A Knowledge Rich Task Planning Framework for Human-Robot Collaboration

In this paper, we position ourselves within the context of collaborative planning. Drawing upon our recent research, we introduce a novel, knowledge rich task planning framework that represents and reasons and effectively addresses agents' first-order false beliefs to solving a task planning problem with a shared goal. Our contributions to this work are as follows: First, to enhance the reasoning abilities of an existing framework, an intuitive observability model for situation assessment is addressed, and for that, an improved knowledge modeling is considered. This effectively captures agents' predictable false beliefs and motivates us to exploit the power of off-the-shelf knowledge reasoners. Second, a new planning approach that incorporates this improved encoding. And, to show the effectiveness of our planner and present our initial findings and proof of concept, we conduct a thorough use case analysis.