Max Robotics — Global Service Robot Directory + US Certification

About

Chongkai Gao is a rising star in robotics, whose research is pushing the boundaries of how robots learn and adapt over time. His primary focus lies at the intersection of lifelong robot learning, imitation learning, and general-purpose robotic manipulation. Gao’s most impactful contribution is the **LIBERO benchmark** (2023), which has garnered over 22 citations and has become a standard testbed for evaluating knowledge transfer in lifelong robot learning. This work addresses a critical gap: unlike traditional lifelong learning in image or text domains, robotics requires transferring both declarative and procedural knowledge. His earlier work, **CRIL** (2021, 17 citations), pioneered continual imitation learning by combining generative and prediction models, enabling robots to acquire diverse skills sequentially without catastrophic forgetting. More recently, Gao has tackled foundational challenges in dexterous manipulation, introducing a unified representation for cross-embodiment grasping (2025) and a language-guided framework for deformable object folding (MetaFold, 2025). His work on the **ManiFoundation Model** (2024) aims to create a general-purpose manipulation system akin to LLMs for robotics. With a rapidly growing citation count and a clear trajectory toward building generalist robot agents, Gao is establishing himself as a key figure in the next generation of robotic intelligence.