A Smart Walker to Understand Walking Abilities

Professor Pascal Poupart, University of Waterloo


Walking is a basic ability that is essential for mobility and many activities of daily living. However, injuries and various diseases common with aging may compromise our walking abilities. To assess walking abilities, various physical and neuro-psychological tests are typically performed at regular check ups with patients. While these assessments give a good characterization of the core physiological abilities of a person, they are done in an artifical setting and provide only point estimates in time. To understand the challenges faced by a person in her daily activities, there is a need to directly assess walking abilities as these activities are performed. Also, to optimize a treatment plan and prevent falls, there is a need to continuously monitor changes in walking abilities.

Wheeled walkers are currently the second most popular walking aid after the cane to improve balance. In this talk, I will describe a "smart walker" instrumented with various sensors (e.g., load sensors, accelerometers, wheel encoder and cameras) that continuously gather statistics about the walking characteristics of a user in a natural setting. In the first part of my talk, I will describe an initial study with residents at the Village of Winston Park (retirement community in Kitchener, Ontario) to understand how traditional neuro-psychological tests relate to walker usage and in particular to the measurements collected by the sensors. In the second part of my talk, I will describe how machine learning and computer vision techniques can be used to track motions of the lower limbs of the user and infer the high level context (e.g. activity/behaviour). More specifically, I will describe how to automatically recognize behaviours related to walker usage (e.g., walking, sitting, standing, etc.) based on non-video sensor measurements with probabilistic temporal models such as hidden Markov models and conditional random fields. I will also describe a monocular vision system to track the pose of the lower limbs.

This work is done in collaboration with a multidisciplinary research team at the University of Waterloo, the Toronto Rehabilitation Institute and UW-Schlegel Research Institute for Aging.


Pascal Poupart is an Associate Professor in the David R. Cheriton School of Computer Science at the University of Waterloo, Waterloo (Canada). He received the B.Sc. in Mathematics and Computer Science at McGill University, Montreal (Canada) in 1998, the M.Sc. in Computer Science at the University of British Columbia, Vancouver (Canada) in 2000 and the Ph.D. in Computer Science at the University of Toronto, Toronto (Canada) in 2005. His research focuses on the development of algorithms for reasoning under uncertainty and machine learning with application to Assistive Technologies, Natural Language Processing and Information Retrieval. He is most well known for his contributions to the development of approximate scalable algorithms for partially observable Markov decision processes (POMDPs) and their applications in real-world problems, including automated prompting for people with dementia for the task of handwashing and spoken dialog management. Other notable projects that his research team are currently working on include a smart walker to assist older people and a wearable sensor system to assess and monitor the symptoms of Alzheimer's disease.

Pascal Poupart received the Early Researcher Award, a competitive honor for top Ontario researchers, awarded by the Ontario Ministry of Research and Innovation in 2008. He was also a co-recipient of the Best Paper Award Runner Up at the 2008 Conference on Uncertainty in Artificial Intelligence (UAI) and the IAPR Best Paper Award at the 2007 International Conference on Computer Vision Systems (ICVS). He is a member of the editorial board of the Journal of Artificial Intelligence Research (JAIR) and the Journal of Machine Learning Research (JMLR). His research collaborators include Google, Intel, AideRSS, the Alzheimer Association, the UW-Schlegel Research Institute for Aging, Sunnybrook Health Science Centre, the Toronto Rehabilitation Institute and the Intelligent Assistive Technology and Systems Laboratory at the University of Toronto.