Distributed Koopman Learning using Partial Trajectories for Control

TL;DR

This paper introduces a distributed deep Koopman learning framework where multi-agent systems learn a shared dynamics model using partial trajectories and local neural networks, enabling effective control without sharing private data.

Contribution

The paper presents a novel distributed Koopman learning method using partial trajectories, allowing agents to reach consensus on dynamics models without sharing training data.

Findings

Agents achieve consensus on global dynamics

Small approximation errors on test datasets

Effective model predictive control demonstrated

Abstract

This paper proposes a distributed data-driven framework for dynamics learning, termed distributed deep Koopman learning using partial trajectories (DDKL-PT). In this framework, each agent in a multi-agent system is assigned a partial trajectory offline and locally approximates the unknown dynamics using a deep neural network within the Koopman operator framework. By exchanging local estimated dynamics rather than training data, agents achieve consensus on a global dynamics model without sharing their private training trajectories. Simulation studies on a surface vehicle demonstrate that DDKL-PT achieves consensus on the learned dynamics, and each agent attains reasonably small approximation errors on the testing dataset. Furthermore, a model predictive control scheme is developed by integrating the learned Koopman dynamics with known kinematic relations. Results on a reference-tracking task indicate that the distributedly learned dynamics are sufficiently accurate for model-based optimal control.