Abstract
This work presents a new approach to tune the parameters of the discontinuous component of the Sliding Mode Generalized Predictive Controller (SMGPC) subject to constraints. The strategy employs Particle Swarm Optimization (PSO) to minimize a second aggregated cost function. The continuous component is obtained by the standard procedure, by Sequential Quadratic Programming (SQP), thus yielding a dual optimization scheme. Simulations and performance indexes for a non minimum linear model result in a better performance, improving robustness and tracking accuracy.

Abstract
A simulation teaching experiment to control a gantry crane system is proposed. The control is performed both in open-loop and closed-loop. The open-loop control is based on the Posicast feedforward technique and the closed-loop control uses a two-degrees of freedom configuration. Posicast control is used as a pre-filter outside the feedback loop to enhance the set-point tracking response and a PID controller is used in the feedback loop to deal with disturbance rejection. Students are required to use a gantry crane animation, to visualize its movement promoting a better perception and control techniques understanding. The experiment was performed by undergraduate feedback control students which provided learning results through a survey questionnaire.

Abstract
This paper proposes the recently introduced Gravitational Search Algorithm (GSA) to tune a Sliding Mode Controller (SMC) applied on the temperature control of a grains drying system. The problem of maintaining the temperature precisely adjusted inside a silo is relevant to avoid thermal damage and spoilage losses, and thus guarantee the right conditions for storage. The objectives of setpoint tracking and disturbance rejection are incorporated into the minimization of the integral of the time-weighted absolute error. Simulation results are presented and compared with PID and with SMC tuned by Particle SwarmOptimization (PSO) and by earlier proposed tuning equations. © Springer International Publishing Switzerland 2017.

Abstract
The gravitational search algorithm (GSA) is reviewed, by presenting a tutorial analysis of its key issues. As any other metaheuristic, GSA requires the selection of some heuristic parameters. One parameter which is crucial in regulating the exploratory capabilities of this algorithm is the gravitational constant. An analysis regarding this parameter selection is presented and a heuristic rule proposed for this purpose. The GSA performance is compared both with a hybridization with particle swarm optimization (PSO) and standard PSO. Preliminary simulation results are presented considering simple continuous functions optimization examples.

Abstract
The purpose of this paper was to examine the characteristics of waves generated when swimming with and without the use of Aquatrainer (R) snorkels. Eight male swimmers performed two maximal bouts of 25 m breaststroke, first without the use of a snorkel (normal condition) and then using a snorkel (snorkel condition). The body landmarks, centre of the mass velocity, stroke rate, stroke length, stroke index, and Strouhal number (St) were quantified. Fourier analysis was conducted to determine the frequency, amplitude, and phase characteristics of the vertical undulations. We also determined the undulation period, the first and second harmonic wave percentage, and the contribution of these components to the power of each of the wave signals. The first wave harmonics had a frequency of 0.76 Hz (normal condition) and 0.78 Hz (snorkel condition), and the second wave harmonics had a frequency of 1.52 Hz (normal condition) and 1.56 Hz (snorkel condition). Under the normal conditions, the wave amplitude was higher on the vertex (0.72 m) and cervical (0.32 m) than that produced under snorkel conditions (0.71 m and 0.28 m, respectively). The lowest values were found in the hip (0.03 m in normal conditions, and 0.02 m in snorkel conditions) and in the trunk (0.06 m in normal conditions, and 0.04 m in snorkel conditions). It can be concluded that snorkel use seems to lead to slight changes in the biomechanical pattern in swimming velocity, as well as several stroke mechanical variables.

Abstract
New times demand new teaching and learning methodologies. A partial flipped classroom methodology was tested in a modelling and feedback control course for undergraduate biomedical engineering. A semester course was divided in two parts: the first half using a flipped approach and the second one using a classical approach. The experience results are reported, presenting the methodology ups and downs. The use of videos as student's primer study supporting element to prepare classes in advance was explored. Using part of the class time to perform quiz group activities proved to be a major enabler to actively involve students in the learning process. The results achieved in the flipped part strongly confirm a much higher engagement and participation level from students in theoretical classes. Moreover, the flipping approach promotes students to continuously study along the semester.