Papers
15
Total Citations
254
H-Index
8
About
Wanxin Jin is a robotics and control researcher whose work sits at the intersection of optimal control theory, robot learning, and human-robot interaction. He is best known for pioneering advances in **inverse optimal control (IOC)**, developing mathematical frameworks that allow autonomous systems to infer underlying objective functions from observed behavior — even when that behavior is incomplete or sparsely sampled. His foundational contributions in this area, spanning from early work on incomplete trajectory observations (2018) to multiphase cost functions (2019), have collectively garnered over 120 citations and established him as a leading voice in the field. Jin's landmark **Pontryagin Differentiable Programming (PDP)** framework provides a unified end-to-end approach to learning and control, elegantly bridging classical optimal control theory with modern machine learning. Extending this, his Continuous PDP method enables robots to learn from just a handful of demonstrated keyframes — a significant practical breakthrough. His broader portfolio also addresses human-guided robot learning, safe control under uncertainty using control barrier functions, dexterous manipulation, and visuo-tactile sensing for legged robots. Together, his contributions reflect a coherent vision: building robots that learn efficiently, safely, and collaboratively from human guidance and real-world interaction.
Research Focus
Key Achievements
Top Papers
- 1Inverse optimal control from incomplete trajectory observations52 citations · 2021
- 2Inverse Optimal Control for Multiphase Cost Functions51 citations · 2019
- 3Learning From Sparse Demonstrations32 citations · 2022
- 4
- 5Learning From Human Directional Corrections18 citations · 2022
- 6
- 7Inverse Optimal Control with Incomplete Observations.14 citations · 2018
- 8Task-Driven Hybrid Model Reduction for Dexterous Manipulation10 citations · 2024
- 9
- 10An Iterative Method for Inverse Optimal Control5 citations · 2022