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

Firmware 8.0

This site may earn commission on affiliate links.
Taking this one step further, a Tesla will also be able to bounce the radar signal under a vehicle in front - using the radar pulse signature and photon time of flight to distinguish the signal - and still brake even when trailing a car that is opaque to both vision and radar. The car in front might hit the UFO in dense fog, but the Tesla will not.
Wow
 
Very interesting bits of information about V8.0 update from a Dutch journalist who attended the webex conference with Elon:

  • There will be a "nag" in V8.0 to remind us to hold the steering wheel based on following criteria:
  1. Under 8mph: No nag!
  2. Up to 45mph: "Hold steering wheel" will apear every 5 minutes when driving straight.
  3. Above 45mph: every 3 minutes depending on how the car drives.
  • The nag was added because according to Elon; expert users are the biggest problem, because they tend to ignore warnings.
  • Tesla will send 3 more beta versions before a general rollout in 1-2 weeks.
Here is his original link: Tesla AutoPilot gets dramatic safety update. Keep your hands on the wheel experts of you will restart (with images, tweets) · Vincenteverts
 
Yeah, that should be interesting. In my experience, "1" isn't close enough to deter everyone, but is close enough to make it dangerous when they come in, so I have been setting it to "3" and just letting things happen even though it encourages and rewards idiots. It'll be interesting to see what the new firmware does.

That's one thing I noticed about myself ever since I started using AP - I don't care that much anymore about the idiots. Or maybe I am just much more relaxed. Either way, I think it's a good thing :)
 
That's one thing I noticed about myself ever since I started using AP - I don't care that much anymore about the idiots. Or maybe I am just much more relaxed. Either way, I think it's a good thing :)

Yeah, I'm certainly much more relaxed in stop and go on AP, even though the heavy traffic is the one place I'm really gripping the wheel, too (usually keep just a couple fingers on it to stay in touch in lighter traffic.)

It's just the principle of the thing - if folks can gain by taking advantage consistently, some folks keep doing it - and causing accidents.
 
Hopelijk net op tijd voor mijn verjaardag...

Wat ik overigens interessant vond waren deze punten:
  • manual mode, alerts driver if about to leave the road and no torque on steering wheel has been detected since Autosteer was deactivated
  • With further data gathering, car will activate Autosteer to avoid collision when probability ~100%
Als ie voelt dat je niet meer bij de les bent, waarschuwt ie je, en als ie bijna 100% zeker denkt dat een ongeluk gaat gebeuren schakelt ie zelf autopilot in om in te grijpen (in een later stadium weliswaar, misschien 8.3 of zo).
Het mooiste zou zijn als ie autonoom de vluchtstrook op zou zoeken en het alarmnummer zou bellen als ie geen input meer krijgt van de bestuurder.
 
Wow, indeed, lijkt inderdaad écht veelbelovend! Ook al omdat iets anders me opviel: het gebruik van de toekomstige tijd in wat Tesla zegt over fleet learning:

"This is where fleet learning comes in handy. Initially, the vehicle fleet will take no action except to note the position of road signs, bridges and other stationary objects, mapping the world according to radar. The car computer will then silently compare when it would have braked to the driver action and upload that to the Tesla database. If several cars drive safely past a given radar object, whether Autopilot is turned on or off, then that object is added to the geocoded whitelist."

Dat AP beter is geworden de laatste maanden is zeker, maar van dat fleet learning (of gewoon van 147-maal-zelf-over-dezelfde-plek-rijden-learning) heb ik toch nog niet veel gevoeld. Wie wil moet eens de test doen op Brusselse Ring Oost komende van Waterloo: precies op de plek waar je Vlaanderen binnenrijdt (toeval natuurlijk, maar het is wel precies die plek, waar het bord "Hoeilaart" verschijnt) begint de auto de zigzaggen (geldt op beide rijstroken, richting noord). Als ik alleen ben en er is geen tientonner naast me laat ik de auto doen, en na 500 meter vindt ie na lang zwabberen zijn weg dan wel terug, maar als ik mijn echtgenote als passagier heb moet ik toch meteen ingrijpen en heb ik telkens recht op ongeveer 15 seconden slechte poëzie over AP. En er zijn nog wel andere plekken in mijn buurt waar ik zelf telkens al 147 keer heb moeten ingrijpen en AP vandaag nog niet begrijpt waarom.

En dus vind ik het gebruik van de toekomstige tijd veelbelovend. Die fleet learning (vermoedelijk al bijna twee jaar geleden aangevat) zal zich nu dus eindelijk in resultaten omzetten :)
 
Last edited:
No 8.0 for Classic Teslas? :(
From the USA Today article: Tesla CEO Elon Musk announces Autopilot upgrades
"The new system will be beamed wireless to Tesla cars made in the last couple of years in the next two weeks, Musk said."
Guess it's time to trade up...

Gosh, I hope not. I guess they will eventually have to stop updating older cars (as Apple does with older phones), but I hope the product cycle window would be longer on a car. Will be watching this closely as I am looking at buying a "classic". If updates are done and there is no more possibility of having the little bugs with NAV and media player, etc fixed then I will have to think long and hard about buying.

Keep in mind that the new AP updates only apply to cars manufactured after October 2014.
All previous cars will still benefit from the Nav and other important updates going forward. :cool:

Tesla makes Autopilot safer, smarter with major update - Roadshow

Tesla's Version 8 software update is coming, and along with it some major enhancements to the company's Autopilot functionality. Autopilot, available on all Model S and Model X cars manufactured since October 2014, is a term that aggregates a suite of features cumulatively enabling the cars to self-steer and adjust their speed on many driving circumstances, relying on a combination of imaging, sonar and radar sensors. Now, with Version 8, the cars will make even greater reliance on the radar sensor built into the car's bumper.

Tesla claims that its Autopilot system will be able to resolve enough detail from the car's radar alone to not rely on any other sensor. The car will, however, rely on the data gathered by other Teslas on the road to help identify things like street signs that, with their reflective nature, can actually be difficult for radar systems to accurately resolve.
 
Elon's quote about expert AP users ignoring the 'nags' is why there should never have been any nagging to begin with. If it only alerted when it needed driver intervention (AP 7.0) the system would be and was, much safer. Tesla is now on the road to requiring hands be detectable at all times on the wheel. Green1 was smart to stay on 7.0.
 
  • Disagree
Reactions: 3s-a-charm
Waiting a couple of quarters means you'll miss out on driving a Tesla for a couple of quarters. Don't let the perfect be the enemy of the good-- especially as it will never be perfect, there will always be a hardware update.

You make a good point, if the advances for AP 2.0 are only going to be merely incremental. After I’ve made a major purchase decision, I generally stick with it and ignore any new announcements. However, AP 2.0 is not just incremental; it would be a major shift in the capabilities of Tesla’s autonomous development platform. That’s hard to ignore.

Elon’s blog today announced some fairly significant advances that have changed my overall perception. His mentioning that they elevated the radar sensors beyond the camera to serve in primary detection, enhanced with some new robust signal processing, suggests that the AP 1.0 platform has actually been underutilized, and maybe there’s still more they can do with it. It further begs the question why people think that AP 2.0 is right around the corner when Tesla is making such major advances on the current AP 1.0 platform. I’ve read half a dozen post-blog accounts today and there’s still no mention of imminent hardware changes.
 
Wie wil moet eens de test doen op Brusselse Ring Oost komende van Waterloo: precies op de plek waar je Vlaanderen binnenrijdt (toeval natuurlijk, maar het is wel precies die plek, waar het bord "Hoeilaart" verschijnt) begint de auto de zigzaggen (geldt op beide rijstroken, richting noord). Als ik alleen ben en er is geen tientonner naast me laat ik de auto doen, en na 500 meter vindt ie na lang zwabberen zijn weg dan wel terug, maar als ik mijn echtgenote als passagier heb moet ik toch meteen ingrijpen en heb ik telkens recht op ongeveer 15 seconden slechte poëzie over AP. En er zijn nog wel andere plekken in mijn buurt waar ik zelf telkens al 147 keer heb moeten ingrijpen en AP vandaag nog niet begrijpt waarom.

Euh… . Heb dit inderdaad zaterdagavond daar aan de lijve ondervonden! :eek: Kwam van Villers richting Gent.
Het begint lichtjes met zwalpen, maar begint dan nogal heftig te "schudden". Ik reed op het middenvak en stak net een vrachtwagen voor.
Heb angstvallig AP maar 'losgetrokken' en overgenomen, maar mijn drie passagiers vroegen onmiddellijk "Wat gebeurt er?"
Hmm, blijkbaar een lokaal probleem daar; ik had het nog niet eerder zo meegemaakt… . :confused:
 
Elon’s blog today announced some fairly significant advances that have changed my overall perception. His mentioning that they elevated the radar sensors beyond the camera to serve in primary detection, enhanced with some new robust signal processing, suggests that the AP 1.0 platform has actually been underutilized, and maybe there’s still more they can do with it. It further begs the question why people think that AP 2.0 is right around the corner when Tesla is making such major advances on the current AP 1.0 platform. I’ve read half a dozen post-blog accounts today and there’s still no mention of imminent hardware changes.

Yeah, it kinda sounds like Tesla is committed to the current approach for a good long time. If they were planning to push out a triple camera version next month that would solve all the issues and extend the capabilities greatly, would they really be invest large amounts of resources into building this geotagged whitelist for radar navigation, which will presumably take months to put together even with all of us contributing?

I'm sure we'll see a second generation AP hardware set at some point, but after seeing the depth of Tesla's commitment here I'm thinking they may skip the triple camera version and go straight to a 360 camera and radar set that will eventually be able to handle full autonomy - sometime next year or even early in 2018, and with no upgrade for current cars. (Full Autonomy won't be available for at least a couple years after that I suspect.)
 
  • Like
Reactions: MarkS22
I'm sure we'll see a second generation AP hardware set at some point, but after seeing the depth of Tesla's commitment here I'm thinking they may skip the triple camera version and go straight to a 360 camera and radar set that will eventually be able to handle full autonomy - sometime next year or even early in 2018, and with no upgrade for current cars. (Full Autonomy won't be available for at least a couple years after that I suspect.)

I think we are seeing implementation of a new sensor architecture which will ultimately be base for AP2.0 on Model 3. They need to start building cars with at least the wiring support so they can get a few hundred S/Xs out to EAP mules to get started on a very significant software rebuild. Needs to be done to weave into model 3 testing. Doubt that current AP1.0 folks will every see an upgrade path to AP 2.0. I also bet that anything produced from here on out will see one.