Paulo Moura Oliveira

Entropy Diversity in Multi-Objective Particle Swarm Optimization

Multi-objective particle swarm optimization (MOPSO) is a search algorithm based on social behavior. Most of the existing multi-objective particle swarm optimization schemes are based on Pareto optimality and aim to obtain a representative non-dominated Pareto front for a given problem. Several approaches have been proposed to study the convergence and performance of the algorithm, particularly by accessing the final results. In the present paper, a different approach is proposed, by using Shannon entropy to analyze the MOPSO dynamics along the algorithm execution. The results indicate that Shannon entropy can be used as an indicator of diversity and convergence for MOPSO problems.

Teaching particle swarm optimization through an open-loop system identification project

The particle swarm optimization (PSO), one of the most successful natural inspired algorithms, is revisited in the context of a proposal for a new teaching experiment. The problem considered is the open‐loop step identification procedure, which is studied as an optimization problem. The PSO canonical algorithm main issues addressed within the proposed open‐loop step identification experience are: the swarm random initialization methodology, the population size variation, and the inertia weight selection. The teaching experience learning outcomes are stated, simulation results presented, and feedback results from students analyzed.

Many-Objective PSO PID Controller Tuning

Proportional, integral and derivative controller tuning can be a complex problem. There are a significant number of tuning methods for this type of controllers. However, most of these methods are based on a single performance criterion, providing a unique solution representing a certain controller parameters combination. Thus, a broader perspective considering other possible optimal or near optimal solutions regarding alternative or complementary design criteria is not obtained. Tuning PID controllers is addressed in this paper as a many-objective optimization problem. A Multi-Objective Particle Swarm Optimization algorithm is deployed to tune PID controllers considering five design criteria optimized at the same time. Simulation results are presented for a set of four well known plants.

Multi-objective Genetic Manipulator Trajectory Planner

This paper proposes a multi-objective genetic algorithm to optimize a manipulator trajectory. The planner has several objectives namely the minimization of the space and join arm displacements and the energy required in the trajectory, without colliding with any obstacles in the workspace. Simulations results are presented for robots with two and three degrees of freedom, considering the optimization of two and three objectives.

A statistical classifier for assessing the level of stress from the analysis of interaction patterns in a touch screen

This paper describes an approach for assessing the level of stress of users of mobile devices with tactile screens by analysing their touch patterns. Two features are extracted from touches: duration and intensity. These features allow to analyse the intensity curve of each touch. We use decision trees (J48) and support vector machines (SMO) to train a stress detection classifier using additional data collected in previous experiments. This data includes the amount of movement, acceleration on the device, cognitive performance, among others. In previous work we have shown the co-relation between these parameters and stress. Both algorithms show around 80% of correctly classified instances. The decision tree can be used to classify, in real time, the touches of the users, serving as an input to the assessment of the stress level.

Mean Arterial Pressure PID Control Using a PSO-BOIDS Algorithm

A new hybrid between the particle swarm optimization (PSO) and Boids is presented to design PID controllers applied to the mean arterial pressure control problem. While both PSO and Boids have been extensively studied separately, their hybridization potential is far from fully explored. The PSO-Boids algorithm is proposed to perform both system identification and PID controller design. The advantage over a standard particle swarm optimization algorithm is the promotion of the diversity of the search procedure. Preliminary simulation results are presented.

Greenhouse Heat Load Prediction Using a Support Vector Regression Model

Modern greenhouse climate controllers are based on models in order to simulate and predict the greenhouse environment behaviour. These models must be able to describe indoor climate process dynamics, which are a function of both the control actions taken and the outside climate. Moreover, if predictive or feedforward control techniques are to be applied, it is necessary to employ models to describe and predict the weather. From all the climate variables, solar radiation is the one with greater impact in the greenhouse heat load. Hence, making good predictions of this physical quantity is of extreme importance. In this paper, the solar radiation is represented as a time-series and a support vector regression model is used to make long term predictions. Results are compared with the ones achieved by using other type of models, both linear and non-linear.

Control Engineering Learning by Integrating App-Inventor Based Experiments

This paper presents a teaching/learning experiment on the use of MITAppI2 as a friendly tool in Automation courses. The goal was to assess if the up-to-date mobile applications can act as promoters in learning automation topics. The experiment took place in two Portuguese universities. The results achieved point towards a successful use of these tools in university classes.

Particle swarm optimization for gantry control: a teaching experiment

The particle swarm optimization algorithm is proposed as a tool to solve the Posicast input command shaping problem. The design technique is addressed, in the context of a simulation teaching experiment, aiming to illustrate second-order system feedforward control. The selected experiment is the well known suspended load or gantry problem, relevant to the crane control. Preliminary simulation results for a quarter-cycle Posicast shaper, designed with the particle swarm algorithm are presented. Illustrating figures extracted from an animation of a gantry example which validate the Posicast design are presented.

Automation and Control in Greenhouses: State-of-the-Art and Future Trends

This paper presents the state-of-the-art in terms of automation and control for protected cultivation in greenhouses. Aspects such as modeling, instrumentation, energy optimization and applied robotics are considered, aiming at not only to identify latest research topics, but also to foster continuous improvement in key cutting-edge problems.