Yunshan Deng

Papers

Total Citations

H-Index

About

Yunshan Deng’s research lies at the intersection of safe reinforcement learning, control barrier functions, and multi-robot coordination, with a focus on ensuring safety and feasibility in real-time autonomous systems. Deng’s most impactful work introduces a differential high-order control barrier function (HOCBF) framework for safe RL, which enforces safety constraints through a quadratic programming-based filter—a contribution that has already garnered early citations and promises to shape the field of safety-critical learning. In parallel, Deng has advanced formation control for non-holonomic mobile robots by proposing a tangential-force-based collision avoidance method that prevents local infeasibility, a practical solution for multi-robot teams. More recently, Deng developed a convex model predictive control (MPC) scheme for affine systems that eliminates the need for terminal components, dramatically improving real-time control for unreachable setpoints. With a total of four citations across three highly targeted papers, Deng’s work is gaining traction for its theoretical rigor and practical applicability. These contributions position Deng as an emerging voice in safe autonomy, with clear potential for high-impact advances in robotics and control theory.