Xinyang Wu

About

Xinyang Wu is a robotics researcher whose work focuses on the intersection of safety, learning efficiency, and intelligent automation in human-robot collaboration (HRC) and robotic systems. His key research areas include deep reinforcement learning (RL), Bayesian neural networks, and additive manufacturing for robotics. Wu’s most impactful contribution, "Towards Safe Human-Robot Collaboration Using Deep Reinforcement Learning" (2020, 62 citations), addresses a critical bottleneck in industrial HRC: the trade-off between safety and productivity. By proposing a deep RL framework that dynamically balances hazard mitigation with task performance, he offers a path to reduce costly, over-emphasized safety measures and lengthy risk assessments during layout reconfigurations. In his more recent work, "Uncertainty-Guided Active Reinforcement Learning with Bayesian Neural Networks" (2023, 5 citations), Wu tackles the dual challenges of low learning efficiency and safety in model-free RL, introducing an uncertainty-aware approach to accelerate training while maintaining robustness. Additionally, his design of a 3D-printed intelligent soccer robot match system (2020, 2 citations) demonstrates his versatility, integrating open-source hardware like Arduino with 3D printing for accessible, customizable robotics. Wu’s research is shaping safer, more efficient autonomous systems for real-world industrial and educational applications.

Research Focus

Key Achievements

Top Papers

1. 1
   Towards Safe Human-Robot Collaboration Using Deep Reinforcement Learning

   62 citations · 2020
2. 2
   Uncertainty-Guided Active Reinforcement Learning with Bayesian Neural Networks

   5 citations · 2023
3. 3
   3D Printing Intelligent Soccer Robot Match System

   2 citations · 2020

Key Collaborators

Contact & Links