I agree that rail strikes are heinously disruptive, but it's a symptom of the elephant in the room, and we can either choose to treat it as an experiment in orderly and fair transition from human to automated labour, or use it as a reason to ban all industrial action, which appears to be where the government is going.
I worry that if we end up with this chaos from the mismanagement of labour transition in one industry we are in for a very rough time when automation starts genuinely threatening much broader categories.
So you're right in the long term, but getting there is hard - you can't just sweep humans under the carpet; it gets unbearably lumpy if you do.
https://tfl.gov.uk/corporate/about-tfl/what-we-do/docklands-...
It serves the financial districts of London. If you're visiting London, it's worth a look; it's always hilarious to watch people in high-powered suits subtly jockeying for the coveted front-right seat, where the driver would normally sit...
The DLR is automated but there is still a human aboard who makes the decision to close the doors and start the train moving. They can also take manual control of the train if needed.
If that is counted as automatic the Victoria, Central, Jubilee and Northern lines are also automated. They have modern ATO systems as well, it's just the train operator gets their own compartment for historical reasons.
The "sub-surface" lines (Metropolitan, District, H&C and Circle) will have similar levels of automation once their long delayed signalling upgrade is complete.
The more interesting thing is the New Tube for London programme to replace trains and signalling for the Picadilly, Central and Bakerloo lines that plans for ATO without anyone aboard the train (at least where possible, various bits where the lines are shared with other services will need human supervision). But that's currently about a decade into the future, they're mostly still at the planning stage, a report about it is at http://content.tfl.gov.uk/ntfl-feasibility-report.pdf (PDF).
BTW there are ~3000 tube train drivers in London. Base pay for that role starts at about 50k, so I'm guessing full cost including overtime, employers' NI, pension contributions and other benefits total 300m-400m GBP per year. So even though the ratio of drivers to passengers is low, there's still a large saving to be made if you can automate the driving and then operate the doors remotely (as you could then have a 1:8 ratio of operators to doors).