Ke Lin

Harbin Institute of Technology, Guangdong Institute of Intelligent Manufacturing, Shenzhen Institute of Information Technology

Papers

Total Citations

H-Index

About

Ke Lin is a rising researcher in artificial intelligence and robotics, whose work bridges the gap between classical planning and modern reinforcement learning (RL). His primary research areas include multi-agent pathfinding (MAPF), safe and agile quadrotor control, and sample-efficient deep RL. Lin’s major contributions are highlighted by his comprehensive 2023 review of graph-based MAPF solvers, which has garnered 34 citations and serves as a key reference for navigating multi-robot systems from classical to beyond-classical approaches. He has also advanced the frontier of aerial robotics with his 2024 work on learning agile quadrotor flight in restricted environments, providing safety guarantees for RL-based controllers—a critical achievement for real-world deployment. Demonstrating versatility, Lin applied average-reward RL to battery management for warehouse robots, optimizing operational efficiency. His most recent 2025 paper on sample-efficient backtrack temporal difference deep RL (18 citations) further underscores his commitment to improving RL’s practicality. With a growing citation impact and a focus on safety and efficiency, Ke Lin is shaping the future of autonomous multi-robot systems and intelligent control.