Why Our Company’s Trucks Won’t Carry Lidar
spectrum.ieee.org
spectrum.ieee.org
> Reliability is also limited. I mentioned that we use components that are automotive grade, that we’re certain will work over the lifetime of our vehicles, in all the various conditions we require. Lidar isn’t there yet.
That makes me think how inefficient it is for every vehicle to carry its own lidar, radar, etc. Why not pay once for high-quality sensors built into the highway infrastructure, and have them feed data to passing vehicles. It would solve the lidar range problem, among other things.
One obvious challenge would be that smaller roads would not be outfitted with the sensors. Perhaps automated driving would be disabled in that case.
Collective action problem.
It's not impossible to do what you say. But do you live in the United States? This isn't a country does compromise and cooperation very well, especially not now.
A more recent example is cellphone technology. Europe forced standardization of cell technology early on and so cell service (especially 2G and 3G) spread much more quickly and consistently than in the U.S. But there was much more innovation in the U.S.[1] because companies had more degrees of freedom. CDMA lost the standardization wars but won the technology wars, and arguably that's because companies could build out CDMA in the U.S. and Asia.
I don't think there's an objective answer to which model is preferable, not for cellphones nor for autonomous driving. But relative to Europe the United States clearly prefers market-driven solutions. The mantra is that the government isn't supposed to "pick winners and losers" in a competitive market place. It may seem obvious that it's better to cooperate on building a shared infrastructure, but unless and until there's a consensus that such shared infrastructure offers not only the most social utility but that its the "best" approach (imagine all the details!), it's not likely to happen here.
And that's without considering who will pay for such infrastructure.
[1] At least, that's the common wisdom. Which is all that matters when discussing political viability.
> There's far less contention regarding the scope and design of those things.
There is less contention now, when they are mature solutions deployed decades ago. I expect that the situation was much different at the time they were proposed.