Different gripper mass but no changing in torques

Hello,

I am currently working with two versions of the UR5 robotic arm: the original version and a modified version where I have only reduced the mass of tool0. However, during simulations using the inverse dynamics block in Simulink, I've observed that the torques do not appear to change between the two versions. This is unexpected because reducing the mass should affect the torques.

Could this behavior be considered normal? If not, what could be the possible reasons for not observing any changes? Is it because the tool0 has no inertia maybe ? The trajectories are generated by CHOMP and then fed into Simulink using minjerk polynomial trajectories to obtain the joint variables (q, qd, qdd), which are subsequently used in the inverse dynamics block. The model is quite basic.

Answer 1

[Karsh Tharyani]

Hi @Denizhan AKINCI,

Inverse Dynamics should be affected by changing mass of a body. It will be helpful to take a look at your whole system to get a sense of what is going on. But, to illustrate, see the changed torque outputs.

ur=loadrobot('universalur5e',dataformat='row'); 
q=rand(1,6);                                                       
qd=q;                                                              
qdd=q;                                                             
inverseDynamics(ur,q,qd,qdd)                                       

ans = 1x6

-0.3293 3.4280 1.8323 0.1584 0.0042 -0.0000

<mw-icon class=""></mw-icon>

<mw-icon class=""></mw-icon>

ur.Bodies{end}.Mass                                                

ans = 0

ur.Bodies{end}.Mass=2;                                             
inverseDynamics(ur,q,qd,qdd)                       

ans = 1x6

-0.2623 5.0726 3.2131 0.5601 0.0065 -0.0000

<mw-icon class=""></mw-icon>

<mw-icon class=""></mw-icon>

Hope that helps!

Karsh