Abstract
The development of an intelligent wheelchair (IW) platform that may be easily adapted to any commercial wheelchair and aid any person with special mobility needs is the main objective of this project. To be able to achieve this main objective, three distinct control methods were implemented in the IW: manual, shared and automatic. Several manual, shared and automatic control algorithms were developed for this task. This paper presents three of the most significant of those algorithms with emphasis on the shared control method. Experiments were performed, using a realistic simulator, with real users suffering from cerebral palsy in order to validate the approach. These experiments enabled to conclude which were the best shared control methods to implement on the IW. The experiments also revealed the importance of using shared (aided) controls for users with severe disabilities. The patients still felt having complete control over the wheelchair movement when using a shared control at a 50% level and thus this control type was very well accepted and should be used in intelligent wheelchairs since it is able to correct the direction in case of involuntary movements of the user but still gives him a sense of complete control over the IW movement.

Abstract
The inclusion of perceptual information in the operation of a dynamic robot (interacting with its environment) can provide valuable insight about its environment and increase robustness of its behaviour. In this regard, the concept of Associative Skill Memories (ASMs) has provided a great contributions regarding an effective and practical use of sensor data, under a simple and intuitive framework [2, 13]. Inspired by [2], this paper presents a data mining solution to the fall prediction problem in humanoid biped robotic locomotion. Sensor data from a large number of simulations was recorded and four data mining algorithms were applied with the aim of creating a classifier that properly identifies failure conditions. Using Support Vector Machines, on top of sensor data from a large number of simulation trials, it was possible to build an accurate and reliable offline fall predictor, achieving accuracy, sensitivity and specificity values up to 95.6%, 96.3% and 94.5%, respectively.

Abstract
Cardiotocography is a diagnostic exam performed from the 28th week of pregnancy that registers the fetus cardiac frequency and uterine contractions. From this exam results a cardiotocogram whose reading and observation of the patterns contained in it allow an evaluation of the baby's condition and the fetal vitality in the maternal womb. This work aims the creation of a classification model using Learning Algorithms/Data Mining using the tool Rapid Miner. The subject of study was a Data Set with information registered from a total of 2126 cardiotograms, with 23 attributes, properly classified by 3 specialized obstetricians as to the baby status, in three possible states, namely: N = Normal; S = Suspect; P = Pathologic. All models tested showed an overall accuracy greater than 80%. Therefore the usefulness of creating predictive models for the classification of this type of diagnosis is great.

Abstract
This article focuses on evaluating the usability of an intelligent wheelchair (IW) in both real and simulated environments. The wheelchair is controlled at a high-level by a flexible multimodal interface, using voice commands, facial expressions, head movements and joystick as its main inputs. A quasi-experimental design was applied including a deterministic sample with a questionnaire that enabled to apply the System Usability Scale. The subjects were divided in two independent samples: 46 individuals performing the experiment with an IW in a simulated environment (28 using different commands in a sequential way and 18 with the liberty to choose the command); 12 individuals performing the experiment with a real IW. The main conclusion achieved by this study is that the usability of the IW in a real environment is higher than in the simulated environment. However, there were not statistical evidences to affirm that there are differences between the real and simulated wheelchairs in terms of safety and control. Also, most of users considered the multimodal way of driving the wheelchair very practical and satisfactory. Thus, it may be concluded that the multimodal interfaces enables very easy and safe control of the IW both in simulated and real environments.

Abstract
Calendars are used daily by those who have an active social life, whether professional or personal. Although the calendar tends to have a personal connotation, there are many families who use it as the best way to manage their activities and organize family time, sharing by all the same calendar. Reconcile all events in one place shared without privacy commitment of each one, it becomes a difficult task. One option is to use a second calendar that puts the events that should not be shared within the family. However, this solution becomes slow for all events consultation. Thus, the proposed solution is the creation of a calendar, with three types of data display, to ensure that one can use it without compromising the privacy of each element. A web platform with a calendar and tasks to support these three types of exhibition was created and tested by families in their day to day over a period of time. The responses allowed to conclude that a calendar is a great memory aid as more than 80% use it to this end. The results obtained showed the needs that a user has when using a calendar as well as the willingness to use it in the family.

Abstract
The role of referees and the decisions they take have, in many cases, a decisive and direct influence on the course of a game. Unfortunately these decisions are not always correct. In a large amount of times, the mistakes are due to referee's bad positioning on the field. So it is fundamental to develop a model that may determine the correct position for a soccer referee. For this research work, data with all the game participants' positions in the field during the match has been used as its basis. The data included, players, ball and referees positions gathered by an automatic system. Referees with large experience in international competitions were used to obtain correct position knowledge. The achieved results, using a data mining approach, are very positive, both for predicting the assistant referee position, regardless of the team he is accompanying or the match with 77% of success, as to predict the referee position with 82% of success, in controlled game situations. In short, despite the limitations, the models are conclusive and can efficiently estimate the correct positioning of the referee in a soccer match.