1) "Control properties of multiagent dynamical systems modelling brain neural networks" by
Professor Dr. Maria Isabel Garcia-Planas, Department of Mathematics at the “Universitat Politecnica de Catalunya, Barcelona, Spain
Abstract: The control of linear dynamical systems is a strategy that the brain uses to control its own intrinsic dynamics. The brain structure can be modelled as a networked system that is expressly interesting system to control because of the role of the underlying architecture, which predisposes some components to particular control motions. The concept of brain cognitive control defined by neuroscientists is related to the mathematical concept of control defined by physicists, mathematicians, and engineers, where the state of a complex system can be adjusted by a particular input. The in-depth study on the controllability and structural controllability character of linear dynamical systems, despite being very difficult, could help to regulate the brain cognitive function. Small advances in the study can favour the study and action against learning difficulties such as dyscalculia, dyslexia or other disturbances like the phenomena of forgetting. Between different aspects in which we can study the controllability we have the notion of structural controllability and exact controllability. In this talk, we revise these concepts for linear dynamical systems and multiagent neural networks.
Brief Biography of the Speaker: Professor Dr. Maria Isabel Garcia-Planas joined the Department of Mathematics at the “Universitat Politecnica de Catalunya” Barcelona, Spain in 1981. Her work had been centered on Linear Algebra, Systems and Control Theory and neural networks. She has authored over a two hundred papers having been cited more than 500 times and serves on the referee on several indexed scientific journals. She has been plenary Speaker in International Conferences.
2) "Using the sequences of linear operators to approximate signals" by
Professor Dr. Octavian Agratini, Faculty of Mathematics and Computer Science, Babeş – Bolyai University
and Tiberiu Popoviciu Institute of Numerical Analysis, Cluj - Napoca, Romania
On the last decades the interest of the study of positive approximation processes have emerged with growing evidence. A special place is occupied by the in-depth study of classical operators. The most eloquent example is Bernstein operator which represents a permanent challenge for the researches in the mentioned field. However, in our talk we focus on two aspects. Firstly, we discuss about a class of operators introduced by G.C. Jain that have long been in a shadowy cone. In recent years many papers have appeared about their properties and many generalizations have been analyzed. In our approach, there is no question of an exhaustive treatment, but only of indicating some new results that prove the importance of this class through the generous possibilities offered by the approximation of signals from different function spaces. Secondly, we are acquainted with the notion of statistical convergence of the approximation processes, revealing its usefulness. The advantage of replacing the ordinary convergence by statistical convergence consists in the fact that the second convergence is efficient in summing divergent sequences which may have unbounded subsequences. Following this direction, new properties of moduli of smoothness and K- functional are obtained, they representing the main tools for measuring the approximation error.
Brief Biography: Visit http://math.ubbcluj.ro/~agratini/
3) "Optimal Control and Observation of Efficient Motions in Electromagnetic Actuators"
Prof. Dr.-Ing. Paolo Mercorelli, Institute of Product and Process Innovation - PPI, Leuphana University of Lueneburg, Lüneburg, Germany
Abstract: Camless internal combustion engines offer improvements over traditional engines in terms of torque performance, reduction of emissions, reduction of pumping losses and fuel economy. Theoretically, electromagnetic valve actuators offer the highest potentials for improving efficiency due to their control flexibility. For real applications, however, the valve actuators developed so far suffer from high power consumption and other control problems. In this sense, in the context of real applications of actuators, one key point is the design of the actuator considering its controllability and observability directly in its initial conception together with the reference trajectory to be tracked by the closed loop controller. In this talk, a design technique aimed at minimizing power consumption is proposed. A constrained optimization problem is formulated also in terms of “soft landing” and its solution is approximated by exploiting local flatness and physical properties of the system. To minimize the number of sensors and to realize an optimal observation of the actuator to be controlled an Extended Kalman Filter is proposed. The performance of the designed trajectory is validated via a real valve actuator.
Brief Biography: Paolo Mercorelli received a Ph.D. degree in Systems Engineering from his Alma Mater, Studiorum University of Bologna, Bologna, Italy, in 1998. In 1997, he was a visiting researcher for one year in the Department of Mechanical and Environmental Engineering, University of California, Santa Barbara, USA. He received an award from the Marie Curie Actions research fellowship program sponsored by The European Commission in year 1998. Thanks to this scholarship, from 1998 to 2001, Paolo Mercorelli was a postdoctoral researcher with ABB (Asea Brown Boveri) Corporate Research, Heidelberg, Germany. During the Heidelberg period he introduced, for the first time in ABB, software structures based on wavelet packets for fault detection and data reconciliation and he developed industrial trading software using these wavelet packets. He was the inventor cited in three patents in which the applicant was ABB Zuerich. Moreover, these wavelet packet structures were implemented and integrated in the Inferential Modelling Platform of the Advanced Control and Simulation Solution Responsible Unit within the ABB industry division. From 2002 to 2005, he was a senior researcher with the Institute of Automation and Informatics, Wernigerode, Germany, where he was the leader of the control group. From 2005 to 2011, Paolo Mercorelli was an Associate Professor of Process Informatics with Ostfalia University of Applied Sciences, Wolfsburg, Germany. In Wolfsburg, he was involved in various projects with the Volkswagen AG Research Center developing different control systems which have been implemented in production series of vehicles as, for instance, the control algorithms for Intelligent Parking Assist System. Since 2012 he has had the position of Full Professor and Head of the Chair of Control and Drive Systems at the Institute of Product and Process Innovation, Leuphana University of Lueneburg, Lueneburg, Germany. Since 2018 he has obtained an international visiting professor fellowship at the Institute of Automatic Control of Lodz University of Technology (Poland) and he is responsible for two courses at the Master’s in “Automation and Robotics“. In Fall 2019 he was visiting Professor at the University of Miskolc (Hungary) giving lectures at the Master’s and PhD programs. His actual interests include Applications with Kalman Filters, Robotics, Wavelets, Geometric Control, Sliding Mode Control and Application with Sliding Mode Control.
4) "Modern Robots are Cyberphysical Systems: Personal Assistants and Online Shopping Warehouses"
Professor Angel P. Del Pobil, Department of Engineering and Computer Science UJI Robotic Intelligence Laboratory Universitat Jaume I Spain
Abstract: An intelligent robot is a perfect paradigm of a cyber-physical system (CPS), since its very nature is based on the seamless integration of computational algorithms and physical components, including embedded sensors, processors and actuators in order to sense and interact with the physical world. In my speech I will address some of the challenges for robots considered as CPS, such as adaptability, autonomy, functionality, resiliency, and safety, with emphasis on the physical interaction with the environment. As test cases I will consider robots as personal assistants, along with robots in online shopping warehouses, as an example towards the 4th industrial revolution, the so-called Industry 4.0, with some lessons learned from the project RoboPicker and our participation in the Amazon Robotics Challenge.
Brief Biography of the Speaker: Angel P. del Pobil is a professor at Jaume I University (Spain), where he was the founding director of the UJI Robotic Intelligence Laboratory. He has been Co-Chair of two Technical Committees of the IEEE Robotics and Automation Society and is a member of the Governing Board of the Intelligent Autonomous Systems (IAS) Society (2012-present) and EURON (European Robotics Research Network of Excellence, 2001-2009). He has over 250 publications, including 13 books, the last three published recently by Springer. Prof. del Pobil was co-organizer of some 50 workshops and tutorials including: five at ICRA (1996, 2000, 2010-2012), ten at IROS (2000, 2004-2013), three at RSS (2008-2010, 2012), ECAI'04, ICAR'05 and ACM/IEEE HRI 2010. He was Program Co-Chair of the 11th International Conference on Industrial and Engineering Applications of Artificial Intelligence, and General Chair of the 13th International Conference on Adaptive Behaviour (SAB 2014) and of five editions of the International Conference on Artificial Intelligence and Soft Computing (2004-2008). He is Associate Editor for ICRA (2009-2015) and IROS (2007-2012) and has served on the program committees of over 130 international conferences, such as IJCAI, ICPR, ICRA, IROS, EUROS, IAS, ICAR, CIRA, etc. The UJI Robotic Intelligence Lab has organized 12 consecutive editions of IURS, the International UJI Robotics School. He has been General Chair for IURS 2005 on Robotics and Neuroscience, IURS 2006 on Humanoid Robots, IURS 2009 on Visuomotor Coordination, IURS 2012 on Perceptual Robotics and the 2015 IEEE-RAS Summer School on Experimental Methodology, Performance Evaluation and Benchmarking in Robotics. He has been involved in robotics research for the last 30 years. Professor del Pobil has been invited speaker of 63 tutorials, plenary talks, and seminars in 14 countries. He serves as associate or guest editor for 12 journals, and as expert for research evaluation at the European Commission, the US National Science Foundation and the Spain Ministry of Science. He has supervised 16 Ph.D. Thesis, including winner and finalists of the Georges Giralt EURON PhD Award and the Robotdalen Scientific Award Honorary Mention. He has been Principal Investigator of 30 research projects. Del Pobil is an active member of the IEEE Robotics and Automation Society and a lifetime member of the Association of the Advancement of Artificial Intelligence (AAAI). Recent projects at the Robotic Intelligence Lab funded by the European Commission include: FP6 GUARDIANS (Group of Unmanned Assistant Robots Deployed In Aggregative Navigation supported by Scent detection), FP7 EYESHOTS (Heterogeneous 3-D Perception Across Visual Fragments), and FP7 GRASP (Emergence of Cognitive Grasping through Emulation, Introspection, and Surprise).
Past Plenary Speeches5) "Improved Governance and Technological Innovations in Cage Culture can Enhance Sustainability of Aquaculture" by
by Professor Dr. Neil Ridler, University of New Brunswick, CANADA.Abstract:
Concern over negative environmental impacts of cage culture has led to boycotts, litigation and even vandalism in certain countries. One country that has avoided social unrest is Norway which is the world's largest producer of Atlantic salmon worth more than US$ 7 billion annually. Its governance of the industry ensures accountability, transparency and community participation, and has encouraged social licence. Its governance of aquaculture can be modified in other countries. A further means of inducing sustainability is technical innovation. This paper examines Integrated multi-trophic aquaculture which replicates the natural environment and can be profitable and socially acceptable.Brief Biography of the Speaker:
Educated at Oxford University and Simon Fraser University and now Emeritus Professor of Economics at the University of New Brunswick, New Brunswick, Canada, Dr. Ridler has published more than 100 refereed articles, book chapters and books. For the last twenty years much of his research has focused on the socio-economics of aquaculture. As consultant and visiting researcher at the Food and Agriculture Organisation of the United Nations, his work has encompassed many countries of Africa, Asia and the Middle East.6) "Evaluation of Coupled Heat, Air and Moisture Transfer Incidence on Building Energy Performance, Indoor Air Quality and Durability of Structure"
by Prof. Rafik Belarbi, University of La Rochelle, FRANCE.
Abstract: Buildings are one of the highest energy consumption sectors, contributing to almost 45% of the world energy consumption and responsible for 36.1 billion tons of CO2 emission. High moisture level can cause metal corrosion, wood decay and structure deterioration. Moreover, moisture transport results in condensation-evaporation processes which accompany energy transfer through building envelopes has a significant influence on indoor air humidity and air-conditioning loads, especially latent cooling load. Consequently, studying the coupled heat and moisture transport behavior of porous building materials have a significant effect on building energy performance and durability of constructions. The present research work aims to understand the influence of geometric parameters of envelope materials on the mechanisms of coupled heat and moisture transfer at different scales in order to predict the long-term behavior of a whole building and to control it. The ultimate goal is to improve the building and its sustainability. The hybrid approach was be implemented in this work. It involves understanding the dominant physical phenomena and their interactions on a microscopic scale. This follows a second phase of modeling based on up scaling technics in order to learn about the macroscopic equivalent behavior of materials. The development of model materials with controlled properties allows to evaluate their intrinsic macroscopic properties and serves as input parameter for the models. Finally, an implementation of the model in the global scale of the building was be undertaken to assess its energy performance and durability and to optimize it. During this stage, a new methodology to predict the overall behavior of buildings, which combines two simulation tools: COMSOL Multiphysics© and TRNSYS. The first software is used for the modeling of heat, air and moisture transfer in multilayer porous walls (HAM model - Heat, Air and Moisture transfer), and the second is used to simulate the hygrothermal behavior of the building (BES model - Building Energy Simulation). The combined software applications dynamically solve the mass and energy conservation equations of the two physical models. Simulation in case studies buildings highlight the effect on heat and moisture transfer on energy performance on building, indoor air quality and durability of building.
Brief Biography of the Speaker: Professor, Dr-Ing. Rafik BELARBI, Received his engineer degree in Building Physics from School of engineers of Poitiers (ESIP) and Master of thermal sciences from University of Poitiers, France in 1993. He obtained his PhD Thesis in civil engineering in La Rochelle University, 1998. In 1999, he joined the LEPTAB research staff laboratory and civil engineering department of La Rochelle University as is, actually, full Professor and Head of Civil engineering and Mechanical Department. His Research field covers wide spectrum and several domains. It cover multi physic and multiscale approaches as: building material for energy and environment applications, urban microclimate modelling and durability aspect; comfort and indoor air quality as well as renewable and energy system. His main expertise is in microstructural, thermal, physical and hydric characterization of porous building material and heat and mass transfers with application in Energy Efficiency in Buildings and Indoor Environment and durability of constructions. Since 1994, he was involved in several National and International projects dealing with Heat and moisture transfer in the building energy conservation. The main projects are: Pascool/Joule and Altener/Sink (passive cooling systems modelling and their impact on the building energy consumption), PDEC/Joule II (Utilisation of Passive Downdraught Evaporative Cooling systems on non-domestics buildings), Joule/Thermie B (Efficient Ventilation Systems for Buildings), Altener/Greencode (Reglementary Frame for Renewable Energy Use in Urban Site Through Vegetation Planting and Strategic Surfacing), Altener/SolVent, (Development of Strategies for Efficient Use of Solar Passive Ventilation in Urban Buildings) and Altener/Cluster (Solar Passive Heating and Cooling), Seventh Framework Programme" Marie Curie (OldMasonryRepair), Erasmus+: programme Capacity Building in Higher Education “Boosting Environmental Protection and Energy Efficient Buildings in Mediterranean Region”. He is author or co-author of more than 140 papers in international journals or international conferences.
7) "Zero Defect Manufacturing Framework for Machining Defects"Prof. Paul Xirouchakis, University of Strathclyde, United Kingdom.
Abstract: A Zero Defect Manufacturing (ZDM) framework for reducing or avoiding workpiece machining defects is presented. At first the ZDM framework is described for the Wire Electrical Discharge Machining (WEDM) process and the following surface defects: surface roughness, recast layer thickness and surface lines. The main ZDM framework components are: (i) an offline process planning system to reduce the average recast layer thickness while maintaining the surface roughness within the desired specification; (ii) an online monitoring and control system to prevent the appearance of surface lines. The components of the offline process planning system are: (i) a Multiple-Input Multiple-Output (MIMO) fuzzy nets system that simultaneously predicts the surface roughness and average recast layer thickness for given input process parameters; (ii) a multi-objective optimization based on the non-dominated sorting genetic algorithm-II (NSGA-II) to select process parameter values for minimizing the average recast layer thickness while respecting the desired surface roughness bound. The components of the online monitoring and control system are: (i) a current signal in the discharge zone monitoring system; (ii) a spark frequency feature extraction component; (iii) an algorithm for real time calculation of the short circuit duration and (iv) a real-time process parameter (pulse-off time) adjustment system. The ZDM framework for WEDM has been implemented and validated in an industry setting. The presentation concludes with a proposal for a generic ZDM architecture for machining defects.
Brief Biography of the Speaker: The focus of Professor Paul Xirouchakis’s research is multidisciplinary at the scientific confluence of design, manufacturing, mechanics, operations research and artificial intelligence:
- Remanufacturing of high value products (molds & dies; oil & gas equipment; diesel engines) using laser metal deposition to reduce lead time, save costs and extend service life
- Laser cutting and repair of CFRP composites based on process development approaches to reduce the heat affected zone and increase productivity
- Zero defect manufacturing using cognitive computation approaches to reduce recast layer thickness and avoid the occurrence of lines & marks in wire Electro-Discharge Machining (EDM) and reduce porosity and micro-cracks in laser metal deposition
- Digital manufacturing for efficient and sustainable milling based on chatter vibrations avoidance to reduce machining time and energy consumption
- Additive manufacturing of aircraft components to realize the Zero Assembly Factory of the Future through component consolidation and topological optimization
- Physics based CAD environment through Medial Axis Transforms to reduce the time to develop a new mechanical product
Professor Paul Xirouchakis was a Professor and Director of the Computer Aided Design and Manufacturing Laboratory at the Swiss Federal Institute of Technology in Lausanne (EPFL) from July 1995 till August 2015. He has 164 international refereed publications and has supervised 21 completed PhD theses. He holds a PhD in Structural Mechanics 1978 from Massachusetts Institute of Technology (MIT), Cambridge, Massachusetts.