Interactive & Emergent Autonomy Lab

Projects

Active Learning and Data-Driven Control Active Perception in Human-Swarm Collaboration Algorithmic Matter and Emergent Computation Control for Nonlinear and Hybrid Systems Cyber Physical Systems in Uncertain Environments Harmonious Navigation in Human Crowds Information Maximizing Clinical Diagnostics Reactive Learning in Underwater Exploration Robot-Assisted Rehabilitation Software-Enabled Biomedical Devices

Harmonious Navigation in Human Crowds

Feedback Diagram

This project aims to enable seamless robot navigation in human crowds; such capability is critical for bringing robots into our daily lives. We develop the theoretical foundation for interactive navigation decision-making when anticipating inherently uncertain human behavior by combining tools from game theory, Bayesian statistics, and active learning. Furthermore, we develop full-stack robot navigation systems to verify the algorithm with dense crowds in unstructured real-world environments, with fully on-board computation and perception.

People

Muchen Sun (Ph.D. Student)

Collaborators

Peter Trautman, Honda Research Institute (USA)

Publications

Move Beyond Trajectories: Distribution Space Coupling for Crowd Navigation
M. Sun, F. Baldini, P. Trautman, T. D. Murphey
Robotics: Science and Systems (RSS), 2021. PDF, Video, Code

Funding

This project is funded by Honda Research Institute (USA).


Other Projects

Active Learning and Data-Driven Control
Active Perception in Human-Swarm Collaboration
Algorithmic Matter and Emergent Computation
Control for Nonlinear and Hybrid Systems
Cyber Physical Systems in Uncertain Environments
Harmonious Navigation in Human Crowds
Information Maximizing Clinical Diagnostics
Reactive Learning in Underwater Exploration
Robot-Assisted Rehabilitation
Software-Enabled Biomedical Devices