Convert to occupancyMap3D

Question

MINJUN SO 2021-3-25

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/783018-convert-to-occupancymap3d

评论： David Meira Pliego 2021-4-12

在 MATLAB Online 中打开

First now, I convert stl file(from CATIA) to .mat file (class: pde.DiscreteGeometry).

And I want to convert this .mat file to 3D occupancy Map.

How can I solve this problem?

My goal is that I want to build real map on 3D occupancy Map.

And I'm trying this example RRT code which is loaded on MathWorks.

mapData = load("uavMapCityBlock.mat","omap");
omap = mapData.omap;
% Consider unknown spaces to be unoccupied
omap.FreeThreshold = omap.OccupiedThreshold;
inflate(omap,1)
figure("Name","CityBlock")
show(omap)

At first line, "uavMapCityBlock.mat" is a occupancyMap3D class. And I want to change that .mat file to my converted file which I'm asking you guys.

pls understanding my short english skill. thanks :)

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

MINJUN SO 2021-3-25

If there is no exist method, pls comment "there is no answer" or let me show another way.

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Aditya Patil 2021-3-30

2
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/783018-convert-to-occupancymap3d#answer_662274

在 MATLAB Online 中打开

There is currently no direct function to convert stl to occupancy matrix/grid.

As a workaround, if you create the stl file using large number of points, you should be able to pass to points to insertPointCloud function, as follows

gm = importGeometry('ForearmLink.stl');
points = gm.Vertices;
omap = occupancyMap3D;
pose = [0 0 0 1 0 0 0];
maxrange = 100;
insertPointCloud(omap, pose, points, maxrange);
show(omap);

However, this won't work very well for stl files with low number of vertices. So instead, you can create an alphashape with points, and then check for occupancy. The accuracy of method below will depend upon how well the alphaShape is created. Try different values from alphaShape parameters to get a better result.

gm = importGeometry('ForearmLink.stl');
points = gm.Vertices;
shp = alphaShape(points, 23);
plot(shp);
[X,Y,Z]= meshgrid(-150:150, -150:150, -150:150);
insideMat = inShape(shp, X, Y, Z);
insideId = find(insideMat);
[Xoc, Yoc, Zoc] = ind2sub(size(X), insideId);
pcshow([Xoc, Yoc, Zoc])