Distributed model predictive control for continuous-time nonlinear systems based on suboptimal ADMM

摘要

The paper presents a distributed model predictive control (DMPC) scheme forcontinuous-time nonlinear systems based on the alternating direction method ofmultipliers (ADMM). A stopping criterion in the ADMM algorithm limits theiterations and therefore the required communication effort during thedistributed MPC solution at the expense of a suboptimal solution. Stabilityresults are presented for the suboptimal DMPC scheme under two different ADMMconvergence assumptions. In particular, it is shown that the requirediterations in each ADMM step are bounded, which is also confirmed in simulationstudies.