Gravity setting for accurate joint force/torque data

Post Reply
jl18
Posts: 1
Joined: Mon Mar 16, 2020 4:55 pm

Gravity setting for accurate joint force/torque data

Post by jl18 »

Hello, I am trying to do robot grasping and collect joint force/torque data in pybullet.

I found that the joint force/torque measurements could be affected by the gravity setting. For example, if I SetGravity(0,0,0), the forces will be close to zero. I found many examples use gravity (0,0,-10) and if I SetGravity(0,0,-10), the forces will be larger.

Is the unit of SetGravity N or N/kg?

Is (0,0,-10) the correct value of gravity to get accurate force/torque reading?

Thanks !
User avatar
Erwin Coumans
Site Admin
Posts: 4221
Joined: Sun Jun 26, 2005 6:43 pm
Location: California, USA
Contact:

Re: Gravity setting for accurate joint force/torque data

Post by Erwin Coumans »

This is expected: joint reaction forces are influenced by gravity acceleration, there is no explicit gravity compensation performed.

The gravity acceleration is N/kg.

Also, for more help you may want to create a small reproduction case. There are various robots available in PyBullet SDK (panda, xArm etc, see https://github.com/bulletphysics/bullet ... let_robots)
Post Reply