Flanders Mechatronics Technology Centre vzw (FMTC) is a research centre, operating since October 2003. It is an initiative of Agoria, the Belgian multisector federation for the technology industry, and fourteen leading mechatronic companies in Flanders. The Flanders Mechatronics Technology Centre wants to be a bridge between academic and industrial know-how in mechatronics and thereby improve the competitiveness of its member companies on the world market. For achieving this, the centre executes industry driven mid-term and long-term research projects. For its operation, FMTC is supported by the Flemisch government via IWT.
The presence of FMTC in the project guarantees a smooth knowledge transfer of the project results to the industry.
The Computational Modeling Lab is one of the research labs at the department of computer science in the faculty of sciences at the Vrije Universiteit Brussel. CoMo is headed by Prof. Dr. Ann Nowé and Prof. Dr. Bernard Manderick.
The research group focuses on the one hand on the modelling of natural phenomena, and on the other hand on developing algorithms for complex problem solving inspired by these natural phenomena.COMO has experience in a wide range of learning techniques such as: Reinforcement Learning, Genetic Algorithms, Neural Networks, Support Vector Machines, Bayesian Networks, Genetic algorithms etc. The research at COMO is organized around two major research tracks:
1) Machine learning techniques for data mining applications
2) Evolution and Learning in multi-agent systems
Regarding this project, the experience of COMO will contribute especially to the understanding of collaboration modelling in dynamical settings.
“ELEC” stands for “Fundamental Electricity and Instrumentation” (in dutch: “Algemene Elektriciteit en Instrumentatie”) and the name corresponds with the educational and research tasks and objectives of the department. The main research activity of this department of the Vrije Universiteit Brussel is the development of new measurement techniques using advanced signal processing methods, embedded in an identification framework.
System Identification constructs mathematical models starting from experimental data. The system identification process consists of the following basic steps and questions: i) Collect experimental data; ii) Select a mathematical model structure that is flexible enough the capture the real-life behavior; iii) Match the model and the experimental data; iv) Quantify the quality of the final model. Although modeling is very closely linked to the specific goal and application field, it turns out that the system identification tools that address these questions are very universal, and this is the basis for system identification as a discipline. The department ELEC is active in the four identification steps, integrating the measurement, modeling and validation aspects in a streamlined, well described methodology. Within this project, ELEC will contribute to the identification of nonlinear systems.
PMA (Production engineering, Machine design and Automation) is, as a division of the Mechanical Engineering Department, a part of the Engineering Faculty of the KU Leuven. PMA carries out research in the areas of production engineering, machine design and mechatronics. PMA pursues a balance between basic or long-term research on the one hand, which is vital in order to maintain an advanced scientific level, and applied or short-term research on the other hand, which is probably the most important mission of an engineering research laboratory.
The activity of the control research group of the division PMA is focused on model based control, and is backed up with research on modelling and experimental identification of complex systems. Their activity on Iterative Learning Control focuses on the one hand on linear techniques for impact noise, periodic noise and on the other hand on vibration reduction in production machines and non-linear techniques for hydraulic vibration simulators and automatic transmissions. Some of these activities run in cooperation with FMTC. This research group recently also started research on Model Predictive Control for machine tools. In this project PMA will develop Iterative Learning Control algorithms for nonlinear systems and fast Model Predictive Control algorithms, dedicated for the control of mechatronic applications.
The division MeBioS (Mechatronics, Biostatistics and Sensors) of the Department of Biosystems of the Katholieke Universiteit Leuven in Belgium investigates the interaction between biological systems and physical processes. The emphasis is on the measurement of properties of biomaterials and process variables, the analysis of the measured signals, and finally process design and control. The basic research in MeBioS is in the area of bionanotechnology, including disciplines such as micro- and nanofluidics, (bio)sensor technology, mechanobiology and biophysical properties of cellular systems. Multiscale modelling, in which biological systems are described at multiple spatial and organisational scales plays an important role. Applications are situated at multiple spatial scales, from agricultural machinery automation over medical diagnostics and predictive microbiology, to postharvest storage of fruit and vegetables and food processing.
MeBioS is one of the leading groups in the area of agricultural machinery automation. In the field of the development of intelligent algorithms to monitor agro-food processes, MeBioS participated and coordinated several European projects. In this project, MeBioS will evaluate local and global learning control methods and extend the Moving Horizon Estination and Stochastic Model Predictive Control frameworks.
SCD’s major research objective is to design and build advanced methods for crucial problems in information processing. It builds on the enormous growth in computer power, communication bandwidth and available data and the various needs in society for effective use of these opportunities. SCD has a number of mutually reinforcing groups that build expertise from the more conceptual and mathematical level downstream until the validation with our partners in practical applications in the field. A major driving force of the research is the diverse use of generic information processing methods based on applied mathematics like linear and multilinear algebra, statistics, discrete mathematics, differential geometry, and optimization. The SCD group has built up considerable expertise in the area of Model Predictive Control, which will be used in this project.
The SYSTeMS group (UGent) has a long tradition in the development and the application of self-adaptive control strategies. These are methods in which the controller learns in real-time the variations which occur in the process characteristics. Experience is also available in the field of modelling nonlinear dynamical systems by means of neural networks and the subsequent application of these Neural Networks as prediction models in nonlinear predictive control. The SYSTeMS research group has also a strong background in the field of Machine Learning, which spans the spectrum of models for learning, including those based on statistics, mathematics, neural structures, information theory, and evolutionary search algorithms. In this project, the SYSTeMS group will be involved both in the activities on Model Predictive Control and on Machine Learning