Haofei Lu

Papers

Total Citations

H-Index

About

Haofei Lu is a robotics researcher whose work lies at the intersection of manipulation, tactile sensing, and generative AI. His primary research areas include dexterous grasping, robotic manipulation of deformable objects, and vision-based tactile perception. Lu's most notable contribution is **DexDiffuser** (2024, 24 citations), a pioneering method that leverages diffusion models to generate, evaluate, and refine dexterous grasps from partial point clouds. This work introduces the conditional diffusion-based grasp sampler DexSampler and the evaluator DexEvaluator, enabling robots to achieve high-quality grasps on complex objects—a significant step toward human-like manipulation. In his earlier work (2023, 11 citations), Lu advanced tactile-enabled manipulation by equipping robot grippers with low-cost vision-based tactile sensors. This algorithm allows robots to handle both soft and rigid objects, bridging the gap between fragile item handling and robust grasping. With a focus on integrating tactile feedback with generative models, Lu's research is shaping the future of adaptive, sensor-rich robotic hands. His work is particularly impactful for students and researchers interested in combining perception, learning, and control for real-world robotic dexterity.