Deployment of the Saddle Space Transformation in Tracking the Base of Support
This work addresses balance impairment issues, particularly for elderly populations at risk of falls, by improving BoS tracking accuracy, though it appears incremental as it builds on existing BoS methods.
The paper tackles the problem of accurately tracking the Base of Support (BoS) for bipedal stability assessment by proposing a method that uses a reference frame to define postural models of BoS geometry, tested on 12 healthy subjects in six postures, showing it can deform the BoS shape to adapt to different postures and potentially remove the need for force/torque sensors.
Balance is the fundamental skill behind human locomotion, and its impairment is the principal indicator of self-perceived disability. Despite significant improvements in balance assessment, there is still large incidence of fall related injuries among elderlies. The Base of Support (BoS) is a popular method for bipedal stability assessment, but its accuracy depends on the accuracy the BoS geometry measurement. This work presents a method to ease the BoS tracking by the identification of a reference frame that allows to define postural models of the BoS geometry. Although we also propose a geometry based on the geometry determined from centre of pressure range of motion within the foot obtained from literature, this methodology can be used with other models (i.e., the feasible base of support). The model has been tested with 12 healthy subjects, which have been asked to explore their stability in six different postures. The results show that the model can accurate deform the geometry of the BoS to adapt its shape to the different postures, which can remove the necessity of force/torque sensors in some application. Potentially the proposed method can be also applied to describe any posture dependent attribute (e.g., gravitational forces), and it can be also applied to bipedal robots. Therefore, it constitutes a novel mathematical tool that can be deployed to develop both better sensors and models for bipeds. For example, it can be used with the Extrapolated CoM model to evaluate dynamic stability from the body kinematics.