This research project is intended to create methods to automatically compute humanlike motions for virtual mannequins. Such methods are integrated into the motion planning software platform Move3D developed at LAAS. Our goal is to produce and combine different behaviors such as locomotion or manipulation to automatically generate complex animations.
We are interested in extending the autonomy capabilities of a virtual mannequin evolving in a 3D virtual environment with new methods and techniques such as:

• Path and motion planning
• Motion-captured based controllers for human locomotion
• Bio-mechanical approaches
• Manipulation planning algorithms
• Dynamic constraints
• Reactive motion planning
• Coordinated motion animation.

   Potential applications include PLM, video games, movies, human-robot interaction, etc. This research action contributes to European projects Movie and Cogniron.

   The first considered level is control. Walk control aims at automatically providing a natural walking sequence from a given configuration to a given goal. Configurations of the mannequin are in 3-dimensions: two parameters for the position, one for the orientation. Our walking controller is based on motion capture data editing techniques. This method is integrated in a randomized motion planning scheme, including a steering method dedicated for human walk. This means that we integrate the human motion control in the main loop of the path planning algorithm. Here, local human navigation is modeled as a third degree Bézier curve, and a walking path as a B-Spline.

Selected Publications:

"Planning Human Walk in Virtual Environments"    
J. PETTRE, T. SIMEON, J.P. LAUMOND,
IEEE / RSJ International Conference on Intelligent Robots and Systems (IROS'02)

high resolution (6.3Mo)     low resolution (1.4Mo)

   The second level of automation is planning. The main addressed issue is how to automatically compute eye-believable human walking sequences while guaranteeing a 3D obstacle avoidance. The solution is based both, on probabilistic motion planning and on motion capture based blending and warping techniques. Here we develop a modular architecture to generate a sequence of human walking from planning a trajectory in a 3D cluttered environment to synthesizing the motions for the virtual character. Our method has the specificity of considering the 3D model of the environment for collision avoidance with the character's upper body.

Selected Publications:

"A 2-Stages Locomotion Planner for Digital Actors"    
J. PETTRE, J.P. LAUMOND, T. SIMEON,
ACM SIGGRAPH / Eurographics Symposium on Computer Animation (SCA'03)

Walking among sheep:  MPG  MOV Complex goal reaching:  MPG  MOV

   We are interested in imposing manipulation constraints to a walking virtual character. For this we extend our motion planner to allow a digital mannequin to carry a bulky object in a cluttered environment. The approach is based on an analysis of the global task (manipulating while walking) according to three types of constraints:

•    3D obstacle avoidance
•    believable locomotion
•    object manipulation

   To address these constraints altogether we combine three types of techniques within the same framework: probabilistic path planning methods to deal with obstacle avoidance; a motion capture based walking controller to provide believable animations and inverese kinematics techniques to deal with object manipulation.

Selected Publications:

"Planning Fine Motions for a Digital Factotum"    
G. ARECHAVALETA, C. ESTEVES, J.P. LAUMOND,
IEEE / RSJ International Conference on Intelligent Robots and Systems (IROS'04)

Eugene in the Factory Video 1 (2.9Mo) :  MPG  MOV Video 2 (1.2Mo) :  MPG  MOV

"El Mariachi" Video (686Ko) :   MPG  MOV

Eugene delivering pizza Video (642Ko) :  MPG  MOV

   We extend our approach to consider coordinated manipulation among two or more virtual mannequins. Here we model the global task within a single system that gathers all the degrees of freedom of the agents and the object. This system is automatically built by computing a so-called "reachable cooperative space". Coordinated motions are produced by applying an algorithm with three stages:

• Plan a collision free trajectory for a reduced model of the system.
• Animate locomotion and manipulation behaviors independently.
• Tune the generated motions to avoid residual collisions.

   These steps are applied based on a geometric and kinematic decoupling of the system and using different techniques such as path planning, locomotion controllers, inverse kinematics and path planning for closed-kinematic mechanisms.

Selected Publications:

"Motion Planning for Human-Robot Interaction in Manipulation Tasks"    
C. ESTEVES, G. ARECHAVALETA, J.P. LAUMOND,
Special session at the IEEE International Conference on Mechatronics and Automation (ICMA'2005)

"Animation Planning for Virtual Mannequins Cooperation"    
C. ESTEVES, G. ARECHAVALETA, J.PETTRE, J.P. LAUMOND
(Extended version of previous paper)

A workout example:

Planning Stage MPG    MOV

Animation Stage MPG    MOV

Tuning Stage MPG    MOV

Coordinated Manipulation Examples:

In the Factory Video 1 (2.07Mo):  MPG  MOV Video 2 (1.25Mo):  MPG  MOV

"Buren's Columns" Video (5.3Mo):  MPG  MOV

"Piano Mover's Problem" Video (1.1Mo):  MPG  MOV

   One of our current research directions is to apply the above techniques to deal with mechanical part assembly planning. The goal is to automatically compute a collision-free path for both, the part to be assembled and the mannequin manipulating it. Here, we increase the difficulty of the part assembly problem by adding a virtual mannequin into the reasoning loop. Two approaches are proposed according to the difficulty of the problem and are currently being evaluated:

• First planning a path for the part alone and then verify the feasibility of the solution by adding the mannequin.
• Consider the part grasped and the mannequin that manipulates it as a single system.

Disassembling a radiator Video (217Ko):   MPG   MOV

Wind-screen wiper disassembly Video (81 Ko): MPG  MOV