Forum » Znanost in tehnologija » Zakaj so avtonomna vozila prava stvar
Zakaj so avtonomna vozila prava stvar
Temo vidijo: vsi
Okapi ::
fikus_ ::
- posnetek mora biti objavljen do konca leta 2034, pred 1. januarjem 2035al
Predlagam traso - slovenska različica "coast-to-coast" - z Goričkega v Piran, ne po avtocesti (v navigacijo daš "brez cestnin"). V desetih letih samo v sanjah.
Odličen predlog.
Učite se iz preteklosti, živite v sedanjosti in razmišljajte o prihodnosti.
BigWhale ::
- posnetek mora biti objavljen do konca leta 2034, pred 1. januarjem 2035al
Predlagam traso - slovenska različica "coast-to-coast" - z Goričkega v Piran, ne po avtocesti (v navigacijo daš "brez cestnin"). V desetih letih samo v sanjah.
S Teslo, kupljeno leta 2022, ki ima vgrajen HW3. :> Ker to je nekaj kar Tesla obljublja.
kow ::
https://arstechnica.com/cars/2024/12/ma...
Novica spominja na sahovske programe in zgodbe kot:
David Levy (chess player) @ Wikipedia
Novica spominja na sahovske programe in zgodbe kot:
David Levy (chess player) @ Wikipedia
Gregor P ::
A2RL seems to understand and talks about the complications of aiming for an AI-powered vehicle. It would be nice if those companies testing on our streets did the same.
... predvsem opozori ostale na zahtevnost tega podviga preden testirajo na resničnih ljudeh v realnem prometu.
The main failure in computers is usually located between keyboard and chair.
You read what you believe and you believe what you read ...
Nisam čit'o, ali osudjujem (nisem bral, a obsojam).
You read what you believe and you believe what you read ...
Nisam čit'o, ali osudjujem (nisem bral, a obsojam).
kow ::
Tole sem cez vikend nasel:
Uporabljali so verzijo 12.3.6, ki tece na HW3 (avto je iz leta 2022).
Primer verzije 13.2.2 na HW4 (model Y):
Na avtocesto zavije cirka 22:40
Uporabljali so verzijo 12.3.6, ki tece na HW3 (avto je iz leta 2022).
Primer verzije 13.2.2 na HW4 (model Y):
Na avtocesto zavije cirka 22:40
Zgodovina sprememb…
- spremenil: kow ()
kow ::
A2RL seems to understand and talks about the complications of aiming for an AI-powered vehicle. It would be nice if those companies testing on our streets did the same.
... predvsem opozori ostale na zahtevnost tega podviga preden testirajo na resničnih ljudeh v realnem prometu.
Iscejo izgovore, ker so zasrali. Ce beres clanek, vidis da niso sposobni implementirati ogrevanja gum.
Zgodovina sprememb…
- spremenil: kow ()
kow ::
Sem nasel nekoga, ki namerava narediti cannonball z novejsim softwareom (v13):
https://x.com/AlexRoy144/status/1871214...
Rezultati za v12:
Tesla FSD Cannonball Run Record 001 Data
Software: FSD v12.5.6.4
- Engagement Rate: 98.7%
- Disengagements (Total): 32
- Disengagements (Voluntary, Charging): 21
- Disengagements (Total, non-Charging): 11
- Disengagements (Involuntary): 2
- Disengagements (Voluntary, Safety-Related): 7
- Disengagements (Voluntary, Navigation-Related): 2
- Navigation Errors: 3 (VDEs for 2 noted above, 3rd was allowed to play out)
- Other Errors: 1
- System Resets (Navigation-related): 2
The legacy metrics, for those curious about time and EV stats:
- Overall Time: 45 hours, 36 minutes
- Drive Time: 39 hours, 44 minutes
- Charge Time: 5 hours, 52 minutes
- Charge Stops: 21
- Total Energy: 941 kWh
- Driving Speed (Average): 71 mph
- Overall Speed (Average): 62 mph
- Distance: 2,833 miles
The Car
Vehicle: 2024 Tesla Model S
Wheels: Factory 19" Tempest
Tires: Continental ProContact RX
Mileage (post-run): 8,367 miles
Aftermarket Hardware & Software
- 1x Starlink Mini
- 3x GoPro Hero Black (2x Forward, 1x Interior, GPS enabled)
- 1x Teledyne FLIR
Prototype Thermal Pencil Camera (Forward) w/Display
- iPhone 16 running Waze
- iPhone 16 running Highway Radar
- iPhone 16 running Glympse
(for live witness tracking)
- iPhone 16 running Speed Tracker
(for GPS recording)
The following were superfluous, as FSD restricted top speed to 85mph, and often much less, depending on the speed limit and conditions:
- Stinger Laser Jamming & Radar Detection
w/3x pair heads front & rear
- Valentine V1 Gen 2
Laser/Radar Detector
https://x.com/AlexRoy144/status/1871214...
Rezultati za v12:
Tesla FSD Cannonball Run Record 001 Data
Software: FSD v12.5.6.4
- Engagement Rate: 98.7%
- Disengagements (Total): 32
- Disengagements (Voluntary, Charging): 21
- Disengagements (Total, non-Charging): 11
- Disengagements (Involuntary): 2
- Disengagements (Voluntary, Safety-Related): 7
- Disengagements (Voluntary, Navigation-Related): 2
- Navigation Errors: 3 (VDEs for 2 noted above, 3rd was allowed to play out)
- Other Errors: 1
- System Resets (Navigation-related): 2
The legacy metrics, for those curious about time and EV stats:
- Overall Time: 45 hours, 36 minutes
- Drive Time: 39 hours, 44 minutes
- Charge Time: 5 hours, 52 minutes
- Charge Stops: 21
- Total Energy: 941 kWh
- Driving Speed (Average): 71 mph
- Overall Speed (Average): 62 mph
- Distance: 2,833 miles
The Car
Vehicle: 2024 Tesla Model S
Wheels: Factory 19" Tempest
Tires: Continental ProContact RX
Mileage (post-run): 8,367 miles
Aftermarket Hardware & Software
- 1x Starlink Mini
- 3x GoPro Hero Black (2x Forward, 1x Interior, GPS enabled)
- 1x Teledyne FLIR
Prototype Thermal Pencil Camera (Forward) w/Display
- iPhone 16 running Waze
- iPhone 16 running Highway Radar
- iPhone 16 running Glympse
(for live witness tracking)
- iPhone 16 running Speed Tracker
(for GPS recording)
The following were superfluous, as FSD restricted top speed to 85mph, and often much less, depending on the speed limit and conditions:
- Stinger Laser Jamming & Radar Detection
w/3x pair heads front & rear
- Valentine V1 Gen 2
Laser/Radar Detector
Zgodovina sprememb…
- spremenil: kow ()
kow ::
Next Steps: Tesla FSD Cannonball Run Drive 002/003
We will publish more detailed data/content/comprehensive analysis in January 2025 after reviewing data and video footage from all three drives.
We will publish more detailed data/content/comprehensive analysis in January 2025 after reviewing data and video footage from all three drives.
kow ::
https://x.com/AlexRoy144/status/1886311...
Tesla FSD Cannonball Run 002 Results: v13 Better, But Hardware Has Limits
Drive 001: The Tesla FSD Cannonball Run Record
On Drive 001 we tested FSD to set what we call the Production Autonomy Cannonball Run record, creating the benchmark for future testing. The key metrics are Engagement Rate and Disengagements:
12/13 - 12/14/2024
Hardware 4
Software v12.5.6.4
Distance (Miles): 2,833
Engagement Rate: 98.52%
Disengagements (Total): 32
Drive 002: Did Not Finish (DNF)
On Drive 002 we tested FSD v13.2.2, which showed improvements over v12, but was unable to operate effectively in bad weather. The drive was cut short for safety 836 miles from the finish line.
12/25 - 12/27/2024
Hardware 4
Software v13.2.2
Distance (Miles): 1997 -- DNF
Engagement Rate: 99.51%
Disengagements (Total): 14
The Good: FSD v13.2.2 Is Better Than FSD v12.6.5.4
FSD v13.2.2 enabled autonomous entry and exit from Tesla Superchargers, dramatically reducing the volume of VDE's for charging.
FSD v13.2.2 required far fewer VDEs for safety, specifically for lane incursions by trucks.
FSD v13.2.2 feels much smoother and more comfortable than v12.5.6.4. Alas, we didn't bring a G-meter for granular measurement of ride quality. We will consider this addition to future runs.
The Par: Involuntary Disengagements
FSD v13.2.2 had twice as many IVDEs as its predecessor, but two of these were related to weather not encountered on Drive 001.
Drive 002 still had two unexplained IVDEs, in clear, low-complexity scenarios.
Both unexplained IVDEs required system reboots for FSD functionality to return, which also occurred on Drive 001.
The Limiting Factor: Tesla HW4 Cameras
Upon hitting the storm in Indiana, IVDEs became so frequent that FSD was effectively unusable, and the drive was cancelled.
These IVDEs were accompanied by camera obstruction and system error alerts.
One IVDE was triggered by manual activation of the windshield wipers.
Tesla is aware of weather-related camera issues, and recommends periodic precision cleaning of the forward-facing cameras
. This may mitigate and/or postpone visibility issues and IVDEs, but will not eliminate them.
Tesla has also been improving camera hardware with almost every new model release, but the performance limitation of current hardware forces clarification of when and where current (and older) Teslas will be capable of unsupervised FSD operations.
Next Steps: Tesla FSD Cannonball Drive 003/004
Our Model S has just completed refitting for Drive 003, with multiple additions/upgrades:
FLIR Prototype Thermal Camera + Display Upgrade
Nexar AI-Enabled Dashcam
Point One Navigation Hi-Resolution GPS
And of course, FSD 13.2.5 has been installed.
Next report in March!
Tesla FSD Cannonball Run 002 Results: v13 Better, But Hardware Has Limits
Drive 001: The Tesla FSD Cannonball Run Record
On Drive 001 we tested FSD to set what we call the Production Autonomy Cannonball Run record, creating the benchmark for future testing. The key metrics are Engagement Rate and Disengagements:
12/13 - 12/14/2024
Hardware 4
Software v12.5.6.4
Distance (Miles): 2,833
Engagement Rate: 98.52%
Disengagements (Total): 32
Drive 002: Did Not Finish (DNF)
On Drive 002 we tested FSD v13.2.2, which showed improvements over v12, but was unable to operate effectively in bad weather. The drive was cut short for safety 836 miles from the finish line.
12/25 - 12/27/2024
Hardware 4
Software v13.2.2
Distance (Miles): 1997 -- DNF
Engagement Rate: 99.51%
Disengagements (Total): 14
The Good: FSD v13.2.2 Is Better Than FSD v12.6.5.4
FSD v13.2.2 enabled autonomous entry and exit from Tesla Superchargers, dramatically reducing the volume of VDE's for charging.
FSD v13.2.2 required far fewer VDEs for safety, specifically for lane incursions by trucks.
FSD v13.2.2 feels much smoother and more comfortable than v12.5.6.4. Alas, we didn't bring a G-meter for granular measurement of ride quality. We will consider this addition to future runs.
The Par: Involuntary Disengagements
FSD v13.2.2 had twice as many IVDEs as its predecessor, but two of these were related to weather not encountered on Drive 001.
Drive 002 still had two unexplained IVDEs, in clear, low-complexity scenarios.
Both unexplained IVDEs required system reboots for FSD functionality to return, which also occurred on Drive 001.
The Limiting Factor: Tesla HW4 Cameras
Upon hitting the storm in Indiana, IVDEs became so frequent that FSD was effectively unusable, and the drive was cancelled.
These IVDEs were accompanied by camera obstruction and system error alerts.
One IVDE was triggered by manual activation of the windshield wipers.
Tesla is aware of weather-related camera issues, and recommends periodic precision cleaning of the forward-facing cameras
. This may mitigate and/or postpone visibility issues and IVDEs, but will not eliminate them.
Tesla has also been improving camera hardware with almost every new model release, but the performance limitation of current hardware forces clarification of when and where current (and older) Teslas will be capable of unsupervised FSD operations.
Next Steps: Tesla FSD Cannonball Drive 003/004
Our Model S has just completed refitting for Drive 003, with multiple additions/upgrades:
FLIR Prototype Thermal Camera + Display Upgrade
Nexar AI-Enabled Dashcam
Point One Navigation Hi-Resolution GPS
And of course, FSD 13.2.5 has been installed.
Next report in March!
Zgodovina sprememb…
- spremenil: kow ()
Gregor P ::
Kako bodo potem stare Tesle s takratno strojno opremo lahko delale z novimi FSD-ji, kaj šele povsem avtonomno vožnjo (FFSD
), kar obljublja Musk?
), kar obljublja Musk? The main failure in computers is usually located between keyboard and chair.
You read what you believe and you believe what you read ...
Nisam čit'o, ali osudjujem (nisem bral, a obsojam).
You read what you believe and you believe what you read ...
Nisam čit'o, ali osudjujem (nisem bral, a obsojam).
RedDrake ::
Nič bati, nam je WarpedOne že pred 9-imi leti napisal, da je samodejna vožnja rešen problem. In da je ves potreben HW že v takratnih 3-kah. Zdaj smo pa itak že skoraj na AGI, čez par let bo verjetno že singularnost, prosto po Thomasu. Po drugi strani pa še vedno ni možno da bi AI sam vozil v nevihti. Ziher zato, ker so kamere povsem dovolj, tako kot 640kB spomina.
Zgodovina sprememb…
- spremenil: RedDrake ()
Okapi ::
Problem s kamerami je, ker so napačno postavljene. Načeloma bi bila dovolj ena (premična) kamera v kabini vozila. Če človek lahko vozi z eno "kamero" (vozijo lahko tudi tisti, ki so slepi na eno oko), ni nobene teoretične ovire, da tega ne bi mogel tudi robot.
Dokler robot ni dovolj "pameten", pa tudi lidar in rada in sonar in sto kamer ni dovolj.
Dokler robot ni dovolj "pameten", pa tudi lidar in rada in sonar in sto kamer ni dovolj.
Zgodovina sprememb…
- spremenil: Okapi ()
DamijanD ::
To je že fino, da je avto sposoben spremljati hkrati celotno okolino. Človek s premično kamero kdaj kaj spregleda, ker je osredotočen na napačno stran...
kow ::
Kaksno je optimalno stevilo kamer, je tezko reci. Vec kamer, vec procesiranja. Ljudje sicer vozimo samo z eno ali dvema "kamerama", ampak imamo pripomocke (ogledalo, senzorje za mrtvi kot, parkirni senzorji itd.).
Tesla trenutno vozi s cirka 100W, za Waymo ne vem, ce je znano. Ko bo enkrat samodejna voznja resena, bodo cifro verjetno hoteli spraviti za kaksen velikostni razred nizje. Vec se verjetno ne splaca.
Tesla trenutno vozi s cirka 100W, za Waymo ne vem, ce je znano. Ko bo enkrat samodejna voznja resena, bodo cifro verjetno hoteli spraviti za kaksen velikostni razred nizje. Vec se verjetno ne splaca.
Zgodovina sprememb…
- spremenil: kow ()
BigWhale ::
Okapi ::
Ja, in to pomeni, da je šibka točka AI softver, ne harver (možgani komarja imajo okoli 200.000 živčnih celic in okoli 50 milijonov povezav med njimi.
BigWhale ::
Meni bolj deluje, da je hardware, na katerem poskusamo implementirati vse skupaj, precej nekompatibilen s samo nalogo. Nekako nisem preprican, da lahko na tem HWju spisemo tak software. Komarjevi mozgani niso binarni.
Ceprav se v osnovi strinjam, da je problem tudi v SW.
Ceprav se v osnovi strinjam, da je problem tudi v SW.
Zgodovina sprememb…
- spremenil: BigWhale ()
kow ::
Cloveski mozgani so precej slab hw. Malo prej sem si hotel zapomniti 8 mestno stevilko, pa v prvo ni uspelo. Popolno sranje. /s
Zgodovina sprememb…
- spremenil: kow ()