Synthesizing animations of human manipulation tasks
ACM SIGGRAPH 2004 Papers
Hierarchical motion controllers for real-time autonomous virtual humans
Lecture Notes in Computer Science
Animation planning for virtual characters cooperation
ACM Transactions on Graphics (TOG)
Interactive motion correction and object manipulation
Proceedings of the 2007 symposium on Interactive 3D graphics and games
Interactive Continuous Collision Detection Using Swept Volume for Avatars
Presence: Teleoperators and Virtual Environments
Animation planning for virtual characters cooperation
ACM SIGGRAPH 2008 classes
Interactive motion correction and object manipulation
ACM SIGGRAPH 2008 classes
SmartBody: behavior realization for embodied conversational agents
Proceedings of the 7th international joint conference on Autonomous agents and multiagent systems - Volume 1
Sampling-based algorithms for optimal motion planning
International Journal of Robotics Research
Synthesis of concurrent object manipulation tasks
ACM Transactions on Graphics (TOG) - Proceedings of ACM SIGGRAPH Asia 2012
ADAPT: the agent development and prototyping testbed
Proceedings of the ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games
Hi-index | 0.00 |
We present a heuristic-based real-time reach planningalgorithm for virtual human figures. Given the start andgoal positions in a 3D workspace, our problem is tocompute a collision-free path that specifies all theconfigurations for a human arm to move from the start tothe goal. Our algorithm consists of three modules: spatialsearch, inverse kinematics, and collision detection. Forthe search module, instead of searching in jointconfiguration space like most existing motion planningmethods do, we run a direct search in the workspace,guided by a heuristic distance-to-goal evaluationfunction. The inverse kinematics module attempts toselect natural posture configurations for the arm alongthe path found in the workspace. During the search,candidate configurations will be checked for collisionstaking advantage of the graphics hardware - depthbuffer. The algorithm is fast and easy to implement. Itallows real-time planning not only in static, structuredenvironments, but also in dynamic, unstructuredenvironments. No preprocessing and prior knowledgeabout the environment is required. Several examples areshown illustrating the competence of the planner atgenerating motion plans for a typical human arm modelwith seven degrees of freedom.