机器人听觉声源定位研究综述

声源定位技术定位出外界声源相对于机器人的方向和位置,机器人听觉声源定位系统可以极大地提高机器人与外界交互的能力.总结和分析面向机器人听觉的声源定位技术对智能机器人技术的发展有着重要的意义.首先总结了面向机器人听觉的声源定位系统的特点,综述了机器人听觉声源定位的关键技术,包括到达时间差、可控波束形成、高分辨率谱估计、双耳听觉、主动听觉和视听融合技术.其次对麦克风阵列模型进行了分类,比较了基于三维麦克风阵列、二维麦克风阵列和双耳的7个典型系统的性能.最后总结了机器人听觉声源定位系统的应用,并分析了存在的问题和未来的发展趋势.     

A novel biologically inspired neural network solution for robotic 3D sound source sensing

This paper presents a novel real-time robotic binaural sound localization method based on hierarchical fuzzy artificial neural networks and a generic set of head related transfer functions. The robot is a humanoid equipped with the KEMAR artificial head and torso. Inside the ear canals two small microphones play the role of the eardrums in collecting the impinging sound waves. The neural networks are trained using synthesized sound sources placed every 5° from 0° to 255° in azimuth, and every 5° from − 45° to 80° in elevation. To improve generalization, the training data was corrupted with noise. Thanks to fuzzy logic, the method is able to interpolate at its output, locating with high accuracy sound sources at positions which were not used for training, even in presence of strong distortion. In order to achieve high localization accuracy, two different binaural cues are combined, namely the interaural intensity differences and interaural time differences. As opposed to microphone-array methods, the presented technique, uses only two microphones to localize sound sources in a real-time 3D environment. This work is fully supported by the German Research Foundation (DFG) within the collaborative research center SFB453 “High-Fidelity Telepresence and Teleaction”.     

一种半监督K均值多关系数据聚类算法

提出了一种半监督K均值多关系数据聚类算法.该算法在K均值聚类算法的基础上扩展了其初始类簇的选择方法和对象相似性度量方法,以用于多关系数据的半监督学习.为了获取高性能,该算法在聚类过程中充分利用了标记数据、对象属性及各种关系信息.多关系数据库Movie上的实验结果验证了该算法的有效性.     

Harmony K-means algorithm for document clustering

Fast and high quality document clustering is a crucial task in organizing information, search engine results, enhancing web crawling, and information retrieval or filtering. Recent studies have shown that the most commonly used partition-based clustering algorithm, the K-means algorithm, is more suitable for large datasets. However, the K-means algorithm can generate a local optimal solution. In this paper we propose a novel Harmony K-means Algorithm (HKA) that deals with document clustering based on Harmony Search (HS) optimization method. It is proved by means of finite Markov chain theory that the HKA converges to the global optimum. To demonstrate the effectiveness and speed of HKA, we have applied HKA algorithms on some standard datasets. We also compare the HKA with other meta-heuristic and model-based document clustering approaches. Experimental results reveal that the HKA algorithm converges to the best known optimum faster than other methods and the quality of clusters are comparable.     

A clustering method combining differential evolution with the K-means algorithm

The present paper considers the problem of partitioning a dataset into a known number of clusters using the sum of squared errors criterion (SSE). A new clustering method, called DE-KM, which combines differential evolution algorithm (DE) with the well known K-means procedure is described. In the method, the K-means algorithm is used to fine-tune each candidate solution obtained by mutation and crossover operators of DE. Additionally, a reordering procedure which allows the evolutionary algorithm to tackle the redundant representation problem is proposed. The performance of the DE-KM clustering method is compared to the performance of differential evolution, global K-means method, genetic K-means algorithm and two variants of the K-means algorithm. The experimental results show that if the number of clusters K is sufficiently large, DE-KM obtains solutions with lower SSE values than the other five algorithms.     

An effective web document clustering algorithm based on bisection and merge

To cluster web documents, all of which have the same name entities, we attempted to use existing clustering algorithms such as K-means and spectral clustering. Unexpectedly, it turned out that these algorithms are not effective to cluster web documents. According to our intensive investigation, we found that clustering such web pages is more complicated because (1) the number of clusters (known as ground truth) is larger than two or three clusters as in general clustering problems and (2) clusters in the data set have extremely skewed distributions of cluster sizes. To overcome the aforementioned problem, in this paper, we propose an effective clustering algorithm to boost up the accuracy of K-means and spectral clustering algorithms. In particular, to deal with skewed distributions of cluster sizes, our algorithm performs both bisection and merge steps based on normalized cuts of the similarity graph G to correctly cluster web documents. Our experimental results show that our algorithm improves the performance by approximately 56% compared to spectral bisection and 36% compared to K-means.     

基于熵加权K-means全局信息聚类的高光谱图像分类

目的 高光谱图像波段数目巨大，导致在解译及分类过程中出现“维数灾难”的现象。针对该问题，在K-means聚类算法基础上，考虑各个波段对不同聚类的重要程度，同时顾及类间信息，提出一种基于熵加权K-means全局信息聚类的高光谱图像分类算法。方法 首先，引入波段权重，用来刻画各个波段对不同聚类的重要程度，并定义熵信息测度表达该权重。其次，为避免局部最优聚类，引入类间距离测度实现全局最优聚类。最后，将上述两类测度引入K-means聚类目标函数，通过最小化目标函数得到最优分类结果。结果 为了验证提出的高光谱图像分类方法的有效性，对Salinas高光谱图像和Pavia University高光谱图像标准图中的地物类别根据其光谱反射率差异程度进行合并，将合并后的标准图作为新的标准分类图。分别采用本文算法和传统K-means算法对Salinas高光谱图像和Pavia University高光谱图像进行实验，并定性、定量地评价和分析了实验结果。对于图像中合并后的地物类别，光谱反射率差异程度大，从视觉上看，本文算法较传统K-means算法有更好的分类结果；从分类精度看，本文算法的总精度分别为92.20%和82.96%， K-means算法的总精度分别为83.39%和67.06%，较K-means算法增长8.81%和15.9%。结论 提出一种基于熵加权K-means全局信息聚类的高光谱图像分类算法，实验结果表明，本文算法对高光谱图像中具有不同光谱反射率差异程度的各类地物目标均能取得很好的分类结果。     

一种基于语料特性的聚类算法

为寻求模型不匹配问题的一种恰当的解决途径,提出了基于语料分布特性的CADIC(clustering algorithm based on the distributions of intrinsic clusters)聚类算法。CADIC以重标度的形式隐式地将语料特性融入算法框架,从而使算法模型具备更灵活的适应能力。在聚类过程中,CADIC选择一组具有良好区分度的方向构建CADIC坐标系,在该坐标系下统计固有簇的分布特性,以构造各个坐标轴的重标度函数,并以重标度的形式对语料分布进行隐式的归一化,从而提高聚     

Auditory scene reproduction for tele-operated robot systems

ABSTRACT

In a tele-operated robot environment, reproducing auditory scenes and conveying 3D spatial information of sound sources are inevitable in order to make operators feel more realistic tele-presence. In this paper, we propose a tele-presence robot system that enables reproduction and manipulation of auditory scenes. This tele-presence system is carried out on the basis of 3D information about where targeted humans are speaking, and matching with the operator's head orientation. We employed multiple microphone arrays and human tracking technologies to localize and separate voices around a robot. In the operator site, separated sound sources are rendered using head-related transfer functions (HRTF) according to the sound sources' spatial positions and the operator's head orientation that is being tracked real time. Two-party and three-party interaction experiments indicated that the proposed system has significantly higher accuracy when perceiving direction of sounds and gains higher subjective scores in sense of presence and listenability, compared to a baseline system which uses stereo binaural sounds obtained by two microphones mounted on the humanoid robot's ears.     

一种面向实时交互的变形手势跟踪方法

变形手势跟踪是基于视觉的人机交互研究中的一项重要内容.单摄像头条件下,提出一种新颖的变形手势实时跟踪方法.利用一组2D手势模型替代高维度的3D手模型.首先利用贝叶斯分类器对静态手势进行识别,然后对图像进行手指和指尖定位,通过将图像特征与识别结果进行匹配,实现了跟踪过程的自动初始化.提出将K-means聚类算法与粒子滤波相结合,用于解决多手指跟踪问题中手指互相干扰的问题.跟踪过程中进行跟踪状态检测,实现了自动恢复跟踪及手势模型更新.实验结果表明,该方法可以实现对变形手势快速、准确的连续跟踪,能够满足基于视觉的实时人机交互的要求.     

A Novelty-based Clustering Method for On-line Documents

In this paper, we describe a document clustering method called novelty-based document clustering. This method clusters documents based on similarity and novelty. The method assigns higher weights to recent documents than old ones and generates clusters with the focus on recent topics. The similarity function is derived probabilistically, extending the conventional cosine measure of the vector space model by incorporating a document forgetting model to produce novelty-based clusters. The clustering procedure is a variation of the K-means method. An additional feature of our clustering method is an incremental update facility, which is applied when new documents are incorporated into a document repository. Performance of the clustering method is examined through experiments. Experimental results show the efficiency and effectiveness of our method.     

Statistical,Evolutionary, and Neurocomputing Clustering Techniques: Cluster-Based vs Object-Based Approaches

Modern day computers cannot provide optimal solution to the clustering problem. There are many clustering algorithms that attempt to provide an approximation of the optimal solution. These clustering techniques can be broadly classified into two categories. The techniques from first category directly assign objects to clusters and then analyze the resulting clusters. The methods from second category adjust representations of clusters and then determine the object assignments. In terms of disciplines, these techniques can be classified as statistical, genetic algorithms based, and neural network based. This paper reports the results of experiments comparing five different approaches: hierarchical grouping, object-based genetic algorithms, cluster-based genetic algorithms, Kohonen neural networks, and K-means method. The comparisons consist of the time requirements and within-group errors. The theoretical analyses were tested for clustering of highway sections and supermarket customers. All the techniques were applied to clustering of highway sections. The hierarchical grouping and genetic algorithms approaches were computationally infeasible for clustering a larger set of supermarket customers. Hence only Kohonen neural networks and K-means techniques were applied to the second set to confirm some of the results from previous experiments.     

Interactive clustering and merging with a new fuzzy expected value

Major problems exist in both crisp and fuzzy clustering algorithms. The fuzzy c-means type of algorithms use weights determined by a power m of inverse distances that remains fixed over all iterations and over all clusters, even though smaller clusters should have a larger m. Our method uses a different “distance” for each cluster that changes over the early iterations to fit the clusters. Comparisons show improved results. We also address other perplexing problems in clustering: (i) find the optimal number K of clusters; (ii) assess the validity of a given clustering; (iii) prevent the selection of seed vectors as initial prototypes from affecting the clustering; (iv) prevent the order of merging from affecting the clustering; and (v) permit the clusters to form more natural shapes rather than forcing them into normed balls of the distance function. We employ a relatively large number K of uniformly randomly distributed seeds and then thin them to leave fewer uniformly distributed seeds. Next, the main loop iterates by assigning the feature vectors and computing new fuzzy prototypes. Our fuzzy merging then merges any clusters that are too close to each other. We use a modified Xie-Bene validity measure as the goodness of clustering measure for multiple values of K in a user-interaction approach where the user selects two parameters (for eliminating clusters and merging clusters after viewing the results thus far). The algorithm is compared with the fuzzy c-means on the iris data and on the Wisconsin breast cancer data.     

基于选择聚类集成的相似流形学习算法

流形学习是当今最重要的研究方向之一.约简维度的选择影响着流形学习方法的性能.当约简维度恰好是本征维度时,更容易发现原始数据的内在性质.然而,本征维度估计仍然是流形学习的一个研究难点.在此基础上,提出了一种新的无监督方法,即基于选择聚类集成的相似流形学习(SML-SCE)算法,避免了对本征维度的估计,并且性能表现良好.SML-SCE利用改进的层次平衡K-means(MBKHK)方法生成具有代表性的锚点,高效地构造相似度矩阵.随后计算得到了多个不同维度下的相似低维嵌入,这些低维嵌入是对原始数据的不同表示,而且不同低维嵌入之间的多样性有利于集成学习.因此,SML-SCE采用选择性聚类集成方法作为结合策略.对于通过K-means聚类得到的相似低维嵌入的聚类结果,采用聚类间的归一化互信息(NMI)作为权重的衡量标准.最后,舍弃权重较低的聚类,采用基于权重的选择性投票方案,得到最终的聚类结果.在多个数据集的大量实验结果表明了该方法的有效性.     

半监督的改进K-均值聚类算法

K-均值聚类算法必须事先获取聚类数目,并且随机地选取聚类初始中心会造成聚类结果不稳定,容易在获得一个局部最优值时终止。提出了一种基于半监督学习理论的改进K-均值聚类算法,利用少量标签数据建立图的最小生成树并迭代分裂获取K-均值聚类算法所需要的聚类数和初始聚类中心。在IRIS数据集上的实验表明,尽管随机样本构造的生成树不同,聚类中心也不同,但聚类是一致且稳定的,迭代的次数较少,验证了该文算法的有效性。     

基于正四棱锥结构的机器人声源定位系统研究

声源定位成为机器人智能研究的重要方向。针对当前声源定位精度不理想、实时性不佳等问题,提出了一种正四棱锥麦克风阵列声源定位结构。采用时间延迟估计的声源定位方法,并提出时延值的快速搜索策略;推导了该结构的基于信号时延的时空映射关系,建立了声源目标位置的几何计算模型,并依据正四棱锥结构特点及冗余的时延值对值域划分,缩小求解范围,运用迭代算法得到声源的位置坐标,并通过双重筛选机制剔除错误的定位结果。实验结果证明了该结构及定位算法在提高系统定位精度和实时性能的有效性,能满足机器人应用中对声源定位的需求。     

Parallel bisecting <Emphasis Type="Italic">k</Emphasis>-means with prediction clustering algorithm

In this paper, we propose a new parallel clustering algorithm, named Parallel Bisecting k-means with Prediction (PBKP), for message-passing multiprocessor systems. Bisecting k-means tends to produce clusters of similar sizes, and according to our experiments, it produces clusters with smaller entropy (i.e., purer clusters) than k-means does. Our PBKP algorithm fully exploits the data-parallelism of the bisecting k-means algorithm, and adopts a prediction step to balance the workloads of multiple processors to achieve a high speedup. We implemented PBKP on a cluster of Linux workstations and analyzed its performance. Our experimental results show that the speedup of PBKP is linear with the number of processors and the number of data points. Moreover, PBKP scales up better than the parallel k-means with respect to the dimension and the desired number of clusters. This research was supported in part by AFRL/Wright Brothers Institute (WBI).     

Sound Localization During Head Movement Using an Acoustical Telepresence Robot: TeleHead

Using our developed acoustical telepresence robot, TeleHead, we have so far confirmed that not only stationary binaural features, but also dynamic cues from head movement play important roles in sound localization. In this study, aiming towards the realization of an ideal acoustical telepresence robot, we clarify the relation between the head movement and the accuracy of sound localization in sound localization experiments. We examined two factors related to head movement that should have an impact on sound localization accuracy: observation from multiple postures and dynamic information during head movement. The results suggest that both factors improve the accuracy of sound localization in experiments. Moreover, even when we can use only one of these factors, the accuracy of sound localization is almost the same as the subject's original accuracy. The results confirm that even under very bad communication, control and head-shape conditions, the synchronization of head movement is important for building an acoustical telepresence robot. They also point to the possibility of building an acoustical telepresence robot with a dummy head of a general shape. This is meaningful from the viewpoint of engineering. In addition, it suggests the strong robustness of the human sound localization function.     

A real-time super-resolution robot audition system that improves the robustness of simultaneous speech recognition

This study addresses a framework for a robot audition system, including sound source localization (SSL) and sound source separation (SSS), that can robustly recognize simultaneous speeches in a real environment. Because SSL estimates not only the location of speakers but also the number of speakers, such a robust framework is essential for simultaneous speech recognition. Moreover, improvement in the performance of SSS is crucial for simultaneous speech recognition because the robot has to recognize the individual source of speeches. For simultaneous speech recognition, current robot audition systems mainly require noise-robustness, high resolution, and real-time implementation. Multiple signal classification (MUSIC) based on standard Eigenvalue decomposition (SEVD) and Geometric-constrained high-order decorrelation-based source separation (GHDSS) are techniques utilizing microphone array processing, which are used for SSL and SSS, respectively. To enhance SSL robustness against noise while detecting simultaneous speeches, we improved SEVD-MUSIC by incorporating generalized Eigenvalue decomposition (GEVD). However, GEVD-based MUSIC (GEVD-MUSIC) and GHDSS mainly have two issues: (1) the resolution of pre-measured transfer functions (TFs) determines the resolution of SSL and SSS and (2) their computational cost is expensive for real-time processing. For the first issue, we propose a TF-interpolation method integrating time-domain-based and frequency-domain-based interpolation. The interpolation achieves super-resolution robot audition, which has a higher resolution than that of the pre-measured TFs. For the second issue, we propose two methods for SSL: MUSIC based on generalized singular value decomposition (GSVD-MUSIC) and hierarchical SSL (H-SSL). GSVD-MUSIC drastically reduces the computational cost while maintaining noise-robustness for localization. In addition, H-SSL reduces the computational cost by introducing a hierarchical search algorithm instead of using a greedy search for localization. These techniques are integrated into a robot audition system using a robot-embedded microphone array. The preliminary experiments for each technique showed the following: (1) The proposed interpolation achieved approximately 1-degree resolution although the TFs are only at 30-degree intervals in both SSL and SSS; (2) GSVD-MUSIC attained 46.4 and 40.6% of the computational cost compared to that of SEVD-MUSIC and GEVD-MUSIC, respectively; (3) H-SSL reduced 71.7% of the computational cost to localize a single speaker. Finally, the robot audition system, including super-resolution SSL and SSS, is applied to robustly recognize four sources of speech occurring simultaneously in a real environment. The proposed system showed considerable performance improvements of up to 7% for the average word correct rate during simultaneous speech recognition, especially when the TFs were at more than 30-degree intervals.     

A hybrid clustering procedure for concentric and chain-like clusters

K-means algorithm is a well known nonhierarchical method for clustering data. The most important limitations of this algorithm are that: (1) it gives final clusters on the basis of the cluster centroids or the seed points chosen initially, and (2) it is appropriate for data sets having fairly isotropic clusters. But this algorithm has the advantage of low computation and storage requirements. On the other hand, hierarchical agglomerative clustering algorithm, which can cluster nonisotropic (chain-like and concentric) clusters, requires high storage and computation requirements. This paper suggests a new method for selecting the initial seed points, so that theK-means algorithm gives the same results for any input data order. This paper also describes a hybrid clustering algorithm, based on the concepts of multilevel theory, which is nonhierarchical at the first level and hierarchical from second level onwards, to cluster data sets having (i) chain-like clusters and (ii) concentric clusters. It is observed that this hybrid clustering algorithm gives the same results as the hierarchical clustering algorithm, with less computation and storage requirements.