Conference Program

In operating industrial facilities, alarm systems are configured to notify operators about any abnormal situation. The industrial standards (EEMUA and ISA) suggest that on average an operator should not receive more than six alarms per hour. This is, however, rarely the case in practice as the number of alarms each operator receives is far more than the standard. There exist strong industrial needs and economic benefits for better interpreting and managing alarms, and redesigning alarm systems to reduce false and nuisance alarms and increase the alarm accuracy. In this talk, we plan to summarize our recent work in this new area, targeting an intelligent and data-based approach, called “alarm data analytics,” and presenting a new set of advanced tools for alarm visualization, performance evaluation and analysis, alarm rationalization design, alarm flood classification, and root cause diagnosis, thereby to help industrial processes to comply with the new standards. The tools have been tested with real industrial data and used by process engineers in Canada and elsewhere.

In recent years, Japan has increasingly emphasized the importance of envisioning the society we wish to realize by 2050 and then backcasting from that vision to identify the technologies we should develop today. Two prominent embodiments of this perspective are the Robot Industry Vision 2050 recently issued by the Japan Robot Association and the Moonshot R&D Program led by the Japanese Cabinet Office, both of which place humans, society, and the environment at the core of future robotics and automation. Building on this philosophy, I will first introduce our work in the Moonshot R&D Program (Goal 3), where we are developing a collective of adaptable AI-enabled robots that co-evolve with humans and nurture users' self-efficacy, illustrated by recent outcomes such as the Robotic Nimbus — a conceptual vision of future robots that flexibly change their form and function to support diverse users. I will then present our research on garment handling conducted under the InnoHK initiative in Hong Kong, focusing on dual-arm robotic manipulation for flexible clothing manufacturing. Finally, I will reflect on what unites these domains — the human at the center — and discuss how our community can co-design a vibrant, inclusive future society.

At the cutting edge of artificial intelligence and biomimetic robotics, why does the combination of Large Language Model (LLM) plus humanoid robots still fall short of expectations and fail to be effectively deployed in practical applications? Drawing on the speaker’s forty years of innovation experience and study in the field, this lecture reviews the past to gain new insights. It explores the current difficulties and obstacles facing AI and discusses how to maximize the enormous dividends brought by AI and biomimetic robots, before looking ahead to the future development trends of intelligent robots as well as the opportunities for innovation and the strategies and tactics to seize them.