this post was submitted on 01 Nov 2024
27 points (100.0% liked)
[Dormant] Electric Vehicles
3206 readers
1 users here now
We have moved to:
A community for the sharing of links, news, and discussion related to Electric Vehicles.
Rules
- No bigotry - including racism, sexism, ableism, casteism, speciesism, homophobia, transphobia, or xenophobia.
- Be respectful, especially when disagreeing. Everyone should feel welcome here.
- No self-promotion.
- No irrelevant content. All posts must be relevant and related to plug-in electric vehicles — BEVs or PHEVs.
- No trolling.
- Policy, not politics. Submissions and comments about effective policymaking are allowed and encouraged in the community, however conversations and submissions about parties, politicians, and those devolving into general tribalism will be removed.
founded 1 year ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
I might be wrong, but my impression is that Tesla is years behind Waymo as of now.
Tesla and Waymo are trying to solve the same problem but two different ways.
Waymo chose the more expensive but easier option, but it also limits their scope and scalability.
Tesla chose the cheaper option, but it's much harder to solve, if they can even solve it with todays technology.
Ignoring that, even if Tesla had a viable solution, they would still be years behind Waymo as even in a best case scenario there's going to be a lot of hoops to jump through before they can be operating like Waymo is today.
The difference is that IF Tesla can solve the problem they are trying to solve, they'll be able to operate anywhere in North America and it'll happen at the flip of a switch for either all their cars with HW3/HW4, only cars with HW4, or in the next few years, only cars with HW5, but all future cars from that point forward would have the capability.
Tesla also makes their own cars so can do this at cost, while Waymo has to purchase them from a partner which means there's a markup, and their sensor suite is very expensive.
Someone on another thread mentioned how they wish their car has radar since it can detect everything, to which I replied, it can't detect stationary objects at high speed. If he ever replies, I imagine he's going to say, thats what Lidar is for, except Lidar doesn't work well in Rain, Fog, Snow or Dust. That leaves vision.
Waymo's current tech/fleet won't ever be able to operate as a L5 vehicle (everywhere all the time) unless they solve vision the same way Tesla has to solve vision, or we have some breakthroughs in radar/lidar tech to bypass their weaknesses, or we create a new sensor technology entirely that solves the problems of the others.
Waymo can operate as a L4 though today due to all these extra sensors.
Tesla may never solve the problem, but they are supposedly the furthest ahead in the vision game which is crucial.
I don't buy it. The lidar data is useful for training the vision models, so there's plenty of reason to believe that Waymo can solve the vision issues faster than Tesla.
Tesla uses lidar to help calibrate and validate the vision as well, it's just not needed on the consumer vehicles to do that part.
You'll occasionally see people post photos of them though.
Edit: just to clarify, it's helpful to ensure things match, but once your confident it matches, it's not something every vehicle needs, it's just something you need to keep an eye on with some test vehicles. Waymo is completely reliant on the lidar though as it's used as a primary sensor, but yes it can also help validate their future vision.