Abstract
This paper describes the development of a multi-agent system (MAS) to support the implementation of zero-defect manufacturing strategies in multi-stage production systems. The MAS infrastructure, combined with on-line inspection tools, data analytics and knowledge generation, constitutes a suitable approach to integrate process and quality control in multi-stage environments. This will allow the early detection of product defects, the adaptation to operating condition changes and the optimisation of manufacturing processes. This type of integrated management structure is aligned with a zero-defect manufacturing production model which is of paramount importance in the actual state-of-the-art manufacturing paradigms. As a proof of concept, the devised manufacturing supervision model was deployed into an experimental multi-stage system that run a set of several tests on electrical motors. The agent-based solution was implemented using the JADE framework and the exchange of information structured by proper data models and industrial based Internet-of-Things and Machine-to-Machine technologies, such as OPC-UA, REST and JSON. The obtained results demonstrate the suitability of the devised integrated management model as a vehicle to achieve dynamic and continuous system improvement in multi-stage manufacturing environments.

Abstract
Interface design flaws are often at the root cause of use errors in medical devices. Medical incidents are seldom reported, thus hindering the understanding of the incident contributing factors. Moreover, when dealing with a use error, both novices and expert users often blame themselves for insufficient knowledge rather than acknowledge deficiencies in the device. Simulation-Based Medical Education (SBME) platforms can provide appropriate training to professionals, especially if the right incentives to keep training are in place. In this paper, we present a new SBME, particularly targeted at training interaction with medical devices such as ventilators and infusion pumps. Our SBME functions as a game mode of the PVSio-web, a graphical environment for design, evaluation, and simulation of interactive (human-computer) systems. An analytical evaluation of our current implementation is provided, by comparing the features on our SBME with a set of requirements for game-based medical simulators retrieved from the literature. By being developed in a free, open source platform, our SBME is highly accessible and can be easily adapted to specific use cases, such a specific hospital with a defined set of medical devices.

Abstract

Abstract
This paper presents an artificial cognitive system tightly integrating object perception and manipulation for assistive robotics. This is necessary for assistive robots, not only to perform manipulation tasks in a reasonable amount of time and in an appropriate manner, but also to robustly adapt to new environments by handling new objects. In particular, this system includes perception capabilities that allow robots to incrementally learn object categories from the set of accumulated experiences and reason about how to perform complex tasks. To achieve these goals, it is critical to detect, track and recognize objects in the environment as well as to conceptualize experiences and learn novel object categories in an open-ended manner, based on human-robot interaction. Interaction capabilities were developed to enable human users to teach new object categories and instruct the robot to perform complex tasks. A naive Bayes learning approach with a Bag-of-Words object representation are used to acquire and refine object category models. Perceptual memory is used to store object experiences, feature dictionary and object category models. Working memory is employed to support communication purposes between the different modules of the architecture. A reactive planning approach is used to carry out complex tasks. To examine the performance of the proposed architecture, a quantitative evaluation and a qualitative analysis are carried out. Experimental results show that the proposed system is able to interact with human users, learn new object categories over time, as well as perform complex tasks.

Abstract
Distributed Smart Systems (DSS) should operate and restore discontinued service to consumers. In order to the system gain theses ability it is necessary to replace the manual switches for remotely controlled switches, improving the system restoration capability having in view the Smart Grids implementation. This paper aims to develop a new model, determining the minimal set of switches to replace in order to automate the system, along with a senility analysis on the position of the new switches, whether it should be placed in the same place as the manual switch or in a new location. The optimization of the system is made considering the renewable energy sources (RES) integration in the grid and energy storage systems (ESS), simultaneously, in order to improve the system reliability. The computational tool is tested using the IEEE 119 Bus test system, where different types of loads are considered, residential, commercial and industrial.

Abstract
Over time, the electricity price and energy consumption are increasingly growing their weight as prime foundations of the electrical sector, with their analysis and forecasts being targeted as key elements for the stable maintenance of electricity markets. The advent of smart grids is escalating the importance of forecasting because of the expected ubiquitous monitoring and growing complexity of a data-rich ever-changing milieu. So, the increasing data volatility will require forecasting tools able to rapidly readjust to a dynamic environment. The Generalized Regression Neural Network (GRNN) approach is a solution that has recently re-emerged, emphasizing good performance, fast run-times and ease of parameterization. The merging of this model with more conventional methods allows us to obtain more sturdy solutions with shortened training times, when compared to conventional Artificial Neural Networks (ANN). Overall, the performance of the GRNN, although slightly inferior to that of the ANN, is suitable, but linked to much lower training times. Ultimately, the GRNN would be a proper solution to blend with the latest smart grids features, which may require much reduced forecasting training times.