Cooperative Path Integral Control for Stochastic Multi-Agent Systems
This work addresses control challenges in stochastic multi-agent systems like UAVs, but it is incremental as it builds on existing path integral and distributed control methods.
The paper tackles distributed stochastic optimal control for cooperative multi-agent systems by partitioning agents into factorial subsystems and designing local control actions based on local observations to optimize joint costs, with numerical verification showing effectiveness in a UAV team simulation.
A distributed stochastic optimal control solution is presented for cooperative multi-agent systems. The network of agents is partitioned into multiple factorial subsystems, each of which consists of a central agent and neighboring agents. Local control actions that rely only on agents' local observations are designed to optimize the joint cost functions of subsystems. When solving for the local control actions, the joint optimality equation for each subsystem is cast as a linear partial differential equation and solved using the Feynman-Kac formula. The solution and the optimal control action are then formulated as path integrals and approximated by a Monte-Carlo method. Numerical verification is provided through a simulation example consisting of a team of cooperative UAVs.