Max Robotics — Global Service Robot Directory + US Certification

About

Felix Jonathan’s research lies at the intersection of human-robot interaction, collaborative robotics, and learning from demonstration, with a focus on making robots accessible to non-expert users. His most impactful contribution is the CoSTAR system, which integrates behavior trees and vision to enable end users to intuitively instruct collaborative robots for complex tasks. This work, published in 2017, has garnered 170 citations, underscoring its influence in democratizing robot programming. Jonathan’s approach challenges the perceived trade-off between system generalizability and ease of use, as demonstrated through user studies that validate CoSTAR’s effectiveness for novice operators. In parallel, his work on imitation learning—planning sequences of actions from expert demonstrations—advances the field by grounding high-level actions in new environments via sampling-based motion planning. More recently, he has explored multi-object detection and classification using machine learning with robotic manipulators, broadening the scope of autonomous perception. Jonathan’s research is notable for its practical, user-centered design, bridging the gap between advanced robotics and real-world application. His contributions are essential reading for anyone interested in building robots that truly collaborate with people.