A control-oriented dynamic model for sit-to-stand motion with fixed support. (June 2018)

• Record Type: Journal Article
• Authors: Taghvaei, Sajjad; Tavasoli, Ali; Feizi, Navid; Rajestari, Zahra; Abdi, Mohammad
• Is Part Of: Proceedings of the Institution of Mechanical Engineers. Volume 232:Number 2(2018:Jun.)
• Page Start: 265
• Page End: 273
• Publication Date: 2018-06
• Subjects: Sit-to-stand -- human multi-body dynamics -- optimal control -- embedded controller -- Sit-to-stand fixed support; Machinery, Dynamics of -- Periodicals; Machine design -- Periodicals; 621.815
• DOI: 10.1177/1464419317731059
• Languages: English
• ISSNs: 1464-4193

Abstract

Establishing a validated dynamic model for human daily life movements is a crucial step in designing assistive devices for people with disabilities. Sit-to-stand motion with fixed support is a frequent task in daily life, and in order to provide assistive devices for such task, a validated dynamic model is required. The main contribution of this paper is to present a multi-body dynamic model of sit-to-stand motion with fixed support and to validate the model with experimental results from human joint tracking and force sensors on the support. Dynamic equations governing the body motion during sit-to-stand are derived using a multi-body model interconnected to the fixed support through handling muscle force. It is hypothesized that sit-to-stand motion is performed by optimized support force and control effort by the subject. In this regard, the human body attempt for sit-to-stand motion is distributed among various joints such that an energy-based cost function is optimized, and at the same time, the choice of biomechanical actuation is prioritized through adjusting various weighting coefficients in the objective function. The limitations on joint torques and the hand force are included as inequality constraints in the optimization problem. The developed multi-body dynamic model is evaluated through comparison between the computed support force (theoretical) and the measured force (experimental) obtained from a fixed support equipped with three-axis force sensors. The proposed dynamic model uses the visual data that are recorded to track the skeleton configuration of the subjects as the desired trajectory to follow by the controller. A set of experimental data from five healthy subjects is obtained, and the results show acceptable agreement between simulation and experimental results. The results suggest energy-based optimization solution can be considered as a valid approach is multi-body dynamic modelling of sit-to-stand task.