Max Robotics — Global Service Robot Directory + US Certification

About

Gerald DeJong is a pioneering figure in artificial intelligence and robotics, best known for his foundational work in explanation-based learning and planning under uncertainty. His research bridges the gap between classical planning and real-world robotic execution, focusing on how machines can acquire planning abilities through observation and incremental learning. DeJong’s most cited paper, “Explanation-based manipulator learning: Acquisition of planning ability through observation” (2005, 44 citations), introduces a robot system that improves its problem-solving by observing command sequences, enabling it to tackle tasks beyond its initial capabilities. This work, along with “Real-world robotics: Learning to plan for robust execution” (1996, 19 citations), demonstrates his commitment to making planning systems resilient in uncertain domains. DeJong also developed the concept of permissive planning, which extends classical planning to handle complex, uncertain tasks, as seen in his 1997 paper (5 citations). His incremental learning approach for completable planning (1994, 6 citations) further advances adaptive robotics. With a career spanning decades, DeJong’s contributions have shaped how robots learn from human demonstration and plan for robust, real-world performance, influencing both AI theory and practical robotics.