Continuum Models for Emergent Shepherding Behavior

Develop continuum-level mathematical models that accurately describe emergent shepherding behavior in systems where a population of controller agents influences a population of target agents through local interactions, to enable analysis and macroscopic control design within the harnessing-complex-systems-for-control framework.

Background

The chapter introduces a control paradigm in which one complex system (a controller population) is used to steer another complex system (a target population) via local interaction rules, exemplified by shepherding scenarios. This framework aims to formalize how a relatively small set of controller agents can coordinate and influence the collective dynamics of many target agents.

Within this paradigm, the authors explicitly identify outstanding issues, including the lack of continuum (macroscopic) models capable of capturing emergent shepherding behavior. Such models would facilitate analysis and control synthesis at the population level, complementing microscopic agent-based formulations.