Prof. Arbnor Pajaziti
University of Prishtina, Kosovo

Speech Title: Advances in Intelligent Robotics and Human–Machine Collaboration: Recent Research from the University of Prishtina

Abstract: This talk will provide an overview of the recent research activities carried out at the Department of Mechatronics, Faculty of Mechanical Engineering, University of Prishtina, where our team has been working at the intersection of robotics, artificial intelligence, and intelligent automation. Over the last few years, we have developed several robotic systems and mechatronic devices aimed at addressing real practical needs in industry, healthcare, and everyday life.
Among our major efforts are the development of bio-inspired and prosthetic robotic hands designed to offer more natural movement and control, as well as the creation of an accessible mechanical ventilator that integrates smart sensing for improved reliability in emergency situations. We have also been exploring new concepts for automation lines that focus on adaptability, efficiency, and ease of integration into existing industrial environments. These projects reflect our broader goal of building systems that are technically strong while also being useful and impactful.
A significant part of our current work is dedicated to bringing AI into robotic decision-making. By combining machine learning approaches with multi-criteria decision models such as our modified TOPSIS framework, we are enabling robots to analyse complex environments, make informed decisions, and operate with greater autonomy. We are also studying sensor fusion and cognitive models to improve how robots perceive and interpret the world around them.
The final part of the talk will look toward the future of robotics, especially how robots can move from simply operating near humans to actively cooperating with them. This includes research on shared autonomy, safe interaction, and adaptive learning, all of which are essential for robots that work alongside people in healthcare, manufacturing, and service applications.
This presentation will offer a clear and integrated view of our ongoing contributions, while highlighting the exciting possibilities ahead for collaborative and intelligent robotic systems.

Biography: Prof. Dr. Arbnor Pajaziti received the B.Sc. degree in Mechanical Engineering from the University of Prishtina, Kosovo, the M.Sc. degree from the University of Zagreb, Croatia, and the Ph.D. degree in Robotics and Intelligent Control from the Vienna University of Technology, Austria. He has been a Professor at the Faculty of Mechanical Engineering, University of Prishtina “Hasan Prishtina,” since 2010, where he also serves as Head of the Department of Mechatronics.

He has extensive research experience in robotics, mechatronics, artificial intelligence, and intelligent control systems, including neural networks, fuzzy logic, and evolutionary algorithms. Prof. Pajaziti has led several national and international research projects in robotics and has authored numerous scientific publications and textbooks in engineering and technology.

His current research interests include mobile robotics, autonomous systems, intelligent motion control, and AI-driven robotic applications.

 

 

 

 

Assoc. Prof. Ting Zou
Memorial University of Newfoundland, Canada

Biography: Ting Zou received the B.Sc. degree in electrical engineering and the M.Sc. degree in automatic control engineering from Xi’an Jiaotong University, Xi’an, China, and the Ph.D. degree in robotics and mechatronics from McGill University, Montreal, QC, Canada. Afterward, she joined the Centre for Intelligent Machines of McGill University as a Postdoctoral fellow, working on the optimum design of the next-generation multi-speed transmissions for electric vehicles and nonlinear motion control of autonomous tracked vehicles for mining drilling operations. She is currently an Associate Professor with the Department of Mechanical and Mechatronics Engineering, Memorial University of Newfoundland, St. John's, NL, Canada. Her current research interests include mechanism design and control of biologically inspired robots, advanced human–robot interaction, machine learning for robotic applications, soft robots, and MEMS. Dr. Zou is a senior Member of the IEEE, and member of ASME, Canadian Society for Mechanical Engineers, and Canadian Committee for the Theory of Machines and Mechanisms.

 

 

 

 

Assoc. Prof. Ajit Salunke
Don Bosco College of Engineering, India

Speech Title: Intelligent Mechatronic Systems for Physical Property-Based Quality Grading of Agricultural Edible Nuts: Integrating Machine Vision and IoT

Abstract: Quality grading of agricultural edible nuts is a critical post-harvest operation that directly influences market value, processing efficiency, and consumer acceptance. In many developing countries, grading of arecanuts, cashew nuts, hazelnuts etc. is still predominantly performed through manual inspection or surface appearance-based methods, which are labor-intensive, subjective, and incapable of reliably assessing internal quality attributes. Moreover, commonly used color-based machine vision systems are highly sensitive to lighting conditions and fail to capture intrinsic quality characteristics such as moisture-related structural integrity. These limitations highlight the need for an objective, scalable, and non-destructive grading approach based on fundamental physical properties.
This work presents an intelligent mechatronic system for physical property–based quality grading of agricultural edible nuts through the integration of machine vision, Internet of Things (IoT), and microcontroller-based actuation. The proposed system utilizes measurable physical parameters, namely true density, bulk density, and porosity, as primary indicators of kernel quality. These properties are known to correlate strongly with internal attributes such as moisture content, maturity, firmness, and processing suitability across a wide range of edible nuts.
A dual-level grading architecture is implemented to enhance efficiency and throughput. At the first level, IoT-enabled bulk density measurement is used for batch-level preliminary screening. A precision load cell integrated with a microcontroller and cloud platform enables real-time data acquisition, remote monitoring, and batch classification. Batches that satisfy predefined quality thresholds proceed to the second level, where individual kernel assessment is performed. At this stage, kernel mass is measured using a load cell, while kernel volume is estimated in real time using a dual-camera machine vision system. Image segmentation techniques are employed to accurately handle irregular kernel geometries, enabling precise volume computation without physical contact.
True density is calculated from the measured mass and estimated volume, and grading decisions are made autonomously based on threshold values. Actuators controlled by embedded controllers physically segregate kernels into accept and reject categories, thereby completing the sensing–decision–action loop characteristic of intelligent mechatronic systems.
The system is experimentally validated using unboiled arecanut kernels as a representative case study, owing to their irregular shape, wide size variation, and economic importance. Validation results demonstrate a volume estimation accuracy of 97.33% using the segmentation-based image processing method, with overall grading accuracy ranging between 95% and 99%. Statistical analyses, including paired t-tests, regression analysis, and Bland–Altman plots, confirm strong agreement between automated measurements and conventional reference methods, as well as size-independent performance.
The proposed approach overcomes key limitations of manual and color-based grading systems by enabling objective assessment of internal quality attributes, reducing dependence on skilled labor, and ensuring consistent, repeatable grading outcomes. Owing to the universal relationship between physical properties and quality, the framework is readily extendable to other edible nuts such as cashews, almonds, walnuts, hazelnuts, and pistachios with appropriate calibration.
Overall, this work establishes a robust and scalable foundation for intelligent, physical property–based quality grading in post-harvest nut processing, offering significant potential for industrial adoption and digital transformation in agricultural value chains.

Biography: Dr. Ajit Salunke is an Associate Professor in Mechanical Engineering at Don Bosco College of Engineering (DBCE), Fatorda, Goa, India. He has served as HOD for 10 years and as Officiating Principal for a year. He holds a PhD from Visvesvaraya Technological University (VTU), Belagavi, India, along with an M.Tech in Computer Integrated Manufacturing and Bachelor’s degree in Mechanical Engineering. His current research interests include physical property-based quality grading and process optimization of agricultural edible nuts, machine vision, and IoT-based data monitoring and control systems.

With over 26 years of professional experience, Dr. Salunke has organized more than 100 workshops/ seminars/ invited talks on robotics, additive manufacturing, virtual instrumentation, computer-aided engineering analysis, and smart technologies for Industry 4.0. He has published several research papers in peer-reviewed international
journals and reputed conferences, including the ICMCR 2025 held at the National University of Singapore.

A Fellow of the Institution of Engineers (India) and Life Member of ISTE, NIPM, AMIEE, and VIBHA, Dr. Salunke received the CSI TechNext “Best HOD of the Year” Award (2017) at IIT Mumbai. He contributes as a resource person for Vidnyan Dhara program of the Directorate of Higher Education - Government of Goa, and Science Film Festival of India.

 

Assoc. Prof. Kei Fujisawa
Yokohama National University, Japan

Speech Title: Online Regime-Transition Monitoring of Erosion in Nuclear Power Plants Using Kalman Filtering

Abstract: High-speed liquid jet impacts can cause erosion in industrial components such as piping in nuclear power plants. Recent studies indicate that the liquid jet impact force on a target varies with transitions between erosion stages, yet robust identification of these transitions from noisy force signals remains challenging. In the present study, impact force signals were monitored using a force sensor attached to an aluminum specimen under high-speed liquid jet impingement and analyzed using an online regime-transition monitoring framework based on a state-space approach. A local linear trend Kalman filter was employed with a two-dimensional state consisting of the force level and its rate of change. Erosion regime transitions were detected by combining velocity-based rise criteria with statistical significance and a minimum level increase with two-dimensional kernel density estimation. The proposed framework offers an interpretable and robust tool for erosion-stage transition monitoring based on mechatronic sensing and state-space signal processing.

Biography: Kei Fujisawa is a researcher specializing in data-driven condition monitoring, state-space modeling, and statistical signal processing for safety-critical industrial systems. His recent work addresses erosion caused by high-speed liquid jet impacts in industrial components such as nuclear power plant piping. Using force-sensor measurements under jet impingement, he develops an interpretable online regime-transition monitoring framework based on a state-space approach, employing a local linear trend Kalman filter with a two-dimensional state and transition detection using velocity-based criteria and two-dimensional kernel density estimation. At the conference, he will present deployment-oriented results on online erosion monitoring using Kalman filtering.

 

 

 

 

Assoc. Prof. Shuqiong Wu
University of Osaka, Japan

Speech Title: Dual-task–based Software as a Medical Device for Detecting Early-Stage Cognitive Impairment

Abstract: Cognitive impairment has emerged as a major challenge in aging societies worldwide. Dementia, a common form of cognitive impairment, begins with subtle symptoms and progressively leads to loss of independence, with no curative treatment currently available. However, early detection during the Mild Cognitive Impairment (MCI) stage, an intermediate state between normal cognition and dementia, enables timely intervention that can slow disease progression. Existing diagnostic tools, such as MRI and PET-CT, are costly and unsuitable for frequent monitoring, while paper-based screening tests (e.g., MMSE) suffer from practice effects that limit their effectiveness for continuous assessment. To overcome these limitations, we propose a dual-task-based assessment system combining a motor task (gait) and a cognitive task (calculation). Individuals with MCI exhibit increased gait instability under dual-task conditions due to elevated cognitive load. By analyzing multimodal performance data acquired during dual-task assessment, the proposed system achieves more accurate early-stage cognitive impairment detection than conventional paper-based tests, demonstrating its potential as a practical software-based medical assessment tool.

Biography: Shuqiong Wu was born in Shanxi, China, in 1985. She received her B.E. and M.E. degrees from Beihang University, Beijing, China, in 2008 and 2011, respectively. She obtained her Ph.D. degree in Computational Intelligence and Systems Science from the Institute of Science Tokyo, Tokyo, Japan, in 2015. From 2015 to 2020, she was a research fellow at the Graduate School of Informatics, Kyoto University. From 2020 to 2025, she served as an assistant professor at SANKEN (The Institute of Scientific and Industrial Research), the University of Osaka. She is currently an associate professor at the Graduate School of Engineering Science, the University of Osaka. Her current research topics include dualtask-based cognitive impairment detection, cognitive status monitoring, medical image reconstruction, and contactless biometric sensing. Her research interests include biomedical signal processing, image processing, three-dimensional reconstruction, pattern recognition, and machine learning.

 

 

 

 

Assoc. Prof. Sunilkumar S. Honnungar
SDM College of Engineering & Technology (SDMCET), India

Biography: Dr. Sunilkumar S. Honnungar is an accomplished academic leader and researcher with 23+ years of experience spanning mechanical engineering education, advanced manufacturing research, and strategic industry–academia collaboration. He currently serves as Associate Professor of Mechanical Engineering and Training & Placement Officer at SDM College of Engineering & Technology (SDMCET), Dharwad, India.
His core expertise covers thermal error minimization in CNC machine tools, feed-drive system thermal behavior, CFD-based analysis of precision manufacturing equipment, advanced and bio-compatible materials, automated quality assessment using image processing, and sustainable manufacturing and clean energy integration. Through his work, he has contributed significantly to improving precision, reliability, and sustainability in modern manufacturing systems.
Dr. Honnungar has a strong international research footprint, with 13 international journal publications and multiple international and national conference papers in reputed venues such as African Journal of Food, Agriculture, Nutrition and Development, Materials Today: Proceedings, AIP Conference Proceedings, International Journal of Scientific & Technology Research, IRJET, and Applied Engineering Research. He has also published on topics including thermal behaviour of ball-screw systems, thermal error minimization in CNC machine tools, CFD simulation of spark ignition engines, bio-compatible implants, and work-cell based manufacturing systems.
He has guided three PhD scholars (part-time, VTU Belagavi) in the domains of automated quality grading of areca nuts, welding of dissimilar metals, and bio-compatible material characterization, with one thesis already submitted and others in advanced stages. In addition, he has mentored numerous undergraduate and postgraduate projects aligned with industrial problems and advanced simulation-based research in mechanical and manufacturing engineering.
Dr. Honnungar’s leadership extends to international academic platforms; he has served as Session Chair at the 3rd International Conference on Mechatronics, Controls & Robotics (ICMCR 2025) scheduled from 14–16 February 2025 at the National University of Singapore (NUS), Singapore. This role reflects his recognition as an expert in advanced engineering systems and his active engagement with the global research community.
At the institutional and professional level, he has coordinated and led several major conferences such as the National Conference on Global Emerging Technologies (Electric Mobility, Industry 4.0, Sustainability, 2024) and the PRIME series (Progress and Research Trends in Mechanical Engineering, 2016/2019/2022), along with TEQIP, AICTE and ISTE-sponsored FDPs and STTPs on topics including computational analysis, pedagogy, and advanced manufacturing. He has also been instrumental in designing and executing training and placement strategies that enhance graduate employability.
His contributions have been recognized through multiple awards, including the S.R. Gollapudi Award for Dynamic & Service-Oriented Leadership and Academic Excellence in Industrial Engineering (IIIE, Mumbai, 2019), a Special Award from IIIE for bridging academia and industry (2017), the Catalyze Tech Innovation Challenge Award (Global Center of Excellence in Affordable & Clean Energy, IIT Dharwad & SELCO Foundation, 2022), and the Best Co ordinator Award from the LEAD (Deshpande Foundation, 2013). He currently serves as Managing Committee Member, SAE India – Bangalore Section (2025–2026) and Board of Studies Member, Indian Institution of Industrial Engineering (IIIE, Mumbai, 2024–2026).

 

 

 

Asst. Researcher Kuo-Chin Jong
National Institutes of Applied Research, Taiwan

Speech Title: Patent Landscape Analysis of eVTOL Technologies

Abstract: The rapid development of electric vertical takeoff and landing (eVTOL) aircraft has led to concentrated innovation around specific technical modules. Using a contour-based patent mapping approach, this study identifies clusters of eVTOL-related patents across core categories: flight control, electric propulsion, battery systems, thermal management (radiators), supply systems, fuselage and power assemblies, and pilot input/flight path control. Among these, flight control and electric propulsion emerge as contested domains where most players converge, while thermal management and supply systems appear as relatively underexplored areas, indicating potential blue-ocean opportunities.

Biography: Dr. Kuo-Chin Jong received his Ph.D. degree in Photonics and Optoelectronics from National Taiwan University, Taipei, Taiwan, in 2010. He is currently an Assistant Researcher at the Science & Technology Policy Research and Information Center, National Institutes of Applied Research, Taipei, Taiwan.
His research focuses on patent landscape analysis, innovation strategy, and emerging technology policy. In particular, he is interested in developing purpose-driven innovation models and applying them to strategic foresight, enabling policymakers and industry leaders to better understand the evolving dynamics of advanced technologies such as electric vertical takeoff and landing (eVTOL) aircraft. Dr. Jong actively integrates patent analytics with innovation frameworks to provide actionable insights that bridge science, technology, and policy.