Abstract
The work described in this paper is part of the development of a framework to support the joint execution of cooperative missions by a group of robotic vehicles, in a simulated, augmented or real environment. Such framework brings forward the need for formal languages in which to specify the group of vehicles that compose a team; the scenario in which they will operate; and the definition of the mission to be performed (surveillance, pollution detection, search and rescue operations, and so on). This paper presents the Mission Description Language (MDL), a XML dialect that describes the specifics of a mission to be performed by a team of autonomous vehicles. The MDL definition decomposes a mission into a number of possibly dependent phases and uses high-level concepts that facilitate the specification of a mission by both experts and non-experts. The dialect was defined using XML Schema, with all the required information being integrated in the definition. An interface was developed and incorporated into the framework, as to allow for the creation and edition of XML files following the defined schema. Once the information is correctly specified, it can be used in the framework, thus facilitating the process of mission specification and deployment. A survey answered by both practitioners and researchers shows that the degree of satisfaction with MDL is elevated (the overall evaluation of MDL achieved a 3.97 score (out of 5), with 82.05% of the answers being equal to or above 4); also, the usability of the interface was evaluated, having achieved a score of 82.9 in the SUS scale. These results imply that MDL is flexible enough to represent several types of missions, through a user-friendly interface.

Abstract
One of the research topics on multi-agent systems focuses on the development of mechanisms such as plans to empower a team of agents to cooperate in order to perform complex tasks. In many cases, the definition of these plans are based on a specific and rather complex grammar and stored in structured text files. In the context of the 2D simulated Robotic Soccer domain, a set-play language was proposed to coordinate the execution of teammates' behaviors to improve a team's overall performance. The process of manually writing set-play definitions is hazardous and can benefit from the use of a graphical tool to reach new users and allow typical users to become more productive. This work presents such a tool for which several experiments were run to measure its usability with forty two users by having them perform a set of tasks for which their execution time, number errors and satisfaction were recorded. The tool reduced the previous average time required to completely define a set-play by 90% and enabled even non-expert users to use it Moreover, users were on average satisfied with SPlanner having ranked it with a score of 77 (out of 100) using a System Usability Scale questionnaire.

Abstract
Soccer is a competitive and collective sport in which teammates try to combine the execution of basic actions (cooperative behavior) to lead their team to more advantageous situations. The ability to recognize, extract and reproduce such behaviors can prove useful to improve the performance of a team in future matches. This work describes a methodology for achieving just that makes use of a plan definition language to abstract the representation of relevant behaviors in order to promote their reuse. Experiments were conducted based on a set of game log files generated by the Soccer Server simulator which supports the RoboCup 2D simulated robotic soccer league. The effectiveness of the proposed approach was verified by focusing primarily on the analysis of behaviors which started from set-pieces and led to the scoring of goals while the ball possession was kept. One of the results obtained showed that a significant part of the total goals scored was based on this type of behaviors, demonstrating the potential of conducting this analysis. Other results allowed us to assess the complexity of these behaviors and infer meaningful guidelines to consider when defining plans from scratch. Some possible extensions to this work include assessing which plans have the ability to maximize the creation of goal opportunities by countering the opponent's team strategy and how the effectiveness of plans can be improved using optimization techniques.

Abstract
Collaborative filtering techniques have been used for some years, almost exclusively in Internet environments, helping users find items they are expected to like by using the user's past purchases to provide such recommendations. With this concept in mind, this research uses a collaborative filtering technique to automatically improve the performance of a simulated soccer team. Many studies have attempted to address this problem over the last years but none has shown meaningful improvements in the performance of the soccer team. Using a collaborative filtering technique based on nearest neighbors and the FC Portugal team as the test subject (in the context of the RoboCup 2D Simulation League), several simulations were run for matches against different teams with much better, better and worse performance than FC Portugal. The strategy used by FC Portugal was to combine 8 set-plays and 2 team formations. The simulation results revealed an improvement in performance between 32% and 384%. In the future, there are plans to expand this approach to other contexts, such as the 3D Simulation League.

Abstract
The work described in this paper is part of the development of a framework to support the joint execution of cooperative missions by a group of vehicles, in a simulated, augmented, or real environment. Such a framework brings forward the need for formal languages in which to specify the vehicles that compose a team, the scenario in which they will operate, and the mission to be performed. This paper introduces the Scenario Description Language (SDL) and the Team Description Language (TDL), two Extensible Markup Language based dialects that compose the static components necessary for representing scenario and mission knowledge. SDL provides a specification of physical scenario and global operational constraints, while TDL defines the team of vehicles, as well as team-specific operational restrictions. The dialects were defined using Extensible Markup Language schemas, with all required information being integrated in the definitions. An interface was developed and incorporated into the framework, allowing for the creation and edition of SDL and TDL files. Once the information is specified, it can be used in the framework, thus facilitating environment and team specification and deployment. A survey answered by practitioners and researchers shows that the satisfaction with SDL+TDL is elevated (the overall evaluation of SDL+TDL achieved a score of 4 out of 5, with 81%/78.6% of the answers 4); in addition, the usability of the interface was evaluated, achieving a score of 86.7 in the System Usability Scale survey. These results imply that SDL+TDL is flexible enough to represent scenarios and teams, through a user-friendly interface.

Abstract
It is recognised that the imbalanced data problem is aggravated by other difficulty factors, such as class overlap. Over the years, several research works have focused on this problematic, although presenting two major hitches: the limitation of test domains and the lack of a formulation of the overlap degree, which makes results hard to generalise. This work studies the performance degradation of classifiers with distinct learning biases in overlap and imbalanced contexts, focusing on the characteristics of the test domains (shape, dimensionality and imbalance ratio) and on to what extent our proposed overlapping measure (degOver) is aligned with the performance results observed. Our results show that MLP and CART classifiers are the most robust to high levels of class overlap, even for complex domains, and that KNN and linear SVM are the most aligned with degOver. Furthermore, we found that the dimensionality of data also plays an important role in explaining performance results. © Springer Nature Switzerland AG 2018.