Abstract
Artificial intelligence techniques have been applied to diverse business and governmental areas, in order to take advantage of the huge amount of information that is generated within specific organizations or institutions. Business intelligence can be seen as the process of converting such information into actionable knowledge, which is the basis for data-driven decision making. With this in mind, this work is framed in a project that seeks to improve the activity of the Portuguese Food and Economic Safety Authority, regarding prevention in the areas of food safety and economic enforcement. More specifically, this paper focuses on the generation and optimization of flexible inspection routes. An optimal inspection route seeks to maximize the number of targeted Economic Operators, or the utility gained from the set of Economic Operators that are actually inspected. For that, each Economic Operator is assigned an inspection utility value. The problem was then modelled as a Multi-Depot Periodic Vehicle Routing Problem with Time Windows, and solved using both exact and meta-heuristic methods. The comparison of the meta-heuristic algorithms showed a versatile Hill Climbing implementation in different test cases that explored the effect of the Economic Operators dispersion and density.

Abstract
This work seeks to design and implement a humanoid robotic kick for situations where the robot is moving for the RoboCup simulation 3D robotic soccer league. It employs Reinforcement Learning (RL) techniques, namely the Proximal Policy Optimization (PPO) algorithm to create fast and reliable skills. The kick was divided into 6 cases according to initial conditions and separately trained for each of the cases. A series of kicks, both static and in motion, using two different gaits were developed. The kicks obtained show very high reliability and, when compared to state of the art kicks, displayed a very high time performance improvement. This opens the door to more dynamic games with faster kicks in the RoboCup simulation 3D league.

Abstract
Multiple real-world problems are naturally modeled as cooperative multi-agent systems, ranging from satellite formation to traffic monitoring. These systems require algorithms that can learn successful policies with independent agents that rely solely on local partial-observations of the environment. However, multi-agent environments are more complex, due to their partial-observability and non-stationarity from an agent's perspective, as well as the structural credit assignment problem and the curse of dimensionality, and achieving coordination in such systems remains a complex challenge. To this end, we propose a multi-agent actor-critic algorithm called Asynchronous Advantage Actor Centralized-Critic with Communication (A3C3). A3C3 uses a centralized critic to estimate a value function, decentralized actors to approximate each agent's policy function, and decentralized communication networks for each agent to share relevant information with its team. The critic can incorporate additional information, like the environment's global state, when available, and optimizes the actor networks. The actor networks of an agent's teammates optimize that agent's communication network, such that each agent learns to output information that is relevant to the policies of others. A3C3 supports a dynamic amount of agents, noisy communication mediums, and can be horizontally scaled to shorten its learning phase. We evaluate A3C3 in two partially-observable multi-agent suites where agents benefit from communicating local information to each other. A3C3 outperforms state-of-the-art multi-agent algorithms, independent approaches, and centralized controllers with access to all agents' observations.

Abstract
In general, games pose interesting and complex problems for the implementation of intelligent agents and are a popular domain in the study of artificial intelligence. In fact, games have been at the center of some of the most well-known achievements in artificial intelligence. From classical board games such as chess, checkers, backgammon and Go, to video games such as Dota 2 and StarCraft II, artificial intelligence research has devised computer programs that can play at the level of a human master and even at a human world champion level. Planning and learning, two well-known and successful paradigms of artificial intelligence, have greatly contributed to these achievements. Although representing distinct approaches, planning and learning try to solve similar problems and share some similarities. They can even complement each other. This has led to research on methodologies to combine the strengths of both approaches to derive better solutions. This paper presents a survey of the multiple methodologies that have been proposed to integrate planning and learning in the context of games. In order to provide a richer contextualization, the paper also presents learning and planning techniques commonly used in games, both in terms of their theoretical foundations and applications.

Abstract
There has been an increasing attention to the study of stress. Particularly, college students often experience high levels of stress that are linked to several negative outcomes concerning academic functioning, physical, and mental health. In this paper, we introduce the EuStress Solution, that aims to create an Information System to monitor and assess, continuously and in real-time, the stress levels of the students in order to predict burnout. The Information System will use a measuring instrument based on wearable device and machine learning techniques to collect and process stress-related data from the students without their explicit interaction. In the present study, we focus on heart rate and heart rate variability indices, by comparing baseline and stress condition. We performed different statistical tests in order to develop a complex and intelligent model. Results showed the neural network had the better model fit.

Abstract
Current packaging machine vendors do not provide any automated mechanism for blank feeding and the state of the art is to have a human operator dedicated to feed the blank piles to the packaging machine. This is a tedious, repetitive and tiring task. This also results in problems with unintentional errors, such as using the wrong pile of blanks. An alternative solution is the use of a fixed robotic arm surrounded by a protective cage. However, this solution is restricted to a single packaging machine, a unique type of blank shapes and does not cooperate with humans. TIMAIRIS is a joint effort between IMA S.p.A., Italy, (IMA) and the Universidade de Aveiro, Portugal, (UAVR), promoted by the European Robotics Challenges (EuRoC) project. Together, we propose a system based on a mobile manipulator for flexible, autonomous and collaborative blank feeding of packaging machines on industrial shop floor. The system provides a software architecture that allows a mobile robot to take high level decisions on how the task should be executed, which can depend on variables such as the number of packaging machines to feed and the rate of blank consumption at each one. Through a computer vision system, blanks of different shapes and sizes are correctly identified for adequate manipulation. The manipulation of the piles of blanks is performed using a single arm using compliant modes of operation to increase manipulation safety and robustness. Additionally, it has a safe navigation system that allows the robot to be integrated in an industrial environment where humans are present. Finally, it provides an enhanced multimodal interaction between human and robot that can be adapted to the environment and operator characteristics to make communication intuitive, redundant and safe. © 2020, Springer Nature Switzerland AG.