从向量获得的值的总和

Question

Let Q be a nx2 vector, which contains n (2-dimensional) coordinates. I want to calculate the distance between the i and i+1-th coordinate for all integers i in the vector and then add it up.

I came up with the following:

syms d

symsum(sqrt(sum((Q(d,:)-Q(d+1,:).^2))),d,1,n-1);

However, I get the following error:

??? Error using ==> sym.sym>notimplemented at 2653
Function 'subsindex' is not implemented for MuPAD symbolic objects.

Error in ==> sym.sym>sym.subsindex at 1359
            notimplemented('subsindex');

Error in ==> test4 at 4
symsum(sqrt(sum((Q(d,:)-Q(d+1,:).^2))),d,1,n-1);

Alternatively I could of course use a for loop, but I suspect that this is probably not the neatest and quickest solution.

Answer 1

I think it is the way you are implementing the syms variables. You are using the symbolic objects d to reference the "d-th" element in the vector Q (your error) and then use (steal) the embedded loop from symsum to go through the vector.

Also symbolic objects are time consuming. Just do the for/while loop it will be easier and faster.

gl.

Answer 2

If I am reading this correctly you just want the Euclidean distance between pairs of points in your nx2 matrix. So all you have to do is this:

>> Q = rand(10,2)  %Define Q

Q =

    0.6557    0.7060
    0.0357    0.0318
    0.8491    0.2769
    0.9340    0.0462
    0.6787    0.0971
    0.7577    0.8235
    0.7431    0.6948
    0.3922    0.3171
    0.6555    0.9502
    0.1712    0.0344

>> distPairs =  sum(diff(Q,1,1).^2,2); %Distance between adjacent coordinates
>> totalDist = sum(distPairs)  %Sum all of the pairwise distances.

totalDist =

    4.0486