Papers
10
Total Citations
488
H-Index
8
About
Ziyi Yan is a robotics and computational intelligence researcher whose work centers on the motion planning and control of redundant robot manipulators, with a particular emphasis on neural network-based optimization methods. His research addresses some of the most persistent challenges in robot kinematics, including joint-angular drift, nonrepetitive motion, obstacle avoidance, and fault tolerance — problems that, if left unsolved, can compromise task execution accuracy or even cause physical damage to robotic systems. Yan's most influential contributions involve the development of innovative recurrent neural network architectures tailored to real-time quadratic programming problems. His varying-parameter convergent-differential neural network (VP-CDNN), introduced in 2018, has garnered 124 citations and established a new benchmark for addressing joint-angular-drift in redundant manipulators. Building on this, his adaptive fuzzy recurrent neural network (AFRNN), with 114 citations, further reduced end-effector position errors in nonrepetitive motion scenarios. Collectively, his top publications have accumulated over 480 citations, reflecting strong community recognition. Beyond theoretical contributions, Yan has also advanced practical robotics applications, including real-time whole-body imitation and teleoperation for humanoid robots using analytical mapping methods. His body of work makes him a significant voice in intelligent robot control and neural computation research.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10