Learning and Control for Intelligent Interaction Between Humans and Autonomous Systems
Project Description
Robots (especially humanoid robots) hold immense promise for disaster response, home assistance, space exploration, and manufacturing. These robots, when equipped with intelligent interaction capabilities, can effectively collaborate with humans in various settings, such as warehouses and military missions. While significant advances have been made in enabling robots to navigate complex environments using both physics-based and data-driven approaches, the development of intelligent and physically safe human-robot interactions remains an unresolved challenge. This challenge is primarily due to two factors: the highly complex dynamical behaviors involved in physical human-robot interaction and the intricate cognitive behaviors of humans, such as trust towards robots and preferences in their usage, which are both difficult to model and predict.
Motivated by these practical demands and research gaps, this project aims to integrate model-based and data-driven approaches to facilitate intelligent interactions between humans and autonomous robots. The project will enhance our understanding of human trust and preferences in physical collaboration with autonomous robots. Building on this understanding, we will develop novel learning control schemes that foster appropriate levels of trust and accommodate human preferences. Experimental validation will be conducted using a Digit humanoid robot available in Dr. Gu’s lab.
Start Date
Spring, Summer, or Fall of 2026
Postdoc Qualifications
Candidates with a strong background and interest in more than one of the following areas are desired:
- Robotics, especially humanoid robots and/or robotic manipulators.
- Controls, Optimization, Learning
- Hands-on/coding experience preferred (not required)
Co-advisors
Yan Gu (primary)
School of Mechanical Engineering
yangu@purdue.edu
Lab website: https://www.thetracelab.com/
Shaoshuai Mou (second)
School of Aeronautics and Astronautics
mous@purdue.edu
https://engineering.purdue.edu/AIMS
Bibliography
1. Gao Y, Yuan C, Gu Y. Invariant filtering for legged humanoid locomotion on a dynamic rigid surface. IEEE/ASME Transactions on Mechatronics. 2022 Aug 2;27(4):1900-9.
2. Jin W, Murphey TD, Kulić D, Ezer N, Mou S. Learning from sparse demonstrations. IEEE Transactions on Robotics. 2022 Aug 3;39(1):645-64.
3. Y. Gao*, V. Paredes*, Y. Gong, Z. He, A. Hereid, and Y. Gu, "Time-Varying Foot-Placement Control for Underactuated Humanoid Walking on Swaying Rigid Surfaces," IEEE Transactions on Robotics, accepted.
4. W. Chen, R. Yeh*, S. Mou*, Y. Gu*, "Leveraging Perturbation Robustness to Enhance Out-of-Distribution Detection," in Proc. of Conference on Computer Vision and Pattern Recognition (CVPR), accepted. (*co-senior advising)
5. Z. He, S. Teng, T.-Y. Lin, M. Ghaffari, Y. Gu, "Invariant Filtering for Full-State Estimation of Ground Robots in Non-Inertial Environments," IEEE/ASME Transactions on Mechatronics, accepted.