Connor Settle

Papers

Total Citations

H-Index

About

Connor Settle is a researcher specializing in robot learning from human demonstrations, with a particular focus on periodic and rhythmic tasks. His major contribution lies in developing methods that enable robots to learn and replicate cyclic movements—such as sweeping, polishing, or assembly motions—directly from visual examples provided by humans. Settle’s work introduces the use of rhythmic dynamic movement primitives (rDMPs) combined with an active learning framework to efficiently optimize policy parameters, allowing robots to adapt to the inherent periodicity of real-world tasks. His most cited paper, "Learning Periodic Tasks from Human Demonstrations" (2022), has garnered 20 citations, reflecting its significance in the field of imitation learning and human-robot interaction. By proposing a structured objective for active learning, Settle’s research reduces the number of demonstrations needed while improving task accuracy, making robot teaching more accessible. His contributions are particularly impactful for applications in manufacturing, healthcare, and domestic robotics, where repetitive, precise motions are essential. Settle’s work bridges the gap between human intuition and robotic precision, advancing the frontier of intuitive robot programming.