Mohammed Diab
Ph.D.
Short biography
Currently, I am an assistant professor at University of Plymouth, and a principle research scientist at Stealth start-up. Prior to these roles, I was a postdoctoral research associate at Imperial College London working on a project which is a part of the UKRI Trustworthy Autonomous Systems Hub. The main role is to build a knowledge-based reasoning framework in order to infer trust between humans and robots, personalisation methodologies, and transfer trust between different agents.
I received my Ph.D. degree in robotics and AI in January 2021. My Ph.D. thesis title is "Perception and Knowledge representation and reasoning for manipulation planning", which uses the TOP-DOWN (TOP: knowledge-driven and DOWN is robotics applications) approach to deal with different types of robotic mobile manipulation problems.
I did my Master's degree in robotics in a topic entitled "Design and Development of 5-DOF Color Sorting Manipulator for Industrial Applications”. Moreover, I studied Mechatronics Engineering for my Bachelor's degree.
My research interests include:
-
Knowledge representation and reasoning
-
Knowledge graphs and ontologies
-
Cognitive robotics
-
Robot manipulation planning
-
Artificial Intelligence in Robotics
-
Human-Robot Interactions
-
Robot-robot collaborations
-
Marine and offshore Robotics
-
Healthcare robotics
-
Generative AI-based model for Explainable AI
Programming Skills
Ph.D. thesis
My Ph.D. thesis title is "Knowledge representation and reasoning for Perception-based manipulation planning", which deals with different types of robotic manipulation problems such as table-top and navigation. During his Ph.D., he developed:
-
A framework called, PMK - Perception and Manipulation Knowledge, which modeled under standardized foundations, Suggested Upper Merged Ontology (SUMO), and Core Ontology for Robotics and Automation(CORA). Also, some reasoning predicates which facilitate the manipulation planning process that involve the integration of symbolic and geometric levels of planning.
-
A framework called, FailRecOnt - An ontology-based approach for failure interpretation and recovery in automated planning and execution, which increases the capabilities of the robot to interpret failures and provide recovery strategies.
-
A framework called, SkillMaN - A Skill-based Robotic Manipulation Framework based on Perception and Reasoning, which uses a multi-model sensory system contains Camera and Radio Frequency Identification (RFID). Moreover, Reasoning, which uses learning-based techniques of adaptation m to increase the use of experiential knowledge in the planning system.
Some of these frameworks were a part of joint works during my 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.
The aforementioned works have been tested in simulation, and also with practical examples in real environments. The robots used are YuMi, TIAGo-PAL.
C
C++
Python
Prolog
RAPID
Web Ontology Language (OWL)
MATLAB
General Tools
CMAKE
GIT
SVN
Robotics Tools
GAZEBO
RVIZ
Robot Operating System (ROS)
Navigation Stack
Sensor Tools
Short stay
Vision
RFID
Speech recognition
Institute for Artificial Intelligence (IAI), Bremen University
Supervisor: Prof. Michael Beetz
City/Country: Bremen - Germany
Type of visit: Short research scholar
Starting date 21-January-2019
Ending date: 07-March-2019
Duration: 45 days
Skills
LaTeX
Microsoft Office
Linux
Windows
Third Institute of Physics Biophysics, Georg-August University Göttingen
Supervisor: Prof. Florentin Wörgötter and Dr. Tomas Kulvicius
City/Country: Göttingen - Germany
Type of visit: Short research scholar
Starting date1-November-2019
Ending date: 21-December-2019
Duration: 50 days
Languages
Arabic (Fluent)
English (Professional)
Spanish (Beginner)
German (Beginner)