Research

Artificial intelligence (AI) is a field of computer science that emerged in the 1950s to systematically study the phenomenon of intelligence and develop useful programs and tools to perform everyday tasks requiring intelligence. The field includes neural networks, machine learning, expert systems, AI for robots, natural language processing, speech processing, computer vision, etc. AI can be applied in many different areas such as education, transportation, industry, finance, marketing, healthcare, management, entertainment, robotics, and telecommunications, among others. 

Robotics is an interdisciplinary field concerned with computer science, mechatronics, electrical engineering, mechanical engineering, nanotechnology and bioengineering. Robotics subfields are numerous and include those that study the more essential characteristics of robots (e.g., robot control, robot locomotives, micro-robotics), as well as the aspects of robots (e.g., aerial robots, underwater robots, industrial robots). 

One recent notable application of AI to robotics is in computer vision, which uses vast amounts of data to assist robots to more accurately navigate, sense, understand, and interact with the real world. For example, AI technology is currently used in cars with driver-assistance functions (self-driving, self-parking, advanced cruise control), or to support a robot performing surgery without intervention from a human doctor. 

We are progressing toward a smart society wherein digital devices and technologies take advantage of the available data to create solutions for both known problems and the potential future challenges of life, leisure and work. Interaction between people and these cyber-physical systems occurs at various stages and in various ways. The outcomes of new cyber-physical-human systems depend largely on the successful integration of human factors into the cyber-physical systems to create human-centered technological designs capable of delivering a “wow experience” on an individual and societal level, yet with a deep understanding of human factors, ergonomics, psychology and anthropology. Understanding the capabilities and limitations of people ensures that more useful, usable and desirable products and services can be designed and implemented anywhere in the world. 

The Industrial Revolution created profound change with new machines that were more powerful than the human body. Similarly, information technology is currently revolutionizing our world, with information collection and processing capabilities far beyond our imagination. Not unlike our material world, the cyber-world is where information is collected, moved, and used by both human and machine intelligence, creating powerful tools to significantly improve our quality of our life. Yet, these tools have brought about significant ethical challenges in terms of privacy and even the nature of humanity.  

Digital technology focuses on the processes of information collection, storage, communication, and computation, and is at the core of modern Electronics Engineering, Computer Engineering and Computer Science. Key research areas include but are not limited to microelectronics, circuits and sensors, IC design, Embedded Systems, Signal Processing, Communication Networks, Data Analytics, machine learning, cloud computing, etc. The future of digital technology research not only lies in the traditional areas, but also in forming innovative interdisciplinary research themes around the cultural and social impacts of the IoT, robotics, machine intelligence, and human-machine interaction.

Design methodology supports design processes to serve design practice in various ways, such as improving creativity, decision-making and problem-solving techniques. Today’s thorny issues require applying new design approaches and new technologies. Design research at ISD is at the intersection of design and technology, exploring new methodological approaches by integrating design and systems thinking to redefine complex social problems and, more importantly, use design and technology to develop aesthetically appealing and technologically enhanced innovative interventions and products. One such research area is design for systems change, which focuses on developing a new strategic design approach for large-scale systems transitions for a sustainable future. Case studies on those large-scale systems transitions research include, the oyster farming of the marine ecosystem and the tram of the mass transit system in Hong Kong.

Systems can be integrated with design to create a better, more innovative product. To avoid fixating on just one—usually the initial—engineering and design solution for a problem, which may eventually be proven false as part of the new product development process, a thorough literature review in the targeted application area is conducted first. The resultant multi-disciplinary knowledge base is shared across all areas of specialized expertise and evaluated via the morphological box method to compare strengths, weaknesses, opportunities and the juxtaposition of design elements in the targeted application area.

After evaluating and clarifying task parameters, multidisciplinary design conceptualization leads to the creation of a detailed, purpose-driven and functional product design. This type of research is both foundational and theoretical, as well as having practical application to comprehensive solution-oriented problem-solving processes. These holistic activities occur concurrently as the project progresses. Computational visualizations can create cross-disciplinary objective models for areas such as ergonomics and human factors, and especially the subjective areas of aesthetics and cultural significance.

Integrative systems are complex. The effective design of such systems relies on the accurate prediction of their behavior. To make these predictions, we use models, including mathematical, behavioral, or even physical.

Models can be observed with a range of stimuli, such as inputs and noise, to predict their behavior.

By definition, models are approximations. The most common outcomes include a reduction in product development time, shrinking development costs, increased product quality and new insights into human or system behavior.

A variety of tools can be used to develop useful models, including for abstract concepts such as timed automatons to predict the behavior of embedded systems, finite element models to predict behavior of buildings subject to extreme typhoons, and human-machine simulations to predict traffic flow to help design a smart city.

Potential graduate students are invited to browse through the research interests and ongoing projects of ISD faculty to look for potential advisors in their areas of interest.

For any species to survive in the world, they must have several basic skills including the ability to understand the environment (perception), to make decisions (thinking), to share information (communication), and to change the environment (manipulation). Human beings conquered the world not only because we have such strong abilities, but more importantly because we are able to create “things” that strengthen our capabilities. For example, we created equipment that can see much farther than our eyes; we designed machines that can compute much faster than us; and we constructed machines that are stronger than us. All these things have been serving us with their unique capabilities, but their powers can be upgraded to a new level if such distributed capabilities can be connected.   

Internet of Things (IoT) is a system that connects billions of smart devices together through the internet. Its true power lies in the connected sensing, storage, computing, and manipulation capabilities distributed over the world, enabling large scale information collection (big data), information sharing (machine to machine communication), automatic decision making (machine learning and AI), and order execution (automation and robotics). IoT has several levels of enabling technologies 

• Applications Level: Smart City, Smart Home, etc.
• Data Level: Big Data, Machine Learning, Data Analytics, etc.
• System Level: Signal processing, Communications, Networking, etc.
• Device Level: Sensors, Embedded systems, Controllers, Actuators, etc.