Communications

The communication presentations (keynote talks, facilitated break-out sessions, posters) will be continuously uploaded in May and June.

Maxime Petit
Industrial Robot Adaptation for Manipulating New Objects in Factory

Maxime Petit received the M.Sc. degree in computer science from the University of Paris-Sud, and an engineering degree in biosciences from the National Institute of Applied Sciences Lyon, France, both in 2010. In 2014, he received a Ph.D. in Neurosciences from the National Institute of Science and Medical Research, in the Stem-Cell and Brain Research Institute in Lyon, within the Robot Cognition Laboratory. He was a Research Associate at the Personal Robotics Lab, Imperial College London (2014-2016) and is now a Postdoctoral researcher at the Ecole Centrale de Lyon, Laboratory of Computer Science for Image and Information Systems (LIRIS).
His research interests include developmental robotics, memory and learning in both industrial and service robotics.

Weblink
https://www.researchgate.net/profile/Maxime_Petit4

Other authors
Amaury Depierre, Xiaofang Wang, Matthieu Grard, Emmanuel Dellandrea and Liming Chen

Abstract

The frequent introduction of new objects to manipulate in factory lines is a big challenge for industrial robotics, where the classical approach requires the intervention of a human expert to program the robot for each new reference. Reinforcement Learning methods allow the robot to adapt its grasping parameters autonomously, however it is known as time consuming, reducing its usefulness in real-world settings. We propose to reduce the training time by providing previously optimized sets of parameters for objects similar to the new ones as starting points of the RL algorithm, using a multi-dimensional based similarity algorithm (e.g. shape, weights, material).

Keywords
Reinforcement Learning, Transfer Learning, Computer Vision, Industrial Robotics, Adaptive Robotics

Wednesday 21/06/2017

Poster

Object recognition and manipulation by robots: data sharing and experiment reproducibility (ORMR)
Moamar Sayed-Mouchaweh
Advanced Decision Support Tools for Cost-effective Based Maintenance and Smart Design of Large Scale Distributed Systems

Moamar Sayed-Mouchaweh received his Master degree from the University of Technology of Compiegne-France in 1999. Then, he received his PhD degree from the University of Reims-France in December 2002. He was working as Associated Professor in Computer Science, Control and Signal processing at the University of Reims-France in the Research center in Sciences and Technology of the Information and the Communication (CReSTIC). In December 2008, he obtained the Habilitation to Direct Researches (HDR) in Computer science, Control and Signal processing. Since September 2011, he is working as a Full Professor in the High National Engineering School of Mines “Ecole Nationale Supérieure des Mines de Douai” at the Department of Computer Science and Automatic Control (Informatique & Automatique IA).

Weblink
https://www.linkedin.com/in/moamar-sayed-mouchaweh-3420ba12/?ppe=1

Abstract

In increasing number of real world applications, e.g., manufacturing systems, electrical energy generation, transmission and distribution grids, data samples arrive continuously online through unlimited streams or flows often at high speed. The generated data streams by these applications possess the 3Vs characteristics of Big Data (i.e., volume, variety and veracity). Therefore, we propose developing innovative advanced decision support tools allowing generating models using these huge volume and ever-growing data streams. These models are used to predict efficiently the behavior of a wide range of real applications in order to optimize their performance, life-cycle or/and to increase their safety and security.
Two case studies can be targeted:
1. Monitoring and Smart Design of Onshore/Offshore Large-Scale Wind Farms in order to obtain a significant cost reduction in wind farm exploitation and maintenance, as well as providing insights for better design of components of wind turbines and foundations (structure, seabed).
2. Active Dynamic Demand Side Management with PV-WT in order to ensure a reliable electrical energy supply by removing the chance of surpluses, shortages and the resulting power failures.

Key words
Data-driven or centric modelling, Evolving intelligent autonomous systems, Big Data challenges, Smart grids

Thursday 22/06/2017

Poster

Industrial big data and process modelling for smart factories (IBDSF)
Samia Oussena
Towards intelligent business process modelling

Dr Samia Oussena is an associate professor in Software Engineering. She has a research background in methodologies and software application development. She has a doctorate from University of Manchester and has been the principal investigator and collaborated in number of research and industrial projects. Prior to academia, she gained an extensive industrial experience in software development. She has lead and been involved in a number of application development projects.

Abstract

Intelligent business processes models are the natural evolution of the business process models, adding more capabilities for greater intelligence within business processes. In order to contribute to the long-term vision of an intelligent business process, MobWEL provided an adaptation to BPEL. This work was undertaken to design a workflow model that defines context-aware content-centric processes. More recently, solutions for executing business processes relying on IoT are becoming more and more common. This allowed us to consider Smart Objects when modeling these processes.  Advanced machine learning is what makes Smart Objects appear “intelligent” by enabling them to both understand concepts in the environment, and also to learn.  The production of intelligent business processes needs to learn from advances and practice of Software Engineering.

Thursday 22/06/2017

Keynote talk

AttachmentSize
PDF icon CHIST-ERA_Oussena.pdf5.35 MB
Industrial big data and process modelling for smart factories (IBDSF)
Wlodzimierz Kasprzak
Variable Structure Controllers for Service Robots performing object recognition and manipulation

Włodzimierz Kasprzak received the Ph.D. in Computer Science in 1987 and the D.Sc. in Automation Control and Robotics in 2002, both from Warsaw University of Technology, Faculty of Electronic Engineering and Information Technology. He also holds the German Dr.-Ing. in Pattern Recognition (received in 1996) from University of Erlangen-Nuremberg. In 2014, W. Kasprzak has been awarded in Poland with the Professor title.
Prof. W. Kasprzak is a permanent staff member of Institute of Control and Computation Enginerring of WUT since 1997 (current director: Prof. Cezary Zieliński). He is teaching courses (at undegraduate, graduate and postgraduate levels) in computer programming, image and speech analysis, computer vision, artificial intelligence and signal processing. Majority of the courses are taught in English (within studies in Electrical and Computer Engineering, Computer Engineering, Automation and Robotics, and European studies EMARO).
Włodzimierz Kasprzak has a rich professional experience gained outside of his University, both in Poland and abroad (mainly in Germany and Japan).

Weblink
robotics.ia.pw.edu.pl

Other authors
C. Zieliński, W. Szynkiewicz, T. Winiarski

Abstract

As  the on-board robot computational resources are limited, but in some cases the demands imposed on the robot by the user are virtually limitless, the solution is to produce a variable structure control system. The task dependent part has to be exchanged, however the task governs the activities of the robot. Thus not only exchange of some task-dependent modules is required, but also supervisory responsibilities have to be switched. Such control systems are necessary in the case of service robot, where the owner of the robot may demand from it to provide many functions. In particular, visual object recognition, single- and two-handed manipulation and autonomous knowledge acquisition are most important services.

Key words
service robots, control framework, object modelling and recognition, two-handed object manipulation, RGB-D images, WWW robot services and repositories

Wednesday 21/06/2017

Poster

Object recognition and manipulation by robots: data sharing and experiment reproducibility (ORMR)