Multi-Robot Task Allocation Games in Dynamically Changing Environments

Multi-Robot Foraging Using the Graphical Multiarmed Bandit Framework

Multirobot Object Pickup

Multirobot Navigation Agreement

Flexible Multi-Agent Decision-Making

Directed Cooperative Multi-Armed Bandit

Searching for Radioactive Material using rad-UCL

Upper-Confidence Limit Length Scale Exploration

This video shows the effects on decision-making of using different length scales in the Upper-Confidence Limit (UCL) algorithm using the underwater Beluga testbed in Naomi Leonard’s DCSL group. It demonstrates how using the correct length scale can dramatically improve performance and shorten the travel distance for agents.