Seems like the only novel part of this is suggesting we use a neural network to plan something related to solar panel tuning? And then it seems like another cas

Seems like the only novel part of this is suggesting we use a neural network to plan something related to solar panel tuning? And then it seems like another case of using the fanciest possible method to get a paper published, because linear regression would probably work better and be easier to implement? And then there's the simple fact that you can't power a car with solar panels on its roof because 1sq-m of solar panels won't produce anywhere near enough power, unless you only plan on driving to the mailbox and back once every other day. Am I missing anything?