Knowledge-Level Planning for Robot Task Planning and Human-Robot Interaction

A robot operating in a real-world domain often needs to do so with incomplete information about the state of the world. A robot with the ability to sense the world can also gather information to generate plans with contingencies, allowing it to reason about the outcome of sensed data at plan time. Moreover, the type of information sensed can vary greatly between domains; for instance, domains involving object ma-nipulation may require sensing information on object location, orientation, and distance, while domains requiring human-robot interaction may involve the sensing of social signals like gaze, facial expression, and language. In this poster, we present some of our recent applications of high-level symbolic planning with incomplete information and sensing actions applied to the problems of task planning and interaction in robotics domains. In particular, buildin