Max Robotics — Global Service Robot Directory + US Certification

About

Qiaoyun Wu's research lies at the intersection of robotic navigation, computer vision, and neural learning, with a focus on enabling intelligent agents to perceive and act in complex environments. Her most impactful work, "Towards Target-Driven Visual Navigation in Indoor Scenes via Generative Imitation Learning" (45 citations), pioneered a mapless navigation system that uses multi-view observations and target images to guide robots without relying on odometry or GPS. She further advanced this direction with "Image-Goal Navigation in Complex Environments via Modular Learning" (11 citations), decoupling planning, collision avoidance, and goal prediction for more robust performance. In 6D pose estimation, Wu's "EANet: Edge-Attention 6D Pose Estimation Network for Texture-Less Objects" (37 citations) introduced edge-attention mechanisms to handle challenging lighting, occlusion, and texture-less objects—critical for robotic manipulation. Her earlier work on neural-fuzzy visual servoing (16 citations) and neural network-based vision guided robotics (9 citations) laid foundational approaches for learning inverse Jacobians in feature-based control. Most recently, Wu is exploring energy-efficient 3D point cloud classification with spiking neural networks (7 citations, 2024), pushing the frontier of neuromorphic computing. Her contributions span from classic visual servoing to modern deep learning, consistently addressing real-world robotic challenges with innovative, modular, and data-efficient solutions.