Read/Search this Article
Abstract
Model based approaches have been widely used for analyzing walking patterns, wheelchair propulsions, and other human movements. One of the approaches is based on the rigid link model of a particular person and the calculation of its inverse dynamics. Moreover, the measurements of the target movement are used with the model to evaluate the consumption energy and the generated tensions of the muscles. The method is attractive because it is noninvasive and powerful to analyze human movements; on the other hand, it is difficult to apply to design problems of assistive devices or to planning problems of human trainings. Simulation studies which solve forward dynamics of human movements are required to estimate the results of the design or the training. Such simulation techniques are useful because the burden of subjects can be reduced during the design or planning phase. However, simulations of human movements are difficult since the degree-of-freedom of the body is high and the mechanism of brain and nervous system which generates command signals to muscles has not been clarified yet. To cope with this difficulty, an optimization method for the trajectory planning of redundant manipulators is applied to wheelchair propulsion problem which achieves a given task with high efficiency. In the method, genetic algorithm is used to optimize the redundant variables of upper extremity. The advantage is that the procedure does not include any forward dynamics computation ; therefore, the computation for optimizing is not heavy.
Journal
- Journal of the Japan Society for Simulation Technology [List of Volumes]
-
Journal of the Japan Society for Simulation Technology 24(1), 23-31, 2005-03-15 [Table of Contents]
Japan Society for Simmulation Technology