JTEHM presents

Automatic Detection and Classification of Unsafe Events during Power Wheelchair Use

Published Articles


Using a powered wheelchair (PW) is a complex task requiring advanced perceptual and motor control skills. We develop technological tools that can be used to characterize a PW user’s driving behavior using recordings from a datalogging platform that records, in real-time, the 3D acceleration of the PW. Our tool applies automatic segmentation to the recorded data to identify events of interest from the data-stream. Next, feature extraction computes a rich set of features from the events of interest. Finally, we apply a machine learning classifier to categorize the events according to type (safe vs. unsafe). Using a similar approach, we could also eventually classify specific activities (35 different classes).
Using a powered wheelchair (PW) is a complex task requiring advanced perceptual and motor control skills. We develop technological tools that can be used to characterize a PW user’s driving behavior using recordings from a datalogging platform that records, in real-time, the 3D acceleration of the PW. Our tool applies automatic segmentation to the recorded data to identify events of interest from the data-stream. Next, feature extraction computes a rich set of features from the events of interest. Finally, we apply a machine learning classifier to categorize the events according to type (safe vs. unsafe). Using a similar approach, we could also eventually classify specific activities (35 different classes).

Using a powered wheelchair (PW) is a complex task requiring advanced perceptual and motor control skills. Unfortunately, PW incidents and accidents are not uncommon and their consequences can be serious. The objective of our research is to develop technological tools that can be used to characterize a wheelchair user’s driving behavior under various settings. In the experiments conducted, PWs are outfitted with a datalogging platform that records, in real-time, the 3D acceleration of the PW. Data collection was conducted over 35 different activities, designed to capture a spectrum of PW driving events performed at different speeds (collisions with fixed or moving objects, rolling on incline plane, rolling across multiple types obstacles). The data was processed using time-series analysis and data mining techniques, to automatically detect and identify the different events. We compared classification accuracy using four different types of time-series features: time-delay embeddings, time-domain characterization, frequency-domain features, and wavelet transforms. In the analysis, we compared the classification accuracy obtained when distinguishing between safe and unsafe events during each of the 35 different activities. For the purposes of this study, unsafe events were defined as activities containing collisions against objects at different speed, and the remainder were defined as safe events. We were able to accurately detect 98% of unsafe events, with a low (12%) false positive rate, using only five examples of each activity. This proof-of-concept study shows that the proposed approach has the potential of capturing, based on limited input from embedded sensors, contextual information on PW use and of automatically characterizing a user’s PW driving behavior.
View full article

Related Articles