Benjamin Seibold - Swarm-Performance of Heterogeneous Multi-Agent Systems Across Scales
Presenter
January 24, 2024
Abstract
Recorded 24 January 2024. Benjamin Seibold of Temple University presents "Swarm-Performance of Heterogeneous Multi-Agent Systems Across Scales" at IPAM's Mathematical Foundations for Equity in Transportation Systems Workshop.
Abstract: Many real-world system are composed on multiple agents whose interactions result in a collective swarm behavior that may be complex, unexpected, and/or unintended. We highlight intriguing cases of interplay between the micro-scale behavior of agents and the macro-scale performance of the swarm, with a particular emphasis on heterogeneous systems composed of different types of agents. In transportation, an important example is vehicular traffic on roadways. Other examples include transportation in 2d/3d environments like vehicles transporting supplies to a disaster area. We demonstrate how heterogeneities matter both at the level of agent behavior as well as for the performance metrics. One example is the role of automation/connectivity on the total energy footprint of urban traffic flow, as well as the system-level and individual benefits if those vehicles are not distributed evenly. Other examples include mixed human/robotic groups, swarms of animals, and/or swarms in which a few agents' capabilities are reduced (or agents behave disruptively) or enhanced (or agents lead). A key objective is to structurally understand and characterize the interplay between the agents' behavior/strategies and emergent swarm performance, contrasting various metrics of success like survival-of-the-fittest versus support-the-weakest.
Learn more online at: https://www.ipam.ucla.edu/programs/workshops/mathematical-foundations-for-equity-in-transportation-systems-january-22-26-2024/?tab=overview