The thing about electromagnets are that they need a source of electricity, thus "electro". You need to have an input to get an output.

Due to the nature of..well, everything I guess, the energy you put in has some amount of loss.

One of the problems surrounding the idea of perpetual energy is that there's no real way to recapture the energy that was lost.

For example, when you charge a battery bank, a certain amount of that energy put into it is lost. When you then charge something using that battery bank, that energy also suffers loss.

Therefore, a major problem and hurdle to overcome with that idea (push/pull electromagnets for perpetual energy) is that eventually, without feeding those magnets energy again, they will stop working.

When it comes to car batteries and EV regenerative braking, we can mitigate or outright generate energy to store in the batteries, but still need another source to get us moving.

When it comes to pumping enough energy into an electromagnet to move something as large and heavy as a car, I wouldn't even begin to pretend to understand the required mathematics or science but what I do know is that there will likely be a greater number of failure rates at some point in the chain, whether that's from insufficient or too much energy or even as simple as the required energy to maintain said electromagnets generating so much heat that something breaks somewhere, possibly even causing the battery to fail itself.

There is also the difficulty of ensuring that electromagnetism can't also damage other components; this is also another hurdle. Tuning could be an issue.

All that said, I'd truly love to move on to newer, better and less destructive battery solutions. Graphene batteries seem pretty promising on that front.

EVs clearly ain't it, at least not the way they've been made.