Tacc annoyance

[Jeffruby]

love the ap2 experience and being patient. But one thing about tacc local driving (lilely tacc should be used more on higheays)- is the following.

If I'm traveling 65 mph with no one in front of me and I approach a stopped car say 0.5 miles ahead of me at a red light- if I continue with tacc- the car will slam into stopped car. I tried it and got so close before manually breaking and swerving.

Same situation- if at approx 0.25 miles away from stoped car I simply push small stalk in to disengage tacc, let car regen for a second- then deactivate tacc- the car stops beautifully

So- it appears that some different line of site or pulse of radar occurs when activating tacc - but not while maintaining tacc. Wish it would work better with stopped cars. Maybe an update?

 

[Adri]

I beleive it is well know, the radar only tracks moving objects......
Stationary objects are assumed not to be cars..

 

[justg0]

Same situation- if at approx 0.25 miles away from stoped car I simply push small stalk in to disengage tacc, let car regen for a second- then deactivate tacc- the car stops beautifully

I beleive it is well know, the radar only tracks moving objects......
Stationary objects are assumed not to be cars..

Yes, but see the point above of disengaging TACC and then activating it again (I assume the OP meant activate tacc) and then it is able to recognize the stationary car and stop. I use this all the time and it works without fail. So looks like it does recognize stationary objects when you start tacc, not when it is already engaged (?).

 

[xsi123]

I have not notice the issue that TACC does not recognize stopped cars in front but I have read multiple threads about it.

One more issue here is that I think the TACC distance setting is currently just ignored on AP2 cars. Hopefully, when they enable it, it will help with the stopped car issue.

 

[ecarfan]

Same situation- if at approx 0.25 miles away from stoped car I simply push small stalk in to disengage tacc, let car regen for a second- then deactivate tacc- the car stops beautifully

This is unclear to me. You first "disengage" TACC, let car slow slightly, then "deactivate TACC"?

Your first action renders TACC inactive. Your second action is to also render it inactive?

Or did you mean to describe your second action as "re-activate"?

 

[justg0]

This is unclear to me. You first "disengage" TACC, let car slow slightly, then "deactivate TACC"?

Your first action renders TACC inactive. Your second action is to also render it inactive?

Or did you mean to describe your second action as "re-activate"?

There was a typo in original post. You disengage TACC, let the car regen and a second later activate TACC again. The car recognizes the stopped car in front every time and comes to a slow steady stop. Works every time for me.

 

[Jeffruby]

Yes. Sorry. Was typo. After regen I re engage tacc and then the stopped car is noticed.

So saying the radar doesn't recognize stopped objects can't be accurate

 

[S4WRXTTCS]

Yes. Sorry. Was typo. After regen I re engage tacc and then the stopped car is noticed.

So saying the radar doesn't recognize stopped objects can't be accurate

The Tesla blog would be some important reading for you to catch up on.

In summary as of the current version of firmware; TACC doesn't react to stopped objects with either AP1 or AP2.

It's not that the radar can't see them, but that it struggles in locating where it came from exactly. Like if it came from an overhead sign, etc. This could potentially change (at least with full on AP) if the white listing of fixed objects is successful.

All of the radar based adaptive cruise control systems on the market today don't react to stopped objects. If it does it's provided by some other sensor (like stereo cameras, lidar, etc).

Personally I prefer TACC to be simple where it isn't prone to false positives (like what's been reported with TACC on AP2).