arXiv subjects

Ricardo Martins

Publications and source records attributed to Ricardo Martins.

8 recordsLinked to original sources

arXiv2026

Collaborative Streaming Anomaly Detection with Interactive Explanations and Ensemble Consensus

We present a collaborative streaming anomaly detection system for high-speed data streams that explicitly integrates human analysts into the decision loop. The system combines heterogeneous detectors and aggregates their outputs through a normalization-based weighted consensus, complemented by artifact-aware rules to stabilize anomaly scoring under deployment. To improve interpretability, it derives surrogate models that approximate the ensemble consensus and expose human-readable sensor conditions associated with anomalous behavior. Analysts can actively intervene by reviewing anomaly episodes, adjusting consensus behavior, and refining surrogate rules used for anomaly prediction, producing a human-adjusted ensemble. We evaluate the approach on an industrial stream with 260\,000 events and 3 anomalous episodes, showing robust detection and actionable human-AI interaction.

cs.LG↗

arXiv2025

Explainable Anomaly Detection for Industrial IoT Data Streams

Industrial maintenance is being transformed by the Internet of Things and edge computing, generating continuous data streams that demand real-time, adaptive decision-making under limited computational resources. While data stream mining (DSM) addresses this challenge, most methods assume fully supervised settings, yet in practice, ground-truth labels are often delayed or unavailable. This paper presents a collaborative DSM framework that integrates unsupervised anomaly detection with interactive, human-in-the-loop learning to support maintenance decisions. We employ an online Isolation Forest and enhance interpretability using incremental Partial Dependence Plots and a feature importance score, derived from deviations of Individual Conditional Expectation curves from a fading average, enabling users to dynamically reassess feature relevance and adjust anomaly thresholds. We describe the real-time implementation and provide initial results for fault detection in a Jacquard loom unit. Ongoing work targets continuous monitoring to predict and explain imminent bearing failures.

cs.LG↗

arXiv2023

Tension Estimation and Localization for a Tethered Micro Aerial Robot

This work focuses on the study of tethered fights of a micro quadcopter, with the aim of supplying continuous power to a small-sized aerial robot. Multiple features for facilitating the interaction between a tethered micro quadcopter and a ground base are described in this paper. Firstly, a tether model based on the catenary curve is presented that describes a quadcopter tethered to a point in space. Furthermore, a method capable of estimating the tension applied to the quadcopter, based only on the inertial information from the IMU sensors and the motor thrusts, is presented. Finally, a novel method for localizing the quadcopter by exploiting the tension imposed by the tether and the shape of the tether is described. The proposed methods are evaluated both in simulation and in real world prototype.

cs.RO↗

arXiv2022

A survey of the European Open Science Cloud services for expanding the capacity and capabilities of multidisciplinary scientific applications

Open Science is a paradigm in which scientific data, procedures, tools and results are shared transparently and reused by society as a whole. The initiative known as the European Open Science Cloud (EOSC) is an effort in Europe to provide an open, trusted, virtual and federated computing environment to execute scientific applications, and to store, share and re-use research data across borders and scientific disciplines. Additionally, scientific services are becoming increasingly data-intensive, not only in terms of computationally intensive tasks but also in terms of storage resources. Computing paradigms such as High Performance Computing (HPC) and Cloud Computing are applied to e-science applications to meet these demands. However, adapting applications and services to these paradigms is not a trivial task, commonly requiring a deep knowledge of the underlying technologies, which often constitutes a barrier for its uptake by scientists in general. In this context, EOSC-SYNERGY, a collaborative project involving more than 20 institutions from eight European countries pooling their knowledge and experience to enhance EOSC's capabilities and capacities, aims to bring EOSC closer to the scientific communities. This article provides a summary analysis of the adaptations made in the ten thematic services of EOSC-SYNERGY to embrace this paradigm. These services are grouped into four categories: Earth Observation, Environment, Biomedicine, and Astrophysics. The analysis will lead to the identification of commonalities, best practices and common requirements, regardless of the thematic area of the service. Experience gained from the thematic services could be transferred to new services for the adoption of the EOSC ecosystem framework.

cs.DC↗

arXiv2019

Development of Head-Mounted Projection Displays for Distributed, Collaborative, Augmented Reality Applications

Distributed systems technologies supporting 3D visualization and social collaboration will be increasing in frequency and type over time. An emerging type of head-mounted display referred to as the head-mounted projection display (HMPD) was recently developed that only requires ultralight optics (i.e., less than 8 g per eye) that enables immersive multiuser, mobile augmented reality 3D visualization, as well as remote 3D collaborations. In this paper a review of the development of lightweight HMPD technology is provided, together with insight into what makes this technology timely and so unique. Two novel emerging HMPD-based technologies are then described: a teleportal HMPD(T-HMPD) enabling face-to-face communication and visualization of shared 3D virtual objects, and a mobile HMPD (M-HMPD) designed for outdoor wearable visualization and communication. Finally, the use of HMPD in medical visualization and training, as well as in infospaces, two applications developed in the ODA and MIND labs respectively, are discussed.

cs.HC↗

arXiv2018

Beyond the Desktop: Emerging Technologies for Supporting 3D Collaborative Teams

The emergence of several trends, including the increased availability of wireless networks, miniaturization of electronics and sensing technologies, and novel input and output devices, is creating a demand for integrated, full-time displays for use across a wide range of applications, including collaborative environments. In this paper, we present and discuss emerging visualization methods we are developing particularly as they relate to deployable displays and displays worn on the body to support mobile users.

cs.HC↗

arXiv2014

Touch attention Bayesian models for robotic active haptic exploration of heterogeneous surfaces

This work contributes to the development of active haptic exploration strategies of surfaces using robotic hands in environments with an unknown structure. The architecture of the proposed approach consists two main Bayesian models, implementing the touch attention mechanisms of the system. The model pi_per perceives and discriminates different categories of materials (haptic stimulus) integrating compliance and texture features extracted from haptic sensory data. The model pi_tar actively infers the next region of the workspace that should be explored by the robotic system, integrating the task information, the permanently updated saliency and uncertainty maps extracted from the perceived haptic stimulus map, as well as, inhibition-of-return mechanisms. The experimental results demonstrate that the Bayesian model pi_per can be used to discriminate 10 different classes of materials with an average recognition rate higher than 90% . The generalization capability of the proposed models was demonstrated experimentally. The ATLAS robot, in the simulation, was able to perform the following of a discontinuity between two regions made of different materials with a divergence smaller than 1cm (30 trials). The tests were performed in scenarios with 3 different configurations of the discontinuity. The Bayesian models have demonstrated the capability to manage the uncertainty about the structure of the surfaces and sensory noise to make correct motor decisions from haptic percepts.

cs.RO↗

arXiv2013

BW - Eye Ophthalmologic decision support system based on clinical workflow and data mining techniques-image registration algorithm

Blueworks - Medical Expert Diagnosis is developing an application, BWEye, to be used as an ophthalmology consultation decision support system. The implementation of this application involves several different tasks and one of them is the implementation of an ophthalmology images registration algorithm. The work reported in this document is related with the implementation of an algorithm to register images of angiography, colour retinography and redfree retinography. The implementations described were developed in the software MATLAB. The implemented algorithm is based in the detection of the bifurcation points (y-features) of the vascular structures of the retina that usually are visible in the referred type of images. There are proposed two approaches to establish an initial set of features correspondences. The first approach is based in the maximization of the mutual information of the bifurcation regions of the features of images. The second approach is based in the characterization of each bifurcation point and in the minimization of the Euclidean distance between the descriptions of the features of the images in the descriptors space. The final set of the matching features for a pair of images is defined through the application of the RANSAC algorithm. Although, it was not achieved the implementation of a full functional algorithm, there were made several analysis that can be important to future improvement of the current implementation.

cs.CV↗