Learning Decentralized Controllers for Robot Swarms with Graph Neural Networks
This addresses the challenge of scalable and efficient control for mobile robot swarms in dynamic environments, representing an incremental improvement by adapting existing graph neural network methods to time-varying conditions.
The paper tackles the problem of learning decentralized controllers for robot swarms with interacting dynamics and sparse communications by imitating centralized policies using graph neural networks adapted for time-varying signals and networks, demonstrating performance in flocking tasks where decreasing communication radius and faster velocities highlight the value of multi-hop information.
We consider the problem of finding distributed controllers for large networks of mobile robots with interacting dynamics and sparsely available communications. Our approach is to learn local controllers that require only local information and communications at test time by imitating the policy of centralized controllers using global information at training time. By extending aggregation graph neural networks to time varying signals and time varying network support, we learn a single common local controller which exploits information from distant teammates using only local communication interchanges. We apply this approach to the problem of flocking to demonstrate performance on communication graphs that change as the robots move. We examine how a decreasing communication radius and faster velocities increase the value of multi-hop information.