I actually work with underwater ROVs, and in general, I think that this sort of a thing is a long way away. For one, most vehicles that small use an onboard bat

I actually work with underwater ROVs, and in general, I think that this sort of a thing is a long way away. For one, most vehicles that small use an onboard battery, and it is really easy to burn that capacity if there's even a slight current. Plus, cables themselves have a large amount of drag, which means that for any distance, the thrusters that can move an ROV under short-distance conditions quickly become under-powered as any flow takes that cable. Even if they could be fully autonomous, there's still a bunch of issues, since caves in general and submerged ones in particular are GPS-denied environments. Acoustics like USBL equipment probably wouldn't work in a cave, instances of underwater VSLAM are still pretty nascent, and inertial navigation system with a low-enough error accumulation rate are almost universally export-controlled by their respective countries. It's definitely an interesting problem, though, and I honestly think that progress in the area is going to be made because folks that aren't experts in caving end up getting interested in it. For example, I'd imagine this kind of a problem is something that space agencies are also really interested in, since there's a lot room in the middle of their respective Venn diagrams.