Planning Motions and Placements for Virtual Demonstrators

In order to deliver information effectively, virtual human demonstrators must be able to address complex spatial constraints and at the same time replicate motion coordination patterns observed in human-human interactions. We introduce in this paper a whole-body motion planning and synthesis framework that coordinates locomotion, body positioning, action execution and gaze behavior for generic demonstration tasks among obstacles. Human-like solutions are achieved with a coordination model extracted from experiments with human subjects. Given an observer location and a target demonstration to be performed, the proposed planner automatically identifies body placements respecting visibility constraints, locomotion accessibility, and action feasibility among obstacles. Actions are modeled with clusters of example motions and a fast collision avoidance procedure in blending space is introduced to avoid nearby obstacles when needed. Locomotion towards new placements integrates planning among obstacles and is based on a motion capture database organized for efficient synthesis of motions with precise path following and arrival constraints. The proposed solution introduces effective approaches for modeling and solving complex demonstrative tasks for interactive applications.