Publications

This paper presents an RGB-D based human re-identification approach using novel biometrics features from the body’s volume. Existing work based on RGB images or skeleton features have some limitations for realworld robotic applications, most notably in dealing with occlusions and orientation of the user. Here, we propose novel features that allow performing re-identification when the person is facing side/backward or the person is partially occluded. The proposed approach has been tested for various scenarios including different views, occlusion and the public BIWI RGBD-ID dataset.

Double-layer hydroxide minerals are part of a very interesting group of natural and synthetic compounds with trigonal or hexagonal symmetry and a flexible layered crystal structure. They are formed extremely frequently in geologic, industrial, and synthetic processes. The ease of formation is related to the possibility of accommodating divalent and trivalent cations in the structure, together with a range of anionic species. Some compounds of the group, namely those based on hydrotalcite chemistry, are invariably found as products of the pozzolanic reaction between lime and clays in ancient mortars and modern binders that serve as alternatives to Portland clinker. The present review wishes to relate the structural properties of hydrotalcitetype compounds to the crystal-chemical mechanisms taking place during long-term pozzolanic processes. The kinetics of CO3 exchange between the hydroxide and the atmosphere has important negative consequences for the radiocarbon dating of ancient mortars.

Human detection and tracking are essential aspects to be considered in service robotics, as the robot often shares its workspace and interacts closely with humans. This paper presents an online learning framework for human classification in 3D LiDAR scans, taking advantage of robust multi-target tracking to avoid the need for data annotation by a human expert. The system learns iteratively by retraining a classifier online with the samples collected by the robot over time. A novel aspect of our approach is that errors in training data can be corrected using the information provided by the 3D LiDAR-based tracking. In order to do this, an efficient 3D cluster detector of potential human targets has been implemented. We evaluate the framework using a new 3D LiDAR dataset of people moving in a large indoor public space, which is made available to the research community. The experiments analyse the real-time performance of the cluster detector and show that our online learned human classifier matches and in some cases outperforms its offline version.

Stress remains a significant social problem for individuals in modern societies. This paper presents a machine learning approach for the automatic detection of stress of people in a social situation by combining two sensor systems that capture physiological and social responses. We compare the performance using different classifiers including support vector machine, AdaBoost, and k-nearest neighbor. Our experimental results show that by combining the measurements from both sensor systems, we could accurately discriminate between stressful and neutral situations during a controlled Trier social stress test (TSST). Moreover, this paper assesses the discriminative ability of each sensor modality individually and considers their suitability for real-time stress detection. Finally, we present an study of the most discriminative features for stress detection.

Recent advances in mobile technology have the potential to radically change the quality of tools available for people with sensory impairments, in particular the blind and partially sighted. Nowadays almost every smart-phone and tablet is equipped with high-resolution cameras, typically used for photos, videos, games and virtual reality applications. Very little has been proposed to exploit these sensors for user localisation and navigation instead. To this end, the “Active Vision with Human-in-the-Loop for the Visually Impaired” (ActiVis) project aims to develop a novel electronic travel aid to tackle the “last 10 yards problem” and enable blind users to independently navigate in unknown environments, ultimately enhancing or replacing existing solutions such as guide dogs and white canes. This paper describes some of the project’s key challenges, in particular with respect to the design of a user interface (UI) that translates visual information from the camera to guidance instructions for the blind person, taking into account the limitations introduced by visual impairment. In this paper we also propose a multimodal UI that caters to the needs of the visually impaired that exploits human-machine progressive co-adaptation to enhance the user’s experience and improve navigation performance.

Technological advances and affordability of recent smart sensors, as well as the consolidation of common software platforms for the integration of the latter and robotic sensors, are enabling the creation of complex active and assisted living environments for improving the quality of life of the elderly and the less able people. One such example is the integrated system developed by the European project ENRICHME, the aim of which is to monitor and prolong the independent living of old people affected by mild cognitive impairments with a combination of smart-home, robotics and web technologies. This paper presents in particular the design and technological solutions adopted to integrate, process and store the information provided by a set of fixed smart sensors and mobile robot sensors in a domestic scenario, including presence and contact detectors, environmental sensors, and RFID-tagged objects, for long-term user monitoring and adaptation.

The automatic detection of anomalies in Active and Assisted Living (AAL) environments is important for monitoring the wellbeing and safety of the elderly at home. The integration of smart domotic sensors (e.g. presence detectors) and those ones equipping modern mobile robots (e.g. RGB-D cameras) provides new opportunities for addressing this challenge. In this paper, we propose a novel solution to combine local activity levels detected by a single RGB-D camera with the global activity perceived by a network of domotic sensors. Our approach relies on a new method for computing such a global activity using various presence detectors, based on the concept of entropy from information theory. This entropy effectively shows how active a particular room or environment’s area is. The solution includes also a new application of Hybrid Markov Logic Networks (HMLNs) to merge different information sources for local and global anomaly detection. The system has been tested with a comprehensive dataset of RGB-D and domotic data containing data entries from 37 different domotic sensors (presence, temperature, light, energy consumption, door contact), which is made publicly available. The experimental results show the effectiveness of our approach and its potential for complex anomaly detection in AAL settings.

We present a system for temporal detection of social interactions. Many of the works until now have succeeded in recognising activities from clipped videos in datasets, but for robotic applications, it is important to be able to move to more realistic data. For this reason, the proposed approach temporally detects intervals where individual or social activity is occurring. Recognition of human activities is a key feature for analysing the human behaviour. In particular, recognition of social activities is useful to trigger human-robot interactions or to detect situations of potential danger. Based on that, this research has three goals: (1) define a new set of descriptors, which are able to characterise human interactions; (2) develop a computational model to segment temporal intervals with social interaction or individual behaviour; (3) provide a public dataset with RGB-D data with continuous stream of individual activities and social interactions. Results show that the proposed approach attained relevant performance with temporal segmentation of social activities.

The Special Issue of Robotics and Autonomous Systems presents papers from the Seventh European Conference on Mobile Robots (ECMR’15). The paper Vision-based Markov Localization for Long Term Autonomy by Benjamin Suger, Michael Ruhnke and Wolfram Burgard describes an approach to localize a mobile robot relative to an image sequence recorded in a different season and its evaluation on several challenging datasets, where the approach outperformed state-of-the-art techniques. The paper Image Features for Visual Teach-and-Repeat Navigation in Changing Environments by Tomas Krajnik, Pablo Cristoforis, Keerthy Kusumam, Peer Neubert and Tom Duckett describes an evaluation of image features for long-term visual teach-and-repeat navigation in outdoor environments with changes in appearance due to seasonal and daily variations. The paper Efficient Retrieval of Arbitrary Objects from Long-Term Robot Observations by Nils Bore, Rares Ambrus, Patric Jensfelt and John Folkesson presents a method for efficient query and retrieval of arbitrarily shaped objects from large amounts of unstructured 3D point cloud data for long-term semantic mapping by mobile service robots. The paper Time-Dependent Gas Distribution Modelling by Sahar Asadi, Han Fan, Victor Hernandez Bennetts and Achim Lilienthal describes an approach for modelling variations over time in gas distributions sensed by mobile robots and predicting future measurements. The paper Probabilistic Ego-Motion Estimation Using Multiple Automotive Radar Sensors by Matthias Rapp, Michael Barjenbruch, Markus Hahn, Juergen Dickmann and Klaus Dietmayer presents a framework for self-motion estimation based on registration of consecutive radar scans and a likelihood model for the Doppler velocity. The paper Integrated Online Trajectory Planning and Optimization in Distinctive Topologies by Christoph Roesmann, Frank Hoffmann and Torsten Bertram presents a novel approach for trajectory planning by mobile robots in the presence of dynamic obstacles such as people.

Stress remains a significant social problem for individuals in modern societies. This paper presents a machine learning approach for the automatic detection of stress of people in a social situation by combining two sensor systems that capture physiological and social responses. We compare the performance using different classifiers including support vector machine, AdaBoost, and k-nearest neighbor. Our experimental results show that by combining the measurements from both sensor systems, we could accurately discriminate between stressful and neutral situations during a controlled Trier social stress test (TSST). Moreover, this paper assesses the discriminative ability of each sensor modality individually and considers their suitability for real-time stress detection. Finally, we present an study of the most discriminative features for stress detection.