Efficient synthesis of physically valid human motion

Fang, Anthony; Pollard, Nancy S.

doi:10.1145/1201775.882286

Cited by 110 publications

(92 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Minimizing (20) with a discrete variational principle leads to an implicit difference equation known as the discrete Euler-Lagrange (DEL) equation 3 :…”

Section: Discrete Lagrangian Dynamicsmentioning

confidence: 99%

See 1 more Smart Citation

Discrete and continuous mechanics for tree representations of mechanical systems

Johnson

Murphey

2008

2008 IEEE International Conference on Robotics and Automation

View full text Add to dashboard Cite

We use a tree-based structure to represent mechanical systems comprising interconnected rigid bodies. Using this representation, we derive a simple algorithm to numerically calculate forward kinematic maps, body velocities, and their derivatives. The algorithm is computationally efficient and scales to large systems very well by using recursion to take advantage of the tree structure. Moreover, this method is less prone to modeling errors because each element of the graph is simple.The tree representation provides a natural framework to simulate mechanical dynamics with numeric computations rather than large symbolically-derived equations. In particular, the representation allows one to simulate systems in generalized coordinates using Lagrangian dynamics without symbolically finding the equations of motion. This method also applies to the relatively new variational integrators which numerically integrate dynamics in a way that preserve momentum and other symmetries. We show how to implement both integration schemes for an arbitrary system of interconnected rigid bodies in a computationally efficient way while avoiding symbolic equations of motion. We end with an example simulating a marionette; a mechanically complex, high degree-of-freedom system. I. INTRODUCTIONEuler-Lagrange simulations are often preferred for robotics and controls applications because they work directly in the generalized coordinates used to analyze the system and specify desired trajectories. Moreover, they provide a certain level of automation to derive correct equations of motion. For complex mechanical systems, however, the necessary symbolic equations become unwieldy and the resulting simulations are typically slow.The popular alternative to Euler-Lagrange simulation is the Newton-Euler force balance approach. Newton-Euler simulations have a larger configuration space than EulerLagrange simulations (typically (R 3 × SO(3)) n vs. generalized coordinates) and hide the mechanical structure in the constraint definitions. However, there are extremely fast and efficient implementations [1] [13]. Much of the performance comes from working with simple, general equations that can be efficiently implemented and evaluated rather than deriving large symbolic equations of motion. These smaller equations can be efficiently implemented and optimized by compilers.Euler-Lagrange simulations [2] [9], on the other hand, typically rely on symbolic equations of motions that are numerically integrated. Our approach avoids such symbolic

show abstract

“…Minimizing (20) with a discrete variational principle leads to an implicit difference equation known as the discrete Euler-Lagrange (DEL) equation 3 :…”

Section: Discrete Lagrangian Dynamicsmentioning

confidence: 99%

“…They are used extensively in motion capture and computer animation and have been applied to dynamics [10] [3]. The use of recursion greatly improves the simulation's efficiency [4].…”

Section: Introductionmentioning

confidence: 99%

Discrete and continuous mechanics for tree representations of mechanical systems

Johnson

Murphey

2008

2008 IEEE International Conference on Robotics and Automation

View full text Add to dashboard Cite

show abstract

“…In [26] the estimation of a character's motion is cast as an optimization problem in a low-dimensional subspace, but one limitation of the method is that for defining this subspace a suitable set of motions must be selected by the user. Fang and Pollard [9] also use optimization as a way to generate new animations and, for this purpose, they consider certain physics constraints which are shown to lead to fast optimization algorithms. In [33], Yamane et al make use of a motion capture database to synthesize animations of a human manipulating an object and they also provide a planning algorithm for generating the object's collision-free path.…”

Section: Related Workmentioning

confidence: 99%

3D visual reconstruction of large scale natural sites and their fauna

Komodakis

Panagiotakis

Tziritas

2005

Signal Processing: Image Communication

View full text Add to dashboard Cite

“…The first one consists in optimizing the joint angles and torques until a correct motion is obtained [12], [13]. Some authors have proposed to control simpler dynamic constraints (such as the linear and angular momentum [14] or the aggregate-force constraints [15]) instead of joint torques leading to lower computation time. However, these methods are clearly designed for off-line simulation.…”

Section: Introductionmentioning

confidence: 99%

Dynamic motion adaptation for 3D acrobatic humanoids

Multon

Hoyet

Komura

et al. 2007

2007 7th IEEE-RAS International Conference on Humanoid Robots

View full text Add to dashboard Cite

Low-cost dynamics is a key issue in planning complex motions. This paper aims at proposing a fast algorithm in order to simulate aerial motions for humanoids while using motion capture data. As the real subject has obviously different anthropometric parameters than the synthetic robot, directly applying motion capture data leads to a non-respect of physical laws. As a consequence, the humanoid can land in a bad configuration. During aerial phase, the acceleration of the center of mass may be different from gravity and the angular momentum may change incorrectly. This so-called retargetting problem is generally addressed in computer animation by solving kinematic constraints. In this paper, we propose a dynamic filter that is able to rapidly adapt motion in order to preserve its physical correctness. If the original motion is changed by a user or just replayed on a different skeleton, the filter is about to correct the postures at each time step in less than 2ms in a common computer. It's consequently possible to change an arm motion during the aerial phase and to immediately check its effect on the global rotation of the humanoid.

show abstract

Efficient synthesis of physically valid human motion

Cited by 110 publications

References 20 publications

Discrete and continuous mechanics for tree representations of mechanical systems

Discrete and continuous mechanics for tree representations of mechanical systems

3D visual reconstruction of large scale natural sites and their fauna

Dynamic motion adaptation for 3D acrobatic humanoids

Contact Info

Product

Resources

About