There are many, many ways of doing this, with all sorts of mixtures of advantages and disadvantages. You have to first choose one and then do what it takes to make it work. So maybe you didn't choose the optimum one, but that's not a problem so long as it's a viable one. Not choosing leaves you stuck, dithering, on the starting line.
You need to be orderly in making your choice.
Make a list of things the switch must do:
XX volts worst case
YY Amps worst case
How fast must it switch? How much time could it take to switch over?
How much voltage drop is allowable?
What power source is available to it?
What control signal does it get? (voltage, current capability)
Then do a list of things it must not do:
Delay too long
Take too long in switching
Take too much power
Unreliability
Excessive cost
Excessive size
Excessive heat
And plenty of other things you can think of.
In this way you build up a spec for the thing. Specs are important. Without one you could create the ideal thing and not be aware.
Only once you've got these things down can you start designing. There's no point running fast if you hadn't first decided on the direction. It looks boring writing a spec, but it's an important step that can save you a lot of waste and anguish.
Wish I'd done it every time
David