Concept issues about initializing avariable in C++ class

Question

I'm translating a script form matlab to C++ and I'm given a code snipet in c++ so I can fill it. My BIG issue is that my knowledge about C++ is almost nothing, I know how classes work in a basic way so now I'm stuck in a variable initialization problem.

So, coming from matlab I have some simple variables to start but, The last one depends on a function for the inizialization, so I dont know how to proceed. Here my matlab code:

For those who are not use to matlab, qs, dqs, ddqs, centrqds, Corrqds are vector of vectors of 6xn size. 
function [PIDOUT,qsinit,dqsinit,ddqsinit,PosError, PosErrorSum,VelErrorSum,IntErrorSum, qs,dqs,ddqs,E2,mRobot,mControl,EM,torque]=initializeLoopVars(fregCentr, fregCor, fregFric,fregG,qc,dqc, ddqc, menu)

PIDOUT=zeros(6,length(qc));
qsinit=qc(:,1);
dqsinit=zeros(6,1);
ddqsinit=zeros(6,1);
mRobot=zeros(6,6,length(qc));
mControl=zeros(6,6,length(qc));
EM=zeros(6,length(qc));


PosError=zeros(6,1);
PosErrorSum=zeros(6,1);
VelErrorSum=zeros(6,1);
IntErrorSum=zeros(6,1);


qs=zeros(6,length(qc));
dqs=zeros(6,length(dqc));
ddqs=zeros(6,length(ddqc));

Centrqds = [dqsinit(1,:);dqsinit(2,:);dqsinit(3,:);dqsinit(4,:);dqsinit(5,:);dqsinit(6,:)];
Corrqds = prodDqiDqj([dqsinit(1,:);dqsinit(2,:);dqsinit(3,:);dqsinit(4,:);dqsinit(5,:);dqsinit(6,:)]);

torque=zeros(6,length(qc));
E2=zeros(6,length(qc));

%initial dynamic compensation
if (menu == 0)
    % V+G+fr
    E2(:,1)=fregCentr(qc(2,1),qc(3,1),qc(4,1),qc(5,1),qc(6,1))*Centrqds+fregCor(qc(2,1),qc(3,1),qc(4,1),qc(5,1),qc(6,1))*Corrqds+fregG(qc(2,1),qc(3,1),qc(4,1),qc(5,1),qc(6,1))+fregFric(dqsinit(1,1),dqsinit(2,1),dqsinit(3,1),dqsinit(4,1),dqsinit(5,1),dqsinit(6,1));
    
    % V+G
    %E2(:,1)=fregCentr(qc(2,1),qc(3,1),qc(4,1),qc(5,1),qc(6,1))*Centrqds+fregCor(qc(2,1),qc(3,1),qc(4,1),qc(5,1),qc(6,1))*Corrqds+fregG(qc(2,1),qc(3,1),qc(4,1),qc(5,1),qc(6,1));
    
    % V
    %E2(:,1)=fregCentr(qc(2,1),qc(3,1),qc(4,1),qc(5,1),qc(6,1))*Centrqds+fregCor(qc(2,1),qc(3,1),qc(4,1),qc(5,1),qc(6,1))*Corrqds;
    
    % G
    %E2(:,1)=fregG(qc(2,1),qc(3,1),qc(4,1),qc(5,1),qc(6,1));
elseif (menu == 1)
    % V+G+fr
    E2(:,1)=fregCentr(qc(2,1),qc(3,1),qc(4,1),qc(5,1),qc(6,1))*Centrqds+fregCor(qc(2,1),qc(3,1),qc(4,1),qc(5,1),qc(6,1))*Corrqds+fregG(qc(2,1),qc(3,1),qc(4,1),qc(5,1),qc(6,1))+fregFric(dqsinit(1,1),dqsinit(2,1),dqsinit(3,1),dqsinit(4,1),dqsinit(5,1),dqsinit(6,1));
    
    % V+G
    %E2(:,1)=fregCentr(qc(2,1),qc(3,1),qc(4,1),qc(5,1),qc(6,1))*Centrqds+fregCor(qc(2,1),qc(3,1),qc(4,1),qc(5,1),qc(6,1))*Corrqds+fregG(qc(2,1),qc(3,1),qc(4,1),qc(5,1),qc(6,1));
    
    % V
    %E2(:,1)=fregCentr(qc(2,1),qc(3,1),qc(4,1),qc(5,1),qc(6,1))*Centrqds+fregCor(qc(2,1),qc(3,1),qc(4,1),qc(5,1),qc(6,1))*Corrqds;
    
    % G
    %E2(:,1)=fregG(qc(2,1),qc(3,1),qc(4,1),qc(5,1),qc(6,1));
end

I'm in process of translating so some variables are missing in the c++ code!!

Notice that Corrqds depends on prodDqiDqj function to be inizialized, that is a simple algebraic calculation.

function [dqidqj] = prodDqiDqj(dq)
    dqidqj = zeros(15, 1);
    cont = 1;

    for i = 1:6
        for j = i + 1:6
            dqidqj(cont, 1) = dq(i) * dq(j);
            cont = cont + 1;
        end
    end
end

Now the c++ part. I have a RobotParameters class in RobotParameters.h where my variables are public. I started creating those variables (and some more, names are maybe changed by convenience):

#pragma once

#include <vector>

    class RobotParameters
    {
    public:
        std::vector<std::vector<double>> posDesired{ 6 };
        std::vector<std::vector<double>> velDesired{ 6 };
        std::vector<std::vector<double>> posRobot{ 6 };
        std::vector<std::vector<double>> torqueRobot{ 6 };
        std::vector<std::vector<double>> ErrorRobot{ 6 };
        std::vector<std::vector<double>> velRobot{ 6 };
        std::vector<std::vector<double>> qc{ 6 };
        std::vector<std::vector<double>> dqc{ 6 };
        std::vector<std::vector<double>> ddqc{ 6 };
        std::vector<double> errorAnt = { 0.0,0.0, 0.0, 0.0, 0.0, 0.0 };
        std::vector<std::vector<double>> P{ 6 };
        std::vector<std::vector<double>> D{ 6 };
        std::vector<std::vector<double>> I{ 6 };
        std::vector<std::vector<double>> Tc{ 6 };
        std::vector<double> errorInt = { 0.0,0.0, 0.0, 0.0, 0.0, 0.0 };
        std::vector<double> posRobotAnt = { 0.0,0.0, 0.0, 0.0, 0.0, 0.0 };
        std::vector<double> Centrqds = { 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 };
        
    
    
        RobotParameters();
    
    };

My issue is, for Corrqds I need to somehow declare the latter prodDqiDqj function somewhere but I'm not sure how to do it.

Someone could help me and give an explanation about how should I do it? As i told, I'm new to C++ and I'm struggling here.

edit: Added more code as requested!

Please provide a [mcve]. Its not clear why you need 17 vectors when in the matlab code there are 5 — 463035818_is_not_an_ai, Feb 28 '23 at 10:24
Is the prodDqiDqj function supposed to be a method of RobotParameters or seperate? — muhmann, Feb 28 '23 at 10:25
In matlab there are as well more vectors, but thought it was pointless to throw a lot of code if they are all the same. I guessed that with a small example was enough. And yes, prodDqiDqj it is supossed to be a method of RobotParameters! — mikel lasa, Feb 28 '23 at 10:43
Remove all the extra details from your code and provide a [minimal reproducible example](https://stackoverflow.com/help/minimal-reproducible-example). Note the "minimal" part. — Jason, Feb 28 '23 at 10:55
so what exactly is stopping you from running the nested loops over the c++ vector just like in your matlab sample ? — nick, Feb 28 '23 at 13:24
It is unclear to me what you’re asking. Where to put a new function? How to write a constructor? In either case I would suggest you read a book or tutorial on the basics of C++. — Cris Luengo, Feb 28 '23 at 15:45

Concept issues about initializing avariable in C++ class

0 Answers0