Papers
2
Total Citations
11
H-Index
2
About
Jia Bi is a researcher whose work bridges the critical intersection of motion planning, robotics, and artificial intelligence, with a particular focus on enabling autonomous systems to navigate complex, dynamic environments. Bi’s major contributions lie in advancing path planning algorithms, notably through the integration of deep reinforcement learning (DRL) with formal methods. Their 2022 paper, “GR(1)-Guided Deep Reinforcement Learning for Multi-Task Motion Planning under a Stochastic Environment,” which has garnered 8 citations, introduces a novel framework that combines reactive synthesis with DRL to solve multi-task motion planning problems under uncertainty. This work addresses a key limitation of standard DRL approaches—sparse reward signals—by using formal specifications to guide the learning process, thereby improving both efficiency and safety in stochastic settings. Earlier in their career, Bi contributed foundational work on soccer robot path planning, as seen in their 2010 paper (3 citations), which proposed an improved grid-based potential field method for dynamic, multi-agent environments. By fusing grid-based environmental representation with potential field navigation, this work provided a practical solution for real-time robot coordination. Bi’s research is notable for its applied focus on real-world robotic systems, offering scalable solutions for autonomous navigation in unpredictable settings.
Research Focus
Key Achievements
Top Papers
- 1
- 2