UASTRO & com.
KEYNOTE SPEAKERS 2026
Dr. Usman Qayyum, Director (Artificial Intelligence) @ RESOLVE NUST
Dr. Usman Qayyum is an experienced researcher, a skilled Artificial Intelligence and roboticist, with a proven track record in the research industry with 18+ years of experience. He obtained his MS degree from National University of Science & Technology (NUST) in 2006 as a gold medalist. He holds a Ph.D. with a focus in Robotics and Artificial Intelligence, from the Australian National University in 2014. During the PhD, his research was focused on autonomous aerial vehicles. Later on, he did his post doctoral on Self-driving cars and was awarded $15000 for his extraordinary work. He is privileged to demonstrate the first self-driving car to the Prime minister of Singapore (Lee Hsien Loong) in July 2015.
Currently he is working as Director of Artificial Intelligence at Research and Solution Ventures (RESOLVE) Pakistan. He has contributed 35+ research papers in the reputed international journal and conferences and has written a book in the field of computer vision. He has supervised various students of BE/MS/PhD in their project & thesis. He is a member of the Artificial Intelligence and Robotics board from the Ministry of Science and Technology (MOST) Pakistan. He is also a part of the evaluation board for the research fund at National Center of Artificial Intelligence (NCAI), Pakistan.
Dr. Usman is chair of Artificial Intelligence and Software technologies track in IEEE International Bhurban Conference on Applied Sciences and Technology (IBCAST) conference and he is also the IEEE deputy publication chair of IBCAST at IEEE Islamabad section. He is privileged to make Pakistan’s First Electro-medical device of Pakistan approved by Pakistan Engineering Council and Drug Regulatory Authority (DRAP) of Pakistan. He has led a team of AI engineers to won the PASHA ICT award in the area of AI healthcare in 2021. He has a solid experience in the field of AI, Deep Learning, Robotics, Computer Vision, Predictive Analytics and healthcare.
Talk: Beyond LLMs: Swarm Intelligence and Decentralized Agentic Orchestration
Abstract: The artificial intelligence landscape is undergoing a massive paradigm shift. We are rapidly moving away from monolithic, prompt-dependent Large Language Models (LLMs) and entering the era of Autonomous Agentic Systems. In this session, we will chart the evolutionary path of AI from the foundational mechanics of deep learning and transformers to the deployment of dynamic, multi-agent ecosystems capable of advanced reasoning, long-term memory, and complex tool execution.
Drawing on two decades of applied AI experience esp spanning highly regulated healthcare diagnostics to collaborative robotics, this talk explores the architectural leap from single-agent bottlenecks to decentralized, multi-agent orchestration.
By bridging decentralized LLM orchestration with principles of swarm intelligence i.e. multi-rover navigation. Moreover, we will uncover how decentralized, peer-to-peer AI networks are setting the stage for the next generation of autonomous digital and physical systems.
Mohammed CHADLI, Ph.D.Full Professor,University Paris-Saclay Evry, IBISC-UEVE Evry France, mchadli20@gmail.com
Mohammed Chadli received his M.Sc (DEA) from the Engineering School INSA-Lyon (France, 1999) and from “Ecole Normale Sup.” (Mohammedia, Morocco), the Ph.D. thesis in Automatic Control from the University of Lorraine (UL), CRAN-Nancy in 2002. He was Lecturer and Assistant Professor at the “Institut National Polytechnique de Lorraine” (UL, 2000-2004). Since 2004, he was Associate Professor at the University of Picardie and is currently a Full Professor at the University Paris-Saclay Evry, IBISC Lab., France. He was a visiting professorship at the TUO-Ostrava (Czech Rep.), UiA (Norway), SMU-Shanghai (2014-2017), NUAA-Nanjing (2018-2024), and the University of Naples Federico II (Italy, 2019).
Dr. Chadli’s research interests include filtering and control problems (FDI, FTC) and applications to vehicle systems, intelligence systems, network systems, and cyber-physical systems. He is the author of books and book chapters (Wiley, Springer, Hermes), numerous articles published in international refereed journals and conference proceedings.
Dr. Chadli is a senior member of IEEE. He is on the editorial board (Editor, Associate Editor) of several international journals, including the IEEE Transactions on Fuzzy Systems, Automatica, the IET Control Theory and Applications, the Franklin Institute Journal, Asian Journal of Control … and was a Guest Editor for Special Issues in international journals and the Vice Dean of the Faculty of Sciences and Technologies (Univ Evry Paris-Saclay). He now serves as the Chair of the IEEE France Section Control Systems Society Chapte, and listed in “100 000 Leading Scientists in the World”
Talk:Filter Design and Fault Diagnosis in Finite Frequency Domain
Abstract - This lecture proposes some results on methods of fault detection filter synthesis, and observer design, for a class of nonlinear systems. Observer-based LMI synthesis methods for T-S systems subjected to unknown inputs are presented. Subsequently, multi-objective synthesis problem is discussed in FDI framework. When we are interested in these problems in finite frequency domains, these classic techniques (in infinite frequency domains) become quite restrictive. Indeed, the problem of multi-objective synthesis in the finite frequency domain is addressed. In a fault diagnosis context, the generated residue must be as sensitive as possible to faults and as robust as possible against unknown perturbations by means of two finite frequency performance indices such as the H_ and H∞ indexes.
JJinan CHARAFEDDINE -Professor Researcher — Artificial Intelligence, Medical Computer Vision, Biomechanics
Professor Researcher at De Vinci Higher Education, Paris La D ́efense, specializing in Artificial Intelligence for Medical Computer Vision and Biomechanics. My research bridges artificial intelligence, robotics, telecommunications, and clinical research to develop next-generation intelligent healthcare technologies. I have extensive experience in academic teaching, interdisciplinary research, and supervision of engineering, master’s, and doctoral projects. • Professor Researcher, De Vinci Higher Education — De Vinci Research Center (DVRC), France • Associate Researcher, University of the Basque Country (UPV/EHU), Spain Associate Researcher, Hospital Raymond-Poincare (AP-HP), Garches, France
Member, Societe de Biomecanique (France). Teaching: Computer Vision, Deep Learning; Head of Mechatronics Engineering Projects (PIX2 – ESILV); Supervision of engineering and research projects in AI, robotics, and medical imaging. Research in medical robotics, biomechanics, and AI-based control systems.Teaching at Bachelor and Master levels.Research collaboration with AP-HP hospitals. Industrial Institute of Tripoli (IUT), Lebanon, Industrial electronics, automation, energy systems, telecommunications; Supervision of applied projects in robotics, smart systems, and sensors. Ph.D., Paris-Saclay University, France Motion Science and Control of Mechatronic Systems (2021). Dissertation: Characterization and Integration of Muscle Signals for the Control of a Lower-Limb Rehabilitation Exoskeleton. M.Sc. (Research), Lebanese University, Lebanon Biomedical Engineering (BioMEMS). Engineering Diploma, Lebanese University, Lebanon Instrumentation and Industrial Computing. B.Sc., Lebanese University, Lebanon Applied Physics. Research Interests: Artificial Intelligence for Medical Data Analysis and • Telecommunication Data Analysis and Intelligent Sys- Computer Vision tems; Author and co-author of 62 peer-reviewed publications in high-impact journals and international conferences, including Information Fusion, Neural Networks, Expert Systems with Applications, Artificial Intelligence Review, IEEE journals, and The Lancet. Full list available on Google Scholar; Reviewer for IEEE and Elsevier journals; Invited Keynote Speaker at international conferences; Rapporteur, examiner, and president of Ph.D. defense juries; Active involvement in international research collaborations and projects. Languages: Arabic (native),French (fluent) English (fluent) .
Talk: Artificial Intelligence–Driven Modeling and Control of Medical Robotic Systems
Abstract: This keynote presents recent advances in artificial intelligence–driven modeling and control of medical robotic systems, with a particular focus on rehabilitation exoskeletons and human–robot interaction. The talk emphasizes simulation-based frameworks combining biomechanical modeling, data-driven learning, and adaptive control strategies to design safe, robust, and personalized assistance systems. Through numerical simulations and comparative studies, the presentation illustrates how optimization and machine learning techniques can be integrated into control architectures prior to real-world deployment.
The proposed approaches aim to bridge control theory, optimization, and artificial intelligence, providing a solid foundation for future experimental and clinical validation.
Dr. Swaminath VENKATESWARAN- Professor in Mechatronics, Leonardo de Vinci Engineering School (ESILV), Paris, France Robotics, Bioinspiration, Mechanical design, Optimization techniques, Researcher in Robotics & Industrial engineering Mar. 2022 - present Research group: New Materials, Intelligent Systems & Innovative Companies, De Vinci Research Center (DVRC)
ESILV, Paris, France ;Design and analysis of hybrid parallel mechanisms; Bio-inspired robotics for industrial applications; Application of cobots for circular economy and remanufacturing; Design & Analysis of flexible mechanisms for upper body exoskeletons; Collaboration with external researchers; Writing and reviewing scientific articles: Conferences, Journals, Book chapters. Researcher in Industrial engineering Oct. 2020 - Aug. 2021 Research group: Product-Process Design (CPP), G-SCOP Laboratory, Grenoble Institute of Technology (Grenoble-INP), Grenoble, France -Integration of a Tulip platform with the cobot Panda; Ergonomic study for identifying ideal posture sof human whileinteracting with a cobot; Trajectory planning for Universal robot UR16e; Doctoral researcher (Ph.D. candidate) Oct. 2017 - Sept. 2020 Research group: Robotics & Living (ReV), Laboratory of Digital Sciences (LS2N)Ecole Centrale de Nantes, Nantes, France, Thesis topic: “Design of a bio-inspired robot for inspection of pipelines” ; Bibliography of bio-inspired locomotion techniques; Numerical and experimental validations on a rigid pipeline inspection robot • Design and analysis of a tensegrity mechanism; Development of control laws based on kinematic architecture of robot. Doctor of Philosophy (Ph.D.) in Robotics -Ecole Centrale de Nantes, France, Specialization: Robotics & Mechanical design, Grade: Honors.
Talk: Neuro-Adaptive Predictive Control for Upper-Limb Rehabilitation Using EMG and Quantum-Enhanced Intention Decoding
Abstract: Upper-limb motor impairments associated with neurological disorders such as cerebral palsy (CP) are characterized by abnormal muscle coordination, spasticity, fatigue, and highly variable electromyographic (EMG) patterns, posing significant challenges for rehabilitation exoskeleton control. While EMG-driven and adaptive controllers have shown promise, existing approaches often rely on fixed gains or address physiological state estimation, prediction, and decision making in isolation, limiting robustness and clinical relevance—particularly in pediatric populations. This paper proposes a unified neuro-adaptive predictive control framework for upper-limb rehabilitation exoskeletons that integrates physiologically interpretable state estimation, task-aware assistance, predictive intelligence, and intention-aware supervisory control within a closed-loop architecture. The framework combines a NMI that fuses muscle co-contraction and joint tracking error, fuzzy EMG-based load estimation for task context awareness, hybrid support vector regression–long short-term memory (SVR–LSTM) trajectory prediction, and robust kernel-based motor intention decoding. NMI and task context are used to adapt admittance parameters in real time, enabling assistance modulation that reflects the user’s evolving neuromechanical condition. The proposed approach is evaluated exclusively in a simulation environment using OpenSim, driven by real pediatric EMG datasets and clinically inspired perturbations including fatigue-like degradation, spastic bursts, noise, and variable external loads. Compared with no assistance and fixed admittance control, the proposed controller achieves substantial reductions in muscular effort (45– 60%) and pathological co-contraction (up to 50%), while significantly improving trajectory tracking accuracy, movement smoothness, and robustness under disturbance conditions (p < 0.001). These results demonstrate that physiologically interpretable, predictive, and intention-aware control can enhance both efficiency and robustness of upper-limb robotic assistance in conditions representative of pediatric CP. Although the present study is simulation-based, it provides a reproducible pre-clinical foundation for subsequent experimental validation and future translation to real-time rehabilitation exoskeletons.
Prof. Zaki SARI, PhD, MSc, Ing. SITIS (Smart Innovative & Training Solutions) citizenship: algerian age: 60 (birthday 9/14/1964) marital status: married (4 children) language profeciency: fluent in english, french, and arabic (writing,readingandspeaking) education: university of missouri rolla, usa university of tlemcen, algeria, national polytechnic school, algiers, algeria inelec boumerdes, algeria inelec boumerdes, algeria professional history: System Engineering Postgraduate Certificate in 2003 Industrial & Manufacturing Engineering PhD in 2003, Senior Consultant (Freelance) for industry services and higher education Full Professor at ESSAT, Tlemcen, Algeria, Full Professor, University of Tlemcen, Algeria, Member of Board of Trustees of Commercial Port of Ghazaouet, Short Term Lecturer at PAUWES, Tlemcen, Algeria, Senior Consultant on Operational Planning at PAUWES, Tlemcen, Algeria Sabbatical leave at Izmir University of Economics, Turkey, Associate Professor, University of Tlemcen, Algeria, Visiting Scholar,University of Missouri Rolla, USA, Assistant Professor (Lecturer), University of Tlemcen, Algeria, Assistant Professor, University of Tlemcen, Algeria, Assistant Professor, INELEC Boumerdes, Algeria, Assistant (Reader), INELEC Boumerdes, Algeria , Director of the manufacturing engineering Laboratory of Tlemcen (MELT) Member of the expert scientific council of CDTA, Head of the national curriculum in manufacturing & industrial engineering, Head of Manufacturing Engineering Team of Tlemcen Control laboratory (LAT) University vice president (acting) in charge of external relations , Dean of studies (acting) of electronic engineering department, Power Engineering Electrical Engineering, Intensive Technical English, Magister (MSc with thesis) in 1990 Engineer (5 years B.Sc) in 1987 Certificate (700 course hours) in 1984.
Talk: Energy-Efficient Design and Control of AS/RS for Industry 4.0
Abstract: The evolution of Automated Storage and Retrieval Systems (AS/RS) into intelligent, energy-aware cyber-physical systems marks a turning point in intralogistics. This paper presents a structured review of modeling, optimization, and control strategies for AS/RS, focusing on the convergence of kinematic performance, artificial intelligence (AI), and carbon reduction objectives. We first revisit foundational travel-time and retrieval-time models before highlighting how recent advances integrate simulation, metaheuristics, and reinforcement learning for adaptive scheduling. AS/RS are then analyzed as cyber-physical systems enabled by digital twins, supporting real-time decision-making and energy management. Particular attention is given to energy-saving strategies, including regenerative hardware, AI-based routing, and sustainable layout configurations. Finally, the paper outlines key research challenges—interoperability, AI robustness, and system-level sustainability—and proposes future directions to close the loop between design, operation, and environmental performance. By synthesizing over 40 recent studies, this work provides a comprehensive framework for designing next-generation AS/RS aligned with Industry 4.0 and low-carbon logistics goals.
Dr. Ramona-Elena CONSTANTINESCU, MD - Neurology
Neurology, Ozone-Therapy, Doppler Ultrasound of Cervico-Cerebral Vessels, Botulinum Toxin; Founding member, INSG (Integrated Medicine Study Group),Founding member, SSROOT (Romanian Scientific Society of Oxygen Ozone Therapy)Member, SNR (Romanian Society of Neurology)
Member, SRAMMI (Romanian Society of Acupuncture and Integrative Medicine)
Talk: Artificial Intelligence and Machine Learning in Neurorehabilitation
Abstract : Artificial Intelligence (AI) and Machine Learning (ML) are rapidly transforming the field of neurorehabilitation, introducing intelligent solutions that enhance the recovery process for patients with neurological disorders. These technologies enable the personalization of therapy, prediction of functional outcomes, and automation of interventions through robotics, exoskeletons, brain-computer interfaces (BCI), and immersive digital platforms such as virtual reality (VR). AI-driven applications in motor, cognitive, and emotional rehabilitation have shown significant clinical benefits, including improved functional independence, enhanced treatment adherence, and more efficient therapy delivery. Machine learning algorithms analyze movement patterns, response times, and behavioral data to dynamically adapt therapeutic tasks, ensuring a more engaging and tailored recovery experience. Advanced platforms integrate wearable sensors and real-time feedback to monitor progress and optimize outcomes.
Emerging technologies, such as AI-assisted neuroimaging analysis, VR- based cognitive training, and remote rehabilitation systems, open new perspectives for highly individualized and data-driven medical interventions. Despite ongoing challenges related to cost, accessibility, data privacy, and ethical concerns, the integration of AI in neurorehabilitation is paving the way for a future where intelligent, patient-centered care becomes the standard. Collaboration between clinicians, engineers, and data scientists is essential to ensure these innovations are both effective and responsibly implemented.
LuigeVladareanu, Professor , Senior science researcher of the Romanian Academy, Institute Solid Mechanics, Bucharest, since 1990. From 2003, Ministry of Education and Research, executive Department for Financing Superior Education and of Scientific University Research - High Level Expert Consulting for MEC/CNCSIS project, from 2003-2005, member of Engineering Science Committee of Romanian National Research Council, from 2005, Scientific Researcher Gr.I of the Romanian Academy, from 2009 Head of Robotics and Mechatronics Department of Institute of Solid Mechanics, Romanian Academy, Director of the Technological Transfer Center from 2020 . Director and coordinator of over 15 grants of international and national research – development programs in the last 5 years, 15 invention patents, developing 17 advanced work methods resulting from applicative research activities and more 60 research projects. He is the winner of the two Prize and Gold of Excellence in Research 2000, SIR 2000, of the Romanian Government and the Agency for Science, Technology and Innovation. 9 International Invention and Innovation Competition Awards and Gold of World’s Exhibition of Inventions, Geneva 2007 - 2009, and other 9 International Invention Awards and Gold of the Brussels, Zagreb, Bucharest International Exhibition. He received “TraianVuia” (2006) award of the Romanian Academy, Romania’s highest scientific research forum. He is a Corresponding Member of the American Romanian Academy and he is a member of the International Institute of Acoustics and Vibration (IIAV), Auburn University, USA (2006), ABI ́s Research Board of Advisors, American Biographical Institute (2006), World Scientific and Engineering Academy Society, WSEAS (2005), International Association for Modelling and Simulation Techniques in Enterprises - AMSE, France (2004), National Research Council from Romania(2003-2005), etc. He is a PhD advisor in the field of mechanical engineering at the Romanian Academy. He was an organizer of several international conferences such as the General Chair of four WSEAS International Conferences (http://www.wseas.org/conferences/ 2008/romania/amta/index.html), chaired Plenary Lectures to Houston 2009, Harvard, Boston 2010 and Penang, Malaysia 2010, Paris 2011, Florence 2014, Tenerife 2015 to the WSEAS International Conferences, and is serving on various other conferences and academic societies, Vice President ICCMIT 2017, Warsaw, Special-session chair ICCMIT 2017, Warsaw, ICCMIT 2015, Prague, ICAMechS 2012- 2016, Tokyo, Beijing, Melbourne, Editors-in-Chief of the Proceedings of the SISOM 2007-2016, under the Romanian Academy's aegis, member of the Sensors Editorial Board since 2020, Guest Editor of Sensors to the Special Issues 2019, 2021 and 2022 with 39 manuscripts (IF 3,847 each).
Talk: Intelligent Decision Support Systems of the Autonomous Mobile Robot Vectors applied in Search and Rescue Missions.
Abstract: The research on cutting-edge technologies of the Autonomous Mobile Robot Vectors (AMVs) aimed at Intelligent Decision Support Systems for Person Detection in Search and Rescue Missions are presented. The adopted methodology develops the concept of an intelligent decision support system in order to achieve automated detection of the human silhouette regardless of posture or partial occlusion of people. The developed interfaces are Single Shot Detector (SSD) and Faster R-CNN, applied on the USC Pedestrian Dataset database chosen especially for the multitude of images containing children. The intelligent decision support and adaptive communications system architecture (IDSACS) for real-time control of Autonomous Mobile Robot Vectors, terrestrial and aerial is presented. The IDSACS is based on the 3D VЕRО VIPRО Platform derived from the VIPRO platform, an innovative technical solution for the development of intelligent interfaces applicable to autonomous aerial robots through the virtual projection method. SSD analysis is applied to multi-object image identification and compared to regional proposal object recognition (RPN) algorithms or R-CNN series, which require two analyses: one for generating region proposals and another for object detection. As a result, SSD performs significantly faster than RPN approaches. The SSD approach is based on an innovative feed-forward network followed by max-suppression to produce the final detections. The SSD architecture uses a pre-trained Inception V2 model, pre-trained on the Open Image Dataset (OID) dataset, with a learning rate of 0.0007 and a momentum optimizer, which led to the best accuracy. Two object detector networks trained on the USC database were used to identify and localize the person images, applying the COCO metrics, which led to high average model accuracy (mAP) for the Faster R-CNN network and fast response for the SSD network. Future research aims to develop Intelligent Decision Support Systems for people in danger by using autonomous aerial and ground robot vectors in first responder-led missions, essential for saving lives.
Profesor Ph.D.Eng.Adrian Olaru finishes the University Politehnica of Bucharest, the Faculty of Machine-Tools, and now work inRobotics and Production Systems Department. Now, from 1998, he is a university full professor, and he teach the following courses: industrial robots dynamics behaviour, labview application in modelling and simulation of the dynamic behaviour of robots and Personal and social robots. He is a doctor from 1989. He have some important scientific contribution in experimental validation for mathematical models of the kinematic and dynamic of industrial robots; assisted research of the magneto rheological dampers; assisted research of the intelligent dampers; assisted research of the neural networks; optimizing of the robots dynamic behaviour by using the Fourier proper analyser; optimizing the dynamic compliance and global transmissibility by using the assisted research and the proper neural networks.
Talk:Dynamic modelling and simulation for control systems using the transfer multipol functions
Abstract: The presentation contents generality about the theory of the transfer functions with multipol modelling with separately channels of movements and efforts, to say what are the transfer functions between them and how can optimize the final answer. Modelling and simulating servo systems using transfer functions and multipole modules (multiple inputs, multiple outputs and multiple reference quantities) are strictly necessary steps in the analysis of dynamic behaviour and the transition to the next stage: servo system synthesis. These steps include the following activities: (i) writing the mathematical model for each component of the servo system; (ii) eliminating intermediate variables and determining a relationship between input and output quantities; (iii) creating the block diagram using elementary transfer functions of the first and second order proportional derivative type, first and second order inertia integrator, first and second orderproportional with inertia, some multipliers, add and constants; (iv) developing the block diagram using proper subroutines of the LabView software platform for simulation and assisted analysis; (v) determining the real and frequency characteristics with highlighting the parameters and performances of the dynamic behaviour; (vi) the analysis by drawing comparative characteristics with the modification of some constructive-functional parameters and highlighting the comparative dynamic behaviour parameters and performances..The simulation was done by using the LabView proper virtual instruments library what will be shown in the presentation. The Neural Network modelling was used to solving the inverse kinematics. One proper neural network was proposed after were modelling and simulation with LabView soft-ware some of the more know neural networks. The analyze with multi-pol transfer functions was developed and applied in one method of the satellite control of orientation with three wheels inertial system. One mathematical complex equations was developed to solve the inverse kinematics of the mechanical inertial system. The presentation contents also the analyze of the simulation results of the proposed algorithm. The results were show with comparative real and frequencies characteristics. The future work will contents apply all designed virtual LabView instrumentation in only one assisted platform what must be cover all calculus and optimisation of the kinematic and dynamic behavior of the satellite; The new assisted platform for satellite will cover all needed calculus for servo driving, establishing the number of pulses for each motor, animation of the satellite in the space, optimising the used neural network; The new platform will perform all input data of the complex method- Pseudoinverse Jacobian Matrix Method (PIJMM) coupled with Bipolar Hyperbolic Tangent Neural Network with Time Delay and Recurrent Links (BSHTNN(TDRL)).