Nav on AP to get automatic HOV lane detection

[diplomat33]

"Tesla is expected to add an automatic High-Occupancy Vehicle (HOV) recognition feature that will enable Navigate on Autopilot to automatically utilize carpool lanes when the vehicle detects two or more passengers."
Tesla Navigate on Autopilot to get Automatic HOV Lane detection

 

[eli_]

Seem like that could get complicated, some lanes require more than 2 passengers, some have varying time of day rules, etc.

 

[r0xx0r]

They just need to do the simple thing first: Recognize the HOV lane.

Then, later on, they will solve all those different kinds of complicated signs/rules.

 

[Puddles]

Isn't there already a switch saying whether or not you want NoA to use HOV lanes? Implying such recognition exists already?

Although I, and many others, complain incessantly at how badly it works.

Edit... okay, I see what you did here. The car will make the determination for you.

But still, if it can't even tell what an HOV lane is in the first place, I'm not convinced an additional layer of complexity is a good idea.

 

[diplomat33]

Isn't there already a switch saying whether or not you want NoA to use HOV lanes? Implying such recognition exists already?

Yes there is. What the feature will do is automatically flip the " Use HOV lane" option on or off based on how many people are in your car (using the car seat sensors).

 

[boonedocks]

In GA we get to use the HOV lanes with just the driver if our car has an Alternate Fuel Tag. Wouldn’t help us at all

 

[alsetym]

Isn't there already a switch saying whether or not you want NoA to use HOV lanes? Implying such recognition exists already?

Although I, and many others, complain incessantly at how badly it works.

Edit... okay, I see what you did here. The car will make the determination for you.

But still, if it can't even tell what an HOV lane is in the first place, I'm not convinced an additional layer of complexity is a good idea.

I typically turn off NoA in HOV lanes. The car seems to still try to get out of the lane as if it is lane hogging or if the next lane over is going a bit faster (temporarily). And it only recognizes about 75% of the HOV lanes in southern ontario (i.e. 403/QEW eastbound!!). I've generally assumed these are map issues, so here's hoping they either give us some way of telling them what is wrong with their maps or stop using them for NoA! That might solve the tendency to route me onto the collectors rather than the express.

 

[Kmartyn]

In Ontario all green plated vehicles can use the HOV lanes with just a single person in the car.

 

[alsetym]

In Ontario all green plated vehicles can use the HOV lanes with just a single person in the car.

Agreed...which SHOULD make it even easier for NoA to figure out lol!

 

[kaffine]

Agreed...which SHOULD make it even easier for NoA to figure out lol!

I don't think the rear camera can see its own license plate how will it know it has a green plate?

 

[LakeWorthB]

Here in South Florida, my car doesn't like the way the double lines are for HOV lanes. They are dashed lines, but they are 3ft apart, so when it changes lanes over it, most of the time it will either abort, or activate alarm. So I so far just leave HOV lanes off.

 

[run-the-joules]

I'd rather that it understand how time-based HOV lanes work in the Bay Area before they worry about this. Right now if I'm commuting during HOV times, nav on autopilot is 100% unusable because it doesn't know that the left lane is HOV-only and I'm there on purpose, and keeps trying to get out of the lane miles and miles before my exit… which is on the left side if you're in the HOV lane.

 

[eli_]

We have some highways with HOV restrictions in the right-hand lane instead of left, sometimes even on both sides of the highway at once, and a lot of times the map data it relies on doesn't mark those correctly I assume. I tried fixing the lane tagging in OSM, but I have no idea if it will make it to the car, or if it's even using OSM tagging for NoA. They also still load some TomTom map data, it's like this weird mashup of data sources.