individual
What campus are you from?
Daytona Beach
Authors' Class Standing
Ava Neubert, Junior
Lead Presenter's Name
Ava Neubert
Faculty Mentor Name
Bryan Watson
Abstract
Understanding how individual agents within a swarm form, share, and adapt their opinions is essential for developing robust and cooperative multi-agent systems (MAS). This research focuses on visualizing collective swarm behavior through a dynamic RGB color-coding system that represents each agent’s opinion state in real time. The study aims to provide an intuitive and interpretable method for observing how consensus forms and evolves across a distributed network. A simulation was developed in AnyLogic to model swarm interactions under conditions of limited communication and environmental uncertainty. Each agent’s color continuously changes based on its internal state and the influence of neighboring agents. As opinions converge or diverge, emergent color patterns appear—allowing observers to visually track information exchange, agreement formation, and instability within the system. Preliminary observations indicate that the RGB visualization effectively reflects swarm consensus and highlights points of disagreement or isolation. This approach not only enhances understanding of emergent behavior in MAS but also provides a practical visualization framework for identifying communication bottlenecks, instability, or anomalous behavior within the swarm. Overall, this work demonstrates how real-time visual feedback can bridge the gap between algorithmic processes and human interpretation, supporting the future development of more adaptive and transparent autonomous systems for applications in exploration, monitoring, and defense.
Did this research project receive funding support from the Office of Undergraduate Research.
No
Visualizing Swarm Behavior and Opinion Dynamics in Multi-Agent Systems Using RGB Color Feedback
Understanding how individual agents within a swarm form, share, and adapt their opinions is essential for developing robust and cooperative multi-agent systems (MAS). This research focuses on visualizing collective swarm behavior through a dynamic RGB color-coding system that represents each agent’s opinion state in real time. The study aims to provide an intuitive and interpretable method for observing how consensus forms and evolves across a distributed network. A simulation was developed in AnyLogic to model swarm interactions under conditions of limited communication and environmental uncertainty. Each agent’s color continuously changes based on its internal state and the influence of neighboring agents. As opinions converge or diverge, emergent color patterns appear—allowing observers to visually track information exchange, agreement formation, and instability within the system. Preliminary observations indicate that the RGB visualization effectively reflects swarm consensus and highlights points of disagreement or isolation. This approach not only enhances understanding of emergent behavior in MAS but also provides a practical visualization framework for identifying communication bottlenecks, instability, or anomalous behavior within the swarm. Overall, this work demonstrates how real-time visual feedback can bridge the gap between algorithmic processes and human interpretation, supporting the future development of more adaptive and transparent autonomous systems for applications in exploration, monitoring, and defense.