Research Areas

Our research focuses on developing fundamental algorithms that enable large teams of autonomous agents to accomplish collaborative tasks intelligently in dynamic environments.

A summary of our ongoing research can be found here [updated: August 2024].

Multi-Agent Path Finding (MAPF)

We aim to develop principled algorithms to solve challenging MAPF instances via a variety of AI and optimization technologies, such as constraint reasoning, heuristic search, stochastic local search, and machine learning.

Relevant publications



Coordination of Large Robot Teams in Automated Warehouses

We aim to combine task planning, path planning, and execution to coordinate thousands of mobile robots to fulfill delivery tasks in automated warehouses.

Relevant publications



Multi-Robotic-Arm Cooperative Manipulation

We aim to develop combined task planning, motion planning, and execution frameworks to jointly plan safe, low-cost plans for a team of robotic arms to perform cooperative manipulation and assembly.

Relevant publications



Environment Optimization for Fostering Agent Collaboration

While traditional research in multi-agent systems focuses on improving agents’ algorithms under fixed environmental settings, our team takes a complementary perspective: We aim to optimize the environment itself to enhance multi-agent performance.

Relevant publications



Intelligent Traffic Management

We aim to develop intelligent planning systems to coordinate trains, airplanes, autonomous vehicle, etc. on complex road networks under uncertainty.

Relevant publications