A Novel Approach For Assessing Power Wheelchair Users' Mobility By Using Curve Fitting

It is important to assess power wheelchair users' mobility characteristics, which can provide insights about an individual's quality of life and health status. For practicality considerations, we propose to use smartphones (particularly, the built-in accelerometer) to collect wheelchair maneuvering data for assessing power wheelchair users' mobility characteristics. However, accelerometer data demonstrates a wide variety of patterns due to significant noise, which makes it difficult to process using traditional methods. To address these challenges, we developed a novel regression-based curve fitting approach that can transform the un-patterned raw sensor data into a sinusoid-like data curve to facilitate the analysis of wheelchair users' mobility. To evaluate the proposed approach, we have conducted a series of experiments in an indoor setting, which contains various types of terrains. Experimental results showed that our approach is promising in achieving accurate analysis of wheelchair users' mobility.