Autonomous Vehicles News & Discussions

[Yuli Ban]

DMV probing whether Tesla violates state regulations with self-driving claims

Tesla is “under review” by the California Department of Motor Vehicles to determine whether the electric car maker misleads customers by advertising its “full self-driving capability” option, the agency told The Times.

A Tesla equipped with the $10,000 full self-driving package can change lanes, take highway exit ramps, and stop at traffic lights and stop signs on its own, the company says. However, it is not capable of fully driving itself, according to widely accepted engineering standards.

Asked for detail, DMV spokesperson Anita Gore said via email, “The DMV cannot comment on the pending review.” She did list the penalties that might be applied if a company is found to have violated DMV regulations that prohibit misleading advertising concerning automated vehicles.

In small print, Tesla says on its website that full self-driving “does not make the car autonomous” and that “active supervision” is required by the driver. But social media are rife with videos showing drivers, mostly young men, overcoming Tesla’s easily defeated driver-monitoring system to crawl into the back seat and let the Tesla “drive itself” down public highways.

[wjfox]

Self-driving technology company Argo AI makes lidar breakthrough

May 5, 2021

Argo AI, the self-driving technology company backed by Ford and Volkswagen Group, claims to have overcome one of the main challenges holding back the deployment of driverless cars: accurately seeing and identifying objects at long distances.

The company on Tuesday announced its own lidar sensor, aptly named Argo Lidar, which it says can accurately spot and identify objects more than 1,300 feet away, or about 300 feet more than current lidar sensors—something that's vital for highway driving. Argo Lidar is said to be so accurate that it can even correctly identify graffiti on a wall or spot small moving objects such as animals among vegetation and static objects.

Most players in this space rely on a combination of sensors to guide self-driving cars, with the primary ones being cameras and lidar. While cameras work well at long distances, they don't work well in poor light, for example during night driving or inclement weather. While lidar doesn't suffer the same light issues as cameras, the technology until now was limited by its range and accuracy.

Read more: https://www.motorauthority.com/news/113 ... eakthrough

[Yuli Ban]

[wjfox]

Poll options could be better. How about including 2030, 2035.

[wjfox]

Germany gives greenlight to driverless vehicles on public roads

Tue, May 25, 2021

Germany has adopted legislation that will allow driverless vehicles on public roads by 2022, laying out a path for companies to deploy robotaxis and delivery services in the country at scale. While autonomous testing is currently permitted in Germany, this would allow operations of driverless vehicles without a human safety operator behind the wheel.

The bill, which last week passed the Bundestag, Germany’s lower house of parliament, specifically looks at vehicles with Level 4 autonomy. Level 4 autonomy is a designation by the Society of Automobile Engineers (SAE) which means the computer handles all the driving in certain conditions or environments. In Germany, these vehicles will be limited to geographic areas.

“In the future, autonomous vehicles should be able to drive nationwide without a physically present driver in specified operating areas of public road traffic in regular operation,” reads the legislation. “According to the Federal Government, further steps must be taken to introduce corresponding systems into regular operation so that the potential of these technologies can be exploited and society can participate in them.”

https://finance.yahoo.com/news/germany- ... 57434.html

[weatheriscool]

Comma AI Open Source Self Driving Car Software
May 28, 2021 by Brian Wang

Comma AI has two self-driving cars products called Openpilot and Comma two. Comma had about $5 million in annual revenue in 2020. They have 40 million miles of driving data collected from drivers using its inexpensive off-the-shelf electronics. George Hotz is the Founder and President of Comma.ai. They have raised $8.1 million in funding.

The Comma Two is a niche product for enthusiasts. The hardware costs $999 with an additional $200 cable needed. For regulatory and liability reasons, customers have to install the OpenPilot software onto the Comma Two.

https://www.nextbigfuture.com/2021/05/c ... tware.html

[Yuli Ban]

A Self-Driving Truck Got a Shipment Cross-Country 10 Hours Faster Than a Human Driver

Self-driving cars are taking longer to come to market than many expected. In fact, it’s looking like they may be outpaced by pilotless planes and driverless trucks. A truck isn’t much different than a car, but self-driving technology is already coming in handy on long-haul trucking routes, as a recent cross-country trip showed.

Last month TuSimple, a transportation company focused on self-driving technology for heavy-duty trucks, shipped a truckload of watermelons from Arizona to Oklahoma using the truck’s autonomous system for over 80 percent of the journey. The starting point was Nogales, at Arizona’s southern end right on the border with Mexico. A human driver took the wheel for the first 60 miles or so, from Nogales to Tucson—but from there the truck went on auto-pilot, and not just for a little while. It drove itself all the way to Dallas, 950 miles to the east (there was a human safety driver on board the whole time, but not controlling the truck).

[Yuli Ban]

Tesla could dominate $1.2 trillion market with autonomous, Uber-like vehicle fleet

Tesla (TSLA) is looking at a more-than-decent chunk of a market valued at $1.2 trillion with its upcoming Tesla Network, a new Uber-like ride-hailing service with autonomous electric cars, according to Ark Invest.

For years, Tesla has taunted a ride-hailing app to be powered by its self-driving system. It’s been dubbed the “Tesla Network.”

The new product is dependent on Tesla solving computer vision to deploy its full self-driving system.

In case it does happen, which CEO Elon Musk has been claiming it will this year, investors have been trying to value the extremely disruptive new product.

Ark Invest, a Tesla shareholder, has been trying to put some numbers on the value potential of such a service in its Big Ideas 2021 report.

The group believes that autonomous ride-hailing platforms will generate over $1 trillion in earnings by 2030:

[Time_Traveller]

NVIDIA to buy autonomous vehicle mapping company DeepMap

June 11th, 2021

NVIDIA is to acquire DeepMap, a company that makes high-definition mapping technology for self-driving cars. "DeepMap is expected to extend our mapping products, help us scale worldwide map operations and expand our full self-driving expertise," said NVIDIA VP Ali Kani.

DeepMap provides maps with high levels of precision. NVIDIA points out that maps accurate to within a few meters are fine for turn-by-turn GPS directions, but autonomous vehicles require greater accuracy. "They must operate with centimeter-level precision for accurate localization, [so that] an AV can locate itself in the world," NVIDIA wrote in a blog post.

https://www.engadget.com/nvidia-acquire ... 12963.html

[weatheriscool]

Prediction that Tesla Will Solve FSD by Mid-2022
June 16, 2021 by Brian Wang

Mr Know It All has a projection that Tesla should solve FSD (Full self-driving) to a six sigma level within 275 days. This is based upon the improvement of the FSD system matching to the amount of data from the number of miles driven used by Tesla.

Tesla currently has about 35 billion kilometers of driving data. They are adding about 2000 cars every day and this will increase to 5000-6000 cars every day by mid-2022 and 7000-8000 by the end of 2022. Each car will average 50 kilometers of driving data per day.

Mr Know It All’s projection does not include the increase in new cars per day. This could reach 2500 to 3000 cars per day by the end of 2021.

https://www.nextbigfuture.com/2021/06/p ... -2022.html

[weatheriscool]

New algorithm helps autonomous vehicles find themselves, summer or winter
https://techxplore.com/news/2021-06-alg ... inter.html
by California Institute of Technology

Without GPS, autonomous systems get lost easily. Now a new algorithm developed at Caltech allows autonomous systems to recognize where they are simply by looking at the terrain around them—and for the first time, the technology works regardless of seasonal changes to that terrain.

Details about the process were published on June 23 in the journal Science Robotics.

The general process, known as visual terrain-relative navigation (VTRN), was first developed in the 1960s. By comparing nearby terrain to high-resolution satellite images, autonomous systems can locate themselves.

The problem is that, in order for it to work, the current generation of VTRN requires that the terrain it is looking at closely matches the images in its database. Anything that alters or obscures the terrain, such as snow cover or fallen leaves, causes the images to not match up and fouls up the system. So, unless there is a database of the landscape images under every conceivable condition, VTRN systems can be easily confused.

[Time_Traveller]

Lip-Reading By AI Self-Driving Cars Is Either Alarming Or Ingenious

Jun 30, 2021

Spy movies used to include a pivotal moment involving a secret agent that would surreptitiously read the lips of someone seated at a table while behind a pane of glass such as in a distant restaurant or high-rise building. Upon cleverly deciphering the mouthing words of what the person was saying, the secret agent would finally know the truth about a crucial matter.

Such is the glory of being able to do lip-reading.

The ability to read lips is considered a vaunted skill and one that can be challenging to master. This capability to do speech reading is pretty tricky in a myriad of ways.

First, you need to be able to visually see the lips of the person that you are trying to decipher. That might seem like an obvious requirement, but the point is that sometimes the lips are not particularly easy to see. The person doing the speaking might have their hands up toward their mouth, partially covering their lips from view. You see this happen intentionally in baseball games when the catcher goes out to the mound to chat with the pitcher, and they both cover their mouths with their baseball gloves.

Besides trying to see the lips of a speaker, it really makes quite a difference if you can also see their entire mouth, their teeth, their jaw, their head, their neck, and so on. Turns out that the delivery mechanisms of speaking are apt to entail all of those anatomical elements. Lips alone are somewhat limiting in terms of discerning what someone is saying.

https://www.forbes.com/sites/lanceeliot ... ingenious/

[weatheriscool]

Autonomous excavators ready for around the clock real-world deployment

by Baidu
https://techxplore.com/news/2021-06-aut ... world.html

Researchers from Baidu Research Robotics and Auto-Driving Lab (RAL) and the University of Maryland, College Park, have introduced an autonomous excavator system (AES) that can perform material loading tasks for a long duration without any human intervention while offering performance closely equivalent to that of an experienced human operator.

AES is among the world's first uncrewed excavation systems to have been deployed in real-world scenarios and continuously operating for over 24 hours, bringing about industry-leading benefits in terms of enhanced safety and productivity.

The researchers described their methodology in a research paper published on June 30, 2021, in Science Robotics.

"This work presents an efficient, robust, and general autonomous system architecture that enables excavators of various sizes to perform material loading tasks in the real world autonomously," said Dr. Liangjun Zhang, corresponding author and the Head of Baidu Research Robotics and Auto-Driving Lab.

Excavators are vital for infrastructure construction, mining, and rescue applications. The global market size for excavators was $44.12 billion in 2018 and is expected to grow to $63.14 billion by 2026.

Given this projected market increase, construction companies worldwide are facing hiring shortages for skilled heavy machinery operators, particularly excavators. Additionally, COVID-19 continues to exacerbate the labor shortage crisis. Another contributing factor is the hazardous and toxic work environments that can impact the health and safety of on-site human operators, including cave-ins, ground collapses, or other excavation accidents that cause approximately 200 casualties per year in the U.S. alone.

[Yuli Ban]

Elon Musk Didn't Think Self-Driving Cars Would Be This Hard to Make

Tesla CEO Elon Musk said on Saturday that making a self-driving car was harder than he expected, after the company's timeline for its latest Full Self-Driving (FSD) software slipped again.

Self-driving was a "hard problem" that "requires solving a large part of real-world AI," Musk tweeted.

"Didn't expect it to be so hard, but the difficulty is obvious in retrospect," he said.

He also said that the latest version of Tesla's FSD beta software would be shipped soon. "I swear!" he added.

FSD, which doesn't make a Tesla car fully autonomous, currently costs a one-off $10,000, and was released in beta to some Tesla owners in October. It has all the features of Tesla's Autopilot — which brakes, accelerates, and steers automatically — plus it allows cars to change lanes, park themselves, and recognize stop signs and traffic lights.

[Yuli Ban]

[Yuli Ban]

SAIC Motor says it will put Robotaxis capable of L4 self-driving on roads in China this year

Chinese automotive design and manufacturing company SAIC Motor says it will launch 40 to 60 Robotaxis capable of Level 4 self-driving throughout Shanghai and Suzhou before the end of 2021. The bold statement was made by Zu Sijie, vice president and chief engineer of the company, during this year’s World Artificial Intelligence Conference.

SAIC Motor Corporation Ltd. is a Chinese state-owned automotive company based in Shanghai, and one of the largest automakers in the country. Electrek has previously covered SAIC-owned brands like MG Motor, but SAIC also produces vehicles in China through joint ventures with companies like Volkswagen and GM.

Now, SAIC is apparently expanding its automotive portfolio to include digital transportation in the form of autonomous Robotaxis. This news comes as a surprise from a more traditional automotive company, diving into a nascent space flooded with Chinese startups and major tech companies like Baidu and its Apollo Robotaxi.

In the US, startups like Cruise and Waymo paint a similar picture, while companies like Tesla carve their own unique path toward full self-driving.

[Yuli Ban]

Fully Self-Driving Cars Being Tested On Streets Of New York City

Fully self-driving cars are now being tested on New York City streets.

Two Ford Fusion models were specially customized by Mobileye, an Israeli firm owned by Intel.

The company received a special permit allowing its test drivers to keep their hands fully off the steering wheel while the cars drive themselves.

[Yuli Ban]

[weatheriscool]

Ford's vehicles to offer autonomous ride-hailing through Lyft
By Nick Lavars
July 22, 2021

Back in 2017, Ford made a significant investment in an artificial intelligence startup called Argo that harbored big ambitions in the self-driving vehicle space. The pair are now bringing Lyft into the fold in an effort to deploy autonomous ride-hailing on a commercial scale, with the first rides to take place later this year.

Ford teamed up with Argo AI hoping to combine its own expertise in vehicle manufacturing with the startup's specialist knowledge in robotics and artificial intelligence. The automaker sees Lyft and its network as the ideal means to deploy the self-driving vehicle technology it has developed in the interim, and its self-driving cars will start taking passengers on the Lyft network in Miami later this year.

“This collaboration marks the first time all the pieces of the autonomous vehicle puzzle have come together this way,“ says Lyft co-founder and CEO Logan Green. “Each company brings the scale, knowledge and capability in their area of expertise that is necessary to make autonomous ride-hailing a business reality.”

https://newatlas.com/automotive/argo-fo ... iling-yft/

[andmar74]

FSD V 9.1 is out. Looping behavior.



FSDBeta 9.1 - 2021.4.18.13 - Unprotected Left Turns with Drone View - Looping behavior