> Сompanies spend billions on self-driving cars with quite limited luck.

That's in part because it needs to be very reliable to not kill people.

If the worst that can happen is killing a garden gnome or running over someone's toes you can tolerate more error.

Yeah, a "small enough to kick out of your way" autonomous driving project is a one or two semester student project (if you're starting from nothing, Sebastian Thrun's old Udacity course was a good way to bootstrap through the algorithmic parts, then maybe watch James Bruton for "everything you could possibly imagine doing with wheels, motors, and an infinite supply of 3d printer filament")

that is because they run in a very different environment than an RC-size car in your living room would

100%. Arms are hard. Mobility is hard. Mobile arms are double hard.

No, rather a larger version of robotic vacuum.