Abstract

Abstract
FLEXUS unmanned surface vehicle was designed in the context of the Internet of Moving Things. This small catamaran weights less than 15kg and is less than 1m long, making it a very convenient vehicle with reduced logistics needs for operations in real outdoor environments. The present paper describes the resulting system both in terms of design and performances. Based on the requirements for this project, the subsystems composing the vehicle are described. Results obtained from experiments conducted in outdoor conditions have successfully validated this design and are presented in this paper.

Abstract
Virtual Reality offers the possibility of creating the most diverse types of videogames, being possible to have different types of videogames such as calm, eventful or even terrifying. The present study aimed to verify whether different types of videogames have impact on player's emotional responses as wecc as on the self-reported sense of presence. In addition, two distinct display interfaces were considered: conventional computer display and head-mounted display. To carry out this study, an experimental study was designed based on three videogames with distinct natures (exploration, puzzle and horror), where the participants had to fulfill certain tasks within a predetermined period. The data analysis revealed that there no significant emotional changes between the different videogames used. The results also revealed that the use of head-mounted displays increases the sense of presence in scenarios where the field of view is wide.

Abstract
In this paper, we present a comparison between the older DSTV file format and the newer version of the IFC standard, dedicating special attention of its impact in the robotization of welding and cutting processes in the steel structure fabrication industry. In the last decade, we have seen in this industry a significant increase in the request for automation. These new requirements are imposed by a market focused on the productivity enhancement through automation. Because of this paradigm change, the information structure and workflow provided by the DSTV format needed to be revised, namely the one related with the plan and management of steel fabrication processes. Therefore, with this work we enhance the importance of the increased digitalization of information that the newer version of the IFC standard provide, by showing how this information can be used to develop advanced robotic cells. More in detail, we will focus on the automatic generation of robot welding and cutting trajectories, and in the automatic part assembly planning during components fabrications. Besides these advantages, as this information is normally described having as base a perfect CAD model of the metallic structure, the resultant robot trajectories will have some dimensional error when fitted with the real physical component. Hence, we also present some automatic approaches based on a laser scanner and simple heuristics to overcome this limitations.

Abstract
Digital image methodologies related with Melanoma has become in the past years a major support for differential diagnosis in skin cancer. Computer Aided Diagnosis (CAD) systems, encompassing image acquisition, artifact removal, detection and selection of features, highlight Machine Learning algorithms as a novel strategy towards a digital assisted diagnosis in Dermatology. Although the central role played by color in dermoscopic image assessment, Machine Learning algorithms mainly use texture and shape features, derived from gray level images, obtained from the true color images of the skin. Since the acquisition conditions are key for the color characterization and thus, central for the quantification of different colors in a dermoscopic image, this work presents a strategy for color normalization, joint with its use in the calculation of the number of colors of a dermoscopic image. This methodology will contribute to the uniformity in the use of color features extracted from different datasets in CAD systems (acquired by distinct dermoscopes) possibly presenting distinct illumination characteristics. This normalization proposal can also be applied as an image preprocessing step, aimed to achieve higher scores in the standard metrics in ML algorithms.

Abstract
Many stochastic search algorithms are designed to optimize a fixed objective function to learn a task, i.e., if the objective function changes slightly, for example, due to a change in the situation or context of the task, relearning is required to adapt to the new context. For instance, if we want to learn a kicking movement for a soccer robot, we have to relearn the movement for different ball locations. Such relearning is undesired as it is highly inefficient and many applications require a fast adaptation to a new context/situation. Therefore, we investigate contextual stochastic search algorithms that can learn multiple, similar tasks simultaneously. Current contextual stochastic search methods are based on policy search algorithms and suffer from premature convergence and the need for parameter tuning. In this paper, we extend the well known CMA-ES algorithm to the contextual setting and illustrate its performance on several contextual tasks. Our new algorithm, called contextual CMAES, leverages from contextual learning while it preserves all the features of standard CMA-ES such as stability, avoidance of premature convergence, step size control and a minimal amount of parameter tuning.