Keynote 1

Enhancing Multi-Hop Information Scavenging Transmission with Artificial Intelligence

Assoc. Professor Dr. Kenneth Tze Kin Teo, Universiti Malaysia Sabah (UMS)

Abstract

Artificial Intelligence towards Precision Computing is a necessary evolution when a developing society reaches its maturity in the knowledge of science, mathematics, technology, engineering, computational skills, and scientific research breakthroughs. The fulfilment of precision computing is responsible for driving multi-stage digital transformation in society build-up; bringing forth the advancement in almost everywhere from personal lifestyle, national systems to international communication. It also means cheaper and more affordable services and products; comprehensive, transparent and user-interactive information sharing around the world; complicated dynamics and hidden uncertainty increasingly in control. Central to the precision computing is artificial intelligence. With the high end computation ability of current computer processors and programming algorithms, revolutionary changes of implementing precision measurement to obtain on-line estimated model is applicable. As such, the concept of dynamic optimization is likely to be implemented with the dynamic modelling of the process.

Precision computing requires information scavenging transmission that employing a multi-hop protocol among the multi-node of semantic agents, which consists of path discovery algorithm and data management algorithm. Due to the network topology is dynamic and changing from time to time, establishing data transmission becomes very challenging. As stability and connectedness are mandatory conditions to ensure robust and reliable data delivery, it is necessary to discover an energy efficient wireless multi-hop protocol for information scavenging from source to destination. Evolutionary computation can be used to decide the best path for information scavenging while assisting to elect the cluster heads to manage the data collection process during transmission. An objective function is used to determine the best-potential solution towards the determined task. In fact, such objective function is the “main brain” used to evaluate and obtain the near-optimal solution of the path optimization. In addition, the objective or fitness function with dynamic parameters extraction via deep learning algorithm can be used to evaluate the fittest solution to produce the least modelling error. As such, the harnessing of precision computing technology could surmount more existing challenges and bring about new level of community services, while dedicating to interdisciplinary specialization and education of future engineered profession.

Biography

KENNETH TZE KIN TEO received the B.Eng (Hons.), M.Sc. and Ph.D. degrees in Electrical & Electronic Engineering from University of Leicester (UK) and Universiti Malaysia Sabah (Malaysia), respectively. He is currently a faculty member of Engineering in Universiti Malaysia Sabah. His research activities for the past twenty years focusing on precision optimization and artificial intelligence in the field of intelligent transportation, smart energy, precision automation, and analytical mechatronics & biomedical. He pioneered the Modelling, Simulation & Computing Laboratory (mscLab) founded in 2009. With the expertise in dynamic modelling and optimization, his team had evolved from 3C: Computation, Communication & Control towards PC: Precision Computing especially in the area of evolutionary computation, deep learning and semantic agent. He is currently leading Artificial Intelligence Research Group in his university and committed to IEEE activities as chair of IEEE Malaysia Sabah Subsection.

Keynote 2

Decision making support based on Sensing and Simulation Technology for Realizing a Smart Society

Professor Dr. Atsuo Ozaki, Osaka Institute of Technology (OIT)

Abstract

In recent years, many Japanese companies have been warned of the "cliff of 2025," a problem that will greatly hinder business expansion and corporate growth due to the aging of existing IT systems and its becoming black boxes, and the harmful effects of a vertical sectioning system, as well as the loss of know-how due to the retirement of experts and the shortage of IT personnel due to the declining birthrate and aging population. In addition, many companies have begun digital transformation (DX) initiatives in order to keep up with the drastic changes in the business environment caused by changes in consumer behavior and digitalization, but they are struggling. On the other hand, new solutions such as AI/IoT are emerging based on the large amount of data obtained from e-commerce and social networking services on the Internet, IC tags and barcode readers for shopping, surveillance cameras installed throughout the city, and various sensors mounted on moving objects ranging from satellites to vehicles. There is an urgency matter to utilize these technologies and data to improve efficiency, save manpower, and automate operations, as well as to create new value to improve operations and develop businesses to solve the issues mentioned above.

"Society 5.0" is a government-led project to promote such initiatives, and its main theme is the realization of a smart society that integrates cyber space and real space. For example, based on the large amount of data collected by the distributed deployment of various sensors in the real space, decisions are made using AI technology and so on in the cyber space, and the results are reflected to the real space in real time and seamlessly. In order to contribute to the realization of this smart society, my laboratory is researching and developing sensing technology to detect the number of visitors in large-scale events and simulation technology to predict the flow of people based on the detection results. To obtain the number of visitors, detected WiFi signals are used emitted from visitors' smartphones, which do not contain personal information. As for simulation technology, multi-agent simulation technology is adopted that simulates each individual visitor. If the number of people in a target space can be detected and predict the flow of people in real time with high accuracy, this will contribute greatly to improving operations and developing business. In this talk, the contents of the research and development being conducted toward the realization of a smart society described above will be shown.

Biography

ATSUO OZAKI is a professor of information science and technology in Osaka Institute of Technology (OIT). He worked at Information Technology Research and Development center, Mitsubishi Electric Corporation from 1990 to 2018, then transferred as a professor of OIT in 2018. He received his B.E., M.S., and Ph.D. degree in Computer Science from Kyushu Institute of Technology, Japan in 1988, 1990, and 2001 respectively. His research interests are multi-agent simulation, parallel and distributed systems, and AI/IoT applications. He received the Best Paper Awards in 8th European Simulation Symposium, 1996 and in IEICE society for the study of concurrent system technology, 2005 respectively. Prof. Ozaki is a member of IEEE, IPSJ, and IEICE.

Keynote 3

Toward Practical Application of Image AI Research

Professor Dr. Yoshimitsu Aoki, Keiko University

Abstract

In this lecture, I will introduce our efforts toward industrial application and practical application of image AI research. The latest research on image recognition and image generation by deep learning and its application examples are described. Main topics are industrial image recognition systems and image recognition systems for supporting sports activity. I would like to discuss the direction and development potential of image AI research in the future.

Biography

YOSHIMITSU AOKI received the Ph.D. degree in engineering from Waseda University in 2001. From 2002 to 2008, he was an Associate Professor with the Department of Information Engineering, Shibaura Institute of Technology. He is currently a Professor with the Department of Electronics and Electrical Engineering, Keio University. He performs research in the areas of computer vision, pattern recognition, and media understanding.

Keynote 4

Predictive Modeling for Intelligent Integrated Energy System

Professor Dr. Hui Hwang Goh, Guangxi University

Abstract

Traditional power systems are dominated by fossil fuels, such as coal, oil and gas. In recent decade, traditional energy structure hasn’t been able to meet the demand of electricity market with the continuous development of social economy, and brought a series of environmental problems. Modern energy system is undergoing unprecedented important revolution under the joint promotion of clean energy, electrification, electrochemistry, Internet of things and other technologies. The core development trend is from single energy system (such as power system and natural gas system) to intelligent integrated energy system (IIES). The key technologies of IIES are operation optimization strategy considering multiple energy coupling interaction and energy demand prediction considering multi-source and multi-dimensional big data. Generation and demand prediction can help the dispatch department to arrange the power generation plan and operation mode reasonably, and improve the security and economy of power system. This research establishes reasonable and accurate forecasting models to improve the stability and security of modern power system, considering the impact of meteorological information, holidays and other factors on wind power prediction, photovoltaic prediction, users demand prediction and electric vehicle load prediction.

Biography

HUI HWANG GOH (Senior Member, IEEE) received the B.Eng. (Hons.) and M.Eng. degrees in electrical engineering and the Ph.D. degree in electrical engineering from Universiti Teknologi Malaysia, Johor Bahru, Malaysia, in 1998, 2002, and 2007, respectively. He is currently a professor of electrical engineering with the School of Electrical Engineering, Guangxi University, Nanning, China. His research interests include embedded power generation modeling and simulation, power quality studies, wavelet analysis, multicriteria decision-making, renewable energies, and dynamic equivalent. He is also a Fellow of the Institution of Engineering and Technology (IET), U.K., the ASEAN Academy of Engineering and Technology (AAET), and The Institution of Engineers, Malaysia (IEM), a Chartered Engineer under the Engineering Council United Kingdom (ECUK), and a Professional Engineer under the Board of Engineers, Malaysia (BEM). He is also the foreign fellow of Chinese Society for Electrical Engineering (CSEE).