Papers
11
Total Citations
205
H-Index
8
About
Peng Chang is a robotics researcher whose work spans computer vision, autonomous navigation, and robotic manipulation, with particular expertise in bridging the gap between perception and physical interaction. His early contributions in the late 1990s and early 2000s focused on mobile robot navigation, developing omnidirectional camera systems capable of structure from motion (SFM) reconstruction — work that earned over 67 citations and established him as an early innovator in wide-field visual perception for robotics. His investigations into semi-autonomous urban robot control and visual servoing further demonstrated his commitment to practical, real-world robotic systems. More recently, Chang has made significant strides in deformable object manipulation, a notoriously challenging frontier in robotics. His pioneering Sim2Real2Sim framework — accumulating over 36 citations — introduced a cyclical strategy for transferring robot learning between simulation and the physical world, enabling reliable automation of flexible object handling tasks such as wire and cable manipulation. Complementing this, his model-based approaches incorporating visual curvature feedback have advanced how robots perceive and interact with non-rigid objects. His research on shared human-robot control for collaborative assembly tasks further highlights his broad impact across perception, learning, and human-robot interaction — making him a versatile and influential voice in modern robotics research.
Research Focus
Key Achievements
Top Papers
- 1Omni-directional structure from motion67 citations · 2002
- 2
- 3
- 4<title>Experiments with driving modes for urban robots</title>22 citations · 1999
- 5Omni-directional visual servoing for human-robot interaction15 citations · 2002
- 6A Shared Control Method for Collaborative Human-Robot Plug Task13 citations · 2021
- 7
- 8
- 9
- 10