FSD: Tesla better get on it!

[SeminoleFSU]

UPS Quietly Using Self-Driving Trucks For Months

 

[Daniel in SD]

UPS Quietly Using Self-Driving Trucks For Months

There are literally dozens of companies using “self-driving”* vehicles on public roads.

*meaning there is a driver in the car to take over when the system fails.

 

[diplomat33]

There are literally dozens of companies using “self-driving”* vehicles on public roads.

*meaning there is a driver in the car to take over when the system fails.

Yeah, I would say pretty much everybody is trying to do some form of "self-driving" to various degrees of success. They usually put cameras, radar and lidar on their vehicles, coupled with HD maps, and write software to tell the vehicles how to navigate and avoid hitting anything. And since a driver is in the vehicle to take over if necessary, the software does not need to be perfect. At this point, it's pretty easy to put a system together that can act like a self-driving system. Getting a self-driving system that is actually L4/5 reliable, that's the tough part.

 

[Octo]

UPS Quietly Using Self-Driving Trucks For Months

Yeah, I would say pretty much everybody is trying to do some form of "self-driving" to various degrees of success. They usually put cameras, radar and lidar on their vehicles, coupled with HD maps, and write software to tell the vehicles how to navigate and avoid hitting anything. And since a driver is in the vehicle to take over if necessary, the software does not need to be perfect. At this point, it's pretty easy to put a system together that can act like a self-driving system. Getting a self-driving system that is actually L4/5 reliable, that's the tough part.

True L5 FSD is going to happen at businesses like UPS first.

I mean, those guys have money and huge labor expenses. They can shell out the big bucks per vehicle much easier than a Tesla customer who actually is a lot more likely to want to drive the car instead of being robotaxied.

 

[SeminoleFSU]

There are literally dozens of companies using “self-driving”* vehicles on public roads.

*meaning there is a driver in the car to take over when the system fails.

Thanks for emphasizing my point

 

[SeminoleFSU]

If this company can put together a reliable way to retrofit this system into existing trucks it removes one of the big reasons to buy the Tesla semi... Not having to pay a driver. Sure there are other pluses.. lower fuel costs, theoretically lower maintenance costs, etc but the driver is a big one.
Along those same lines, if a company can come up with a retrofit system for consumer cars that is relatively low cost to install.... And provide the retrofit as a service....

 

[Daniel in SD]

If this company can put together a reliable way to retrofit this system into existing trucks it removes one of the big reasons to buy the Tesla semi... Not having to pay a driver. Sure there are other pluses.. lower fuel costs, theoretically lower maintenance costs, etc but the driver is a big one.
Along those same lines, if a company can come up with a retrofit system for consumer cars that is relatively low cost to install.... And provide the retrofit as a service....

A company that could make a self-driving semi would be a hundred billion dollar company. There is no reason to fiddle around with retrofits. You could sell as many as you could produce and shipping companies would drive the old semis straight to the junkyard.

 

[EVNow]

Along those same lines, if a company can come up with a retrofit system for consumer cars that is relatively low cost to install.... And provide the retrofit as a service....

That's what Comma.ai wants to do. Sell a package to keep lane.

It is unlikely NOA can be achieved without side cameras. So, retrofits won't work well for anything but lane keeping.