Peifa Jia
Papers
13
Total Citations
66
H-Index
4
About
Peifa Jia is a robotics and intelligent systems researcher whose work spans biped locomotion, robotic manufacturing, and formal modeling of complex systems. His most recognized contribution lies in applying machine learning to challenging real-world robotics problems: his 2007 work on reinforcement learning-based dynamic walking control (17 citations) demonstrated how quasi-passive biped robots actuated by compliant MACCEPA actuators could achieve more robust and energy-efficient gaits through adaptive learning. In robotic manufacturing, Jia has made notable strides in belt grinding automation — a field critical for replacing human workers in hazardous environments — leveraging support vector machines, echo state networks, and particle swarm optimization to build accurate process models and optimize grinding parameters across multiple studies. His early work also includes pioneering China's first SERCOS-based industrial robot for shipbuilding applications. Beyond control systems, Jia has contributed to formal methods through his development of Fuzzy Timed Object-Oriented Petri Nets and their extensions for modeling multi-robot cooperative systems, providing rigorous frameworks for analyzing timing and uncertainty in dynamic environments. Collectively, his research bridges intelligent control, advanced manufacturing automation, and formal systems modeling, offering practical tools for next-generation robotic applications.
Research Focus
Key Achievements
Top Papers
- 1A Reinforcement Learning Based Dynamic Walking Control17 citations · 2007
- 2
- 3An adaptive modeling approach based on ESN for robotic belt grinding7 citations · 2010
- 4
- 5An open industrial robot based on SERCOS4 citations · 2003
- 6RTOC: A Rt-Linux Based Open Robot Controller4 citations · 2006
- 7Fuzzy Timed Object-Oriented Petri Net4 citations · 2006
- 8
- 9
- 10