Every extra part is extra weight, needing extra fuel, which in turn needs extra-extra-fuel because of the weight of that new fuel, and so on. There is very little redundancy on a rocket.

It's unknown what exactly 'not upright' means just yet.

Fair enough, how about some robotic arms on the drone ship ?

Now you have two problems. But I wonder if something much simpler, like a net, wouldn't make more sense. As soon as the rocket touches down, the net extents and wraps the rocket, keeping it upright.

What would be the two problems?

How would you envision the net though? A net does sound simpler but I have a hard time picturing it

Robot arms are a complex control problem all on their own. To catch the rocket they'd have to be very fast, and very precise. At the same time, the rocket's control loops already have only a very small margin for error. A small miss by either system and you have a much higher chance of failure.

In terms of a net, I imagine something that's lying on the deck under tension between four bendy rods. When the rocket is close enough, the rods snap over to the other side, furling the rocket in the net and holding it more or less steady.

I'm assuming that means it's in 10,000 pieces scattered across the Pacific. I'd be shocked if it didn't topple over, and once it goes over it's blowing up.

> 'emergency' backup legs to address this kind of event?

For the extra weight and complexity a second set of legs would add, I'd expect they'd shore up the primary legs instead, unless they were already at the threshold of being too heavy.