4 ms·
As a total layperson, I think the focus should be on long term car-to-car networking. Skip this visual feedback rut we are in and make every new car broadcast i
by meroes 8y ago
As a total layperson, I think the focus should be on long term car-to-car networking. Skip this visual feedback rut we are in and make every new car broadcast its position, velocity, acceleration, heading, etc.
Problem is how to get manufacturers on board en masse.
- blattimwind 8y ago> As a total layperson, I think the focus should be on long term car-to-car networking. Skip this visual feedback rut we are in and make every new car broadcast its position, velocity, acceleration, heading, etc. knock knock -- Who is it? -- Leafs on the road -- What is it now?! -- Raining :>
- asah 8y agoI'm also concerned about security.
- xGrill 8y agoThat doesn't help when there is a pedestrian crossing the street.
- 0xffff2 8y ago>Problem is how to get manufacturers on board en masse. I think security is a much, much larger problem. Designing a safe, secure and reliable system for vehicles to communicate information with each other on an ad-hoc basis and designing a system to make use of that information in a way that doesn't allow bad actors to cause mayhem may be an equally hard problem to solve as the generalized computer vision that we need to make the current self-driving thrust work.