Haoqiang Fan
Papers
1
Total Citations
9
H-Index
1
About
Haoqiang Fan is an emerging researcher at the intersection of robot learning, imitation learning, and generative modeling, with a particular focus on advancing policy learning for robotic manipulation. His most notable work, *FlowPolicy* (2025), represents a significant contribution to the field by addressing critical limitations in existing diffusion and flow matching-based policy generation methods. By introducing Consistency Flow Matching, Fan developed a framework that enables robots to acquire complex manipulation skills from expert demonstrations with substantially improved speed and robustness — two long-standing challenges in vision-based imitation learning that have hindered real-world deployment. With 9 citations already accumulated shortly after publication, *FlowPolicy* has attracted early attention from the robotics and machine learning communities, signaling its relevance to ongoing efforts to make robot learning more practical and scalable. Fan's research sits at a timely crossroads, as the field increasingly turns to generative models to bridge the gap between demonstration data and deployable robotic behavior. His work contributes meaningful progress toward robots that can reliably learn from human demonstrations and operate effectively in complex, real-world manipulation scenarios.
Research Focus
Key Achievements
Top Papers
- 1