'Error in port widths or dimensions' for using Adaptive MPC controller
14 Ansichten (letzte 30 Tage)
Ältere Kommentare anzeigen
I'm trying to construct the Co-simulation environment of path planning using MPC with CarSim.
but I'm struggle with debugging some errors. In the above simulink code, it makes a one Block error that says right below
"Error in port widths or dimensions. The signal connected to the "Model.B" port of the "Base_Model/Adaptive MPC Controller" block must be a matrix signal of 5 rows and 2 columns."
I don't definitley make sense since I think that the size of matix B is [5x2] as you can see above picture.
In addition, here is the Matlab Function code for generating state space matrix for updating model on adaptive MPC.
function [A,B,C,D,U,Y,X,DX] = fcn(x,u)
% Sample time
Ts = 0.1;
% Continuous-Time model
coder.extrinsic('linmod');
[Ac,Bc,Cc,Dc] = linmod('vehicleModel',x,u);
% Discretize using zero-order hold
nx = 5; nu = 2;
temp = coder.nullcopy(zeros(7,7));
A = coder.nullcopy(zeros(5,5));
B = coder.nullcopy(zeros(5,2));
C = coder.nullcopy(zeros(2,5));
D = coder.nullcopy(zeros(2,2));
X = coder.nullcopy(zeros(5,1));
U = coder.nullcopy(zeros(2,1));
Y = coder.nullcopy(zeros(2,1));
DX = coder.nullcopy(zeros(5,1));
% nx = size(Ac,1);
% nu = size(Bc,2);
temp = expm([[Ac Bc]*Ts; zeros(nu,nx+nu)]);
A = temp(1:nx, 1:nx);
B = temp(1:nx, nx+1:nx+nu);
C = Cc;
D = Dc;
% Nominal conditions of DT plant
X = x;
U = u;
Y = C*x;
DX = A*x + B*u - x;
0 Kommentare
Antworten (3)
Jakobus Louw
am 9 Apr. 2020
See my answer at: https://www.mathworks.com/matlabcentral/answers/427070-adaptive-mpc-dimension-error-of-block-input
Its similar to Dhruv's answer, but the block is called SIgnal Specification.
0 Kommentare
Sai Sri Pathuri
am 25 Feb. 2020
The Bus Creator block you are using in the model combines all the output signals from ML function block. But the expected input of Adaptive MPC block is not the entire bus signal. So, you may use Bus Selector block to select the appropriate signal from the bus signal.
0 Kommentare
Dhruv Thakkar
am 12 Mär. 2020
You can use Signal Dimension block to mention the dimension as [5 2] to ensure that there is no discrepency.
0 Kommentare
Siehe auch
Kategorien
Mehr zu Refinement finden Sie in Help Center und File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!