Mechanical Engineering

With the evolution of industrial automation, there is a need to automate even the simplest processes, to keep industrial companies competitive. The objective of this Master's Thesis is to automate the positioning system of an internal washing process of tanker trucks, with supervision from the Web. In this project work, a robot positioning system was developed with position control through the image of video cameras. The vision algorithm was programmed in Matlab software with the help of computer vision toolboxes and communication with a stepper motor controller developed in Arduino architecture. The development was done within the concept of the industry 4.0, that is, the whole system is totally automatic except for the introduction of the number of holes by the operator. Controllers were tested with communication based on the Ethernet protocol, namely the PC with the Matlab client to communicate with an Apache server so that it is possible to establish the connection between the system and the supervision on the web. The entire system can be supervised through a smartphone or tablet from the installed browser. A prototype was also built to scale, selected to validate the vision algorithm and the simulations performed in Matlab. The prototype consists of a structure in V-slot profiles, as well as the controller and USB camera that although having a lower quality of image, was enough to validate the implemented vision algorithm.

With the new industrial revolution (industry 4.0) under way, there is a need to automate/digitalize even the simplest processes to keep companies competitive, efficient and secure. This work proposes an automatic and robotic positioning system with position control through the image of video cameras. The objective is to propose this type of automatic system to companies aiming to improve robotic systems operations that are often controlled manually and with associated risks. The developed system was tested in simulation (using Matlab  ) and in an experimental scale prototype.

This chapter is concerned with the application of fuzzy neural networks to fault detection and isolation systems. Thus, for readers not familiar with the subject, the background knowledge associated with artificial neural networks and the potential fields of application of this technology is presented in the introduction section. Furthermore, aiming to demonstrate that such a technology is mature enough to be applied in the solution of several kinds of industrial problems, a wide range of industrial applications of classical feedforward artificial neural networks are also reported in section 10.2, as well as applications of different types of fuzzy neural networks. Section 10.3 is concerned with the development of FDI approaches based on fuzzy neural networks and a specific fault isolation system based on a hierarchical structure of several fuzzy neural networks is described in detail. The robustness and performance of such a fault isolation system has been assessed in section 10.4 by using a test bed consisting of a pneumatic servomotor actuated industrial control valve. Different kinds of faults have been considered, which has been assumed to occur in an abrupt or incipient manner, or by affecting the measurement variables in the process under supervision in an abrupt way or, instead, by affecting the process behaviour slowly (incipient faults). The results presented in section 10.4 have shown that, under abrupt faults, the HSFNN provides very accurate results and is characterized by a good generalization capability as a fault isolation system. Under incipient or multiple simultaneous faulty scenarios, the performance of the proposed methodology depends on the fault development speed and/or on the system nonlinearities.

This paper provides the description and presentation of the actuator benchmark used in fault diagnosis studies within the DAMADICS European Research Training Network presented in the papers of this special issue. The benchmark system is openly available, is FDI method-independent and based on an in-depth study of the phenomena that can lead to likely faults in valve actuator systems. The study uses a detailed consideration of the physical and electro-mechanical properties of an intelligent industrial actuator and includes typical engineering requirements of an actuator valve operating under challenging process conditions, e.g. provision of a set of performance indices for evaluating the results. The industrial application is focused on the sugar factory Cukrownia Lublin SA, Poland.

Recent approaches to fault detection and isolation for dynamic systems using methods of integrating quantitative and qualitative model information, based upon soft computing (SC) methods are surveyed. In this study, the use of SC methods is considered an important extension to the quantitative model-based approach for residual generation in FDI. When quantitative models are not readily available, a correctly trained neural network (NN) can be used as a non-linear dynamic model of the system. However, the neural network does not easily provide insight into model behaviour; the model is explicit rather than implicit in form. This main difficulty can be overcome using qualitative modelling or rule-based inference methods. For example, fuzzy logic can be used together with state space models or neural networks to enhance FDI diagnostic reasoning capabilities. The paper discusses the properties of several methods of combining quantitative and qualitative system information and their practical value for fault diagnosis of real process systems.

Collaborative networks in the logistics sector have proven to be a solution that both meets environmental footprint reduction goals and addresses the impact of rising fuel prices on logistics companies, especially for small- and medium-sized enterprises. Despite these benefits, these collaborative networks have not received the desired amount of participation due to reputational risk. This paper develops a framework for assessing and managing reputational risk to encourage logistics companies’ participation in collaborative networks. To this end, customer satisfaction factors were correlated with logistics operations, and this correlation was then modeled using the Bowtie method, fault trees, event trees, reliability theory, and the Monte Carlo model. The results show that it is possible to implement a structured model that can be easily put into practice. Using an illustrative case study, it is also possible to prioritize three companies according to their reputational risk as asse...

Human arm movement control theories are reviewed in the current paper. The motor planning problem stated as a generation of a time plan for the execution of a movement task, is a major concern of the current paper. It will be suggested that computational models of motor control have a strong potential for the use in the area of human motor rehabilitation. Their use can be discriminated in three main areas of application: the generation of correct trajectories to be demonstrated to human subjects during physiotherapy; the assessment of motion disorders and movement quality; and the devising of challenging interaction exercises to promote recovery.

The present paper aims to demonstrate why usually when theoretical mathematical models are used to compute the frequencies corresponding to a faulty rolling bearing a deviation is obtained between the computed values and the real frequencies emitted by such a device. A lab rolling bearing test ring has been developed to perform the current studies.

In this paper a new method for a global selflocalization of mobile robots based on a PCA positioning sensor to operate in unstructured environments is proposed and experimentally validated. Unlike many existing systems that use RGB signals to capture information from the environment, in this work a 2D depth sensor is used, allowing the self-localization to be performed under different illumination conditions. However, depth sensors provide measurements corrupted with missing data, due to limitations on the support physic principles exploited (e.g. light that illuminates surfaces with diffuse reflection or wave fading), which severely degrades the performance of the estimation techniques and limits its use. The main goal of this paper is to present a self-localization system for mobile robots based on a PCA positioning sensor that relies on corrupted depth measurements and the corresponding experimental validation. The proposed method allows for the implementation of a global self-localization system for indoor environments with higher accuracy, that provide a Gaussian estimated position error and work in any illumination conditions.

The main goal of this paper is to present a visual odometry system, for localization of mobile robot in indoor unstructured environments, using only ceiling depth images, captured by a Kinect sensor. The use of odometric sensors is a common practice for localization of mobile robots. The method proposed in this work exploits information from an independent source of depth data and thus allows to complement or substitute the use of classic odometric sensors, like wheels encoders, with well known limitations. The experimental validation of the proposed solution shows that the method is able to accurately compute the attitude and linear velocities that allow a more precise mobile robot localization, even in presence of corrupted data from the sensor. Furthermore, the method works in an extended range of lighting conditions, without the need to perform any specific feature extraction.

This paper presents the experimental validation of a new method for mobile robot global self-localization in unstructured environments, i.e. that does not need any beacons or other artifacts structuring the environment. The method resorts to a PCA-based positioning sensor, filtered in a Bayesian probabilistic grid and combined with linear Kalman filters to estimate the global pose of mobile robots. In the implemented system, the information of the environment is captured only with onboard sensors installed in a differential drive robot: encoders, compass, and 2D depth sensor pointed to the ceiling. The use of PCA in a Bayesian probabilistic grid allows to fuse the highly compressed PCA database information, obtained with the low computational effort, in an environment where repetitive scenarios can occur. To avoid the negative impact in the localization estimate caused by the corrupted data existing in the 2D depth sensor, an extension to the classic PCA algorithm is suggested. Thus, the proposed method allows the self-localization of mobile robots in indoor environments with bounded accuracy and working in a wide range of illumination conditions.

In this paper a new PCA-based positioning sensor and localization system for mobile robots to operate in unstructured environments (e.g. industry, services, domestic.. .) is proposed and experimentally validated. The inexpensive positioning system resorts to principal component analysis (PCA) of images acquired by a video camera installed onboard, looking upwards to the ceiling. This solution has the advantage of avoiding the need of selecting and extracting features. The principal components of the acquired images are compared with previously registered images, stored in a reduced onboard image database, and the position measured is fused with odometry data. The optimal estimates of position and slippage are provided by Kalman filters, with global stable error dynamics. The experimental validation reported in this work focuses on the results of a set of experiments carried out in a real environment, where the robot travels along a lawnmower trajectory. A small position error estimate with bounded co-variance was always observed, for arbitrarily long experiments, and slippage was estimated accurately in real time.

A bstract In the field of appearance-based robot localization, the mainstream approach uses a quantized representation of local image features. An alternative strategy is the exploitation of raw feature descriptors, thus avoiding approximations due to quantization. In this work, the quantized and non-quantized representations are compared with respect to their discriminativity, in the context of the robot global localization problem. Having demonstrated the advantages of the non-quantized representation, the paper proposes mechanisms to reduce the computational burden this approach would carry, when applied in its simplest form. This reduction is achieved through a hierarchical strategy which gradually discards candidate locations and by exploring two simplifying assumptions about the training data. The potential of the nonquantized representation is exploited by resorting to the entropy-discriminativity relation. The idea behind this approach is that the non-quantized representation facilitates the assessment of the distinctiveness of features, through the entropy measure. Building on this finding, the robustness of the localization system is enhanced by modulating the importance of features according to the entropy measure. Experimental results support the effectiveness of this approach, as well as the validity of the proposed computation reduction methods.

It is important to understand and forecast a typical or a particularly household daily consumption in order to design and size suitable renewable energy systems and energy storage. In this research for Short Term Load Forecasting (STLF) it has been used Artificial Neural Networks (ANN) and, despite the consumption unpredictability, it has been shown the possibility to forecast the electricity consumption of a household with certainty. The ANNs are recognized to be a potential methodology for modeling hourly and daily energy consumption and load forecasting. Input variables such as apartment area, numbers of occupants, electrical appliance consumption and Boolean inputs as hourly meter system were considered. Furthermore, the investigation carried out aims to define an ANN architecture and a training algorithm in order to achieve a robust model to be used in forecasting energy consumption in a typical household. It was observed that a feed-forward ANN and the Levenberg-Marquardt algorithm provided a good performance. For this research it was used a database with consumption records, logged in 93 real households, in Lisbon, Portugal, between February 2000 and July 2001, including both weekdays and weekend. The results show that the ANN approach provides a reliable model for forecasting household electric energy consumption and load profile.

In this paper a new PCA-based positioning sensor and localization system for mobile robots to operate in unstructured environments (e.g. industry, services, domestic,. . .) is proposed and experimentally validated. The positioning system resorts to principal component analysis (PCA) of images acquired by a video camera installed onboard, looking upwards to the ceiling. This solution has the advantage that the need of selecting and extracting features is avoided. The principal components of the acquired images are compared with previously registered images, present in a reduced onboard image database and the position measured is fused with odometry data. The optimal estimates of position and slippage are provided by a Kalman filter, with global stable error dynamics. The experimental validation reported in this work focus on the results of a set of exhaustive experiments carried out in a real environment, where the robot travels along straight lines. A small position error estimate was always observed, for arbitrarily long experiments, and slippage was estimated accurately in real time.

The automation of a building requires a huge number of cables that highly increase the installation costs. At the physical level, wireless plays an important role towards flexibility and selfconfiguration in building automation. With the use of wireless technologies it is expected to reduce total costs of ownership and maintenance in new building automation systems. Wireless systems could easy provide old buildings with relatively low cost, automation functionalities. The paper presents the state-of-art in this knowledge domain discussing the advantages and handicaps of wireless approaches in building automation.

Nowadays, companies in global markets have to achieve high levels of performance and competitiveness to stay “alive”.Within this assumption, the building maintenance cannot be done in a casual and improvised way due to the costs related. Starting with some discussion about lean management and building maintenance, this paper introduces a model to support the Lean Building Maintenance (LBM) approach. Finally based on a real case study from a Portuguese company, the benefits, challenges and difficulties are presented and discussed.

In recent years, the globalization of markets combined with the Industry 4.0 revolution has brought increasing competition among enterprises. Nowadays, organizational success is measured by leadership in the market share, revenue growth, level of competitiveness supported by customer satisfaction, and innovation capacity. Therefore, companies need to develop new products and/or services faster and better than their competitors to stay "alive." In order to respond quickly to market needs, enterprises should operate more efficiently reducing and/or eliminating activities that make up the actual processes that do not add value but also they need to develop a "healthy" cultural environment within the organization that allows the implementation of new ways of work. In this context, the innovation capacity assumes a decisive importance in the level of competitiveness of an organization, and the entrepreneur plays an important role in launching new business. Based on this perspective, the innovation processes within an organization are obtained through entrepreneurial initiatives that have their origin in the existence of an organizational culture that promotes an entrepreneurial behavior.

This chapter aims to (i) understand the functioning and importance of incubators in supporting the creation of new business and local economic development, as they help locally engage young entrepreneurs and thus retain expertise, for example, in regions farthest from large urban centers, (ii) realize the importance of incubators and science parks in the development of start-ups, (iii) understand the role of education in promoting entrepreneurship, and (iv) understand the competitive environment as it becomes more complex and dynamic, as there are opportunities to serve customers who are dissatisfied; thus the threats that large companies experience offer small business opportunities.