Hello, i am currently experimenting with the follwing example: https://de.mathworks.com/help/mpc/ug/solve-custom-mpc-quadratic-programming-problem-and-generate-code.html Now i am interested how i can, for example make the system follow a reference trajectory. I tried to change the 2nd line of: for ct = t_constrained [u,status,iA] = mpcActiveSetSolver(Linv,F*x,Ac,b0,[],zeros(0,1),iA,opt); uMPC(ct+1) = u(1); x = A*x+B*uMPC(ct+1); yMPC(ct+1) = C*x; end to [u,status,iA] = mpcActiveSetSolver(Linv,F*(x-set),Ac,b0,[],zeros(0,1),iA,opt); I chose this line, because it seems it is the only point where the actual state is coming into the optimization process and i studied some literature where that made sense when looking at the block chart of general mpc controls. My idea was to input the deviation of the state x with the setpoint set with same dimensions (and later on a moving vector).

Set-Point Control in MPC example

Sebastian Starke

2021 5 5

0 个回答

10 次查看（30 天）

0 个投票

Hello,

i am currently experimenting with the follwing example:

https://de.mathworks.com/help/mpc/ug/solve-custom-mpc-quadratic-programming-problem-and-generate-code.html

Now i am interested how i can, for example make the system follow a reference trajectory. I tried to change the 2nd line of:

for ct = t_constrained
    [u,status,iA] = mpcActiveSetSolver(Linv,F*x,Ac,b0,[],zeros(0,1),iA,opt);
    uMPC(ct+1) = u(1);
    x = A*x+B*uMPC(ct+1);
    yMPC(ct+1) = C*x;
end

[u,status,iA] = mpcActiveSetSolver(Linv,F*(x-set),Ac,b0,[],zeros(0,1),iA,opt);

I chose this line, because it seems it is the only point where the actual state is coming into the optimization process and i studied some literature where that made sense when looking at the block chart of general mpc controls. My idea was to input the deviation of the state x with the setpoint set with same dimensions (and later on a moving vector).