A direct optimization algorithm for input-constrained MPC
One challenge of running a model predictive control (MPC) algorithm in a production-embedded platform is to provide the certificate of worst-case computation complexity, that is, its maximum execution time has to always be smaller than sampling time. This paper proposes for the first time a direct optimization algorithm for input-constrained MPC: the number of iterations is data-independent and dependent on the problem dimension n, with exact value ⌈log(2n/ϵ)/-2log(1-1/4√(2n))⌉+1, where ϵ denotes a given stopping accuracy.
READ FULL TEXT