Approximate explicit model predictive control incorporating heuristics
Tóm tắt
Explicit piecewise linear state feedback solutions to the constrained linear model predictive control problem have recently been characterized and computed numerically using multiparametric quadratic programming. The piecewise linear state feedback is defined on a polyhedral partitioning of the state space, which may be quite complex. Here we suggest an approximate multi-parametric quadratic programming approach, which has the advantages that the partition is structured as a binary search tree. This leads to real-time computation of the piecewise linear state feedback with a computational complexity that is logarithmic with respect to the number of regions in the partition. The algorithm is based on heuristic rules that are used to partition the state space and estimate the approximation error.
Từ khóa
#Predictive models #Predictive control #Piecewise linear techniques #State feedback #Piecewise linear approximation #Quadratic programming #State-space methods #Binary search trees #Computational complexity #Approximation algorithmsTài liệu tham khảo
borrelli, 2001, Efficient on-line computation of explicit model predictive control, Proc IEEE Conf Decision and Control, tupll-2
10.1109/9.871769
10.1080/002071799220560
10.1145/361002.361007
10.1007/978-3-662-04245-8
tøndel, 2002, Complexity reduction in explicit model predictive control, Preprints of 13th World Congress of IFAC
10.1016/S0167-6911(96)00057-6
10.1109/CDC.2000.912749
10.1109/CDC.2000.912842
10.1109/CDC.2001.981048
10.1016/0098-1354(90)87012-E
bemporad, 2001, Suboptimal explicit Mpc via approximate quadratic programming, Proc IEEE Conf Decision and Control, frp08-5
johansen, 2002, Explicit suboptimal linear quadratic regulation with input and state constraints, Automatica, 38, 10.1016/S0005-1098(02)00004-3
10.1109/ACC.2000.876624
10.1016/S0005-1098(01)00174-1
johansen, 0, Approximate explicit model predictive control implemented via orthogonal search tree partitioning, IFAC World Congress Preprints Barcelona Spain 2002