Papers
6
Total Citations
86
H-Index
5
About
Dezhi Yu is an emerging researcher specializing in intelligent robotics, autonomous navigation, and machine learning applications for warehouse automation. His work sits at the intersection of reinforcement learning and practical logistics systems, addressing real-world challenges in modern automated warehouses. Yu's most significant contribution is the development of the Proximal Policy-Dijkstra (PP-D) algorithm, a novel hybrid approach combining Proximal Policy Optimization with Dijkstra's classical pathfinding method to enable efficient, real-time decision-making in complex warehouse layouts — a paper that has already garnered 31 citations since its 2024 publication. Complementing this, his research on deep reinforcement learning for obstacle avoidance introduces an improved value function network that accounts for pedestrian interactions and historical state importance, earning 17 citations and demonstrating practical safety considerations for human-robot shared environments. Yu has also made notable strides in optimizing picking and packing operations through machine learning-enhanced control systems, with related publications accumulating over 30 citations collectively. With multiple highly cited works published within just two years, Dezhi Yu is rapidly establishing himself as a promising voice in intelligent warehouse robotics, offering solutions with direct industrial applicability.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6