search

Gepetto / example-robot-data

Set of robot URDFs for benchmarking and developed examples.
BSD 3-Clause "New" or "Revised" License
143 stars 51 forks source link

Release/3.13.1 #117

Closed nim65s closed 2 years ago

nim65s commented 2 years ago

@cmastalli @spykspeigel : The links ground and Tip doesn't have any associated weight. I guess they are not real links, are they ? Wouldn't you prefer frames ?

cmastalli commented 2 years ago

@cmastalli @spykspeigel : The links ground and Tip doesn't have any associated weight. I guess they are not real links, are they ? Wouldn't you prefer frames ?

What exactly do you mean with frame? I don't understand the proposition.

nim65s commented 2 years ago

I mean these are not real links on the robot, as they don't have a body, with shape and inertia, right ? edit: ground as a mesh. But what about Tip ?

cmastalli commented 2 years ago

I mean these are not real links on the robot, as they don't have a body, with shape and inertia, right ? edit: ground as a mesh. But what about Tip ?

That's correct. I think ground can be removed and tip might be useful for being able to define the robot's tip. @spykspeigel -- do you agree? could you handle this issue asap?

spykspeigel commented 2 years ago

ground link has a mesh and Tip link can be useful. So as I understand I shouldn't modify anything?

nim65s commented 2 years ago

I don't see how the Tip link could be useful. It doesn't have anything more than the EE link, has it ? On the other hand, I can see how this Tip link can break simulations: it doesn't define a mass, so the pybullet simulator will by default assume it has a mass of 1.0 kg, which is plain wrong.

edit: ok, EE is a joint, not a link. Then I think we should only add a 0 mass to the Tip link. I'll handle this, and finish the release if you are fine with that.

spykspeigel commented 2 years ago

@nim65s Thanks a lot for handling this.