Tesla FSD BETA obstacle testing #2

[@bergonius]

This is a stress test and a half. Terrific work, great editing. You the only one who can make a car driving video entertaining for everyone.

[@freezy5367]

The perfect amount of love for Carol

[@tcfs]

But not too much! He is watching us...🤣

[@donaldowensiii7478]

Read this just as he said it😂

[@wizzyno1566]

The perfect amount for you or for him?

[@chickenhawk212]

Glad it chose the box over Carol in your backroad trolley dilemma. That’s actually pretty sophisticated

[@devindykstra]

I am continuously amazed by your ability to keep pushing your videos. This one had me on the edge of my seat the whole time, absolutely terrified by what you were gonna attempt next. An absolute emotional rollercoaster, but also quite educational.

[@graceshotdown]

Thank you to carol for being the behind the scenes glue that brought this together! 😊

[@nowteslafy]

Another stellar video and thank you Carol for helping!! Looks like some great improvements. I love that you attempted to approach this with some scientific rigor 😊

[@barrydaugherty5528]

Wow you are brave!! And Carol is very patient! What a good sport. Thanks for sharing all the data from your experiments. It’s nice to see the progress. Now if only I would get v11! My wife got it last Sunday!

[@medienkonsument4]

What a throwback to times when FSD folded the mirrors at each corner! It didn't even know it still has this move in its options. Thanks for the video.

[@Anything_Anthony]

Not too much love, but also love the channel

[@aaronb7990]

The Inst 360 cams are freaking awesome! I really like how you incorporate the 3rd person view in your videos and your editing in general.

[@barefootID]

Just the right amount of love! 🙂

[@CharlesBurnsPrime]

Thank you, Carol! You aren't just helping your awesome husband, but also 80,000 or so other Tesla drivers around the world.

[@theodorehaskins3756]

So if anyone is unimpressed by this version of FSD beta, I would like to know what they’re smoking or drinking, because this is terrific stuff here, and before I go further, my hats off to you and your wife for doing these videos, because it really shows the progress that Tesla continues to make, and this demonstrates FSD beta’s capability. So what will it be capable of in the very near future, because I don’t think most people truly understand any of this stuff and as you said, I don’t understand it myself, but I do know this, no other car manufacturer that has this level of capability without all of that external apparatus that you see some of these other so called self driving car companies like, CRUISE, LYFT, WAYMO, Uber and others who have equipped their leased vehicles, none of which they manufactures with, LiDAR, extra cameras, Geo fenced routes, and all that kind of stuff, which adds an additional amount of expense to the vehicle’s cost, so when Tesla make this work, just think of the advantage that they will have, because no one wants to buy a car that looks like some kind of weird looking Frankenstein monster. So it just doesn’t look like something that I would wanna buy, and so let them do their thing and Tesla will certainly continue to pursue their version of FSD beta to perfection, because they understand just how valuable this technology truly is, and I think the other guys know it too. So when I think about the fact that the AI is thinking and figuring things out, that’s what AI is about, and so I think too many people don’t truly appreciate just how incredible this stuff is. So I don’t have the update yet, and can’t wait to get it, but I can say this though, it would not surprise me that by the end of this year, FSD beta will be fully capable of driving itself. Cheers 🥂

[@larsnystrom6698]

This is both funny and enlightening! And your wife is brave... and put a lot of trust in you.

[@gonzaloenrique8741]

What trust is needed for this? There's no conflict of interest going on here.

[@scpdatabase969]

It’s not even my car and my butthole clenches when your car gets close. Love these torture tests. Seeing the new visualizations for gates auto opening and closing made me realize the car really can detect objects in the way and the road, but still has difficulty making a decision on how to behave

[@mobiuscoreindustries]

This is the reason why Tesla is pushing for the decision and planning system to also use neural networks in their core instead of only partially. Because as much as we think the road is logical, it is anything but that. You can't simulate a driver with if-else statements. You can make a lane following algorithm with it, you can make an auto park with it, and you can make emergency breaking assists with it. But you cannot make a unconstrained level 3-5 system with simple statements. With a neural network you can train it to account for nuance and probable outcomes (as opposed to definitive ones) but if item detection already needed and extensive dataset, actually making a planning AI for the entirety of the road's situations will require a level of data never before seen.

[@profcpbl]

Nice work. You two are hilarious!! I can't wait for the next iteration of this... and for FSDbeta to catch up with hopes and expectations.

[@tedbaxter5234]

I’m using 11 too. Yesterday, I noticed it recognised speed bumps with painted chevrons on the 25 mph street. It didn’t slow quite as much as I would have but it dropped 10-12 mph - reasonably comfortable. I don’t know if it recognises unmarked speed bumps… Thank you for the video and the time and effort you give!

[@robertwhite3503]

KZbin shows me, it seems to think 17 mph is right for speed humps that it recognises.

[@BlopFlop123]

Can't wait for 3rd test when FSD 12 comes out

[@DiscoveryOwners]

You are the real star of this show! Extremely useful information that I am very grateful for. I am 75 and feel like eventually my wife and I will need all the driving help we can get, hence the Tesla MY purchase. I haven’t subscribed to FSD Beta yet, waiting for the V11 stack plus $200/month is a lot for us that needs to be justified by increased safety. My big question now is how well FSD with HW3 will perform at speed, say 35-70 mph.

[@haz226]

You probably shouldn’t. It is kind of more dangerous than driving yourself IF you aren’t paying attention. In your case, this can give you a false sens of confidence that is more dangerous. You should pay as much attention with FSB than without it. It’s far from perfect, and multiple models got recalled in the last couple of month. It is not safer, for the moment it is more of a gadget. Please drive safe! ❤

[@formytots0128]

Awesome video. Thanks AIDRVR and Carol for all of your hard work. We truly appreciate the great content. We all learn from your tests and get entertained at the same time. :)

[@Mrbfgray]

OMG he owns an ICE truck! 😲 How impure. 😄 Nice tests that almost no one else is doing for YT. I'm upset you'd risk a Model S like this, I mean wives are replaceable but...

[@TheLastMoccasin]

Amazing work as always!! Can't wait to watch V11 evolve!! 🤯

[@GuardianTiger]

Folding the mirrors squeeze through that truck was very smart and pretty cute.

[@marshall5]

Thank you for doing these tests, I do not know of anyone else who does it and I think it is a fantastic way to get some insight into what's going on "under the hood". But like you, I have no idea what FSD is thinking in some of these situations. Looking forward to the next ones!

[@saabnico]

I've been watching al your videos since my last comment like 10 videos ago, and im glad that you have managed to deal with your stress. al videos are top notch and are so enjoying to watch!! 😊

[@benjaminsandeen9241]

Not having any idea why it's making decisions it does is an inherent feature to how AI/neural networks currently work. Not even the people who develop them really have a good idea of why it makes certain decisions. This is also why self-driving cars are still terrifying, even though on average they're extremely safe

[@w0ttheh3ll]

People generally don't have a good idea of why they made certain decisions. Does that mean that humans driving are terrifying for this reason?

[@rogergeyer9851]

Benjamin: Yes. It's like we're literally growing "alien intelligence" when advanced AI work results in things like Alpha-Go or Alpha-Star, etc. This is a significant issue even for games of skill. Imagining it for something as complex and serious as a general driving solution is just mind boggling.

[@rogergeyer9851]

@@w0ttheh3ll: But we do have a good general concept for the way human minds work, and we're adding to that all the time. We might have no real idea AT ALL of the way AI works, given lots of parameters. It's something VERY different, IMO.

[@benjaminsandeen9241]

@@w0ttheh3ll I don't think so, because I think the premise flawed. People might not understand why they make certain decisions in life, but the basic goals of driving are usually pretty straightforward: Arrive at your destination, and do so safely and on time. The rules of the road are vastly simpler than the rules of life. Additionally, I think you're conflating the meta-knowledge I'm referring to. Generally, we can assume that any drivers that see a pedestrian on a road will come to the determination that they are in fact a pedestrian. Since we don't understand how an AI system "thinks", we have no way to intuit how likely that system is to identify that same pedestrian correctly. Humans typically have a huge portion of their cognitive abilities dedicated to understanding and predicting the behaviors of their fellow humans. We have built that up over tens or hundreds of thousands of years. We have no corresponding facilities for understand an AI's behavior, and what's more, we will be very unlikely to develop such abilities as long as AI remains an inscrutable black box. All that being said, please note that I deliberately talked about whether human drivers who *see* a pedestrian come to determine that it's a human; an AI system is much more likely to perceive a pedestrian than a human driver, which is certainly a massive point in its favor. On the other hand, if the AI system makes an incorrect decision about how to react to the pedestrian, it may well decide to speed up, for some unknowable reason due to its AI. It's incredibly unlikely that a human driver would react in the same way; the risks with a human driver almost exclusively lie with the lack of perception. So, when I say it's terrifying, I suppose I'm saying that I'm scared about the changing risk structure of existing around multi-ton missiles that are controlled by robots, the decisions of which I have a reduced risk of predicting as I attempt to navigate myself around them.

[@thedishbench]

Great video! Thank you to Carol for making the challenges possible.

[@briansaddress]

Your videos are light years ahead of everyone else. I’m watching the future unfold before my very eyes. I will now be waiting for the next one…

[@scanzas]

Your videos never fail to amaze me, and this one was no exception! The production value and attention to detail were truly incredible, and it's clear that your content is always improving. I'm eagerly looking forward to the upcoming updates and can't wait to see what you have in store for us. Keep up the great work!

[@AIDRIVR]

Thank you so much for the kind words Bernardo!

[@JeffMcNeal]

I actually thought that it was very impressive that beta caught the wheel in the road, particularly when it was in the shade through so much of the testing. It would have been easier to detect had it been in the sun, making a higher contrast image.

[@gfbrnby]

Scary stuff! Thanks to you and to Carol for another outstanding test of FSD Beta.

[@DorkmasterFlek]

Great work! Very interesting test results! On the truck turning towards your lane test, actually I'm kind of okay with that change. I actually think in my experience the current v10 build freaks out and slows down *too* easily IMO. Like if someone is coming up to turn out of a parking lot or driveway, the car will slow down quite a bit when it really shouldn't. The default assumption appears to be (and should be) that even if it's moving towards your lane in that moment, they're not just going to go straight into your lane like a crazy person. Not saying it doesn't happen, but I think waiting until it actually enters your path is appropriate and more natural IMO.

[@rogergeyer9851]

Andrew: Yes. It's difficult. If it is too cautious, then you get a bad passenger experience. I suspect the "lots of data" and TIME is going to really matter for things like this. (Or I hope so, because having to hard code solutions is somewhere between bad and risky). Maybe the full blown occupancy network, coupled with training and Dojo can really make a difference for issues like this.

[@JohnVanderbeck]

Unfortuntely in my experience v11 is just like v10 in this respect. I don't know why it acted like this in the video but in actual driving for me it still freaks out and damn near slams the b rakes whenever a car even thinks about entering my lane.

[@Anthony1337]

These are the best videos you make! Thank you!! Any chance you could try a baby stroller next time?

[@Singularitarian]

Still the best FSD Beta youtuber! Love the dramatic opening.

[@russell2449]

IMO, if you're going to have someone roll/push an obstacle onto the road, they need to be HIDING behind something because with FSD detecting the person, it's going to default to the best way to avoid hitting the human.

[@PenkoAngelov]

It's amazing to see so much progress over so little time.

[@stonesonfire]

GIVE A LIKE AND LEAVE A COMMENT ( To help boost algorithm)

[@kyleb8117]

You and Carol are amazing for doing these tests. It's so cool to see how FSD reacts and I was pleasantly surprised how it did. I almost wonder if the other vehicle needs to be perpendicular (or nearly) to be considered. You'd expect how much they're in your lane to matter. Hopefully with the next major version.

[@bluesmoke8714]

Thank you Carol.

[@JohnVanderbeck]

I find the turn in test fascinating because in my experience with the beta, anytime another car even hints at entering the road, or hell is even just sitting in the media pointing at the road, FSD Beta for me loves to slam on the brakes.

[@FermiBubbleGummyBear]

These are great, very thorough test cases that are incredibly insightful. Thank you.

[@jamaicantesladriver]

HOLY CRAP that first test with the truck made me yell out loud.I was terrified how close your car was. I do not have the cohones for that, more power to you!

[@jhagen4850]

Awesome work! I can't believe how advanced this is! Absolutely incredible. Great testing!

[@mattesrocket]

Carol knows what we want. 👍👍👍 Carol! And you, for your courage! I don't understand people who say "FSD will be sooo safe". I have avoided a couple of times really tiny* objects on the road, that were maybe only 2 inches high but very dangerous like metal rods or sheet metal parts or nails on parking lots. Avoiding quite small but extremly deep pot holes also prevented a lot of damage to my undercarriage. And if not dangerous I go around even very small animals like toads, squirrels or small cats. Who am I that I could decide that small animals are worth nothing? I safed my life also once by avoiding a very, very unexpected object falling from very high in front of me, I could because I saw it starting to roll. That all needs understanding the whole situations and what these objects really are and what they will do seconds later. FSD will never calculate how a small animal typically will behave or if construction objects can start rolling or not, FSD will not be able to distiguishe between the behaviour of a tractor or digger etc. etc.

[@mattesrocket]

an example: the second biker predicts something, FSD would not be able to predict it because it has no clue what a straw bale is. This situation at high speed and a much bigger trailer happened to me on a country road and the bale came from very high where FSD anyway ignores all objects (except traffic lights). kzbin.info/www/bejne/m3OWo6Wsm9uZaNk

[@johannel8104]

Don't compliment her too much bcz I'm watching you 😂😂😂 I think the Beta reacts to a vehicle crossing over the lines into your lane IF said vehicle also is on a collision course with the Beta vehicle. If it's just in your lane with room for you to pass and it's path is not heading to cross yours Beta will continue as normal. By the way. Great vid and thanks for all the effort. Your vids are always fun!

[@thespacepeacock]

I love these obstacle tests, and damn do i respect you for putting *both* your cars at risk for them. Thanks AIDriver, and (respectfully) thanks Carol!😉

[@will0499]

Your wife is incredible for literally putting her self in harms way for this test! ❤

[@dada373678229]

I think it's all about the speed, 25 mph is fine, 28 or 30 mph is more likely to not react in time.

[@ericelfner]

Insta360 is sponsoring you because your use of their camera is off the charts great. I would not be surprised to your video clips in their sales material.

[@2nd3rd1st]

The daylight running lights on the S are so fucking sexy. Like literally.

[@johnwurts3953]

Love your channel. I hope you continue to challenge the beta.

[@CYBRLFT]

Love these so much. If you can think of some crazy obstacles for a double trouble video I’ll drive out to ya 😎

[@piyushkothari3]

Thanks Carol for showing up again, you are the love of my life 😛

[@robertwhite3503]

😲

[@Muhahahahaz]

9:45 I think there must be some sort of inflection point involved… Like it allows the other driver to mess about within their own lane, under the assumption that the other driver will typically choose to self-correct and/or not leave their lane. But as soon as they cross over the Tesla is ready to react

[@DLWELD]

FSD needs to binge watch dash cam vids -

[@minicirco]

9:18 Guzzler didn’t want to enter a road, as it‘s correctly interpreted by the path planner, so no intention to stop or going slower. The guzzler is „just“ standing in the road. So, if this would be an official entrance of road, then it would slow down, I guess.

[@crusadeagainsttomatoes2518]

Thanks, Carol! I’m glad you could help to make this video possible!

[@TomHarrisP]

I’d love to see an additional metric in future videos: a G-Force meter. It would complement the wheel stripe.

[@130rapid]

6:03 The woman with wheel = a motorbike!

[@reggiebald2830]

Excellent stress test and good editing. Kudos to the wife. I appreciate you weren't kicking the can down the road!

[@greenfrog58]

Great test Chris!👍👍🇸🇪

[@raywflam]

Wow! I will be too scared to do what you did. Thank you, thank you! BTW, I really like your steering wheel cover that covers the steering wheel pretty completely making the steering wheel a bit fatter. Please tell me where you get it from. Since my fingers are suffering from arthritis, I can’t really wrap my fingers around the steering wheel on my Model Y.

[@Hamachingo]

Whenever I see a self-driving car prototype, I try to make the engineers inside happy by giving them solid data by scaring it.

[@brettlynch78]

Giving not too much love to Carol for making this video possible.

[@Mochu_s_Junkyard]

that starlink box has seen things

[@aditroy57]

"Better not be too much love, I'm watching you" 😂😂😂

[@Gelp]

Superb video AGAIN! Loving your pushing! 😊

[@ChristianBlueChimp]

You did good, but Carol did better! ❤

[@lylestavast7652]

I have friends who have had unusual objects that flew into their path in the past - one died from something that bounced off the truck ahead of them, penetrated the windsheild and crushed their chest as driver following. You could maybe rig some of this sort of thing since you have a pickup - have someone sit low in the back and throw out soft but solid objects - like pool noodles, nerfballs of various sizes and shapes - i think they also make fake concrete blocks and the like - they'll bounce right off a vehicle and not mark. You could dump a box of them at once just for grins... Good video and I think they're making serious progress. Maybe they can take those 3d files and ingest for direct AI training / labeling use ?

[@lylestavast7652]

might also take some different colored boxes or some spray paint to do patterns on them or totally change the full box color - like away from grey which matches the asphalt there...

[@adriancires4889]

Awesome as always brother. Thanks for the content! P.S. Much love (but not too much) to the channel

[@1969kodiakbear]

Obstacle testing #2. By the way, I have difficulty communicating because I had a stroke in Broca’s area, the part of the brain that controls speech. 2/8/2021 but I lived again. (My wife helped me compose this.)

[@AIDRIVR]

Wow, glad you are still here and okay! And thank you for watching my videos, truly means a lot

[@BeaverInSpace]

It seems the orientation of the vehicle is taken into account while deciding to mark it “blue.”

[@mandoleg]

Giving some love for Carol, but not too much. :D

[@curlytoezz]

New sub! Love your videos man, especially ones like these!!

[@lovefrogsalot]

Wouldn't the truck turning mimic someone waiting their turn to cross the road? Maybe it would be unsafe if Beta slowed down every time someone creeped toward its lane?

[@JeromeDemers]

Your wife honking was funny. Next time ask wife to pull a string to bring obstacles on the road.

[@anshulg14]

Great stress test, amazing how much decisions it's making, even harder for humans. I have to say, it decided the way I would do...and path planner is thinking like humans, left or right better and then final decision. We just don't see our mind decision path

[@Zedus-rl9hp]

I assume that just objects that potentially cross your own lane are displayed in blue. So if a vehicle protrudes slightly over the lane line and there is enough remaining width, but the lanes don't cross, then it stays white and FSD uses the remaining lane width to create enough clearance.

[@rogergeyer9851]

Zedus: But when it's OVER the line, ignoring it is just wrong. There needs to be awareness and balance, re managing risk from bad drivers. It's something humans (who pay attention) do well.

[@Zedus-rl9hp]

@@rogergeyer9851 Just because it's not blue doesn't mean it doesn't get attention. I referred only to when it is displayed in blue. The coloring is primarily for the driver, who then also sees that this vehicle is crossing his line. If a vehicle is slightly above the center line because it might want to turn but does not want to block the entire lane for following traffic, then a human driver will also drive past it, assuming that the vehicle will stop there. At least that's my assumption based on the videos I've seen so far.

[@jonasweber9408]

Hello I’m sending the adequate amount of love to Carol ❤ and thank you for this amazing channel ❤

[@JacobAuthier]

Can you make a video going over your setup? For example what mounts you use and so forth? :)

[@victorfrombcn]

Looking forward for your vision only park assist testing.

[@nielsnb11]

Could a put a version stamp on the video, since I'm sure it will be grabbed by someone your wants to show how bad the FSD is?

[@MyTeslaTrek]

Strong Work, Carol!!

[@tommyvuitton3113]

Nice work Carol!

[@teslarex]

Great job! Important improvements! Invisible stick works great-I always wondered how you did it!

[@nelsonmachupa]

Peeped the Valentine One. Good taste 👌🏻

[@health101DOTorg]

Always wondered why the "behind the car" shots bounced. Now I know.

[@Thierry.g38]

Thanks for this video. These tests are always very interesting Just a comment about the test with the wheel on a fix place 😊´ the road. May be the autopilot choose during one test to avoid tu on the right (instead on the left) is due to the fact you were not exactly align on the same position when you started the test (a little bit more on the right compare to the other tests) It’s just a possible explanation 😉

[@rogergeyer9851]

Grenier: One of the real challenges with this is how many factors the models end up using as the AI evolves, and it can get far beyond human intuition re things it's actually considering. I saw James Douma, who has some AI expertise re FSD concepts discussing that just yesterday on a long FSD video with Tesla Motor Club.

[@gridcoregilry666]

woah, insane tests, great job, please stay safe!!!

When the speed of the turning truck is too low to actually intersect with your path and position in time the network might not actually see it further downstream. It might indicate that the margin of error is not big enough for the estimation framework

[@glowcube]

Fun to watch! Thanks :)

[@LunnarisLP]

I think like with the box test with the wheel your wife standing at the side could be a reason on which side it would pass the obstacle, but would need more testing to see. I def. noticed her being shown and shown walking sometimes possibly towards the road in some of these tests. Then on the first truck tests, I'd say this looks like a common situation where someone else is making a left turn through your lane, which you would usually not want to slow down for as you have the right of way. It's a very difficult problem to see if the guy is actually slowing down and waiting as you have the right of way or if he isn't paying attention. I can't tell you how many motorcycle incidents happened just like this with some car driver not paying attention and pulling into the opposite lane exactly like this, but again, obviously in most cases the driver just prepares his move and give you the right of way, so strong breaking probably would often give you phantom breaking in real life scenarios.

[@andyspry3030]

See if you can find some water - of various depths, for it to drive through...

[@Omie334]

I wonder if the reason why it didn't react to the wheel being rolled in front of the car (the second set of tests where the wheel was rolled a bit later) was mainly due to the blind spot in front of the car. Beta has a blind spot right in front of the car. Seems like the car just forgets about the wheel when it gets in front of it in that camera blind spot. I wonder how the HD radar version would do with this. I bet it would avoid it like how its supposed to... I have been told that this doesn't matter but the visualizations do show the other car a bit off. On the truck test, the truck was in your lane and the car didn't see it as in the lane... It should react to it if its heading straight for you but it doesn't which is weird but then as part of the body crosses into your lane, the car doesn't even recognize it. I think Tesla is doing something a bit cheeky here. The car (from what I have noted with my testing results so far) is a bit too scared when cars are coming towards it from the other side (cross traffic on a 2 lane road). since its a 2D picture for the car, the car has a bit of an issue figuring out if the car coming up is in your lane or not. My previous update would always slow down and swerve away from cross traffic on turns because of this. Tesla is either intentionally or unintentionally moving the cross traffic recognition box a bit further away from the car to avoid a crazy reaction when the cross traffic gets closer. Either they moved the boxes or the car is having difficulties figuring out where cross traffic is. This is where radar matters...

[@MultiVfc]

Thanks Carol!

[@dmacm]

I would guess the issue with the truck not turning blue as the car goes into your lane is due to the fact that as humans we have a huge margin of error in this situation and the car doesn't. If the car within a few feet of our lane we become paranoid and think its in our lane.

[@kottuning3294]

It looks like once something crosses over the line into your space it enters the path planner. But its odd because when on a normal road it'll highlight cars in the opposite lane in blue. Ego has a mind of its own lol

[@vling4312]

Awesome video. Thanks!

[@theycallmeabe]

Nice job! My wife also will humor me in my shenanigans. Good job to your wife.

[@AE-dn8es]

Carol gurl u r 2 gud for him baybee call me wink wink Lmao love your videos and can't wait to watch this torture test, the last one was awesome.