Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Summon seems like a silly party trick to me

Economite

Member
Dec 19, 2016
490
589
Los Angeles
I understand since I agree that Smart Summon will almost certainly result in additional people getting injured or killed than if it wasn't released. I'm not exactly sure why I'm not as upset as you about it. I guess I see that people do get enjoyment out of it and that's got to be worth something.

Here's a question. If General Motors somehow caused all of their cars on the road to have this same version of Smart Summon active, would you be ok with that from a road safety perspective? 'Cause if it would be irresponsible for GM, it is also irresponsible for Tesla.

Tesla's "advantage" over the rest of the auto industry here isn't that it is technologically more advanced, but that it is willing to put onto the road features that all of the other manufacturers would (correctly to my mind) consider to be not safe enough to release.

If all the manufacturers acted the way Tesla did with AP, FSD and their subcomponants, we'd have a mess on the road (and in the parking lot). Tesla only gets away with this because it is relatively small. But putting cars on the road that imitate toddlers is a menace to everyone else on the road.
 
  • Like
Reactions: croman and am_dmd

alsetym

Member
Apr 28, 2018
276
349
hamilton
Actually, no.... this is rather bad.

There are 268.8 million cars in the United States.
There are 50,000 parking lot crashes per year in the United States.
That's a 0.02% chance of parking lot accident per car-year for the entire population of cars.

There are ~500,000 Teslas in the United States.
There have been ~10 Smart Summon accidents in just five days.
That's ~730 Smart Summon accidents in a prospective year.
That's 0.1% chance of Smart Summon accident per car-year for Teslas.

Even with just a few days of data, it looks like it could be as much as 5x worse. This is NHTSA is investigating it already.

In order to get this back on track I was wondering how long we'll have to go before we determine 'today is the day that SS hasn't hurt or killed anyone for long enough it is roughly equivalently as safe as a human'. Using Sandia's base of 268.8M cars (which is a high estimate of both cars registered and also of cars that are actually used...I have a car in my garage that is registered but got used twice this year and 0 times last year for instance).

I haven't found EXACT numbers but plenty of references to 60,000 injuries (911 type injuries) and at least 500 deaths per year in the US as of 2016, with a sharp trend upwards due to distracted driving so likely much higher now. I'm always wary of round numbers, if anyone has more reliable...feel free. So that means 268.8/60,000 gives us 1 injury per 4,480 cars registered per year and one death per 537,600 cars registered.

What does that mean to SS? A true apples to apples is impossible, we don't have all the right data. But lets do some estimating.

I looked around a bit but couldn't find any official estimates on how many parking lot trips we take in a year. I'd guess rural populations use parking lots much less than once a day and sub-urban populations use parking lots a bit more than once a day. I've lived in both, that's my experience. If the average registered car uses a parking lot once per day that is 4,480 cars/injury/year x 365 parking lot retrievals/year or... 1 injury for every 1.6m times someone goes back to a car in the parking lot and leaves (i.e. a summon opportunity). Lets assume twice that to be generous and reduce counter argument...so 1 injury for every 3.2m times someone goes back to a car in a parking lot by my figuring.

From the other direction. Lets assume each summon attempt requires 5 registered attempts. That's high, but I'm feeling generous. This is good will towards @electronblue for staying reasonable. High fives and hugs and stuff.

So, we would need to clock 5x 3.2M = 16.3M summon attempts with no significant injuries to be pretty convinced that Summon is less dangerous (safety, not property damage) than a regular human. At 100k summon attempts per day, (likely more now that V10 is fully rolling out but may taper in time as it gets old) it will take approximately 163days (23 weeks, 5.5 months)...so somewhere between feb and april 2020 (assuming we already have several hundred k attempts before the FSD roll out). I did NOT, this is important, look to see if it is a leap year so that's +0/-1 days range.

that's a while, I was hoping it would be like next week so we wouldn't have to wait lol. But I also think that's a very conservative estimate (assuming my math is right...I didn't use a spread sheet, so, well, you've been warned). I'd bet that tesla (and more importantly NHTSA) will be able to declare this sooner, maybe as little as 3 months. Things will get twisted because updates will make some things better and some worse, so we won't really no for sure of course.

First, please chime in with my mistakes or better estimates. Second, votes on if we make it 23 weeks (or whatever the right number is)? I don't care about property damage for this analysis. That was covered earlier, and yep SS is currently not up to par with a human there, though rates are still low they need improvement to be as good as good 'ol mama meat sack.

@SandiaGrunt you don't get a vote.
 
  • Helpful
Reactions: electronblue

S4WRXTTCS

Well-Known Member
May 3, 2015
5,740
6,897
Snohomish, WA
IRL, people listen when I talk. Online, no one gives a **** about anything I say. It can be pretty frustrating. My wife periodically forces me to delete the Twitter app from my phone.

Ha, you wouldn't survive 2 min on Elons twitter. There is hardly anyone here that is anywhere close to as bad as the "fans" Tesla has on his twitter feed. I can't imagine ever having a productive conversation there.

On here my initial interactions with you didn't go well as you immediately assumed I was an idiot when you accused me of thinking satellites had a live view. This was when mentioned that path planning would be better if Tesla used the satellite map view to figure out the best path. That's the view I use when trying out Smart Summon. But, apparently you didn't connect the dots because you're not an owner. An owner would have immediately known what I was talking about.

Pretty much everything you bring up is things you see or read about, and not things you experience with a Tesla.

It's hard to take that all that seriously without knowing more.

Some non-owners are simply fans, but some are the opposite. Some are genuinely interested in the technology, and they come on here to talk about it. But, there is a small amount of people who come on here just to insult us owners. Where they take pleasure in it.

Most of us on here are owners that are here to exchange experiences, and to get help with stuff. Some of us are hopeful about the future of FSD, and some of us like myself think the whole FSD was a really bad move on the part of Tesla. To commit so early to something no one has really pulled off. Ughh.

But, even those of us like myself want to see things improve. I have hope that HW3 will bring significant improvements. I've pretty much written off Smart Summon for now.
 

Daniel in SD

Well-Known Member
Jan 25, 2018
7,067
10,533
San Diego
Here's a question. If General Motors somehow caused all of their cars on the road to have this same version of Smart Summon active, would you be ok with that from a road safety perspective? 'Cause if it would be irresponsible for GM, it is also irresponsible for Tesla.

Tesla's "advantage" over the rest of the auto industry here isn't that it is technologically more advanced, but that it is willing to put onto the road features that all of the other manufacturers would (correctly to my mind) consider to be not safe enough to release.

If all the manufacturers acted the way Tesla did with AP, FSD and their subcomponants, we'd have a mess on the road (and in the parking lot). Tesla only gets away with this because it is relatively small. But putting cars on the road that imitate toddlers is a menace to everyone else on the road.
I’m not really ok with Tesla doing it so of course not. I agree with you and GM through their ownership of Cruise has far more sophisticated autonomous technology than Tesla and they don’t plan to commercialize it until it’s no longer a beta system. I’m not convinced that Smart Summon is a major disaster since I don’t think people will actually end up using it.
If it were up to me property owners would have to opt in to allowing Smart Summon to be used on their property. If I owned a parking lot I sure wouldn’t want it being used there.
 

bhakan

MSLR MSM/Blk/19/FSD, 6/10 order, EDD 11/18-12/08
Jan 2, 2018
270
260
Bethesda, MD
I have tried it in my small'ish very low traffic parking lot at my office and it failed both time and just stopped weirdly at an angle, like it just gave up. It may have continued after some time, I did not release the hold button for an addition 7-10 seconds with no continued movement, and gave up and ran to the car to get it out of the way.

Pretty simple scenario too:

View attachment 461444
i tried numerous times in few locations since last Saturday. Very simple task and the car has to just turn and come 20-25 ft. It did not complete the task. Constantly I get the message "Summon aborted." It does not matter to me as much as that is not the reason I bought the car. But I want to just give it a try since others talk about it or if friends ask me to do the trick I have to acknowledge that at least for me it is not working.
 

alsetym

Member
Apr 28, 2018
276
349
hamilton
I’m not really ok with Tesla doing it so of course not. I agree with you and GM through their ownership of Cruise has far more sophisticated autonomous technology than Tesla and they don’t plan to commercialize it until it’s no longer a beta system. I’m not convinced that Smart Summon is a major disaster since I don’t think people will actually end up using it.
If it were up to me property owners would have to opt in to allowing Smart Summon to be used on their property. If I owned a parking lot I sure wouldn’t want it being used there.

Keeping in mind of course that GM hid the ignition switch issue and freely admits at least 13 (but could be as many as 153) people died because of it. Either way, any unnecessary deaths are not good but I'd rather Tesla's approach of making it public. I also believe Tesla believes they won't kill anyone for the next 23 weeks or so.
 

Daniel in SD

Well-Known Member
Jan 25, 2018
7,067
10,533
San Diego
Keeping in mind of course that GM hid the ignition switch issue and freely admits at least 13 (but could be as many as 153) people died because of it. Either way, any unnecessary deaths are not good but I'd rather Tesla's approach of making it public. I also believe Tesla believes they won't kill anyone for the next 23 weeks or so.
If it were Tesla people on this forum would be blaming the victims for having too much crap on their keychains.
What happens in 23 weeks?
Edit: I read your post. The statistics on Smart Summon usage are pretty useless. We simply don’t have the data. I racked up a large number of usages just summoning two Teslas towards each other to see what would happen. I bet the vast majority of people aren’t substituting actual parking lot driving with Smart Summon. And since a real use of Smart Summon would bring the car closer to the entrance of a building it would actually increase the number of accidents simply by increasing the distance you drive yourself in the parking lot.
 
Last edited:

cucubits

Active Member
May 17, 2019
1,718
1,749
TX
Smart Summon is laughable compared to the state of the art in autonomous vehicles. In other threads people were claiming that it was being worked on by the B team but Musk confirmed that this is the work of the best minds at Tesla.

This! This is what a lot of people here refuse to see and they get unreasonably upset when they realize this is where Tesla actually is now and nowhere near on the verge of FSD robotaxis.

I can understand it and I get "irked" when I keep seeing unfounded optimistic statements made by fanboys...
 

S4WRXTTCS

Well-Known Member
May 3, 2015
5,740
6,897
Snohomish, WA
In order to get this back on track I was wondering how long we'll have to go before we determine 'today is the day that SS hasn't hurt or killed anyone for long enough it is roughly equivalently as safe as a human'. Using Sandia's base of 268.8M cars (which is a high estimate of both cars registered and also of cars that are actually used...I have a car in my garage that is registered but got used twice this year and 0 times last year for instance).

I haven't found EXACT numbers but plenty of references to 60,000 injuries (911 type injuries) and at least 500 deaths per year in the US as of 2016, with a sharp trend upwards due to distracted driving so likely much higher now. I'm always wary of round numbers, if anyone has more reliable...feel free. So that means 268.8/60,000 gives us 1 injury per 4,480 cars registered per year and one death per 537,600 cars registered.

What does that mean to SS? A true apples to apples is impossible, we don't have all the right data. But lets do some estimating.

I looked around a bit but couldn't find any official estimates on how many parking lot trips we take in a year. I'd guess rural populations use parking lots much less than once a day and sub-urban populations use parking lots a bit more than once a day. I've lived in both, that's my experience. If the average registered car uses a parking lot once per day that is 4,480 cars/injury/year x 365 parking lot retrievals/year or... 1 injury for every 1.6m times someone goes back to a car in the parking lot and leaves (i.e. a summon opportunity). Lets assume twice that to be generous and reduce counter argument...so 1 injury for every 3.2m times someone goes back to a car in a parking lot by my figuring.

From the other direction. Lets assume each summon attempt requires 5 registered attempts. That's high, but I'm feeling generous. This is good will towards @electronblue for staying reasonable. High fives and hugs and stuff.

So, we would need to clock 5x 3.2M = 16.3M summon attempts with no significant injuries to be pretty convinced that Summon is less dangerous (safety, not property damage) than a regular human. At 100k summon attempts per day, (likely more now that V10 is fully rolling out but may taper in time as it gets old) it will take approximately 163days (23 weeks, 5.5 months)...so somewhere between feb and april 2020 (assuming we already have several hundred k attempts before the FSD roll out). I did NOT, this is important, look to see if it is a leap year so that's +0/-1 days range.

that's a while, I was hoping it would be like next week so we wouldn't have to wait lol. But I also think that's a very conservative estimate (assuming my math is right...I didn't use a spread sheet, so, well, you've been warned). I'd bet that tesla (and more importantly NHTSA) will be able to declare this sooner, maybe as little as 3 months. Things will get twisted because updates will make some things better and some worse, so we won't really no for sure of course.

First, please chime in with my mistakes or better estimates. Second, votes on if we make it 23 weeks (or whatever the right number is)? I don't care about property damage for this analysis. That was covered earlier, and yep SS is currently not up to par with a human there, though rates are still low they need improvement to be as good as good 'ol mama meat sack.

@SandiaGrunt you don't get a vote.

6 million accidents occur per year in the US
1/5 of all the accidents happen in parking lots so that's 1.2 million accidents occurring in parking lots.
The average household takes 9.5 trips a day where 82 percent of them are vehicle trips. So that's about 2843 car trips per year per US household.
There are 127.59 million households in the US so that's 362,738,370,000 trips per year.

So on average a parking lot accident happens once every 302,282 trips (approx)

The trickiest part of any kind of AV calculation is you have to determine how much safer it should be.

I would employ a sliding scale that was more okay with accidents/mishaps during the start than at a later date.

I would say an average summons probably covers about 1/4" of a parking lot trip. There is no self park, and you still have to drive out of the parking lot from where it picked you up.

So we're talking an allowance of maybe 1 accident per 1.2 million summon events to be on par with humans.

Now not every summon event is successful as sometimes it stops for apparently no reasons. In my testing its around half of the time. So that 1.2 million is more like 2.4 million.

Basically I'd like to see Smart Summon reach about 2.5 million or so without any accident.

I don't see that happening, but you never know.

I love that Tesla is now in the insurance game so they're going to be directly plugged into actual claims. So we shall see what becomes of that. Would be funny if it got shut down only for Tesla insurance folks. Like "Lower your insurance rate by giving up Smart Summon!!"
 
  • Like
Reactions: alsetym

S4WRXTTCS

Well-Known Member
May 3, 2015
5,740
6,897
Snohomish, WA
This! This is what a lot of people here refuse to see and they get unreasonably upset when they realize this is where Tesla actually is now and nowhere near on the verge of FSD robotaxis.

I can understand it and I get "irked" when I keep seeing unfounded optimistic statements made by fanboys...

I highly doubt it.

Summon looks and feels like what's essentially throw away code. There is absolutely nothing in the way of driving policy, and it doesn't appear to be looking for any objects that it wasn't already trained to.

This has zero resemblance to what they were demoing with reporters and investors in the cars back on autonomy day with HW3.

This also has no HW3 neural networks.

So I'm going to reserve judgement on Karpathy, and others until well after HW3 ships. I don't doubt that they had some involvement, but they had much bigger fish to work on.

I've already said I was going to bail on Tesla if nothing interesting came from their FSD efforts. There are too many other interesting EV's like Rivian, and the Taycan (once they start shipping cheaper ones). I've already been a Tesla owner for FAR longer than I normally stick with a car company. I get bored.
 
Sep 10, 2017
474
463
Outside
In regards to the Birdseye view /360 camera. Has anyone seen what type of images these things see? Coming from what my last two Volvo’s see, the side view images are terribly distorted when it comes to picking out images you can “label” as a “target”. Mostly just blurry blobs if someone is standing next to the vehicle. That being said, Volvo’s 360 cam is one of the “better” ones available. Looks like Audi, BMW, and MB have fancier ones that allow you to manipulate the stitched image and flip it around and whatnot. Maybe they have a better sense of their surroundings/better processing of the images?
 

maverick3n1

Member
Apr 30, 2019
44
27
Oceanside
Davecolene,
Waiting for a better beta is just too long.

It was promised 3 years ago in 2016.

It has been promised coming out soon "next week" such as promised again in April of this year.

It's been almost 6 months since April, so, I don't mind a very bad beta right now because I don't want to wait for it after 3 long years!
That's what gets me.. it was promised 3 years ago.. implying that when they are promising such an item, it should be near ready for release when they announce it. Much like a movie doesn't start advertising trailers until it's most of the way completed.

Instead, it seems more like an upcoming movie deciding "Lets film a few quick action scenes, make a trailer of it and start advertising it on the market.. then we'll start to film the actual movie. Assuming we don't run out of funding or run into licensing issues or actors bouncing out during filming etc etc, we will release in 2-3 years".

If they had an almost ready alpha project to release when they announced it 3 years ago, you'd assume that they have been testing it and improving upon it for the last 3 years. This feels like they announced the project when they started it, not even knowing how viable it would be, how many problems they would run into, or how long it would take to have a working model. Then when they finally had a working model that sort of works, they realized "Crap, we announced it 3 years ago.. we better release something to the public so they don't get mad that we just barely got an alpha version working".

It's a common trend with Tesla and Elon Musk in general. Announce something, then under deliver for years. When the product is finally released, and all of the bugs are worked out of it, it's a great product. I can't however, count how many businesses I've seen that do just as he is doing, and end up going under in the long run because they continue to under deliver. The only reason Elon Musk has been able to keep it up where others fail, is that he has created a cult following, much like Apple, where if it has his brand on it, they want it, no matter how great or bad the product is when it's first released. It's why it's such a volatile stock to invest in. "It's coming... btw, I need more investors. I'm out of money!" We never know whether this time, the numbers he promises the stocks will increase, will happen, or if they will plummet. We never know if he will actually deliver on what he states he will deliver on.
 
  • Like
Reactions: croman and cucubits

diplomat33

Average guy who loves autonomous vehicles
Aug 3, 2017
8,409
11,755
Terre Haute, IN USA
This is my final word on the subject now based on my personal experience with Smart Summon and everything I know about it:

Smart Summon is basic right now. It can handle the ordinary stuff. It will stop for cars. It will stop for pedestrians in the way. It will pause at crossings. It does pause frequently with a message that it is waiting for the path to clear, even when the path seems clear. So it is overly cautious. If you use it for something simple like moving out of a tight spot and come to you 50 ft away, that is an ideal use of the feature. Staying close to the car is a good idea because you are less likely to lose line of sight and more likely to have adequate situational awareness the closer you are to the car. Obviously, I would not recommend using Smart Summon for navigating a busy parking lot 200 ft away. You will almost certainly lose line of sight in that scenario which could potently be extremely dangerous. Smart Summon currently, will not stop for stop signs. Keep that in mind. I agree that the third person view is a problematic because it is difficult to always see the environment from the right angle when you are not inside the car.

Looking towards the future, I do expect Smart Summon to improve. That's a no brainer. Certainly, Tesla should be able to improve the reliability to some degree. And I would expect that Tesla should be able to add additional features like stop sign recognition at some point in the future, if they decide to. But I am not going to speculate on whether it will ever become autonomous. That is above my pay grade.

Anyway, I hope that is a fair review.

Happy Friday!
 

Mojito

Member
Apr 16, 2019
124
47
Orlando
This is a Beta version the owner will be responsible, unfortunate if there continue to increase accident in parking lots in the next few months there maybe laws pass to band summons in private parking lots. People be smart do not ruin it for the system that everyone was pushing Tesla to put out before it was ready. Dont ruin it for the future. Be smart I know we like to show it off cause it is cool but there will be many people that are handicap that truly need this.


REPLY
 

timewasted

Member
Jun 29, 2017
196
566
Las Vegas, NV
The battle cry of the Tesla Fanatic is “it’s still early, give it time!” Nothing is ever really a problem, because it’ll be fixed at some indeterminate time in the future. Always gotta punt the can further down the road because that’s easier than accepting the fact that we got some half baked garbage delivered to us. It’s been a year since the last big banner feature of Navigate on Autopilot was released, and it’s still garbage in exactly the same ways that it was when it was released.

People with domain knowledge come in to have a discussion about things and the people who took classes at the University of YouTube Videos circle the wagons shouting “more data, more time!” until they drive off the person that can damage their safe little view of the world. Yeah, @SandiaGrunt is kind of a dick, but wouldn’t you be if every time you tried to tell someone that 2+2=4 they shouted back about how Elon told them it’s not? It’s disgusting.
 

electronblue

Active Member
Oct 1, 2018
2,325
2,507
Earth
If he works in the AV industry - he is being unethical in attacking Tesla here without disclosing his potential ties to a Tesla competitor.

I worked for MS for over a decade and I didn't make a single comment about any of MS competitors without disclosure.

For what it is worth, given that he has said to have worked on AVs for nine years, it is obvious he works at a company other than Tesla. Whether or not they view Tesla as a competitor is a good question, but at least that was disclosed when he started posting here.
Anyway, his comments should stand on their own - without having to attack other members - esp. for being pro-Tesla in a Tesla centric forum.

True. I’m not sure I’d agree with the pro-Tesla part making any difference, but in general true.
 

electronblue

Active Member
Oct 1, 2018
2,325
2,507
Earth
You're wrong. I never said a goddamn word about my credentials until that asshole @electronblue outed me on like page 7. I actually didn't even want to mention my credentials.

(I'm kidding @electronblue, you have my greatest respect -- you're the only rational voice I've seen on this site in months.)

Here’s the thing: We only learn each others temperament over time. I get it now you are not completely serious in your... shall we say flamboyance. Good to know. :)

Personally I’m cool with that as I’m not a form over fact type of guy but you would be better received here and listened to by toning down the dramatics a little. We do want to hear your insight. So do continue, thanks.
 

electronblue

Active Member
Oct 1, 2018
2,325
2,507
Earth
In order to get this back on track I was wondering how long we'll have to go before we determine 'today is the day that SS hasn't hurt or killed anyone for long enough it is roughly equivalently as safe as a human'. Using Sandia's base of 268.8M cars (which is a high estimate of both cars registered and also of cars that are actually used...I have a car in my garage that is registered but got used twice this year and 0 times last year for instance).

I haven't found EXACT numbers but plenty of references to 60,000 injuries (911 type injuries) and at least 500 deaths per year in the US as of 2016, with a sharp trend upwards due to distracted driving so likely much higher now. I'm always wary of round numbers, if anyone has more reliable...feel free. So that means 268.8/60,000 gives us 1 injury per 4,480 cars registered per year and one death per 537,600 cars registered.

What does that mean to SS? A true apples to apples is impossible, we don't have all the right data. But lets do some estimating.

I looked around a bit but couldn't find any official estimates on how many parking lot trips we take in a year. I'd guess rural populations use parking lots much less than once a day and sub-urban populations use parking lots a bit more than once a day. I've lived in both, that's my experience. If the average registered car uses a parking lot once per day that is 4,480 cars/injury/year x 365 parking lot retrievals/year or... 1 injury for every 1.6m times someone goes back to a car in the parking lot and leaves (i.e. a summon opportunity). Lets assume twice that to be generous and reduce counter argument...so 1 injury for every 3.2m times someone goes back to a car in a parking lot by my figuring.

From the other direction. Lets assume each summon attempt requires 5 registered attempts. That's high, but I'm feeling generous. This is good will towards @electronblue for staying reasonable. High fives and hugs and stuff.

So, we would need to clock 5x 3.2M = 16.3M summon attempts with no significant injuries to be pretty convinced that Summon is less dangerous (safety, not property damage) than a regular human. At 100k summon attempts per day, (likely more now that V10 is fully rolling out but may taper in time as it gets old) it will take approximately 163days (23 weeks, 5.5 months)...so somewhere between feb and april 2020 (assuming we already have several hundred k attempts before the FSD roll out). I did NOT, this is important, look to see if it is a leap year so that's +0/-1 days range.

that's a while, I was hoping it would be like next week so we wouldn't have to wait lol. But I also think that's a very conservative estimate (assuming my math is right...I didn't use a spread sheet, so, well, you've been warned). I'd bet that tesla (and more importantly NHTSA) will be able to declare this sooner, maybe as little as 3 months. Things will get twisted because updates will make some things better and some worse, so we won't really no for sure of course.

First, please chime in with my mistakes or better estimates. Second, votes on if we make it 23 weeks (or whatever the right number is)? I don't care about property damage for this analysis. That was covered earlier, and yep SS is currently not up to par with a human there, though rates are still low they need improvement to be as good as good 'ol mama meat sack.

@SandiaGrunt you don't get a vote.

Interesting math, I certainly have nothing to better offer myself, so unless someone else does let’s look at those numbers of weeks over time.

The biggest open question that remains in my mind about the safety of Level 2 driver’s aids is: how big a part does luck play in however they turn out and how we find out (or not) about them. Luck in both the events that occur as well as luck in how reliably the role of the automation is assessed or publicized or counted when an event occurs (or is not counted). I’m not yet quite convinced we have enough of these cars, their users and enough of the publicity or ability to gather that data reliabily enough to know...

For example, it was widely thought that Joshua Brown was the first Autopilot related fatality (cross traffic). It was only later that we found out about the earlier death in China (stationary object). There have been other suspicious deaths where the role of Autopilot can not have been ruled out... and how many cases we have simply not heard of at all? Who knows how many statistics simply get missed by use since we don’t hear of them or get reliable data.

It certainly is an interesting thought experiment somene presented: What if every GM car suddenly had these features? Their use and probably our abilty to get data on them would probably skyrocket simply due to numbers. Now Tesla is still small enough to fly under the radar a little and it is difficult to get good readings when something is flying under the radar.

Of course Smart Summon may be a little easier to assess because nobody is in the car. With other Autopilot related accidents things get muddy as the role of the driver and the car intertwine more.

The luck question of course plays both ways. One unfortunately unlucky event (for Tesla Joshua Brown probably was somewhat such an event) and the results may be more dire than the statistics would dictate.
 

electronblue

Active Member
Oct 1, 2018
2,325
2,507
Earth
This is my final word on the subject now based on my personal experience with Smart Summon and everything I know about it:

Smart Summon is basic right now. It can handle the ordinary stuff. It will stop for cars. It will stop for pedestrians in the way. It will pause at crossings. It does pause frequently with a message that it is waiting for the path to clear, even when the path seems clear. So it is overly cautious. If you use it for something simple like moving out of a tight spot and come to you 50 ft away, that is an ideal use of the feature. Staying close to the car is a good idea because you are less likely to lose line of sight and more likely to have adequate situational awareness the closer you are to the car. Obviously, I would not recommend using Smart Summon for navigating a busy parking lot 200 ft away. You will almost certainly lose line of sight in that scenario which could potently be extremely dangerous. Smart Summon currently, will not stop for stop signs. Keep that in mind. I agree that the third person view is a problematic because it is difficult to always see the environment from the right angle when you are not inside the car.

Looking towards the future, I do expect Smart Summon to improve. That's a no brainer. Certainly, Tesla should be able to improve the reliability to some degree. And I would expect that Tesla should be able to add additional features like stop sign recognition at some point in the future, if they decide to. But I am not going to speculate on whether it will ever become autonomous. That is above my pay grade.

Anyway, I hope that is a fair review.

Happy Friday!

That is fair. You have good Friday too!
 
  • Like
Reactions: diplomat33

Products we're discussing on TMC...

About Us

Formed in 2006, Tesla Motors Club (TMC) was the first independent online Tesla community. Today it remains the largest and most dynamic community of Tesla enthusiasts. Learn more.

Do you value your experience at TMC? Consider becoming a Supporting Member of Tesla Motors Club. As a thank you for your contribution, you'll get nearly no ads in the Community and Groups sections. Additional perks are available depending on the level of contribution. Please visit the Account Upgrades page for more details.


SUPPORT TMC
Top