How can the plot be divided into 2 using the graph line?

Hello everyone,

I have a graph that I obtained with the help of the RRT algorithm.

In addition, I have an array of points that are on the plot as you can see.

I want to write a function \ use a classification algorithm (or any other machine learning algorithm) to divide the points into 2 using the graph (points on the right side of the graph and points on the left side)

How can this be done? I tried several options but failed.. :/

I would be very happy to help! :)

Answer 1

[Matt J]

Perhaps something like the following:

load roadPoints

[X,Y]=ndgrid(0:2.5:100);

p1=polyshape([roadPoints;min(roadPoints(:,1)), max(roadPoints(:,2))],'Simplify',true);

Warning: Polyshape has duplicate vertices, intersections, or other inconsistencies that may produce inaccurate or unexpected results. Input data has been modified to create a well-defined polyshape.

p2=polyshape([roadPoints;max(roadPoints(:,1)), min(roadPoints(:,2))],'Simplify',true);

Warning: Polyshape has duplicate vertices, intersections, or other inconsistencies that may produce inaccurate or unexpected results. Input data has been modified to create a well-defined polyshape.

class1=isinterior(p1,X(:),Y(:));
class2=isinterior(p2,X(:),Y(:));
noclass=~(class1|class2);

hold on
scatter(X(class1), Y(class1),'g','filled','MarkerEdgeColor','none');
scatter(X(class2), Y(class2),'c','filled','MarkerEdgeColor','none');
scatter(X(noclass),Y(noclass),'MarkerFaceColor',[0.5,0.5,0.5],'MarkerEdgeColor','none')
plot(roadPoints(:,1), roadPoints(:,2),'r','LineWidth',5); 
legend('Class 1', 'Class 2','UnClassified','Location','NorthOutside')
hold off

[Yovel]

It looks great!! I will try it now

Thank you very much!! :))))

[Matt J]

You're welcome, but please Accept-click the answer if you decide it is what you need.

[Yovel]

Hi,

If the route looks like this, the classification of the points is not done correctly..

The calssification shouls be like this:

Thanks for your help! :)

[Yovel]

I fix it!

thanks! <3

[Matt J]

I'm glad, but agan, please Acept-click the answer if it ultimately worked.

Answer 2

[Matt J]

Perhaps you can find the nearest point on the boundary line with pdist2, then make a decision based on whether the nearest point is to the right or to the left.

[Yovel]

There is no algorithm that takes the output of a map and classifies the points using the line on the map?

[Matt J]

Didn't I just propose one?

Answer 3

[Image Analyst]

It looks like all the points on the right have a y value greater than about 6, so just do

rightIndexes = (y >= 6);
xRight = x(rightIndexes);
yRight = y(rightIndexes);
xLeft = x(~rightIndexes);
yLeft = y(~rightIndexes);

[Yovel]

Hey,

Every time I get a different random route. Therefore I cannot say y>6

Is there another way?

[Dyuman Joshi]

Is the route completely random or can a relation/observation be identified?

And you have not defined what is the criteria/logic to be used for dividing points to the left and right side.

[Yovel]

I get the route from the RRT algorithm and I later turn it into a discrete route.

Basically, my goal is to build a walking track for a 2-legged robot that can only step in places where there are dots.

I want to create a track for each leg (for the right leg and the left leg)

That's why I want to choose each time 2 points from each side of the track that are closest to the center point of the body (the point of the track is RRT)