I’m currently working on reverse engineering a yacht for research and development. I’ve been pulling .dwg solids and .e57 point cloud files from Trimble Realworks. The data was aquired with an X7 scanner at minimal point density and maximum time efficiency.
Basically, what is the workflow for reverse engineering a yacht point cloud!
See this sharepoint for the project (incudes everything I’ve loaded in so far):
A couple of questions!
Can I extrapolate mesh intersections?
How do you merge meshes?
I’ve noticed there are multiple types of surfaces, where might I find how each of the following differ and which favours which workflows:
@rhettone Welcme to the Rhino forum. It looks like you are new to Rhino and possibly new to modeling with NURBS.
The first step in reverse engineering an object is to learn to create models of smilar shapes. The User’ Guide, Level 1 Training manual and Level 2 Training manual are good places to start. https://www.rhino3d.com/tutorials/
A polysurface is two or NURBS surfaces joined together into a single surface.
A mesh is not considered to be a surface in Rhino.
A quadmesh is a mesh with every face having four sides. Quadmeshes can be used in Rhino 7 to create SubD objects, and SubD objects can be converted to NURBS polysurfaces.
A triangular mesh is a mesh with every face having three sides. Triangular meshes are of limited use in Rhino.
I create boat hull and occasionally deck and cabin models from point clouds and meshes. My usual workflow involves extracting creating key curves such as sheer line, rabbet line, etc using the point cloud or mesh. I create an initial surface using those key curves, Because the point clouds and meshes I use typically have millions of points I extract a representative subset of those points or section curves. Then I use the Rhino Patch command with the section curves as input and the initial surface as the starting surface. Further refinement of the resulting surface is sometimes required. The details of the workflow depend on the shape being modeled and the accuracy and fairness required.
BTW, you may discover the ToNURBS command or MeshToNURB which convert meshes to NURBS and think that they are the answer to your needs. It is not. Those commands just create each face of the mesh an individual NURBS surface and join the individual surfaces into a polysurface. For the type of meshes you are dealing with the resulting surfaces wll be muc too complex (probably orders of magnitude too complex) to be useful.
A friend from work has mentioned your name before, thank you for getting in contact with me.
I’ve since given up my expedition into reverse engineering and my trial for Rhino has expired. I might consider purchasing the software but I will see if I can get another trial license before purchasing the software.