Date of Award

5-2020

Degree Type

Honors College Thesis

Department

Computing

First Advisor

Bikramjit Banerjee, Ph.D.

Advisor Department

Computing

Abstract

Unmanned Aerial Vehicles (UAVs) are being extensively used in diverse sectors of the society for various tasks ranging from videography to an extremely sensitive situation such as first responders helping during a disaster. It has been seen that if a fleet of UAVs is deployed, they can perform a task quicker and more efficiently than a single UAV. With an increase in the number of UAVs, a problem arises of handling them with proper control structures. It has been studied that the Behavior Trees (BT) can be a better control architecture to handle the autonomous vehicles, as BTs are more reactive to changes. Being a fixed-rule decision model, there is still some bias on how a BT chooses its behavior. There has been some work to solve this issue by including the Reinforcement Learning (RL) technique for BTs so that agents can better learn their BTs and choose actions without bias. But this work only applies to a single agent system. Hence, we have proposed a solution on how multiple UAVs can apply RL in a cooperative scenario to learn their BTs. We show how the inclusion of communication between agents will impact the way an autonomous agent learns its BT.

Included in

Robotics Commons

Share

COinS