Studying Antarctica is different; we already know life is there, so we're studying the minutiae of "where and how". On Mars, we're yet to answer the two most critical questions: whether there is - or was - life, and if so, how it is related to the life on Earth, if at all.
As for figuring this out quicker: an astronaut would be able to leave their ship/base, go find some interesting rock, bag it, bring it back, slice it and observe under microscope, all in time it would take the rover to just drive towards the nearest rock. All it takes for a human to be a better sample collector than most advanced of robots on Earth is a shovel.
Robotic missions scale better and may still be more cost-effective, if we built and sent a crewed science ship to Mars orbit. It doesn't have to land, the crew could pilot dozens of rovers remotely at the same time - but this way, with much higher bandwidth (could actually drive FPV looking through a HD camera) and much less lag (which, for Earth-Mars link, is 3 to 22 minutes, depending on relative position of the planets, and that's one way, not RTT). I'd say this is the perfect trade-off until we make more progress on AI front, but it requires manned flights. And if we'll be at the point of sending science ships to Mars, we won't be that far from landing humans there anyway.
As for figuring this out quicker: an astronaut would be able to leave their ship/base, go find some interesting rock, bag it, bring it back, slice it and observe under microscope, all in time it would take the rover to just drive towards the nearest rock. All it takes for a human to be a better sample collector than most advanced of robots on Earth is a shovel.
Robotic missions scale better and may still be more cost-effective, if we built and sent a crewed science ship to Mars orbit. It doesn't have to land, the crew could pilot dozens of rovers remotely at the same time - but this way, with much higher bandwidth (could actually drive FPV looking through a HD camera) and much less lag (which, for Earth-Mars link, is 3 to 22 minutes, depending on relative position of the planets, and that's one way, not RTT). I'd say this is the perfect trade-off until we make more progress on AI front, but it requires manned flights. And if we'll be at the point of sending science ships to Mars, we won't be that far from landing humans there anyway.