Max Robotics — Global Service Robot Directory + US Certification

About

Chad Esselink is an emerging researcher specializing in robot task planning, open-world reasoning, and the integration of large language models (LLMs) with classical AI systems. His work addresses one of robotics' most persistent challenges: enabling robots to operate effectively in dynamic, unpredictable real-world environments rather than controlled, closed-world settings where perfect perception and execution are assumed. Esselink's most impactful contribution, "Integrating Action Knowledge and LLMs for Task Planning and Situation Handling in Open Worlds" (2023), has garnered 53 citations, reflecting strong community interest in his approach to bridging symbolic planning with modern neural methods. His research explores how vision-language models (VLMs) can be grounded to support classical task planners, and how domain knowledge prompting can enhance VLM performance for open-world scenarios — themes developed across multiple publications between 2022 and 2024. A consistent thread throughout his portfolio is making robotic systems more robust and adaptable by combining structured human knowledge with the flexible reasoning capabilities of LLMs and VLMs. For students and researchers working at the intersection of robotics, AI planning, and natural language processing, Esselink's work offers valuable frameworks for tackling perception uncertainty and real-world situational complexity.