Yes, that's roughly what it does.
There's different ways to not only calculate a final result (in this case distance to vehicle in front of us) but additionally calculate the likelihood that our vehicle distance is true. Additionally there are systems that calculate the amount of error from each input source. So if we had a car with 2 forward facing cameras, two radar and a lidar all looking at the car in front of us (common setup) we'd have five independent sources of information that could feed an estimation engine with data.
If one radar started wigging out (Because for example the terrain on one side of the road was unstable), a lidar started wigging out (because the sun reflected off a bumper into it) or a monocular camera started wigging out (because the car started blending in with the semi-truck in front of it) those systems would detect that discrepancy and start trusting them less.
A good example of that is:
Kalman filter - Wikipedia