Inverse Kinematics with Closed Loop Parallelogram

[coppelia]

Please post the scene again. My first guess is that the issue is directly linked to bad mass/inertia ratios, for the given physics engine. Try with a different engine first to see if the issue is related to the used physics engine. Then you may have to further adjust (i.e. balance-out the mass/inertia ratios).

Cheers

[p3jawors]

I have tried with both ODE and Bullet with some success. The image I posted was with ODE; with Bullet it has a similar effect but slowly drifts until the joint separates.

https://drive.google.com/open?id=0B3Ygg ... XhhZmdFMkE

Thanks

[coppelia]

You have modified the scene we have prepared for you. So you broke many things. Have a look at the scene that we sent you via email. It is important to carefully adjust the masses/inertias when using the Bullet or ODE engines.

Cheers

[p3jawors]

My mistake, you mentioned the approach was to have invisible IK groups and visible dynamic groups. However, I did not see the separate groups in the scene you sent as they are in examples like the uarm. How are these invisible IK groups set up if they are different from that example?

[xiemzuo]

Hello, I want use uarm to realize kinematic simulation(not dynamic). I hope my uarm can follow the path added by me.
And now, my problem is I don't know how to close the parallel mechanism and how to add IK Group. If anyone can help me, I will be very grateful. This is my scene


https://onedrive.live.com/?cid=BCF5C68A ... ot&o=OneUp

[coppelia]

Hello,

did you try to follow first a simple tutorial about IK in V-REP?

Cheers