matlab

Dependencies:

The MATLAB version of the repository makes use of the Multi-Parametric Toolbox 3.0 to handle projections of polytopes: M. Herceg, M. Kvasnica, C. Jones, and M. Morari. Multi-Parametric Toolbox 3.0. In Proc. of the European Control Conference, Zürich, Switzerland, July 17-19 2013, pp. 502-510. http://control.ee.ethz.ch/mpt .

Quick-start:

For more information on the quantities below please refer to ALOT21b. The main wrapper function for this code is computeRCIS(A,B,Gx,Fx,Gu,Fu,E,Gw,Fw,implicit,L,T):

• A, B, and E are matrices defining the discrete-time linear system: x^+ = Ax + Bu + Ew.
• Gx is a matrix and Fx is a vector defining the polyhedral safe set Sx = {x \in \R^n | Gx x <= Fx}.
• Gu and Fu define input constraints Su = {u \in \R^m | Gu u <= Fu}. If no costraints use Gu = [], Fu = [].
• Gw and Fw define the disturbance set Sw = {w \in \R^k | Gw w <= Fw}. If no disturbance use E = [], Gw = [], and Fw = [].
• implicit \in {0,1}.
  • If implicit=0, then explicit RCIS = {x \in \R^n | rcisA x <= rcisb}.
  • If implicit=1, then implicit RCIS = {(x,u,v) \in \R^n x \R^m x \R^{m(T+L)} | rcisA(x,u,v) <= rcisb}.
• If only L is specified, then L : L-th level of hierarchy.
• If both L and T are specified:
  • L (lambda): loop of eventually periodic input sequence.
  • T (tau): transient of eventually periodic input sequence.

Output is a Polyhedron object. If implicit=0, then the output is an explicit RCIS as above. Else, it is an implicit RCIS.