EN - PCCIN - Linha de Investigação de Robótica Móvel
Permanent URI for this collection
Browse
Recent Submissions
- Interoperability in Heterogeneous Team of Search and Rescue RobotsPublication . Marques, Mário Monteiro; Lobo, Victor J. A. S.Search and rescue missions are complex operations. A disaster scenario is generally unstructured, time-varying and unpredictable. This poses several challenges for the successful deployment of unmanned technology. The variety of operational scenarios and Tasks lead to the need for multipe robots of different types, domains and sizes. A priori planning of the optimal set of assets to be depoyed and the definition of their mission objectives are generally not feasible as information only becomes available during mission. The ICARUS project responds to this challenge by developing a heterogeneous team composed by different and complementary robos, dynamically cooperating as an interoperable team. This chapter describes our approach to multi-robot interoperability, understood as the ability of multiple robots to operate together, in synergy, enabling multiple teams to share data, intelligence and resources, which is the ultimate objective of ICARUS project. It also includes the analysis of the relevant standardization initiatives in multi-robot multi-domain systems, our implementation of an interoperability framework and several examples of multi-robot cooperation of the ICARUS robots in realistic search and rescue missions.
- “iCub, clean the table!” A robot learning from demonstration approach using Deep Neural NetworksPublication . Kim, Jaeseok; Cauli, Nino; Vicente, Pedro; Damas, Bruno; Cavallo, Filippo; Santos, Victor JoséAutonomous service robots have become a key research topic in robotics, particularly for household chores. A typical home scenario is highly unconstrained and a service robot needs to adapt constantly to new situations. In this paper, we address the problem of autonomous cleaning tasks in uncontrolled environments. In our approach, a human instructor uses kinestethic demonstrations to teach a robot how to perform different cleaning tasks on a table. Then, we use Task Parametrized Gaussian Mixture Models (TP-GMMs) to encode the demonstrations variability, while providing appropriate generalization abilities. TP-GMMs extend Gaussian Mixture Models with an auxiliary set of reference frames, in order to extrapolate the demonstrations to different task parameters such as movement locations, amplitude or orientations. However, the reference frames (that parametrize TP-GMMs) can be very difficult to extract in practice, as it may require segmenting the cluttered images of the working table-top. Instead, in this work the reference frames are automatically extracted from robot camera images, using a deep neural network that was trained during human demonstrations of a cleaning task. This approach has two main benefits: (i) it takes the human completely out of the loop while performing complex cleaning tasks; and (ii) the network is able to identify the specific task to be performed directly from image data, thus also enabling automatic task selection from a set of previously demonstrated tasks. The system was implemented on the iCub humanoid robot. During the tests, the robot was able to successfully clean a table with two different types of dirt (wiping a marker’s scribble or sweeping clusters of lentils).
- Use of UAV for the Performance Assessment of Visual Aids to NavigationPublication . Conceição, Vítor Fernando Plácido da; Duarte, Filipe; Dias, Vitor; Teles, JorgeCoastal and inshore navigational areas are becoming increasingly congested not only due to the vessels traffic, but also from the more recent economic activities such as offshore wind farms, tidal turbines and aquaculture sites. At the same time the challenges presented by coastal development like “light pollution” or operational requirements of larger vessels or high-speed crafts are imposing more complex design solutions for the Aids to Navigation (AtoN). On the other side, users are calling for higher effectiveness of the service being provided, namely through clear statements of the level of service and performance standard. Over the last decade we have witnessed a large diversity of UAV application solution in several domains. The associated technology is becoming cheaper, easily achievable and with higher levels of performance. This paper presents the results of several tests to validate the conceptual use of UAVs in the performance assessment of AtoN. Results point to the possibility for the definition of more detailed performance indicators of AtoN. UAVs fitted with optical sensors may simulate the perception of observers at several heights and directions. Above all, they provide a systematic methodology to assess or monitor the conspicuity of AtoN at pre-set positions or paths.
- Collaborative Method to Develop an Enterprise Architecture in a Public InstitutionPublication . Roboredo, Nuno Paulo RochaThe growth of organizational complexity degrades business processes efficiency. Enterprise Architecture (EA) is an instrument to manage organizational complexity, through the improvement of organizational self-awareness. EA improves alignment between business and IT to ensure the business value of IT, and enables rationalization of organizational resources. However, depending of organizational culture and characteristics, there are several issues hindering the EA development within an organization. Actual frameworks, like TOGAF, require a significant number of skilled human resources (HR), which some organizations, like public institutions, cannot assign to EA activities. Our research goal is to provide an EA capability to public institutions, enabling these institutions to take advantage of EA benefits. Public institution contexts and stakeholder concerns were explored as well as issues acting as enablers or as inhibitors for an EA development. We propose a collaborative method to develop an EA, applying lean and agile principles, focusing on public institution specificities. Our collaborative method tries to capture organizational knowledge, spread among employees, into an EA model, to map the enterprise cartography of the institution. Our method has been demonstrated and evaluated in the IT sector of the Portuguese Navy.