Yeah, if you read the article you’ll see that this is really quite far from L4 and just operating in a more controlled environment. They even say they’re “stacking the deck” as much in their favor as they can to ensure it works.
focused on on-demand transportation in geofenced, easily mapped business districts, college campuses, and closed residential communities.
“Our strategy here is to really stack the deck in our favor,” Olson said. “I think most people think about rider-only as being like a technological milestone. And it is, but it’s way more than that.”
I’m curious to see if their current research and development will allow them to scale beyond geofenced areas in the future, or whether the challenges are different enough that the technology is not transferable. It would be nice to start replacing human drivers as soon as possible, and the bar AVs need to pass is pretty low.
I would assume it’s not directly scalable as there have been geofenced autonomous vehicles for years now (maybe not explicitly taxis, but I don’t think that matters too much). I think the real issue is that in the real world you just can’t hold as many variables constant, and we just don’t have the means yet to deal with that.
Its also possible that through this experiment they’ll learn something that helps them chip away at what things they need to hold constant.
Most public transit is on uncontrolled public roads and AVs can’t quite deliver L4 autonomy in that sort of environment.
Yeah, if you read the article you’ll see that this is really quite far from L4 and just operating in a more controlled environment. They even say they’re “stacking the deck” as much in their favor as they can to ensure it works.
Yeah, it’s a very limited version of L4.
I’m curious to see if their current research and development will allow them to scale beyond geofenced areas in the future, or whether the challenges are different enough that the technology is not transferable. It would be nice to start replacing human drivers as soon as possible, and the bar AVs need to pass is pretty low.
I would assume it’s not directly scalable as there have been geofenced autonomous vehicles for years now (maybe not explicitly taxis, but I don’t think that matters too much). I think the real issue is that in the real world you just can’t hold as many variables constant, and we just don’t have the means yet to deal with that.
Its also possible that through this experiment they’ll learn something that helps them chip away at what things they need to hold constant.