Andrew Haas

University at Albany, State University of New York, Buckingham Browne & Nichols

Papers

Total Citations

H-Index

About

Andrew Haas is a pioneering researcher in artificial intelligence, with a career spanning foundational work in automated planning and situated robotics. His research focuses on the intersection of formal logic, natural language understanding, and autonomous agents. In his seminal 1985 paper, *"Possible events, actual events, and robots"* (19 citations), Haas introduced a novel modal logic to bridge the gap between a robot's reasoning about hypothetical actions and its need to interpret real-world events—a critical step for early AI planning systems. This work established a framework for integrating first-order theorem proving with modal reasoning, influencing subsequent generations of planners. Two decades later, Haas demonstrated the practical application of these ideas in *"Learning to follow navigational route instructions"* (2009, 52 citations). Here, he developed a simulation model that enables a robot to parse unconstrained natural language instructions and navigate to a destination. By segmenting instructions based on required actions and labeling them, his approach tackled the core challenge of grounding language in physical action. This highly cited work remains a touchstone for research in human-robot interaction and instruction following. Haas’s contributions are notable for their rare combination of deep theoretical rigor and direct applicability to embodied AI, marking him as a key figure in the evolution of intelligent robotics.