作者: Walter Alt , Christopher Schneider
DOI: 10.1002/OCA.2126
关键词: Control variable 、 Optimal control 、 Linear quadratic 、 Lipschitz continuity 、 Euler's formula 、 Mathematical optimization 、 Discretization 、 Regularization (mathematics) 、 Linear programming 、 Mathematics
摘要: Summary We analyze a class of linear-quadratic optimal control problems with an additional L1-control cost depending on parameter β. To deal this nonsmooth problem, we use augmentation approach known from linear programming in which the number variables is doubled. It shown that if for given bang-zero-bang and switching function has stable structure, solutions are Lipschitz continuous functions We also show case controls β * with | β − β * | sufficiently small coincide except set measure . Finally, to derive error estimates Euler discretizations. Copyright © 2014 John Wiley & Sons, Ltd.