Papers
15
Total Citations
89
H-Index
4
About
Yaran Chen is a researcher specializing in embodied AI, visual navigation, deep learning, and robotic decision-making — a body of work that spans from efficient object detection to cutting-edge large language model integration for autonomous agents. His most-cited contribution, "MGRL: Graph Neural Network Based Inference in a Markov Network with Reinforcement Learning for Visual Navigation" (2020, 34 citations), established him as a serious voice in graph-based reinforcement learning for robot navigation. His early work on hybrid deep learning for moving object detection (2018) demonstrated a keen interest in real-time, resource-efficient perception systems, a theme that continued through his network pruning research with ABCP (2022), which addressed the critical challenge of deploying complex models on constrained hardware. More recently, Chen has pushed into the frontier of LLM-driven robotics, with RoboGPT (2025) exploring long-horizon task planning for embodied agents following natural language instructions. His NeuronsGym framework and the Neurons Perception Dataset reflect a commitment to building rigorous benchmarks that bridge simulation and real-world robot deployment. Collectively, his work represents a coherent vision: making intelligent, adaptive robots that can perceive, plan, and act reliably in complex, unstructured environments.
Research Focus
Key Achievements
Top Papers
- 1
- 2Hybrid Deep Learning Based Moving Object Detection via Motion prediction11 citations · 2018
- 3
- 4
- 5Neurons Perception Dataset for RoboMaster AI Challenge4 citations · 2022
- 6
- 7
- 8
- 9
- 10