The problem is that such prediction and control is ultimately limited because we ourselves are an inextricable component of that "reality" we aspire to control.
It's not control, we are limited by our influence, and there is nothing we can do, in almost all cases.
Essentially, in our interaction with a matter that we are focused on studying, we change it from its natural state. So our data is after the interaction is not per se accurate, relative to the "reality" that we want to observe.
You hold the old dogmatic view of science, and that it is the goal of science to know and study things as themselves. Such a view is too simple.
I see science is knowing the relations among things, because outside such relations, we can not know the acutal reality. As mentioned above, it isn't reality anymore, since we have interacted, and in minute or major ways such an interaction by us (contraptions, sensors, devices, etc) is not the reality that was a moment ago before our interaction.