Dingchang Hu

Papers

Total Citations

H-Index

About

Dingchang Hu is a robotics researcher focused on advancing robotic manipulation through reinforcement learning, affordance reasoning, and multimodal human-robot interaction. His work addresses critical challenges in enabling robots to grasp and interact with objects in dynamic, unstructured environments. Hu’s most cited paper, "GAP-RL: Grasps as Points for RL Towards Dynamic Object Grasping" (2024, 5 citations), introduces a novel reinforcement learning framework that treats grasps as points, allowing robots to robustly grasp moving objects in continuous motion—a significant leap over prior static-grasping methods. In "Variation-Robust Few-Shot 3D Affordance Segmentation for Robotic Manipulation" (2025, 3 citations), he pioneers a few-shot learning approach that enables robots to segment functional parts of novel 3D objects with minimal training data, overcoming the limitations of traditional affordance models. His work "Target-Oriented Object Grasping via Multimodal Human Guidance" (2025, 2 citations) further explores how combining visual and language cues can improve grasping precision. With a growing citation impact, Hu’s research is shaping the next generation of adaptive, learning-driven robotic systems capable of operating in real-world, dynamic settings.