Max Robotics — Global Service Robot Directory + US Certification

About

Junyi Geng is a robotics researcher whose work spans robot learning, autonomous navigation, aerial manipulation, and physics-based optimization. His research bridges deep learning and classical robotics principles, tackling fundamental challenges in robot autonomy across dynamic, real-world environments. Geng's most impactful contribution — "Intention Aware Robot Crowd Navigation with Attention-Based Interaction Graph" (76 citations) — advances safe reinforcement learning-based navigation by modeling nuanced social interactions and pedestrian intent in dense crowds. His co-development of PyPose (35 citations), a library integrating deep learning with physics-based optimization, has provided the robotics community with a powerful open-source tool for perception and state estimation tasks. His aerial manipulation research, including visual servo control for fully-actuated UAVs and the versatile "Flying Hand" teleoperation framework, pushes the boundaries of what autonomous drones can physically accomplish in high-altitude environments. Beyond these flagship works, Geng has explored bio-inspired landing strategies, autonomous exploration in challenging terrains, robotic depowdering for additive manufacturing, and self-supervised neuro-symbolic learning through his Imperative Learning framework. Collectively accumulating over 180 citations, his research reflects a consistent drive to unify data-driven intelligence with physical reasoning — making robots more capable, generalizable, and autonomy-ready in complex, ever-changing settings.