Reputation: 190
I am trying to solve a differential equation system using bvp4c (boundary conditions). I am using two for loops to solve the functions of bvp4c but the problem is one for loop is done first and then only the last value is used with the iterations of the second for loop, is there a way to make them work simultaneously ? meaning the first iteration of the first for loop uses the first iteration values of the second for loop (not the last as it happens) ? thanks
function RTrajfoll(X,Y)
clf;
for i = 1:length(X)-1
init = bvpinit(linspace(X(i),X(i+1),10),[0 0]);
sol = bvp4c(@Kpath1,@bcpath,init);
x = linspace(X(i),X(i+1),100);
BS = deval(sol,x);
plot(x,BS(1,:),'linewidth',2)
axis([-2 6 -2 6])
hold on
end
function bv = bcpath(L,R)
for j = 1:length(Y)-1
bv = [L(1)-Y(j) R(1)-Y(j+1)];
end
end
end
%Differential equations dy/dx and dtheta/dx
function dx = Kpath1(~,c)
L = 0.12;
r = 0.1;
WL = 0.25;WR = 0.25;
y = c(1);th = c(2);
dy = tan(th);
dth = (2*((r*WR)-(r*WL)))/(L*cos(th)*((r*WR)+(r*WL)));
dx = [dy;dth];
pose = [y;th];
end
Upvotes: 1
Views: 170
Reputation: 114250
Looking at @Lutz's answer, you probably don't need a fully-fledged function for bv
. You may find it more convenient to provide an anonymous function handle, in which case you won't need the third argument:
sol = bvp4c(@Kpath1,@(L,R)[L(1)-Y(i) R(1)-Y(i+1)],init);
Upvotes: 0
Reputation: 25992
The observed behavior is exactly the expected behavior, you assign repeatedly to bv
, the return value is the last assigned value.
You will need to pass the index i
to the boundary condition function, either as a parameter
sol = bvp4c(@Kpath1,@(L,R)bcpath(L,R,i),init);
with
function bv = bcpath(L,R,j)
bv = [L(1)-Y(j) R(1)-Y(j+1)];
end
or by redefining bcpath
in every iteration, using the index as global variable,
for i = 1:length(X)-1
function bv = bcpath(L,R)
bv = [L(1)-Y(i) R(1)-Y(i+1)];
end
init = bvpinit(linspace(X(i),X(i+1),10),[0 0]);
sol = bvp4c(@Kpath1,@bcpath,init);
...
Upvotes: 1