Hi, and thank you in advance for the answers.
I am currently working on the control of a robot gripper, and where I am facing issues is having the object (spherical ball) remain in place to be picked up by the gripper. I have modelled the ball as a 6DOF joint however placing it on a surface or in free space, sees gravitational effects moving the ball before the gripper can ever get ahold of it. In the case where the ball was placed on a surface to be picked up, no external forces were applied however its motion quickly became irratic for no apparent reason, thus I applied a high friction coefficient (sphere to plane contact force between the ball and surface it was placed on) in hopes of keeping it still until the gripper picks it up. This only resulted in non existant converging, getting stuck at a timestamp of 0.014s. Reverting the friction coefficients back to practical values, and instead increasing damping and spring constants in the sphere to plane contact force yielded the exact same converging problem. I then decided to do away with the surface as it seems like aspherical body on a plane seems to roll without any external forces and this may be something inherent to the contact forces library. Thus, I simply imposed PX, PY and PZ damping and spring constants on the 6DOF ball to keep the ball fixed in free space such that the gripper can grab the ball and make contact with it (through sphere to sphere contact forces between the ball and locations on the gripper) and pick it up. This gives unrealistic torque requirements because the PX, PY and PZ internal mechanics act as additional load. I am truely stuck with a situation where I need the ball to be mobile (6DOF) but remain in place until the gripper makes contact with it. Gravitational forces cannot be neglected and the problem is 3 dimesional. Essentially I want to determine the torque required at the joints in order to grip the ball.
Please see the attached image of the problem.