There are several million truckers in the U.S., so it's somewhere around 1% (but it's probably less than 2%).
I guess there are probably lots of delivery drivers, but those vehicles will still need tenders (at least until robots can do the last 100 feet of the delivery).
The "issue" is probably not for the general workforce, but it will significantly reduce the number of not-so-qualified jobs. According to a quick googling "1 in 15 worker in a long truck driver according to the ATA" [1]. But this is expected that automation tend to reduce these kind of jobs first.
Your link says that 1 in 15 workers is employed in the trucking industry. It says there are 3.5 million truck drivers, which is consistent enough with my 1%.
I'm sure it will create lots of turmoil if truckers are put out of work. But there is already an enormous amount of turmoil in the employment market (for lots of reasons). Especially in the part of the employment market that truckers participate in.
I don't understand why subways and trains aren't implementing. The universe of possibilities is so incredibly limited in these situations that it seems like a piece of cake, compared to city driving.
"Can you see that the track is clear of obstacles? If not, stop!"
They are, often on a huge scale. The DLR[1] in London is entirely automated (although sometimes the ticket inspector will sit at the front with the backup control panel open, presumably so they can pretend to drive a train), and the Central, Northern, and Jubilee lines also run semi-automated, and I'm pretty sure the bits that aren't automated are due to union lobbying rather than any particular need.
