Lab

The Berkeley Intelligent Control (ICON) lab develops algorithms for autonomous systems to interact with other agents safely and intelligently. Our goal is to enable autonomous systems to become integrated into the fabric of human life and act in the favor of society. To this end, we draw from control theory, game theory, robotics, and machine learning.

news

Jan 31, 2026 Our paper titled UDON: Uncertainty-weighted Distributed Optimization for Multi-Robot Neural Implicit Mapping under Extreme Communication Constraints has been accepted to the IEEE International Conference on Robotics and Automation (ICRA) 2026!
Jan 31, 2026 Our paper titled MIMIC-D: Multi-modal Imitation for MultI-agent Coordination with Decentralized Diffusion Policies has been accepted to the IEEE International Conference on Robotics and Automation (ICRA) 2026!
Jan 26, 2026 Our paper titled Matching Multiple Experts: on the Exploitability of Multi-Agent Imitation Learning has been accepted to the International Conference on Learning Representations (ICLR) 2026!
Dec 04, 2025 Our PhD student, Kanghyun Ryu, passed his qualifying exam and became a PhD candidate. Congratulations Kanghyun!
Nov 20, 2025 Our PhD student, Kartik Nagpal, passed his qualifying exam and became a PhD candidate. Congratulations Kartik!