Dr Jan ROSELL,
Associate Professor,
Institute of Industrial and Control Engineering
Universitat Politècnica de Catalunya (SPAIN)
E-mail: jan.rosell@upc.edu
Phone: +34 – 934 017 162
Dr Mohammed Diab,
Research associate
Personal Robotics Lab. (PRL)
Imperial College London (ICL)
E-mail: mohammed.diab@upc.edu
Homepage: https://mdiabphd.wixsite.com/mdiab
Talk Title
Reasoning for TAMP
ABSTRACT of the TALK
This talk will present an ontology-based knowledge framework, called Perception and Manipulation Knowledge (PMK), designed to enhance Task and Motion Planning (TAMP) capabilities for autonomous indoor service robots. PMK integrates perception models with cognitive knowledge, providing reasoning predicates related to perception, object features, situation analysis, and geometric reasoning, which are used guide both planning levels. Here we will illustrate the use of PMK to reason on the state of the environment, the robot features and the goal of the task, to automatically generate the Planning Domain Definition Language (PDDL) files needed to solve a given task, thus flexibly configuring robot task planning problems. This procedure has been integrated in a behavior tree task manager, allowing the real execution of robot manipulation tasks in possibly uncertain and changing environments.
BIO
Jan Rosell received the BS degree in Telecomunication Engineering and the PhD degree in Advanced Automation and Robotics from the Universitat Politècnica de Catalunya (UPC), Barcelona, Spain, in 1989 and 1998, respectively. He joined the Institute of Industrial and Control Engineering (IOC) in 1992 where he has developed research activities in robotics. He has been involved in teaching activities in Automatic Control and Robotics as Assistant Professor since 1996 and as Associate Professor since 2001. His current technical areas include task and motion planning, mobile manipulation, and robot co-workers. He is currently the research assistant director at the School of Industrial Engineering of Barcelona (ETSEIB-UPC).
Mohammed Diab is a research associate in the Personal Robotics Lab. (PRL), Imperial College London (ICL). He has been awarded his Ph.D. degree in Automatic Control, Robotics, and Computer Vision at the Institute of Industrial and Control Engineering (IOC), Universitat Politècnica de Catalunya (UPC). During his Ph.D., he developed cloud-based frameworks that, some of them, were a part of joint works during his visits to other labs. The visited labs are IAI (Institute for Artificial Intelligence), Bremen, Germany, and Third Institute of Physics Biophysics, Georg-August University Göttingen. He is a voting member of the "Autonomous Robots" sub-group of the IEEE WG ORA that works on the standardization of ontologies for robotics and automation. Mohammed Diab's research interests include Intelligent Robotics, particularly in intelligent perception, planning that includes manipulation skills, multi-scale user modelling/reasoning, and cognitive robotics to generate human-like behaviour. He is currently working for the UK Research Innovation (UKRI ) Trust Node project funded by Innovate UK, in which he develops human behaviour inference methods for Trustworthy Human-Robot Interaction. More information can be found on his personal web page https://mdiabphd.wixsite.com/mdiab.