Tesla is crazy not to use maps

Topic: 
Tags: 

Every time I compare Tesla with other contenders, people will say things to the effect of "they don't count, they use maps so they can only drive in tiny regions" because they think that's a bug rather than a feature.

Maps are super useful, and a car that can try to drive without a map is a car that can make a map, and so maps scale just fine and help you drive more roads, rather than fewer.

So I made a video outlining the virtues of maps, why they are cheap, and why it's OK if construction changes the road after you map it.

This is also accompanied by a text version of the transcript which can be found at Tesla is crazy not to use maps

FAQ

But Tesla does have maps!

Yes, the video states that Tesla makes use of lane-level maps in many (though not all) places and even uses more detailed maps in some places. They deny they use highly detailed maps anywhere though there is evidence they do use them in some places.

The key is to use maps with good detail, and use them everywhere they are needed. They definitely don't do that yet, even in Silicon Valley, where every road has Teslas going down it constantly and where the developers live.

Humans drive without a map

This is covered in the video. Humans can drive without maps but prefer to have them and do better when they remember the road. They are more likely to make mistakes on roads they have never seen compared to roads they are very familiar with. But most of all, robots are not humans, and there is no sign they will be as smart as humans any time soon, or that the best way to solve the problem of understanding the road is making them think like humans. Any more than the best way to fly is to flap your wings like birds, though that works if you're a bird or fictional plane in Dune.

How doe the maps get to the car?

99.99% of map data does not need to be sent real time, and is updated when connectivity is available (in particular if the car parks somewhere with wifi as they often do.)

The only updates that need be sent in real time are

  1. Data from the car if it's the very first to encounter a change to the road, as soon as it gets a connection
  2. To the car if it has plotted a route that eventually goes along a road that has an update

In the worst case if connectivity is very spotty, you can send the location of a map change to a car that is planning to drive that street so that it knows to avoid it or at least is not surprised. In addition, changes located near the car which it might decide to drive down will come as data is available. Needed data is not very large and low speed (3G) connectivity would be far more than is needed.

Comments

deepmap was very expensive several years ago. 5000 dollars per km turn key. easy to search.

you never provide an estimate of the density distribution of vehicles per paved lane mile to support your assistive map long term is never provided. We all do not live in cali.

Here TomTom Mobileye Atlatec Carmera Momenta Ushr DeepMap Zenrin Mapbox DynamicMap NavInfo

way too fragmented

lol - Infosysbpm presentation $32 billion for US

talk about meaningless research with no smidgen of reality.

*Source: (DEEPMAP data) 2020

duplicative costs of every oem building their own maps is insanity.

just for one country, $32 billion to build 3d point maps with lidar for localization? yes it is probably less expensive today, but by how much.

and all that $ for just one country.

even if other lidar 3d point maps are 50% less expensive than DeepMap fees, the 10 largest oems duplication of what the other oems do is beyond a substantial spend if using lidar maps.

and the observation that lidar currently is mostly used for localization and not object detection (james douma) is numbing when realizing most automotive lidar is above $500 per unit.

carmera, toyota, ford, tesla, nissan and mobileye should get together and build a global map for pennies.
put a camera and a chip in the windscreen at $150 per vehicle and build the maps for mere pennies.

source: Mapping an autonomous driving future - Infosys BPM

While that is an overestimate, and the cost keeps dropping as tech improves and gets cheaper, it actually would be a quite affordable cost if it had to be paid. $100 per American? Under $200 per driving American? Who drives 12,000 miles/year on average and spends $8,000/year doing it? Also it's not a cost per year, it's the one time cost of the initial map with a lower cost to update it.

Even if you thought it would cost this much, it's a tiny drop in the cost of car travel if it makes it safer and more reliable. But it doesn't cost this much.

Admin note: I do prefer if people post more than a few times here and want to be Anonymous that they make up a pseudonym and use it to tell one anonymous poster from another. Regular users are also welcome to create an account which allows many useful features, like editing your comments, putting in HTML if needed, and authenticated userid.

please help with EFF's pursuit for anonymity if possible

I don't deny people the ability to be anonymous here (though EFF feels that the owner of an online space is free to deny that, and even can be wise to deny it, but defends the existence of anonymous communication where it is wanted or needed, and does not want governments to forbid it.)

However, I have a user here who has an account but I don't know the name of. I could make effort to track his or her IP addresses but I have not done so, and it might not be fruitful anyway if the person uses TOR or a VPN. Usually most people are not that worried.

I request a pseudonym just because it makes it easier for people to understand and track threads. As the other participant said, there is somebody here posting anonymously who is fairly confusing and mildly chaotic. I welcome almost all views here, as long as they are useful and interesting, but I will shut down flamewars or ad hominem.

You can certainly be anonymous if you'd like. But what Brad is suggesting is that you pick a consistent pseudonym and use it for all of your postings here. That way, Brad, and his readers, can better understand if a single person is making all of these short comments dumping incomplete thoughts without any conversational niceties, or if a whole bunch of different people are just filling up the threads with hard-to-understand snippets.

If you are trying to engage in a conversation, it will be a lot more productive if you explain something about the points you are trying to make, and perhaps even ask questions. Capitalization, punctuation, and complete sentences would help, too.

capable of generating High Definition Maps ?.

So it would start from maps, though it could have a map-making module to make the maps the simulator needs.

I commented, tongue in cheek, on the YouTube video that Tesla was deliberately making their data more interesting by refusing to learn from what they had seen before, therefore increasing the number of difficult scenarios their autopilot has to deal with.
Running red lights and getting stuck in tramways or dead ends gives their learning algorithm so many more "interesting" scenarios to work compare to boring self driving stacks that used high definition maps.

Everybody certainly tries to do this in sim, to find the craziest and most interesting scenarios.

One of the controversies today is "how much testing should you do on public roads." One answer is "no so much that you're finding bugs faster than you can fix them." Tesla clearly is way beyond that. Aurora feels they can find bugs in sim so fast that they are not on the roads much yet.

E.W. Niedermeyer

Aug 22
Developing a "general solution to autonomy" by validating a single intersection so an influencer's videos look better, you literally couldn't make this shit up

Warren Craddock never cites any body of scientific research in the sensor choice and sensor fusion discussion.

Deep Learning For Autonomous Driving Workshops on youtube 3D-DLAD-v4

Dr. Eric Richter from BASELABS as an example.

previous years also available
IEEE sensor fest

"This year, we'll see approximately 1 million V2X cars in Europe. The expected penetration will be in the 10 million-ish range by around 2024–2025"

What does it mean to be a v2x car? Is this a car regularly transmitting BSMs? A car that uses them? Some other type of V2X?

10 million of the 250M cars in Europe means that the chance that any two in an interaction could use V2X is 0.0016 (about 1/6th of one percent.)

Ars Technica yesterday in NTSB report.

"30 MHz is enough for basic safety features," said Balázs Tóth-Pintér, a communications specialist with V2X company Commsignia, who pointed out that the EU has only allocated a total of 40 MHz for V2X. And unlike in the US, deployment is happening in Europe. "This year, we'll see approximately 1 million V2X cars in Europe. The expected penetration will be in the 10 million-ish range by around 2024–2025," he told Ars.

Just what are those million cars actually doing with their V2X?

NXP blog explains much.
Some insight as well at Conti and Siemens.

In 2022, the company plans to map the entirety of Europe, which will require the system to scale up to 200,000 concurrent vCPUs for 20 days—96 million vCPU hours in total.

The REM team updates the map in near real time: accessing, changing, rebuilding, and stitching together more than 2 million kilometers of drivable paths with detail down to the level of a single stop sign.

aws dot amazon dot com/solutions/case-studies/mobileye-video-case-study/

electrek dot com Oct 24 2022

XNGP (navigation-guided pilot) was announced as XPeng’s last step before achieving fully autonomous driving. XNGP combines all scenarios of the automaker’s existing ADAS capabilities (highway, city, parking) into one holistic system that will soon no longer require high-precision maps to function – essentially opening up its availability to any and all areas.

ISA (intelligent speed assist) content for Europe from HERE manages the non-posted speed limit challenge by delivering speed limit information based on a combination of data sources. Reliance on vehicle camera data alone, for example, is prone to quality challenges without a base map and validated mapping system.

So if China pursued ISA, what does XPeng do?

Overture Maps Foundation initiative launched, founded by Amazon Web Services (AWS), Meta, Microsoft, TomTom and the Linux Foundation.

AWS stated that "maintaining accurate and comprehensive map data is cost prohibitive and complex."

“Mapping the physical environment and every community in the world, even as they grow and change, is a massively complex challenge that no one organization can manage. Industry needs to come together to do this for the benefit of all,” said Jim Zemlin, executive director for the Linux Foundation.

This could imply the development of the Mobileye REM Roadbook is underrated from both a technical and an economic perspective.

NYC alone maintains nearly one million street signs, 13,250 signalized intersections, over 315,000 street lights, and over 350 million linear feet of markings.

There were likely to be around 4.57m traffic signs in England in 2013, and in 2020, an estimated 5.11 million traffic signs.

Imagine the size of the REM Roadbook

The REM Roadbook identification algorithm could have identified over 630 million elements for Europe, Japan and the US combined.

REM Roadbook algorithms are utilized to identify elements such as -
Lane markings
Road markings
Road edge
Crosswalks
Traffic Signs
Traffic Lights
Telco Poles
Construction area
Road surface
Power Poles
Lamps
Manholes
Drainage
Telco cabinets
Electricity boxes
Cracks and road surface quality. Etc

There are about 15,812,406 intersections in the US.

There is an estimated 340,000 signalized intersections in the US.

Some intersections in the US can host as many as 24-30 individual signal housings for both vehicular and pedestrian signaling.

There is over 180 million utility poles around the US.

NY City alone maintains nearly 1 million street signs, 13,500 signalized intersections and 40,000 intersections total, over 315,000 street lights, and over 350 million linear feet of markings.

Phoenix has 1,152 Signalized Intersections and 1,629 Left turn Arrows, 2,775 Speed Humps, 19,366 Stop and Yield Signs, 93,000 Street Lights, 2,400 Parking Meters, 2,000,000 Curb Ramps, and 21,600 Potholes filled each year.

San Francisco has about 1200 traffic signals.

Dallas maintains about 160,000 traffic control and street name signs in Dallas.

Miami Dade County maintains nearly 3,000 traffic signals and 27,000 streetlights.

Seattle has over 204,000 signs
in the Seattle City GIS dataset map.

Los Angeles Department of Public Works administers 97,033 street lights and 4200 traffic signals.

There are nearly 2000 traffic lights in Las Vegas.

Atlanta DOT maintains roughly 100,000 traffic and street name signs and about 950 traffic signals.

Wayne County Micigan maintains approximately 100,000 signs and 1,458 traffic signals on the state and county roads.

Chicago has nearly 300,000 street and alley lights and more than 3,000 signalized traffic intersections.

There were likely to be around 4.57m traffic signs in England in 2013, and in 2020, an estimated 4.9 million traffic signs.

The REM Roadbook identification algorithm could have identified over 630 million elements for Europe, Japan and the US combined.

Intel won the 2022 Ministry of Economic Affairs electronic information international partner excellent manufacturer award.

Intel won recognition from the Ministry of Economic Affairs for its two major projects, the 'Intel DevCup' competition and the 'Intel-Mobileye Smart Transportation'.

Intel won the 2022 Ministry of Economic Affairs Electronic Information International Partner Award, IPO Awards, and was recognized by the Ministry of Economic Affairs with the 'Intel DevCup' competition and the 'Intel-Mobileye Smart Transportation' project. Liu Jingci, general manager of Intel Taiwan Branch Received the Innovative Application Partner Award from Minister of Economic Affairs Wang Meihua.

Intel and Mobileye are working with ecosystem partners in Taiwan to provide innovative big data analytics on traffic flow, road asset inventory, and road conditions. The results of this effort will not only enable more effective implementation of public road maintenance and traffic management, but also provide a powerful Geographical Information System platform for the deployment of autonomous vehicles in the future, helping to create smart cities and smart transportation ecosystems system.

Mobileye will never sell over 10 million Roadbook harvesting capable vehicles in the US in the three years of 2023 through 2025 to satisfy the 1 billion kilometers per day Roadbook collection goal between the US and Europe and PRDC by the end of 2025. Ford, VW, BMW and maybe Nissan won't be enough even if there is 500,000 to a million vehicles already in the pool in 2022. And to think the company needs 25 million vehicles a day harvesting data to meet the goal.

"1B KM of roadside data we expect to collect daily by 2024" - currently on a main section headline page of the Mobileye website.

WTH? 25 million by the end of 2024 and company last stated 1.5 mil vehicles were harvesting data. The addition of 12 mil harvesting vehicles per year for 2 years?

A third of that would be a major accomplishment.

City 3: Shanghai
Baidu has secured the first HD map permit for urban ADAS in Shanghai. It is now the first in China to receive HD map licenses for urban ADAS in three major Chinese cities - Shanghai, Guangzhou, and Shenzhen.

According to Questel_Group, Baidu is a leading applicant of autonomous driving-related patents, ranking No.1 globally with 1,193 filings. Baidu’s portfolio value outshines competitors in not only quantity but also in value and impact.

www bizjournals com/sanjose/news/2022/11/09/heres-why-velodynes-founder-says-lidar-is-bad-bet.html

Baidu's camera only Apollo Navigation Pilot ANP 3.0 intelligent driving system to launch in middle of 2023.

Baidu unifies L4 and L2+ autonomous driving technology in ANP 3.0.
ANP 3.0 is derived from Baidu's L-4 AVP autonomous driving technology.

Baidu and Mobileye follow the same approach.

SIP-adus Workshop 2022 presentation material for Dynamic Maps explains in detail AV maps

Dynamic Map

Dynamic Map-1: Satoru Nakajo
Dynamic Map-2: Andras Csepinszky
Dynamic Map-3: Jean-Charles Pandazis
Dynamic Map-4: Jens Vogelgesang
Dynamic Map-5: Yoshiaki Tsuda
Dynamic Map-6: Hiroyuki Inahata

Nissan selects Ushr maps for ProPILOT Assist 2.0 debut in North America on the Ariya

Very Confusing
Mobileye 2023 CES presentation shows Nissan Ariya using EyeQ technology.

Does the Ariya then use Ushr HD Lidar maps with EyeQ camera for Pro-Pilot?

Is the Mobileye system responsible for any localization or does it simply hand over an object list to Nissan?

Would that not be more expensive to Nissan to be responsible for localization?

Very confusing.

Also, DMP maps were used by GM for localization in 2021 CT4 abd CT5 yet Mobileye 2023 CES presentation shows 2022 CT4 and CT5 product launches.

can Apple afford to take 3-4 years to build broad coverage AV maps in-house of North America and Europe?

Mobileye CEO Jan 26 2023 call

Vertical handle of an end-to-end system, I think, is crucial, because you're talking about perception you're talking about integrating with a map. The map is built together with the teams that are building the perception.

So if you try to separate the map from perception to two different suppliers, you get into a sea of issues, either it will be over-engineered or be under-engineered, cost-wise, it could be crazy. The fact that now the same team (SuperVision) is integrating both the sensing, both the perception and the way the map is being built and served is crucial. Then you have driving policy. The driving policy is also integrated with the perception.

But again, if you try to separate that into a supplier doing the driving policy and other supply doing the perception, you end up with an over-engineered system and in some places it will be under-engineered be too conservative and too slow. So I think in such a complex system an end-to-end where everything is done by one supplier has a lot of advantages and has also not only performance advantages, but also cost advantages.

Everything under one house under one chip is - it offers incredible cost advantages. But we are not shy from cooperating in other ways. For example, there are OEMs that would like to take control of the driving policy, where Mobileye provides only the perception, we're open to that.

This is why we offer the EyeQ Kit, which enables the OEM or a supplier to write code on to our chip on top of our software, whether it is fusion with other sensors, whether it's driving policy, we don't resist that.
But having an end-to-end system can be much more efficient than we can get down to different suppliers.

wth?
Why was Mobileye capacity constrained at just 90,000 units on their in-house developed ECU when Zeekr announced back over a year ago in January 2022 of an aggressive delivery goal of 70,000 vehicles for 2022. Zeekr can't be too happy about it but Mobileye spins it as a strong growth indicator rather than poor supply chain management. Other OEMs have to be scratching their head on that one. Quanta Computer develops and assemble Mobileye ECUs. Blame the supplier and not internal supply chain management.

mobileye market cap hovers 32 billion today

NavInfo to provide BMW with autonomous driving map service in China
February 13 , 2023

Beijing - China’s navigation service provider NavInfo announced on February 13 it inked the agreement with BMW China to offer the latter autonomous driving related map services.

According to the agreement, NavInfo will provide products and services such as ADAS (Advanced Driver Assistance System) maps, HD (High Definition) maps, and LBS (Location Based Services) for BMW Group's next-generation autonomous driving functions in the Chinese market.

Add new comment