Distributed Koopman Operator Learning from Sequential Observations
This work addresses distributed nonlinear system identification for multi-agent systems, offering an incremental improvement with a focus on resource-limited scenarios.
The paper tackles the problem of modeling unknown nonlinear dynamics from sequential observations by multiple agents, presenting a distributed Koopman operator learning framework that achieves exponential consensus and demonstrates convergence and predictive accuracy in simulations under sensing and communication constraints.
This paper presents a distributed Koopman operator learning framework for modeling unknown nonlinear dynamics using sequential observations from multiple agents. Each agent estimates a local Koopman approximation based on lifted data and collaborates over a communication graph to reach exponential consensus on a consistent distributed approximation. The approach supports distributed computation under asynchronous and resource-constrained sensing. Its performance is demonstrated through simulation results, validating convergence and predictive accuracy under sensing-constrained scenarios and limited communication.