The Visual Computer

, Volume 21, Issue 5, pp 289–302

Simulating the task-level control of human motion: a methodology and framework for implementation

  • Vincent De Sapio
  • James Warren
  • Oussama Khatib
  • Scott Delp
original article

DOI: 10.1007/s00371-005-0284-4

Cite this article as:
De Sapio, V., Warren, J., Khatib, O. et al. Vis Comput (2005) 21: 289. doi:10.1007/s00371-005-0284-4

Abstract

A task-level control framework is proposed for providing feedback control in the simulation of goal-directed human motion. An operational space approach, adapted from the field of robotics, is used for this purpose. This approach is augmented by a significant new extension directed at addressing the control of muscle-driven systems. Task/posture decomposition is intrinsically exploited, allowing human musculoskeletal properties to direct postural behavior during the performance of a task. This paper also describes a simulation architecture for generating musculoskeletal simulations of human characters. The evolving capabilities of the collective environment are directed toward autonomously generating realistic motion control for virtual actors in interactive computer graphics applications, as well as synthesizing the control of human-like motion in robotic systems.

Copyright information

© Springer-Verlag 2005

Authors and Affiliations

  • Vincent De Sapio
    • 1
  • James Warren
    • 1
  • Oussama Khatib
    • 1
  • Scott Delp
    • 2
  1. 1.Artificial Intelligence Laboratory, Computer Science DepartmentStanford UniversityStanfordUSA
  2. 2.Neuromuscular Biomechanics Laboratory, Mechanical Engineering and Bioengineering DepartmentsStanford UniversityStanfordUSA

Personalised recommendations