Verwijderd schreef op zondag 2 juni 2019 @ 19:26:
We gaan het zien, ik zie met name door de hoge investeringen maximaal 2 of 3 spelers in level 4 of 5 autonomie en degenen die voor lidar hebben gekozen zijn zo goed als af.
Ik bedoelde écht dat het neurale netwerk elke dag bijleert, niet dat je elke dag met nieuwe software rijdt.
Lidar heeft anders hele grote voordelen om ernaast te hebben draaien. Misschien is het voor nu nog te duur. Maar uiteindelijk zal het toch erbij moeten. Dingen die heel simpel lijken: Welke kant wijst een gezicht op, kan wel met Lidar, maar niet met Radar. Dus om voorspellend gedrag te bepalen is Lidar best wel een fijne techniek.
In de mist of regen heb je wee weinig aan Lidar, wel aan radar. Vooral in het donker is Lidar echt een toevoeging.
Het één sluit het ander niet uit.
Ik merk dat veel mensen Lidar afschrijven, omdat Tesla het niet gebruikt. Als Lidar goedkoper was geweest, zat het ook op iedere Tesla, het is echt een kostenoverweging.
Weer even een stapje terug doen, los van wat Tesla wel of niet doet, Lidar en Radar samen, levert het beste resultaat. Een "neuraal netwerk" klinkt allemaal leuk en aardig, uiteindelijk is het gewoon machine learning en worden daar verschillende AI modellen voor gebruikt. Wees maar niet bang, partijen als Google en Microsoft zijn daar al best ver mee. Niet specifiek op verkeer, maar heel veel andere vlakken. Dat is binnen no-time toegepast op verkeer.
kdeboois schreef op zondag 2 juni 2019 @ 19:50:
[...]
Musk stelt dit zeer duidelijk ter discussie en dat wordt hier 1 op 1 overgenomen.
LIDAR is duur, maar dat is een tijdelijk iets.
Ik ben het deels met Musk eens, maar LIDAR heeft een aantal voordelen. Ik denk zelf dat wat geavanceerdere radar ook goed werk kan doen. De eindoplossing van een stevige sensor fusion implementatie worden met communicatie tussen alles wat om je heen beweegt en de infrastructuur.
Het detail is echter veel minder groot dan bij Lidar. Lidar is gewoon een hele goede toevoeging op zelfrijdende auto's. Gelukkig hoef je niet OF de één OF de ander te kiezen!
Verwijderd schreef op zondag 2 juni 2019 @ 20:02:
Ik ben benieuwd wat de toekomst gaat brengen, of het nou Tesla is die het als eerste voor elkaar gaat krijgen of niet. Het moet ook nog schaalbaar zijn en op dit moment is LIDAR dat niet voor personenauto's.
Er zijn toch al lang zelfrijdende auto's... Genoeg projecten van Google en Waymo... Tesla kan daar niet meer de eerste mee worden. Maar ik zie ook geen toegevoegde waarde in de eerste ergens mee zijn.
Het zal een lerend proces zijn.
Waymo is ook gewoon nog aan het uitbreiden hoor:
https://www.theverge.com/...ing-ride-hail-app-phoenixEindelijk 200kWh accu's!!!
Iedere keer als ik vraag om 100-150kWh wordt er genoemd dat het onzin is en nergens op slaat. Maar ik neem aan dat als Tesla het doet, dat het dan wel perfect is, toch!?

Matszs schreef op zondag 2 juni 2019 @ 22:31:
Elon Musk heeft als reden gegeven dat het onnodige kosten zijn (of die nou 1000 euro of 10.000 euro is), omdat je met camera's hetzelfde kan bereiken en je die sowieso nodig hebt om een object te identificeren. Met LIDAR weet je dat er iets is, maar niet wat.
Hoe kom je daarbij?
Verschil tussen high res radar en lidar:
Camera's hebben licht nodig om te "kijken".
Hij is natuurlijk wel heel erg bevooroordeeld!
Hij kan ook nu niet noemen dat het wel handig is voor zelfrijdende auto's, dan zegt hij namelijk indirect dat zijn huidige auto's niet goed genoeg zijn. Kan niet.
Ook al weet hij dondersgoed dat het een welkome toevoeging is als het goedkoper is, hij zal en kan het nu niet roepen!
Deveon schreef op zondag 2 juni 2019 @ 23:00:
[...]
Zonder lidar, want als die niet voldeed had ik hem niet kunnen bestellen.
Een goed voorbeeld blijft een plastic zak. Lidar zegt stoppen, camera’s zeggen rijden. Wat is dan het voordeel van een dubbel systeem? Uiteindelijk moet er één overrulen.
Dat heb je nu ook al met radar en camera. Het is niet zo moelijk, zodra één van de systemen zegt stoppen: geen risico: stoppen! Het maakt niet meer uit wat de andere systemen dan zeggen.
MeZZiN schreef op maandag 3 juni 2019 @ 11:32:
Nee klopt alleen bij mensen halen we onze schouders op en bij AI niet. Zie hoe elke AP incident breed uit gemeten wordt in de pers AI wordt gewoon aan een veel te hoge standaard gehouden.
Niet helemaal... Bij level 5 zelfrijden, kun je gewoon op de achterbank plaatsnemen, zonder iemand achter het stuur te zetten...
Durf je dat straks aan met autopilot?
Voor en nadelen LIDAR:
Voordelen:
One of the primary advantages of LiDAR is accuracy and precision. The reason why Waymo is so protective over its LiDAR system is because of its accuracy. In December 2017, The Drive reported that Waymo’s LiDAR is so incredibly advanced, it can tell what direction pedestrians are facing and predict their movements. Chrysler Pacificas outfitted with Waymo’s LiDAR can see hand signals that bicyclists use to predict which direction the cyclists should turn.
Another advantage of LiDAR is that it gives self-driving cars a three-dimensional image to work with. LiDAR is extremely accurate compared to cameras because the lasers aren’t fooled by shadows, bright sunlight or the oncoming headlights of other cars.
Luminar LiDAR
Finally, LiDAR saves computing power. LiDAR can immediately tell the distance to an object and direction of that object, whereas a camera-based system must first ingest the images and then analyze those images to determine the distance and speed of objects, requiring far more computational power.
Currently only Audi is deploying front-facing LiDAR in consumer vehicles (with the 2019 A8 and A6). Otherwise LiDAR is still primarily the domain of ride-hailing service vehicles.
Nadelen:
Historically, cost was one of the major disadvantages of LiDAR. Google’s system originally ran upwards of $75,000. Today startups are bringing the costs of LiDAR units down to below $5,000, but that’s still relatively expensive considering you might need several units around the car. Until LiDAR costs come down to more reasonable levels, self-driving cars based on these systems may remain expensive for the ordinary public to purchase.
Interference and jamming are another potential issue with LiDAR as these system roll out more broadly. If there are a large number of vehicles all generating laser pulses (photons) at the same time, it could cause interference and potentially “blind” the vehicles. Manufacturers will need to develop methods to prevent this interference.
It’s also not clear what so much constant laser activity would do to other mechanical and biological systems in the environment. For example, Velodyne claims that LiDAR operating at 1550 nm, like their competitor Luminar uses, could actually damage the human cornea under some circumstances. At CES 2019, someone reported their camera sensor being damaged from LiDAR. There are still many open questions around the environmental safety of LiDAR at scale.
LiDAR also has a limitation in that many systems cannot yet see well through fog, snow and rain. Ford, another leader in race to develop viable self-driving cars, developed an algorithm that helps its LiDAR system differentiate between individual raindrops and snowflakes. Otherwise, autonomous vehicles would interpret a mass of falling snowflakes as a wall in the middle of the road. Ford proved its concept while undergoing testing in Michigan, but its program still has a lot to learn.
Additionally, LiDAR doesn’t provide information that cameras can typically see – like the words on a sign, or the color of a light. Cameras are better suited for this type of information.
Finally, LiDAR systems are currently very bulky since they require spinning laser systems to be mounted around the vehicle, as seen above. Whereas camera systems, as implemented on current Tesla vehicles are almost invisible. That said, Audi, has dipped its toes into the water and placed a front-facing LiDAR (the Valeo SCALA unit) into its vehicles such as the A8 in such a way that it’s fairly hidden. However, it’s only front-facing LiDAR at this point.
[
Voor 24% gewijzigd door
BlakHawk op 03-06-2019 11:42
]