You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have a software for checking collisions within the working area of a machine tool. Atm I use the old library opcode to do the collision check.
I also included FCL to give it a try and to have a newer software design. In principle, it works, but I noticed a much slower performance compared to the Opcode library.
I have followed the instructions from the Git homepage. Here is an excerpt of the main parts for using FCL:
auto model = std::make_shared<fcl::BVHModel<fcl::Bounding_box>>();
model->beginModel();
model->addSubModel(vertices, triangles);
model->endModel();
coll_geo = fcl::Collision_object(model);
This is done for every collision object (I have a lot of different geomtries which needs to be checked. It is handled in the struct Node_fcl.
And this is the actual collision check of two nodes:
Hi,
I have a software for checking collisions within the working area of a machine tool. Atm I use the old library opcode to do the collision check.
I also included FCL to give it a try and to have a newer software design. In principle, it works, but I noticed a much slower performance compared to the Opcode library.
I have followed the instructions from the Git homepage. Here is an excerpt of the main parts for using FCL:
This is done for every collision object (I have a lot of different geomtries which needs to be checked. It is handled in the struct Node_fcl.
And this is the actual collision check of two nodes:
Perhaps I use the library in an inefficient way? Is there some kind of best practice for FCL if performance is crucial?
Best regards
The text was updated successfully, but these errors were encountered: