Orateur
Description
We apply the recently proposed Coupled Adaptable Backward-Forward-Backward Resolvent Splitting Algorithm (CABRA) to the continuous relaxation of the multi-stage stochastic weapon target assignment problem. Our formulation allows decentralized optimization across weapon platforms with minimal data exchange requirements. The CABRA formulation also allows us to adapt the splitting structure to match the available communication paths between weapon platforms, relying on direct connectivity only between platforms which share a target. Unlike other recently developed adaptable forward backward methods, CABRA takes direct advantage of the structure of the nonanticipativity constraints in the lifted problem, thereby reducing memory requirements and accelerating convergence. We demonstrate this in a set numerical experiments which validate the performance of the formulation.
