Team Bunraku

Members
Overall Objectives
Scientific Foundations
Application Domains
Software
New Results
Contracts and Grants with Industry
Other Grants and Activities
Dissemination
Bibliography

Section: Scientific Foundations

Virtual humans


Avatar: 

it is the representation of the user in the virtual world. This representation can be either anthropomorphic or metaphoric.

Autonomous Agent: 

An autonomous agent is a virtual human situated within and a part of an environment that senses that environment and acts on it, over time, in pursuit of its own agenda and so as to effect what it senses in the future.


Human motion is a very challenging field because it is the result of numerous complex processes partially studied in biomechanics, neuroscience or physiology. One of the main outcomes is to understand the laws that capture the naturalness of human motion: what is the role of physical laws and of neurological and physiological processes? For example, neuroscientists studied single arm reaching tasks and suggested that the central nervous system uses an optimality criterion called “minimum Jerk” to calculate the trajectory along which to move:

Im19 $\mtable{...}$(6)

where n is the number of degrees of freedom and $ \theta$i is the ith degree of freedom. In addition to general mechanical laws, many other criteria were proposed in the literature but they are all subject to controversy because they are linked to dedicated protocols that are far from real natural behaviors. Coupling motion analysis and simulation is a promising issue for understanding the subtle relations between all the parameters and laws involved in natural motion control. In computer animation, inverse kinematics is often used to calculate the relations between angular trajectories and Cartesian constraints (such as commanding the position of a given point of the skeleton, including the character's center of mass):

Im20 $\mtable{...}$(7)

where J is the Jacobian of the kinematic function returning the position of the task X according to the angles $ \theta$ , $ \alpha$ is a weight, I is the identity matrix and z is called secondary task. This secondary task can embed the laws suggested in human movement sciences and allows evaluating their effect on the resulting calculated motion. Synthesizing a controller for a human-like dynamic system could also be considered as inverting the function linking the forces and the motion while minimizing a set of criteria intrinsically dealing with generic mechanical laws.

Modeling the human behavior requests to take into account a certain number of topics such as understanding mechanisms underlying functions such as natural language production and perception, memory activity, multi-sensory perception, muscular control and last but not least the production of emotions.

Our problem is not to reproduce the human intelligence but to propose an architecture making it possible to model credible behaviors of anthropomorphic virtual actors evolving/moving in real time in virtual worlds. The latter can represent particular situations studied by psychologists of the behavior or to correspond to an imaginary universe described by a scenario writer. The proposed architecture should mimic all the human intellectual and physical functions.


previous
next

Logo Inria