Are Teslas talking to each other?

[enemji]

In an ideal world, all these FSD cars talking to each other would be the best way to ensure that the intentions of each car is clearly expressed to others on the road.

That being said, the reason I raised this topic is because I was driving on FSDs yesterday through the slow rush hour traffic, and typically the car keeps at least a car length of space ahead. But yesterday, there was a new Model Y ahead of me, and my car on FSDs was driving like 2 feet behind it as if my car was linked like a train. It was either that sure and confident that moment in time or it was clearly in-sync with the car ahead of me. What say you?

 

[uscbucsfan]

No.

I drive behind my wife often enough with either both on FSD or just her...it acts like it always does, sometimes too close, sometimes too far.

My cars get really close in rush hour with FSD, regardless of the car.

 

[ewoodrick]

No, they don't talk to each other.

 

[Joboo7777]

In an ideal world, all these FSD cars talking to each other would be the best way to ensure that the intentions of each car is clearly expressed to others on the road.

That being said, the reason I raised this topic is because I was driving on FSDs yesterday through the slow rush hour traffic, and typically the car keeps at least a car length of space ahead. But yesterday, there was a new Model Y ahead of me, and my car on FSDs was driving like 2 feet behind it as if my car was linked like a train. It was either that sure and confident that moment in time or it was clearly in-sync with the car ahead of me. What say you?

I’ve thought about this concept as well. As more and more cars start to leverage FSD capabilities I would think we could solve some major issues like traffic jams, accidents, etc by understanding and adjusting how the car drives and responds based on increased awareness and partnership with other FSD cars on the road. This opens up an entirely new way of thinking g about how we commute from one place to another. In fact I think this type of awareness might be needed to really get to L5. It would take a major shift in how people think of driving. We as a society are too selfish for this concept. But who knows, maybe at some point we can get there.

 

[ewoodrick]

I’ve thought about this concept as well. As more and more cars start to leverage FSD capabilities I would think we could solve some major issues like traffic jams, accidents, etc by understanding and adjusting how the car drives and responds based on increased awareness and partnership with other FSD cars on the road. This opens up an entirely new way of thinking g about how we commute from one place to another. In fact I think this type of awareness might be needed to really get to L5. It would take a major shift in how people think of driving. We as a society are too selfish for this concept. But who knows, maybe at some point we can get there.

There have been numerous studies and one or two companies, maybe Volvo, have implemented it on a vehicle or two, AFAIK.

The NHTSA was expected at one time to make it a requirement, but I think that it has basically been scrubbed because of the possibilities of hackers injecting bad information.

 

[Vines]

There have been numerous studies and one or two companies, maybe Volvo, have implemented it on a vehicle or two, AFAIK.

The NHTSA was expected at one time to make it a requirement, but I think that it has basically been scrubbed because of the possibilities of hackers injecting bad information.

I had often thought a visual code transmitted by the blinking/refresh speed of LED taillights would be an awesome way to implement some simple wireless vehicle-to-vehicle communication.

Even if that communication only contained the magnitude of the brake press, that would be useful information. However, it would be super easy to hack as well.

 

[Joboo7777]

There have been numerous studies and one or two companies, maybe Volvo, have implemented it on a vehicle or two, AFAIK.

The NHTSA was expected at one time to make it a requirement, but I think that it has basically been scrubbed because of the possibilities of hackers injecting bad information.

security is definetly important, but I would think this is obtainable. We have major dependencies on aviation and they have figured out how to maintained a high level of security. You could argue that Tesla’s own environment roday, if compromised, could do some major damage.

 

[ewoodrick]

I had often thought a visual code transmitted by the blinking/refresh speed of LED taillights would be an awesome way to implement some simple wireless vehicle-to-vehicle communication.

Even if that communication only contained the magnitude of the brake press, that would be useful information. However, it would be super easy to hack as well.

Teslas can see brake and turn lights. The magnitude is easily determined by the change in gap.

 

[Vines]

Teslas can see brake and turn lights. The magnitude is easily determined by the change in gap

Though this is true, direct communication is better. How otherwise can you differentiate between a car in full brake lockup that is panic stopping but not actually stopping very quickly compared to the magnitude of the brake pressure.

Also when just one brake light is showing you don't have the same distance measurements between the 2 lights with which to make these split second decisions.

Direct communication is better imo.

 

[enemji]

Though this is true, direct communication is better. How otherwise can you differentiate between a car in full brake lockup that is panic stopping but not actually stopping very quickly compared to the magnitude of the brake pressure.

Also when just one brake light is showing you don't have the same distance measurements between the 2 lights with which to make these split second decisions.

Direct communication is better imo.

My first car was the best. The harder I pressed the brake, the brighter the brake lights were. Unfortunately that feature was removed from the cars following for they went the route of relays and switches.

 

[Vines]

My first car was the best. The harder I pressed the brake, the brighter the brake lights were. Unfortunately that feature was removed from the cars following for they went the route of relays and switches.

Cool what kind of car was that?

 

[ewoodrick]

Though this is true, direct communication is better. How otherwise can you differentiate between a car in full brake lockup that is panic stopping but not actually stopping very quickly compared to the magnitude of the brake pressure.

Also when just one brake light is showing you don't have the same distance measurements between the 2 lights with which to make these split second decisions.

Direct communication is better imo.

It only takes the car about 1/15th of a second to determine that the gap is closing.
It doesn't measure between the lights, it uses many other bits of information.

But let's say that the car in front was broadcasting it's information. You are on an 8 lane Interstate in stop and go traffic. You hear "Panic Stop" being transmitted!

Which car was it? The one in front? or back? The one immediately to the left? or right? The ones in front of those? or behind? Or the ones past even those.

And let's just say that the brake lights send the information by quick flashing, More than likely, before the information can be sent, collision has occurred.

And the pièce de résistance? It's a kid in the back seat with a Raspberry Pi sending it.

 

[enemji]

Cool what kind of car was that?

A Morris Minor

 

[enemji]

But let's say that the car in front was broadcasting it's information. You are on an 8 lane Interstate in stop and go traffic. You hear "Panic Stop" being transmitted!

Which car was it? The one in front? or back? The one immediately to the left? or right? The ones in front of those? or behind? Or the ones past even those.

First rule of communications: Use of Unique identifier.

This will allow every vehicle to identify the position of other cars in their vicinity. They would not need vision anymore as each one would be transmitting their GPS signal, in addition to other standard operating parameter and every other vehicle can identify their own position and status in relation to them.

 

[ewoodrick]

First rule of communications: Use of Unique identifier.

This will allow every vehicle to identify the position of other cars in their vicinity. They would not need vision anymore as each one would be transmitting their GPS signal, in addition to other standard operating parameter and every other vehicle can identify their own position and status in relation to them.

GPS isn't that accurate, especially at 60 mph.

 

[enemji]

GPS isn't that accurate, especially at 60 mph.

Yes. Hoping that gets better as demand drives it

 

[PianoAl]

typically the car keeps at least a car length of space ahead. But yesterday, there was a new Model Y ahead of me, and my car on FSDs was driving like 2 feet behind it

I've seen the same thing but with a non-Tesla. That is, following distance on FSD is fine, but once in a while, the car tailgates.

I hope engineers are thinking ahead as to how to implement car-to-car communication. Do airplanes do that at all?

 

[gsmith123]

Airplanes do yes, but I kind of think you all are wildly overthinking this. Or at least designing complex solutions when there’s 5 - 10 years of more straightforward optimization to do first.

 

[jgleigh]

 

[ewoodrick]

GPS isn't that accurate, especially at 60 mph.

There's some really basic physics here.
When you are travelling at 88 feet per second, two vehicles calculating their position only a half second apart is the length of a school bus.

And that's if GPS is accurate to cm. It's not. And those solutions that able it to be take time.