Keynote Speakers

Prof. Igor Škrjanc

University of Ljubljana, Slovenia

Title: Evolving Systems in Mobile Robotics

Abstract: In mobile robotics, localization plays the key role in most applications. In most cases advanced sensor systems are required to estimate the robot pose, mostly due to the fact that there is no single and effective sensor that would directly measure the robot pose in indoor environment. A very popular sensor for this purpose is a laser range finder (LRF), which has good coverage, dense information, high accuracy and a high sampling rate. It can be used for localization purposes, map building or SLAM. Using a LRF the robot pose can be estimated by comparing a locally sensed map given by a cloud of reflection points and a known map of the environment. This comparison is usually made by comparing simple geometric features that are extracted from the LRF reflection points. In this talk a new approach, which is based on evolving paradigm, called evolving principal component clustering is applied to the laser data stream. The main reason for implementing the evolving algorithm is the unknown number of features and an on-line nature of the problem. The idea is to fit the model using recursive principal component analysis (PCA), which is very easy to implement and computationally effective. In the proposed methodology only a recursive covariance matrix needs to be evaluated. From the covariance matrix the model parameters that optimally fit the measured data are defined. This means that the method recursively estimates the linear model parameters for each cluster of data. It enables good performance, robust operation, low computational complexity and simple implementation on embedded computers. The proposed approach is demonstrated on real and simulated examples from laser-range-finder data measurements. The performance, complexity and robustness are validated through the comparison with the popular classical localization algorithms.

Short Bio: Prof. Igor Škrjanc received B.S., M.S. and Ph.D. degrees in electrical engineering, in 1988, 1991 and 1996, respectively, at the Faculty of Electrical and Computer Engineering, University of Ljubljana, Slovenia. He is currently a Full Professor with the same faculty and Head of Laboratory for Autonomous and Mobile Systems. He is lecturing  the basic control theory at graduate and advanced intelligent control at postgraduate study. His main research areas are adaptive, predictive, neuro-fuzzy and fuzzy adaptive control systems. His current research interests include also the field of autonomous mobile systems in sense of localization, direct visual control and trajectory tracking control. He has published 81 papers with SCI factor and 27 other journal papers. He is co-author and author of 11 chapters in international books and co-author of scientific monograph with the title Predictive approaches to control of complex systems published by Springer. He is also author and co-author of 226 conference contributions, 31 lectures at foreign universities. He is also mentor at 5 PhD thesis, 3 Msc thesis and 34 diploma works. And co-mentor of 2 PhD thesis and 1 MSc thesis. He is author of 6 university books, 24 international and domestic projects and 4 patents. In 1988 he received  the award for the best diploma work in the field of Automation, Bedjanič award, in 2007 the award of Faculty of Electrical Engineering, University of Ljubljana, Vodovnik award, for outstanding research results in the field of intelligent control, in 2012 the 1st place at the competition organized by IEEE Computational Society, Learning from the data in the frame of IEEE World Congress on Computational Intelligence 2012, Brisbane, Avstralija: Solving the sales prediction problem with fuzzy evolving methods, and in 2013 the best paper award at IEEE International Conference on Cybernetics in Lausanne, Switzerland. In 2008 he received the most important Slovenian research award for his work in the area of computational intelligence in control – Zois award. In year 2009 he received a Humboldt research award for long term stay and research at University of Siegen. He is also a member of IEEE CIS Standards Committee, IFAC TC 3.2 Computational Intellignce in Control Committee and Slovenian Modelling and Simulation society and Automation Society of Slovenia. He also serves as an Associated Editor for IEEE Transaction on Neural Networks and Learning System, IEEE Transaction on Fuzzy Systems, the Evolving Systems journal and International journal of artificial intelligence.

Dr. Edwin Lughofer

Johannes Kepler University Linz, Austria

Title: Recent Advances in Evolving Fuzzy Systems and in Their Application to On-line Quality Control and Predictive Maintenance
+ Discussion on Open Challenges and Future Research Directions

Abstract: The panel talk will provide a round overview picture of the developments in the field of evolving fuzzy systems (EFS) achieved during the last decade since their first time appearance at the beginning of this century.

Opposed to conventional fuzzy systems, EFS can be learnt from data (streams) on the fly during (fast) on-line processes in an incremental and mostly single-pass manner. They enjoy a flexible model structure that is able to automatically self-evolve and self-adapt to changes in the process, as e.g. caused by system drifts, new operation modes or dynamic environmental conditions. As being equipped with specific structural components in form of linguistically readable rules, they are able to offer some sort of interpretability and thus to gain insights for operators and experts into system behaviors and dependencies. This may be a great support for any form of supervision, reason finding and annotation processes as well as may motivate the users to interact with the system on a higher level.

Therefore, they stand for a very important topic in the field of Soft Computing to address modeling problems in nowadays real-world applications with quickly increasing complexity, more and more implying a shift from batch off-line model design phases (as conducted since the 80ties) to permanent on-line (active) model teaching and adaptation cycles toward enriched human-machine interactions. Furthermore, they are widely used in the context of on-line data stream mining and incremental extraction of models and knowledge from huge data bases and Big Data and therefore a fruitful contribution to the research fields of Evolving Adaptive Intelligent Systems and incremental, on-line Machine Learning.

A particular emphasis in this talk will be placed on the application of EFS in the field of quality control and predictive maintenance, which are key applications within the context of “Industry 4.0” and “Factories of the Future (FoF)” Objectives within the EU framework program Horizon 2020, to discuss recent advances and to point out still open problems. Thereby, the author will address three variants for on-line monitoring and supervision of industrial (production) systems:

The talk will be concluded with a substantial discussion about challenges, emerging trends and future research directions regarding important and necessary methodological improvements in evolving adaptive intelligent systems, especially for addressing open problems in on-line quality control systems.

Short Bio: Dr. Edwin Lughofer received his PhD. degree from the Department of Knowledge-Based Mathematical Systems, Johannes Kepler University Linz, where he is now employed as key researcher. During the past 10-12 years, he has participated in several research projects on European and national level. In this period, he has published around 150 journal and conference papers in the fields of evolving fuzzy systems, machine learning and vision, data stream mining, active learning, classification and clustering, fault detection and diagnosis, condition monitoring as well as human-machine interaction, including a monograph on ’Evolving Fuzzy Systems’ (Springer, Heidelberg) and an edited book on ’Learning in Non-stationary Environments’ (Springer, New York). He is associate editor of the international journals IEEE Transactions on Fuzzy Systems (IEEE press), Evolving Systems (Springer), Information Fusion (Elsevier), Complex and Intelligent Systems (Springer) and Soft Computing (Springer), the general chair of the IEEE Conference on Evolving and Adaptive Intelligent Systems 2014 and Area chair of the FUZZ-IEEE 2015 conference in Istanbul. He serves as program committee member of several international conferences and of the IEEE students research grant, acts as a peer-reviewer for 20+ international journals and co-organized several special sessions and issues in international conference and journals with main focus on data-stream mining and on-line, evolving learning techniques. In 2006, he received the best paper award at the International Symposium on Evolving Fuzzy Systems, and in 2013 the best paper award at the IFAC conference in Manufacturing Modeling, Management and Control Conference (800 participants).

He is currently key researcher in the national K-Project imPACts and the associated PAC network, see

Dr. Andre Lemos

Federal University of Minas Gerais (UFMG), Brazil

Title: Adaptive Fault Detection and Diagnosis Using Evolving Intelligent Systems

Abstract: Currently, one of the main challenges of the industrial automation is the automation of the so-called Abnormal Events Management (AEM), i.e., the automation under faulty operating conditions, where much of the work remains manual. The first important step towards the automation of AEM is the Fault Detection and Diagnosis (FDD). Recently, FDD methods based on historical process data have received great emphasis, since the acquisition of data through sensors is widely common in modern automation systems. In several cases, although there are a lot of data representing the normal operation condition of a process, the same is not true for abnormal conditions. It is likely that a given process abnormality has never even been seen in practice. However, most of the conventional history based FDD methods require prior data and information from all operation modes during training, i.e., from normal and all abnormal conditions of the monitored process. In this talk, we present an overview of existing evolving intelligent systems (EIS) applied to FDD. We focus on methods that are able to handle this prior data scarcity issue. We present existing solutions for FDD, which can start monitoring a process based only on data from normal operation condition and, as new abnormal conditions are detected, they are able to extract information about these new operation modes to identify them in the next occurrence. We start by presenting a literature review followed by a detailed description of a solution proposed by our research group. Finally, we identify the current challenges and present open research lines for further developments.

Short Bio: Andre Paim Lemos received a B.S. degree in Computer Science from Universidade Federal de Minas Gerais (UFMG) in 2003, M.S. and Ph.D. degrees in Electrical Engineering from the same university in 2007 and 2011, respectively. Since 2011, he has been working as an Assistant Professor in the Electronic Engineering Department at UFMG. His current research interests are industrial applications of machine learning and computational intelligence, including fault detection and diagnosis and system modeling and identification. Since 2007, he has been working in several R&D projects funded by major Brazilian Companies, such as Petrobras, Gerdau, CEMIG, and CHESF. In 2010 he received the Meritorious Paper Award at the 29th International Conference of the North American Fuzzy Information Processing Society (NAFIPS). In 2011 he received the Best Paper Award at the 30th International Conference of NAFIPS, and in 2012 he received the Best Special Session Paper Award (Adaptive and Dynamic Modeling in Non-Stationary Environments) at the 11th International Conference on Machine Learning and Applications (ICMLA).

Dr. Daniel Leite

Federal University of Lavras (UFLA), Brazil

Title: Evolving Granular Modeling from Uncertain Data Streams

Abstract: In recent years, there has been increasing interest in computational modeling approaches to deal with real-world data streams. Methods and algorithms have been proposed to uncover meaningful knowledge from very large data sets. Evolving granular modeling comprises an array of online learning methods inspired by the way in which humans deal with complexity. These systems explore the information on dynamic environments and derive from it models that can be linguistically understood. A particular challenge faced in stream modeling concerns how to handle uncertainty. Uncertain (granular) data arise from imprecise perception or description of the value of a variable, expert judgement, readings from unreliable sensors, and summaries of numeric data over time periods. Of particular concern to this speech are numerical, interval, and fuzzy types of granular data; and interval, fuzzy, and neurofuzzy modeling frameworks. We focus on the role of interval, fuzzy, and neurofuzzy computing in processing uncertain data and providing a set of rules and accurate and granular solutions.

Short Bio: Daniel Leite received his PhD degree in Electrical Engineering from the University of Campinas, UNICAMP, Brazil. He was a lecturer and a postdoctoral fellow at the Department of Electronics Engineering, Federal University of Minas Gerais. Currently, he is an assistant professor at the Federal University of Lavras, UFLA, Brazil. Daniel was the recipient of the Best PhD Thesis Award in Artificial and Computational Intelligence from the Brazilian Computer Society in 2014, and the recipient of the Best Thesis Award from the North American Fuzzy Information Processing Society, NAFIPS, in 2015. His current research interests are adaptive and evolving systems, intelligent control, nonlinear system modeling, machine learning, and granular computing. Daniel has contributed as a reviewer of several journals and conferences in his research fields. He is a member of the IEEE, of the European Society for Fuzzy Logic and Technology (EUSFLAT), and of the International Federation of Automatic Control.