Professor Chuan Wu, Professor of Department of Computer Science
It was about seven years ago when Professor Chuan Wu from the Department of Computer Science began researching on a smart walker for the elderly, following her father’s diagnosis with Parkinson's disease.
Indeed, not just her but many others understand the serious mobility issue suffered by the elderly, which undermines their quality of life. The walker developed by Professor Wu with the help of students and another professor from the Faculty of Engineering has proven to be a valuable tool.
In 2023, it won a Bronze Medal in the 48th International Exhibition of Inventions in Geneva, Switzerland, in recognition of its significant contribution in the field of mobility and elderly care.
Putting her knowledge and strengths to good use
Given her computer science background, Professor Wu decided to leverage her expertise and knowledge to create a device that could help the elderly move around safely and easily, both indoor and outdoor. The invention – the Smart Robotic Walker – combines advanced Artificial Intelligence (AI) with a home-friendly design, creating a cosy companion for seniors to stay healthy, active, and independent.
They could have potentially received a silver award in Geneva if a high-capacity battery had been used during the demonstration to avoid unstable power supply, according to Professor Wu. Unfortunately, it was not brought along on the trip.
Prototype of The Smart Elderly Walker
Offering simultaneous walking and standing support, the device protects the user from falling and can also be used for the purpose of recovery, by engaging in walking exercises safely, for those who have suffered a fall or walking impairment, for example at the early stage of Parkinson’s disease.
All-round protection offered by the special device
The Smart Elderly Walker is equipped with three fundamental functions. Firstly, it offers sound source localisation and voice command capabilities. With multiple microphones, the walker can receive the user's summoning and move autonomously towards them. Additionally, its advanced AI system effectively detects voice commands and vocalised requests, making interaction with the device convenient and intuitive.
Secondly, the walker has a front-following feature, which detects the user's lower-limb gait using privacy-friendly sensors. This enables the walker to provide additional walking support and anti-fall protection, making it ideal for everyday trips around town or navigating through tight spaces like hallways.
The soft-robotic handle of the walker provides the necessary support during emergencies, with the embedded haptic monitor detecting at-risk events in real time. After detecting abnormal force pressure, the soft interface would activate brakes to provide safety support.
“The walker automatically follows the user,” said Professor Wu. “It can move with you even if you do not hold onto it. Our purpose is to keep you from falling by providing stable support. There are a number of sensors in the device and our AI model, with data of the user’s past movements and gestures, can make predictions about his or her moving intentions.”
“When low in battery, it can move automatically to its charging dock for recharging. You can also summon it to come over to you. If you happen to fall, you can send a message to your caregiver by shouting out, making an emergency call.”
Because of its soft handle and three-side protection offered by the walker, the user should be safe even in the event of falling, said Professor Wu. She added it is safer than conventional walkers, which are not as stable and may fall together with the user.
Highlight functions of The Smart Elderly Walker
Professor Wu did her undergraduate and master’s studies at Tsinghua University. Having completed her doctoral study in Electrical and Computer Engineering at the University of Toronto, she joined the University of Hong Kong (HKU) in 2008. She has since dedicated herself to teaching and research, specialising in the areas of cloud computing, distributed machine learning systems, and data centre networking.
She is also keeping herself busy with basic research in areas including modelling, analysis, design, and implementation of large-scale distributed systems based on solid theoretical foundations, including optimisation, auction theory and machine learning methods.
Professor Wu joined HKU in 2008
Meeting modern-day needs
Professor Wu’s research is highly relevant in today’s world of aging population. Whether it be dementia patients, or the young old – defined by the World Health Organisation as those aged between 60 and 69, the middle old (aged between 70-79) or those aged above 80, the demand for walking assistance has burgeoned.
Professor Wu is keen to improve the elderly’s overall quality of living, in line with the Hong Kong government’s drive to help the local population age in place, rather than being put under institutionalised care.
Hence the walker is only one of a few other projects she is working on related to elderly care, including a smart/robotic speaker being developed together with the Faculty of Social Science of HKU, which could provide cognitive stimulation for older people through chat functions.
The device would allow an elderly person to have chats anytime he wants. Responses would be generated in alignment with cognitive stimulation principles embraced by social science research. “Our goal is to slow down the decline in one’s cognitive ability as he/ she ages,” said Professor Wu.
The speaker could also address the issues of loneliness or symptoms of minor depression. Even if a user lives with their family members or is accompanied by a helper, the speaker can be therapeutic as it offers more than mere causal chats. “We aim at providing guided dialogues for the users,” she explained. “According to social science researchers, including psychologists, 18 principles can be incorporated in the dialogues to help promote cognitive stimulation, one of which is to bring up the user’s memory,” said Professor Wu.
She acknowledges, however, that embedding those principles into the speaking device remains a novel area of research. It is far more challenging than developing the chatbots now commonly used across industries mainly for commercial purposes. “The usual chatbots only provide general dialogues rather than incorporating medical functions.”
One of her research goals is to enrich the dialogue with emotions, and empathy. “Implementing the social science principles is important because people with cognitive decline can be depressed; a reason they see a doctor is because they want to seek empathy.”
Increasing access to the walker
Back to the smart walker, a prototype has already been done in preparation for commercialisation. Professor Wu has also developed a foldable version of the walker in light of the usual limited sizes of flats in Hong Kong.
But despite support from the Innovation and Technology Fund, efforts to turn the prototype into a commercially viable product have been held back by limited manpower.
With almost 20 doctoral students under her supervision on top of her research projects, Professor Wu can hardly afford the attention needed for commercialisation. “I do hope we can take this project forward. We are still trying to persuade doctoral students to be in charge of that but the current situation is computer science PhD graduates can get very good jobs out in the market. It takes much effort to persuade them to do the work,” she said.
Professor Wu delivering a lecture
More work might be needed to cut down on the cost of the walker. Similar, though not as advanced, models of the walker are sold at an exorbitant cost overseas, to the tune of HK$200,000, said Professor Wu. Her model is much less expensive, but the basic manufacturing cost inclusive of the sensors, CPU and chips inside could still be beyond the reach of many.
The AI model used to perform the key function of preventing falling is large, Professor Wu explained, and to run it there is a need for computation and storage capacity. “Storing those computation and memory capacity in the walker will incur significant costs.”
Tapping into her expertise, what she opted for instead was to deploy the AI model in a centralised cloud, then use networked functions to send commands/data to the cloud in order to run the AI models. The results are then returned to the walker through the cloud. When used outdoor, sim cards can be installed on the walker to access wi-fi connection. “We have also tested the walker on uneven surfaces in an outside environment. The results are positive even when going up on a slope.”
A long-term goal
Juggling the various projects has not been a problem for her. After all, she regards her work as conducive to achieving a wider goal – that of creating a holistic home system for the elderly. It is one in which various kinds of technology and devices are connected together providing convenience and practical support for older people, for example, through combining the empathy-filled speaking device and the robotic walker. This would remove the need for seniors to be placed in a residential care home.
She added: “I am thinking of a non-intrusive home system which does not rely on the camera type of sensors; practically speaking the system we are working on is for the young old or people at an early stage of dementia. Many do not want to have intrusive monitoring devices like cameras installed at home,” said Professor Wu.
She said confidently, “I do see the needs of the elderly and I think the current technological development can help them in different areas.”