Simulating Multi-Agent Collaboration in a Classroom Environment
Research Mentor(s)
Wesley Deneke
Description
The current research for multi-agent simulation mainly focuses on sensor-based agents with limited visuals. There is a gap in current research for realistic 3D collaborative multi-agent simulations. Producing realistic simulations in an extensible manner is a challenging problem as there are many components to realism. However, as the sim to real gap continues to close, we will keep seeing more advancements. There are many benefits to producing more realistic simulations, such as reducing the bias in deep learning models, increasing accuracy of activity recognition for the advancement of more helpful AI, and a significant cost reduction over producing simulations with real people in the real world. Our research addresses multi-agent collaboration by creating a realistic classroom simulation platform using Unity 3D that offers rapid authoring of an agent’s characteristics and simulation settings, which in turn diversifies the agent’s actions, creating more unique simulation data and video outputs. We accomplish this with human models and realistic animations for each action the agents perform.
Document Type
Event
Start Date
May 2022
End Date
May 2022
Location
Carver Gym (Bellingham, Wash.)
Department
CSE - Computer Science
Genre/Form
student projects; posters
Type
Image
Rights
Copying of this document in whole or in part is allowable only for scholarly purposes. It is understood, however, that any copying or publication of this document for commercial purposes, or for financial gain, shall not be allowed without the author’s written permission.
Language
English
Format
application/pdf
Simulating Multi-Agent Collaboration in a Classroom Environment
Carver Gym (Bellingham, Wash.)
The current research for multi-agent simulation mainly focuses on sensor-based agents with limited visuals. There is a gap in current research for realistic 3D collaborative multi-agent simulations. Producing realistic simulations in an extensible manner is a challenging problem as there are many components to realism. However, as the sim to real gap continues to close, we will keep seeing more advancements. There are many benefits to producing more realistic simulations, such as reducing the bias in deep learning models, increasing accuracy of activity recognition for the advancement of more helpful AI, and a significant cost reduction over producing simulations with real people in the real world. Our research addresses multi-agent collaboration by creating a realistic classroom simulation platform using Unity 3D that offers rapid authoring of an agent’s characteristics and simulation settings, which in turn diversifies the agent’s actions, creating more unique simulation data and video outputs. We accomplish this with human models and realistic animations for each action the agents perform.