Max Robotics — Global Service Robot Directory + US Certification

About

Michael Burke is a robotics researcher whose work spans robot learning, manipulation, human-robot interaction, and medical robotics. His research is particularly distinguished by innovative applications of Dynamic Movement Primitives (DMPs), a framework he has extended substantially to address contact-rich manipulation tasks such as peg-in-hole insertions and soft-tissue cutting. His 2022 paper on residual learning from demonstration (48 citations) exemplifies this contribution, demonstrating how robots can adapt learned motor primitives to handle the complexities of friction and contact. Burke has also made notable advances in reward learning and imitation learning, developing probabilistic and disentangled approaches that reduce the need for large demonstration datasets. His work on intelligent robotic sonography (31 citations) showcases a compelling application of these methods to autonomous ultrasound imaging, directly addressing real-world clinical challenges posed by operator variability. Earlier contributions to gesture-based human-robot interaction (22 citations) reflect the breadth of his research interests. Across his body of work, Burke consistently bridges theoretical machine learning with practical robotics, producing methods applicable to surgical robotics, aerial systems, and industrial automation, making his research valuable to both academic and applied audiences.