search

RethinkRobotics / baxter_common

ROS Messages and robot description files for the Baxter Research Robot
www.rethinkrobotics.com/sdk
BSD 3-Clause "New" or "Revised" License
56 stars 105 forks source link

basic_hard_tip.DAE and basic_hard_tip.STL are messed up #78

Open Algomorph opened 8 years ago

Algomorph commented 8 years ago

basic_hard_tip.DAE and basic_hard_tip.STL are completely messed up on the bottom of the mesh, where there should be an opening. It looks as if some mesh optimizer was run on all the meshes, but tripped up most badly on this particular one. In general, all the meshes appear to be messy, likely for the same reason. I'm currently fixing the hard tip file, I'll try to post it here once I'm done.

Algomorph commented 8 years ago

I got the official model files for the tip. I have converted one to Blender format, so you can edit and re-export it if need be, STL and DAE (Collada). I noticed your models use 1 unit = 1 meter scale, the STL I generated uses common 3D-printing scale, i.e. 1 unit = 1 mm scale. Just in case I'm also providing the "downscaled" 1unit=1meter version in DAE format. algomorph.com/storage/GripperTipOfficial.blend algomorph.com/storage/GripperTipOfficial.dae algomorph.com/storage/GripperTipOfficial.stl algomorph.com/storage/GripperTipOfficial_small.dae

IanTheEngineer commented 8 years ago

Really interesting! Thank you so much for fixing these meshes. We typically use them for visualization in RViz rather than for 3D printing, but having a valid mesh is always helpful. How were you able to determine that the mesh was invalid?

Algomorph commented 8 years ago

No problem, glad to help. A colleague of mine opened it up in MeshLab, I did the same in Blender, we both found the same problems. We have the actual gripper & tip on hand, so it was easy to tell the difference. It seems like converters from STEP and IGS CAD formats to mesh files, like STL and DAE, sometimes do a messy job, so that might have been the reason.