A Minimum-Effort Motion Algorithm for Digital Human Models DaimlerChrysler Corporation

Format:

Conference/Event

Author/Creator:

Gu, Edward, author.

Conference Name:

Digital Human Modeling for Design and Engineering Conference and Exhibition (2003-06-16 : Montreal, Canada)

Language:

English

Physical Description:

1 online resource

Place of Publication:

Warrendale, PA SAE International 2003

Summary:

A new realistic motion control algorithm for digital human models is presented in this paper based on the principle of effort minimization. The proposed algorithm is developed through an innovative mathematical model to make the applications more flexible and more global, especially for the visualization of human motions in automotive assembly operations. The central idea of this unique model is to interpret the solution of the homogeneous Lagrange equation for a mannequin as the origin of dynamic motion. Furthermore, a digital human possesses about 42 joints over the main body except the head, fingers and toes, and offers a large room of kinematic redundancy. We have found 14 new 3-D independent motion markers assigned over the human body to constitute a Cartesian coordinate system, under which a minimum-effort based dynamic control scheme is developed using a state-feedback linearization procedure. The digital human can then be controlled and maneuvered to perform a given desired task while automatically generating realistic postures. The computational complexity is reduced significantly in comparison with the algorithms using joint angles as the generalized coordinates

Notes:

Vendor supplied data

Publisher Number:

2003-01-2228

Access Restriction:

Restricted for use by site license