Expecting a technology to be released only after it can handle *ALL* the edge cases one can think of, is silly and detrimental. If it handles a specific set of use cases, and that is clearly communicated then we should be good to go.
It's going to be a long time before cars can handle every edge situation safely. Aircraft have been on the autonomous flying track much longer than anyone has been working on it for cars. The joke at Boeing was the new flight deck was one pilot and a dog, the pilot's job was to feed the dog and the dog's job was to bite the pilot if he touched anything. Planes can largely fly themselves, but they still run into edge situations where the pilot needs to step in and take control. Many aircraft accidents in recent years were due to the pilots stepping in during an edge situation and making mistakes they shouldn't have made.
The edge situations will be reduced, but don't expect them to be completely eliminated for a very long time. I expect the edge situations will be reduced from around 0.1% now to 0.01% or even 0.001% in the next few years. The best they can do with the remaining edge conditions is to fail as safely as possible.
The driver of the Tesla must have been completely engaged in something else, or possibly asleep. Most people if they are aware they are about to go under a semi trailer are going to duck if they know they can't stop. Nothing indicated the driver did anything to react.