Shumin Feng

Papers

Total Citations

H-Index

About

Shumin Feng’s research lies at the intersection of autonomous navigation, deep reinforcement learning, and modular robotics, with a focus on enabling intelligent, adaptive behavior in mobile robotic systems. Her most impactful work, “A Collision Avoidance Method Based on Deep Reinforcement Learning” (2021, 41 citations), demonstrates how reinforcement learning can solve complex collision avoidance in tight, unknown environments like narrow corridors—a significant step beyond traditional path-planning models. Feng further advanced the field by developing STORM (Self-configurable and Transformable Omni-Directional Robotic Modules), a novel self-reconfigurable modular system capable of autonomous alignment and docking control (2024). This work addresses the critical challenge of module-to-module coordination in reconfigurable robotics. Her earlier research on deep reinforcement learning for mobile robot obstacle avoidance (2019) laid foundational groundwork for these later contributions. By combining theoretical rigor with practical robotic applications, Feng’s work has direct implications for warehouse automation, search-and-rescue operations, and adaptive manufacturing systems. Her growing citation record reflects the increasing relevance of learning-based approaches to real-world robotic navigation and modular system control.