Synthesizing Decentralized Controllers With Graph Neural Networks and Imitation Learning
Synthesizing Decentralized Controllers With Graph Neural Networks and Imitation Learning
复制标题
DOI：
10.1109/tsp.2022.3166401
复制
发表时间：
2020-12
影响因子：
5.4
通讯作者：
Fernando Gama;Qingbiao Li;Ekaterina V. Tolstaya;Amanda Prorok;Alejandro Ribeiro
中科院分区：
文献类型：
--
作者：
Fernando Gama;Qingbiao Li;Ekaterina V. Tolstaya;Amanda Prorok;Alejandro Ribeiro
由一组自治智能体组成的动态系统面临着必须仅依靠局部信息完成全局任务的挑战。虽然集中式控制器很容易获得，但它们在可扩展性和实施方面面临限制，因为它们不尊重代理网络系统强加的分布式信息结构。考虑到寻找最优分散控制器的困难，我们提出了一种利用图神经网络(GNN)来学习这些控制器的新框架。GNN非常适合这项任务，因为它们是自然分布的体系结构，并表现出良好的可扩展性和可转移性。我们证明了GNN通过模仿学习来学习适当的分散控制器，利用它们的排列不变性成功地扩展到更大的团队，并在部署时转移到看不见的场景。通过对群体问题和多智能体路径规划问题的研究，说明了神经网络在学习分散控制器方面的潜力。
Dynamical systems consisting of a set of autonomous agents face the challenge of having to accomplish a global task, relying only on local information. While centralized controllers are readily available, they face limitations in terms of scalability and implementation, as they do not respect the distributed information structure imposed by the network system of agents. Given the difficulties in finding optimal decentralized controllers, we propose a novel framework using graph neural networks (GNNs) to learn these controllers. GNNs are well-suited for the task since they are naturally distributed architectures and exhibit good scalability and transferability properties. We show that GNNs learn appropriate decentralized controllers by means of imitation learning, leverage their permutation invariance properties to successfully scale to larger teams and transfer to unseen scenarios at deployment time. The problems of flocking and multi-agent path planning are explored to illustrate the potential of GNNs in learning decentralized controllers.