)證移動(dòng)機(jī)器人目標(biāo)跟蹤的魯棒性。通過(guò)實(shí)驗(yàn)結(jié)果可以得到,本文提出的目標(biāo)跟蹤算法可以有效的幫助移動(dòng)機(jī)器人在目標(biāo)遮擋、光線變化等情況下進(jìn)行魯棒的跟蹤。
[Abstract]:With the continuous development and application of artificial intelligence technology, it has risen to the national strategic level. Robot, as the integration of artificial intelligence technology, is being paid more and more attention by more and more researchers in practical use. The autonomous target recognition and tracking application of mobile robot is also a core problem to be solved in its intelligent technology. How to migrate the improved machine learning algorithm to mobile robots and make its methods to deal with the problems of light change, occlusion and complex background are more robust, which is a very challenging research technology. In this paper, the autonomous target tracking and motion control of the robot are realized on the MT-R wheeled mobile robot platform through the improved machine learning algorithm, so as to further realize the intelligent application of the wheeled mobile robot. The main research contents of this paper can be summarized as follows: firstly, this paper reviews the research status of visual target tracking, enumerates the methods used in different visual target tracking algorithms, and analyzes the shortcomings of these algorithms. The tracking method based on multi-case learning and the algorithm combined with collaborative training are emphasized and analyzed. The foundation of the method theory in this paper is completed. At the same time, the research and development of mobile robots at home and abroad are briefly discussed. Then the internal target tracking algorithm of mobile robot is introduced in detail. The target tracking algorithm based on detection usually relies on the classifier to distinguish the target from the background to achieve the goal of tracking. When the classifier is learned, the image will be divided into two separate steps: sample sampling and tagging. However, the sample selected in this way is purposeless, which leads to the instability of the effect of the classifier. In this paper, based on the active learning model, a new sample selection algorithm is proposed. Based on the framework of multi-case learning algorithm, the active sample selection strategy is added between sample sampling and label allocation. In this way, the samples which are helpful to the learning of classifiers can be selected, and then the collaborative training method can be combined to prevent the drift caused by error accumulation and further improve the performance of the algorithm. Compared with the other six algorithms on the standard video sequence, the results show that the proposed method has good performance and robustness under the complex conditions of target occlusion, light change and so on. Finally, the target tracking algorithm is proposed on MT-R mobile robot, and the autonomous target tracking of mobile robot is realized by combining the hardware driving strategy of the robot. The robustness of mobile robot target tracking is verified by experiments in different practical scenarios. The experimental results show that the target tracking algorithm proposed in this paper can effectively help the mobile robot to track the target effectively under the condition of target occlusion, light change and so on.
【學(xué)位授予單位】：浙江理工大學(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6條

1 王麗佳;賈松敏;李秀智;王爽;;基于改進(jìn)在線多示例學(xué)習(xí)算法的機(jī)器人目標(biāo)跟蹤[J];自動(dòng)化學(xué)報(bào);2014年12期

2 陳思;蘇松志;李紹滋;呂艷萍;曹冬林;;基于在線半監(jiān)督boosting的協(xié)同訓(xùn)練目標(biāo)跟蹤算法[J];電子與信息學(xué)報(bào);2014年04期

3 譚民;王碩;;機(jī)器人技術(shù)研究進(jìn)展[J];自動(dòng)化學(xué)報(bào);2013年07期

4 王素玉;沈蘭蓀;;智能視覺(jué)監(jiān)控技術(shù)研究進(jìn)展[J];中國(guó)圖象圖形學(xué)報(bào);2007年09期

5 胡斌;何克忠;;計(jì)算機(jī)視覺(jué)在室外移動(dòng)機(jī)器人中的應(yīng)用[J];自動(dòng)化學(xué)報(bào);2006年05期

6 侯志強(qiáng);韓崇昭;;視覺(jué)跟蹤技術(shù)綜述[J];自動(dòng)化學(xué)報(bào);2006年04期

相關(guān)博士學(xué)位論文 前2條

1 武玉偉;視覺(jué)目標(biāo)跟蹤中的表觀建模研究[D];北京理工大學(xué);2014年

2 邱雪娜;基于視覺(jué)的運(yùn)動(dòng)目標(biāo)跟蹤算法及其在移動(dòng)機(jī)器人中的應(yīng)用[D];華東理工大學(xué);2011年

相關(guān)碩士學(xué)位論文 前6條

1 韓信;基于雙目視覺(jué)的輪式機(jī)器人動(dòng)態(tài)避障研究[D];浙江大學(xué);2016年

2 姜衛(wèi)琳;智能移動(dòng)機(jī)器人均值漂移目標(biāo)跟蹤算法研究[D];西北大學(xué);2014年

3 張全喜;單目視覺(jué)自主移動(dòng)機(jī)器人的目標(biāo)跟蹤及誤差分析[D];南京理工大學(xué);2014年

4 周秋紅;基于多示例學(xué)習(xí)的運(yùn)動(dòng)目標(biāo)跟蹤算法研究[D];大連理工大學(xué);2010年

5 耿盛濤;基于雙目視覺(jué)的機(jī)器人目標(biāo)檢測(cè)與跟蹤研究[D];江南大學(xué);2010年

6 張亞楠;復(fù)雜背景下目標(biāo)跟蹤算法研究及其在移動(dòng)機(jī)器人中的應(yīng)用[D];華南理工大學(xué);2010年

，

本文編號(hào)：2496445