Ekaitz Jauregi

Histogram distance-based Bayesian Network structure learning: A supervised classification specific approach

In this work we introduce a methodology based on histogram distances for the automatic induction of Bayesian Networks (BN) from a file containing cases and variables related to a supervised classification problem. The main idea consists of learning the Bayesian Network structure for classification purposes taking into account the classification itself, by comparing the class distribution histogram distances obtained by the Bayesian Network after classifying each case. The structure is learned by applying eight different measures or metrics: the Cooper and Herskovits metric for a general Bayesian Network and seven different statistical distances between pairs of histograms. The results obtained confirm the hypothesis of the authors about the convenience of having a BN structure learning method which takes into account the existence of the special variable (the one corresponding to the class) in supervised classification problems.

Classifier Subset Selection to construct multi-classifiers by means of estimation of distribution algorithms

Abstract This paper proposes a novel approach to select the individual classifiers to take part in a Multiple-Classifier System. Individual classifier selection is a key step in the development of multi-classifiers. Several works have shown the benefits of fusing complementary classifiers. Nevertheless, the selection of the base classifiers to be used is still an open question, and different approaches have been proposed in the literature. This work is based on the selection of the appropriate single classifiers by means of an evolutionary algorithm. Different base classifiers, which have been chosen from different classifier families, are used as candidates in order to obtain variability in the classifications given. Experimental results carried out with 20 databases from the UCI Repository show how adequate the proposed approach is; Stacked Generalization multi-classifier has been selected to perform the experimental comparisons.

Fusing multiple image transformations and a thermal sensor with kinect to improve person detection ability

This paper proposes a novel approach to combine data from multiple low-cost sensors to detect people in a mobile robot. Robust detection of people is a key capability required for robots working in environments with people. Several works have shown the benefits of fusing data from complementary sensors. The Kinect sensor provides a rich data set at a significantly low cost, however, it has some limitations for its use on a mobile platform, mainly that people detection algorithms rely on images captured by a static camera. To cope with these limitations, this work is based on the fusion of Kinect and a thermical sensor (thermopile) mounted on top of a mobile platform. We propose the implementation of an evolutionary selection of sequences of image transformation to detect people through supervised classifiers. Experimental results carried out with a mobile platform in a manufacturing shop floor show that the percentage of wrong classified using only Kinect is drastically reduced with the classification algorithms and with the combination of the three information sources. Extra experiments are presented as well to show the benefits of the image transformation sequence idea here presented.

Humanizing NAO robot teleoperation using ROS

The work presented here proposes two different ROS packages to enrich the teleoperation of the robot NAO: speech-based teleoperation (in Basque) and gesture-based teleoperation together with arm control. These packages have been used and evaluated in a human mimicking experiment. The tools offered can serve as a base for many applications.

K Nearest Neighbor Equality: Giving equal chance to all existing classes

The nearest neighbor classification method assigns an unclassified point to the class of the nearest case of a set of previously classified points. This rule is independent of the underlying joint distribution of the sample points and their classifications. An extension to this approach is the k-NN method, in which the classification of the unclassified point is made by following a voting criteria within the k nearest points. The method we present here extends the k-NN idea, searching in each class for the k nearest points to the unclassified point, and classifying it in the class which minimizes the mean distance between the unclassified point and the k nearest points within each class. As all classes can take part in the final selection process, we have called the new approach k Nearest Neighbor Equality (k-NNE). Experimental results we obtained empirically show the suitability of the k-NNE algorithm, and its effectiveness suggests that it could be added to the current list of distance based classifiers.

Bertsobot: The first minstrel robot

We describe a robot capable of composing and playing traditional Basque impromptu verses - bertsoak. The system, called Bertsobot, is able to construct improvised verses according to given constraints on rhyme and meter, and to perform it in public. Towards this end, several tools and applications have been developed and integrated in Bertsobot, including: speech-based communication system, text applications for verse generation, and robot behaviours to interact with the environment in a public performance. We describe the tools and processes behind our approach, present some early experimental results and illustrative verses, and finally, remark the conclusions and future steps.

Loop-closing: A typicality approach

Loop-closing has long been identified as a critical issue when building maps from local observations. Topological mapping methods abstract the problem of how loops are closed from the problem of how to determine the metrical layout of places in the map and how to deal with noisy sensors. The typicality problem refers to the identification of new classes in a general classification context. This typicality concept is used in this paper to help a robot acquire a topological representation of the environment during its exploration phase. The problem is addressed using the INCA statistic which follows a distance-based approach. In this paper we describe a place recognition approach based on match testing by means of the INCA test. We describe the theoretical basis of the approach and present extensive experimental results performed in both a simulated and a real robot-environment system; Behaviour Based philosophy is used to construct the whole control architecture. Obtained results show the validity of the approach.

Textual Coherence in a Verse-Maker Robot

The Bertsobot project aims to develop an autonomous robot capable of composing and playing traditional Basque impromptu verses –bertsoak. The system should be able to construct novel verses according to given constraints on rhyme and meter, and to perform it in public. The Bertsobot project, at the intersection of Autonomous Robotics, Natural Language Generation and Human Robot Interaction, works to model the human abilities that collaborate in the process that enables a verse-maker to produce impromptu verses. This paper provides a general overview of the system, specially focusing on the description and evaluation of different semantic similarity methods for predicting the textual coherence of the generated verses.

DOOR HANDLE IDENTIFICATION: A THREE-STAGE APPROACH

Abstract The aim of the work presented here is to develop a door identification subsystem based on door handle recognition. The problem is stated as deciding whether a door handle is present or not in the images taken by the robot while navigating. And it is solved from a supervised classification point of view, combining Machine Learning segmentation with the Hough transform and statistical measures.

Visual Approaches for Handle Recognition

Objects can be identified in images extracting local image descriptors for interesting regions. In this paper, instead of making the handle identification process rely in the keypoint detection/matching process only, we present a method that first extracts from the image a region of interest (ROI) that with high probability contains the handle. This subimage is then processed by the keypoint detection/matching algorithm. Two methods for extracting the ROI are compared, Circle Hough Transform (CHT) and blobs, and combined with three descriptor extraction methods: SIFT, SURF and USURF.