Extraction and Deployment of Human Guidance Policies

Robust and adaptive human motion performance depends on learning, planning, and deploying primitive elements of behavior. Previous work has shown how human motion behavior can be partitioned at subgoal points, and primitive elements extracted as trajectory segments between subgoals. An aggregate set of trajectory segments are described by a spatial cost function and guidance policy. In this paper, Gaussian process regression is used to approximate cost and policy functions extracted from human-generated trajectories. Patterns are identifying in the policy function to further decompose guidance behavior into a sequence of motion primitives. A maneuver automaton model is introduced, simplifying the guidance task over a larger spatial domain. The maneuver automaton and approximated policy functions are then used to generate new trajectories, replicating original human behavior examples.