Max Robotics — Global Service Robot Directory + US Certification

About

Leslie Pack Kaelbling is a pioneering researcher at the intersection of robotics, reinforcement learning, and artificial intelligence, whose decades of work have fundamentally shaped how autonomous systems perceive, learn, and act in uncertain environments. Her early contributions — particularly *Learning in Embedded Systems* (1993, 586 citations) — established foundational frameworks for adaptive behavior in robots and intelligent controllers, while her 1995 work on scaling reinforcement learning to partially observable environments (662 citations) remains a landmark in the field. Kaelbling has made enduring contributions to probabilistic robotics, demonstrating how discrete Bayesian models and partially observable Markov decision processes can guide principled decision-making under uncertainty (468 citations). Her research progressively tackled increasingly complex challenges: from practical reinforcement learning in continuous spaces to integrated task-and-motion planning in belief space, bridging low-level perception with high-level symbolic reasoning. Her 2018 paper on learning symbolic representations for abstract planning (244 citations) and the PDDLStream framework (174 citations) exemplify her commitment to unifying classical AI planning with modern machine learning. Across her career, Kaelbling has produced work that is both theoretically rigorous and practically impactful, making her one of the most influential figures in autonomous robotics research.