I'm having difficulty understand Newton's laws of motion in practice, and how to model true dynamic systems. There are two examples below, where I cannot quite figure out what the true forces and motion would be.
In my first example, a rotating actuator (blue) is causing a "finger" (yellow) to move an object (red) along a flat surface. There is a Torque applied to the actuator, which cause the finger to apply an Impact force on the object. The object also experiences Friction and an upwards contact force equal to its Weight. Now, let's say that the finger is of length 1 m, and the Torque is 1 Nm. This means that the force at the end of the finger is equal to 1 N. This force is then applied to the object, causing an Impact force of 1 Nm. Now, let's also say that the object's weight is 1 N, and it's dynamic coefficient of friction is 0.5. This means that the resultant right-to-left force on the object is 0.5 N (= 1 - 0.5). Due to Newton's second law, this means that there is an acceleration of the object in the left direction. However, the actuator is only rotating at a constant angular velocity, so the object cannot be moving with any acceleration. What am I negelecting to model?
In my second example, there are now two rotating actuators, causing two fingers to grip onto the object, which is in free space. The actuators have equal and opposite torque (Torque1 and Torque2). There are two friction forces keeping the object within the fingers (Force1 and Force2). Each finger also causes an impact force upon the object (Impact1 and Impact2). So, the resulting left-to-right force on the object is 0, because the forces from the two fingers cancel each other out. This makes sense to me in theory, but in practice, Torque1 will never be exactly equal to Torque2. If either of these two torques is even slightly different to the other, then this will cause a resulting force to be applied to the object. However, with a robot gripper for example, the fingers would close upon the object, and then maintain the position at the point of contact, even if the torques are slightly different. The object would not slowly start to accelerate in one direction. What am I neglecting to model?