BRN Discussion Ongoing
- Thread starter TechGirl
- Start date
Stable Genius
Regular
The article is quite dated. May 2022. I recall there was some excitement at the time re Moschips but very little info since.
Good article by the way!
InterestingNow let revisit AI-generated routines soon to be in the new Mercedes.
View attachment 32906
"Just like a fine wine" the car's software will gets better over time.
So just think about all that's for a minute.
I do understand that's Mercedes is talking about software and update overtime to make improvements of it software to become intelligence. BUT, to run the software, hardware needs to be in the Mercedes; so we have the main CPU (Nvidia).
But remember this?
"BrainChip already is seeing some success. It’s Akida 1000 platform is being used in Mercedes-Benz’s Vision EQXX concept car for in-cabin AI, including driver and voice authentication, keyword spotting and contextual understanding."
Neuromorphic Computing Will Need Partners To Break Into The Datacenter
The emerging field of neuromorphic processing isn’t an easy one to navigate. There are major players in the field that are leveraging their size and amplewww.nextplatform.com
Brainchip's Akida technology is to make "sensor smarter". So with Mercedes AI-generated routines. Being implement in the new Mercedes. I do think AKIDA is processing driver and voice authentication, keyword spotting and contextual understanding (learning driver behaviours) at the sensor (process relevant data). Sent it to the main CPU (Nvidia) and the CPU sent it to the cloud (Data Center). Data center process the data and make improvements on the software to make the software more intelligent. Then software updates in the car. So that how it's 'gets better over time.
Just my intoxicated thought (DYOR)
-------
On another note. In the Next Platform article;
"BrainChip also is working with SiFive to integrate the Akida technology with SiFive’s RISC-V processors for edge AI computing workloads and MosChip, running its Akida IP with the vendor’s ASIC platform for smart edge devices."
Good old MosChip...
Learning 🏖
MAR 22, 2023
Robocar No Longer Drives Nvidia GTC
At Nvidia’s GPU Technology Conference (GTC) this week, the shiniest new object was ChatGPT. Autonomous driving is already an old story.What’s at stake:
For years, Nvidia has hung its hat on autonomous driving as the linchpin of its AI technologies. However, that revenue stream is waning, not because AV is a solved problem but because it is just too hard a problem to solve. What’s the next big AI application?
GTC, put together by Nvidia, is one of the world’s premier conferences dedicated to AI developers. Nvidia has used the venue to showcase its AI prowess built on GPU technology.
For several years, AI-enabled autonomous driving has highlighted every GTC. Nvidia presented its AI solutions — deployed in data centers for AI training and its multi-thousand teraflops SoC inside vehicles’ central brains doing AI inference.
However, it was evident in a pre-GTC briefing this week that Nvidia has begun singing a markedly different tune on fully automated driving.
Offering a more measured and somewhat skeptical view on autonomous driving – unprecedented for Nvidia – Ali Kani, vice president and general manager of the automotive team, called fully autonomous vehicles “the next decade’s problems.”
Given that “amazing companies with multibillion dollar investment, like Argo.ai, have gone under,” Kani said that it is “very reasonable” for OEMs to focus on making ADAS better and treating L4 as a long-term goal.
Further, he cautioned that autonomous driving “isn’t like AI development demo – as a kind of a fun little thing I can show.” People’s lives are at stake, and “there is no room for error,” he added.
“Robotaxi companies, trying to make it work in one city, are having challenges … and they are the best tech companies in the world,” he said. To make matters worse, “they’re trying to solve the problem worldwide in a passenger car … You can’t map that world every week. It’s impossible. What they must solve is an entirely different problem an order of magnitude harder.”
In other words, reality bites – even at GTC where AI always seems invincible.
ChatGPT, here it comes
Fast forward to the keynote speech Nvidia founder and CEO Jensen Huang delivered Tuesday.
Huang stressed, “the impressive capabilities of generative AI have created a sense of urgency” among companies eager to reimagine products and business models.
With computing advancing at “lightspeed,” Huang discussed a broad set of partnerships crucial to build an ecosystem for everything from training to deployment for cutting-edge AI services. These encompass Google, Microsoft, Oracle and a range of leading businesses that bring new AI, simulation and collaboration capabilities to every industry.
In other words, shelve autonomous driving, think ChatGPT.
Nvidia’s automotive business
Put simply, Nvidia is no longer counting on AVs to sell its automotive SoCs.
Instead, it has thrown an arsenal of technologies at automakers, urging them to use AI to support software development and plan new automotive plants.
The pitch at GTC to the automotive industry was Nvidia’s Omniverse platform and AI. Nvidia is now saying that automakers can leverage both Omniverse and AI to improve their “product lifecycle from concepts and styling, design and engineering, software and electronics, smart factories, autonomous driving and retail.”
According to announcements at GTC, Nvidia is working with the BMW Group – not on the vehicle front, but on vehicle production infrastructure. By building and operating industrial metaverse applications across BMW Group’s production network, Nvidia claims that it is helping BMW on an electric vehicle plant in Debrecen, Hungary, projected to open in 2025.
As for the central compute SoC, Nvidia’s automotive SoCs (such as Orin or Thor) won’t be powering BMW’s next-generation models. Instead, it will be Qualcomm’s automotive SoC platform.
Nonetheless, one big design win Nvidia announced at GTC this week is BYD. China’s leading EV company is building its mainstream vehicles on Nvidia’s Drive Orin SoC.
Other tier ones and OEMs are also adopting Nvidia’s platform for next-gen vehicle development, claimed Nvidia. They include DeepRoute, Pony.ai, Lenovo, Rimac Technology and smart. However, not every customer is embracing both hardware and software designed by Nvidia.
Nvidia said that carmakers are taking advantage of Nvidia’s flexible offerings. Nvidia’s platform is “open,” with parts of it available in “modules.” Some customers are building their own cockpit application software. Others are even developing their own autonomous driving stacks.
All said and done, Nvidia estimated that its automotive design win pipeline has increased to $14 billion over the next six years (2023 – 2028 inclusive). What matters to its automotive revenue isn’t just accumulating design wins inside vehicle models, but the support infrastructure Nvidia’s AI and Omniverse can extend to automakers.
Bottom line:
More automakers are backing off autonomous vehicles and opting for improved ADAS. In this environment, Nvidia’s automotive SoCs could prove to be overkill. Nvidia’s revised AI strategy for the auto industry appears to be cultivating an alternative path – supporting OEMs’ production and product lifecycles. It is hardly clear, however, if this will prove more lucrative than pursuing design wins for Nvidia’s automotive SoC design in every AV model.
DingoBorat
Slim
Yeah it's been posted before and straight off the bat, these knobs don't know anything about the technology, to be able to comment..
AKIDA1000 was "one size fits all"?
Not sure what their credentials are, but they're knobs..
Wags
Regular
Hoppy, Growing old is mandatory, Growing up is optional.........."with SP going down on many companies"
Thank you for your lovely thoughts there Mr Pope and even though I'll turn 66 years old later this year if I survive that long, there still lives within me a sniggering 12 year old. Is it just me or do some of us just never grow up?
Someone gave me a T-shirt with:
"When is this 'old enough to know better' thing supposed to kick in?'
Last edited:
jtardif999
Regular
Maybe we can deduce a few things with whether or not Akida IP lurks in the Renesas MCU based on the Cortex M85 with Helium:
1) The Helium tech performs vector extensions and together with the M85 acts in a similar way to SiFives X280 series chip.
2) Both the X280 and the M85 with Helium do better with a dedicated AI processor performing the inference.
3) The Renesas demo only mentions the M85, the Helium tech and the Plumeai model being used as the components for the demo. There is no mention of any AI hardware.
4) If Renesas were using their own AI hardware such as the DRP for example, they would have been eager to tell all about it - I would think there would no detail spared. Besides I’m not sure they could embed the DRP as IP with the M85 - they may only be able to utilise it as a self contained chip (like the AKD1000 for example).
5) The week before Renesas showcase AI working on the M85, ARM announce the addition of the custom instructions (ACI) to optimise data processing on the M85.
6) As has been already said, the coincidence of BrainChip announcing compatibility of the Akida product family IP with the M85 in the same week as the Renesas announcement.
As others have said nothing is proven, but gee there’s a bit of feel to it don’t you think? If this turns out to be true then we probably have the announcement we are looking for for some kind of a rerate in the share price. This would be amazing. AIMO.
It’ll be interesting to know that Mercedes will let it’s drivers easily take their experienced cockpit with them when they flip their cars.Now let revisit AI-generated routines soon to be in the new Mercedes.
View attachment 32906
"Just like a fine wine" the car's software will gets better over time.
So just think about all that's for a minute.
I do understand that's Mercedes is talking about software and update overtime to make improvements of it software to become intelligence. BUT, to run the software, hardware needs to be in the Mercedes; so we have the main CPU (Nvidia).
But remember this?
"BrainChip already is seeing some success. It’s Akida 1000 platform is being used in Mercedes-Benz’s Vision EQXX concept car for in-cabin AI, including driver and voice authentication, keyword spotting and contextual understanding."
Neuromorphic Computing Will Need Partners To Break Into The Datacenter
The emerging field of neuromorphic processing isn’t an easy one to navigate. There are major players in the field that are leveraging their size and ample
Brainchip's Akida technology is to make "sensor smarter". So with Mercedes AI-generated routines. Being implement in the new Mercedes. I do think AKIDA is processing driver and voice authentication, keyword spotting and contextual understanding (learning driver behaviours) at the sensor (process relevant data). Sent it to the main CPU (Nvidia) and the CPU sent it to the cloud (Data Center). Data center process the data and make improvements on the software to make the software more intelligent. Then software updates in the car. So that how it's 'gets better over time.
Just my intoxicated thought (DYOR)
-------
On another note. In the Next Platform article;
"BrainChip also is working with SiFive to integrate the Akida technology with SiFive’s RISC-V processors for edge AI computing workloads and MosChip, running its Akida IP with the vendor’s ASIC platform for smart edge devices."
Good old MosChip...
Learning 🏖
This also opens up the possibility of a subscription based model, where, for a fee, a driver can “own” his/her trained Akida cockpit and plug it into any other Mercedes they buy in the future…. Rather than having to retrain from scratch with every new car they buy!
cosors
👀
Here is the Renesas booth presentation in full length.
Here are more videos. A total of almost 20 hours of exhibition tours and interviews. Unfortunately, Brainchip is not featured. But maybe someone is bored or even discovers something interesting. There are exciting things in it even if I've only had time to skim through it so far. Maybe something for the thousand eyes.
Last edited:
No mention of Brainchip nor Valeo... is Merc going with Luminar and ditching Valeo..?Now let revisit AI-generated routines soon to be in the new Mercedes.
View attachment 32906
"Just like a fine wine" the car's software will gets better over time.
So just think about all that's for a minute.
I do understand that's Mercedes is talking about software and update overtime to make improvements of it software to become intelligence. BUT, to run the software, hardware needs to be in the Mercedes; so we have the main CPU (Nvidia).
But remember this?
"BrainChip already is seeing some success. It’s Akida 1000 platform is being used in Mercedes-Benz’s Vision EQXX concept car for in-cabin AI, including driver and voice authentication, keyword spotting and contextual understanding."
Neuromorphic Computing Will Need Partners To Break Into The Datacenter
The emerging field of neuromorphic processing isn’t an easy one to navigate. There are major players in the field that are leveraging their size and ample
Brainchip's Akida technology is to make "sensor smarter". So with Mercedes AI-generated routines. Being implement in the new Mercedes. I do think AKIDA is processing driver and voice authentication, keyword spotting and contextual understanding (learning driver behaviours) at the sensor (process relevant data). Sent it to the main CPU (Nvidia) and the CPU sent it to the cloud (Data Center). Data center process the data and make improvements on the software to make the software more intelligent. Then software updates in the car. So that how it's 'gets better over time.
Just my intoxicated thought (DYOR)
-------
On another note. In the Next Platform article;
"BrainChip also is working with SiFive to integrate the Akida technology with SiFive’s RISC-V processors for edge AI computing workloads and MosChip, running its Akida IP with the vendor’s ASIC platform for smart edge devices."
Good old MosChip...
Learning 🏖
Euks
Regular
Last edited:
Frangipani
Top 20
Now this U-turn by Elon Musk sounds rather intriguing… 
Looks as if he finally realized Tesla had manoeuvred itself into a fully autonomous driving dead end? Literally a blind alley?
https://www.handelsblatt.com/techni...en-tesla-setzt-wieder-auf-radar/29052020.html
I was just about to copy the article when this paywall thingy came up.
Any Handelsblatt subscribers that can help out?
Edit: Problem solved. See a couple of posts below for an English translation…
Looks as if he finally realized Tesla had manoeuvred itself into a fully autonomous driving dead end? Literally a blind alley?https://www.handelsblatt.com/techni...en-tesla-setzt-wieder-auf-radar/29052020.html
I was just about to copy the article when this paywall thingy came up.
Any Handelsblatt subscribers that can help out?
Edit: Problem solved. See a couple of posts below for an English translation…
Last edited:
Sirod69
bavarian girl ;-)
Himax Technologies, Inc. brought endpoint AI to the Arm booth at #EW23, showing:
A facial landmark demo, using Himax’s WiseEye solution
️ High precision 3D sensing for spatial realityWe’re excited to continue to work with our partners to build AIoT #onArm with our low power solutions.
LinkedIn Login, Sign in | LinkedIn
Login to LinkedIn to keep in touch with people you know, share ideas, and build your career.
Sirod69
bavarian girl ;-)
Tothemoon24
Top 20
Millimeter Wave Radar System Keeps Drivers Safe on the Road
This startup’s sensor checks vital signs to detect impaired driving
JOANNA GOODRICH23 MAR 2023
The WISe system measures the driver's vital signs including heart rate and breathing to detect fatigue and possible medical emergencies. The readings are displayed on the car's infotainment screen.
PONTOSENSE
A new sensor could help reduce the number of accidents caused by impaired driving and could protect children left in hot cars. The Wireless Intelligent Sensing millimeter-wave radar system, developed by startup Pontosense, monitors vehicle occupants’ vital signs, and it can detect the presence of passengers in the vehicle and where they are seated.
The WISe system measures the driver’s vital signs including heart rate and breathing to detect fatigue and possible medical emergencies. WISe sends out signals with wavelengths short enough to measure the tiny adjustments in a person’s body from breathing and pulse. The signal echo is then analyzed by an artificial intelligence (AI) algorithm.
“There is an urgent need for this kind of technology,” says cofounder Alex S. Qi, the startup’s CEO. Arecent study on the causes of motor vehicle accidents found that fatigue and medical emergencies were the top reasons.
WISe is the first in-market millimeter-wave wireless sensor used for that purpose in the automotive industry, Qi says. The system is expected to be installed in several vehicle models in the near future, he says.
Monitoring drivers to make roads safer
Current in-car driver-monitoring systems require either cameras or contact sensors, but WISe works wirelessly. The system uses RF sensing to capture echos of the micromovements made by the driver’s or passenger’s body caused by heartbeats and breaths. WISe detects changes in the reflected signal’s phase—the relationship between radio signals that share the same space and frequency—to read the micromovements and vital signs.“Think of the sensor as bouncing waves off of something, like how whales and bats use sonar and echolocation to gauge where objects are located around them. That’s the basis of our technology,” Qi says.
About Pontosense
FOUNDED2021
HEADQUARTERS
Toronto
EMPLOYEES
120
The sensor transmits millimeter waves from an antenna designed by cofounder and CTO Yihong Qi(Alex and Yihong are related). When the waves are reflected back to the sensor, the received echo allows WISe to “see” the small movements and gather data about the person’s health status.
The system checks heart rate variability and respiration rate. WISe takes the driver’s readings daily to discern what the person’s normal range is and displays the data on the car’s infotainment screen. The information is also stored in a microcomputer in the vehicle as well as external servers so it can be accessed later. The system is encrypted to protect the driver’s personal data from hackers.
The sensors can be installed in different locations in the vehicle, including behind the rearview mirror, behind the instrument panel, and between the driver and passenger seats. The device—about the size of a coin, with a 40-millimeter diameter—uses less than 10 watts of power.
The antenna Yihong designed is 10 nanometers to keep the sensor small and flexible, he says. It can transmit millimeter waves at 24, 60, or 77 gigahertz, depending on the vehicle.
“The wavelength at a higher frequency is very short,” Yihong says, “so the sensor is able to more accurately read the driver’s movements.”
The biggest challenge during the sensor’s development, Yihong says, was figuring out how to filter out external noise to ensure accurate readings.
“Take heart rate variability, for example,” he says. “In the hospital, when a patient is having their heart rate measured, sensors are directly attached to them for an accurate reading. But a contactless wireless sensor needs extra help because noises in the environment—such as seatbelts, the car’s engine, and even breathing—are affecting the sensor’s ability to differentiate between the person’s heart rate and irrelevant noise.”
To filter out the noise, Yihong developed signal processing software that uses AI to analyze the data. Algorithms clean up the data, generating a clear radar image for biometrics and communicating with the vehicle and the infotainment system if it suspects an issue with the driver or a passenger. The company says it takes 5 to 10 seconds for WISe to detect a change in the driver, such as an irregular heartbeat, and notify the vehicle.
Each car manufacturer will use a different system for alerting the driver if the measurements deviate from the norm, Alex says. It could involve setting off an alarm, slowing down the car or, for autonomous cars, safely pulling over to the side of the road.
The driver can override the system, the company says, if the alert is incorrect or unnecessary.
WISe can detect the presence of passengers in the vehicle and where they are seated.PONTOSENSE
Protecting against accidental deaths
The current way to count the number of passengers in a car is through pressure sensors installed under each seat. But sometimes a sensor is inaccurate because it was triggered by a heavy object such as luggage. Pressure sensors also can fail to count children weighing less than 29 kilograms, according to a 2020 article published in IEEE Access.Compared with a pressure sensor, WISe can more accurately differentiate whether a passenger is an adult, child, or pet through vital signs and size.
Determining a passenger’s size also matters for airbag deployment. After a crash, airbags eject out of the steering wheel, dashboard, or another location at about 27 kilometers per hour, according to the Insurance Institute for Highway Safety. People who weigh less than 68 kg or are shorter than 1.5 meters can be killed or severely injured because of the airbag’s deployment speed, Alex says. But WISe can communicate to the car to slow down the bag’s deployment by notifying the vehicle that a passenger doesn’t meet the minimum requirements.
The system can also warn whether children or pets have been left in the rear seat by reading their vitals. That could prevent heat stroke or death for those left behind in a hot vehicle. WISe can tell the vehicle to open windows or turn on an alarm to notify those in the area of a forgotten child or pet while the car is not running.
“In-cabin sensing will enable personalized, human-centered enhanced in-car experiences,” Yihong says.
Developing a hyperintelligent wireless sensor
The idea for Pontosense came about while Yihong and Alex were walking the showroom floor at the 2019 Consumer Electronics Show in Las Vegas.“We saw many companies showcasing devices that could measure vital signs, but all needed two or more sensors to work, were invasive, and were not user-friendly,” Alex says. “They also relied on Wi-Fi.”
Finding a Professional Home
Yihong has been an IEEE member for more than 30 years. He joined in 1992 and is a member of the IEEE Electromagnetic Compatibility Society.“I’ve benefited a lot from joining IEEE,” he says. “The organization has a large, global network of talented engineers.” Several of his IEEE colleagues helped him develop WISe, providing him with their expertise and referring him to IEEE journals they thought would help.
Yihong said he enjoys exchanging ideas with other members at IEEE conferences because he always learns something new.
“The sensors used low-frequency bandwidth, which couldn’t make accurate readings,” Yihong adds. “They only detected a person if they walked one meter out in front of the sensor, and they couldn’t differentiate between multiple people in a group.”
Yihong’s work in antenna development, radio frequency and electromagnetic compatibility measurement, and digital signal processing algorithms spans decades. An IEEE Fellow, he holds more than 500 patents in China and the United States.
He started to think about how to make a more accurate, more user-friendly, wireless intelligent sensing device. By the time the trade show ended, he already had several ideas. The two decided to team up and develop a technology to be used in cars.
“Both of us have experienced driving while tired, and we thought a sensor would be a great tool to help decrease the number of accidents caused by impaired driving,” Alex says. He adds that they also wanted to reduce the number of deaths caused when people accidently leave a young child behind in a car on a hot day.
They succeeded in creating the wireless sensor in 2021 and founded Pontosense that year to bring the system to market. The startup, based in Toronto, employs more than 120 people, 25 of whom are on its research and development team.
Pontosense received an IEEE Hyper-Intelligence Technical Committee Award for Excellence in Hyper-Intelligence last year for “contributions on wireless intelligent sensing systems for human safety in automobiles,” in the industrial impact category.
Hyperintelligence uses interdisciplinary technologies that work together to accomplish complex tasks. The award is sponsored by the IEEE Sensors Council.
Benefiting from other technology startups
Pontosense participated in VentureLab’s capital investment program in 2021. The program helps participants write pitches, connect with investors, and develop long-term capital strategies. But raising funds wasn’t difficult, Alex says, as Pontosense isn’t his or Yihong’s first company. They’ve founded several other startups, which helped pave the way for the development and manufacturing of the sensors. Their companies include wireless communication device manufacturers Mercku and General Test Systems.“We have the privilege to be able to mass-produce WISe with our network of companies,” Alex says. “Pontosense can produce hundreds of thousands of modules a month. Our goal for this year is to get the device in as many vehicles as possible so we can save as many lives as possible.”
Attachments
Frangipani
Top 20
I just worked it out somehow and then clicked on “translate“ for the non-German-speaking readers - amazing how translation engines have improved over the years!Now this U-turn by Elon Musk sounds rather intriguing…
https://www.handelsblatt.com/techni...en-tesla-setzt-wieder-auf-radar/29052020.html
I was just about to copy the article when this paywall thingy came up.
Any Handelsblatt subscribers that can help out?
New technology with sensors: Tesla relies on radar again
As the only car manufacturer, the US company has so far relied entirely on cameras for autonomous driving. A wrong path, as Elon Musk now also had to realize.
03/24/2023 by Felix Holtermann and Joachim Hofer
Tesla manufacturing © dpa
New York, Munich It doesn't work without radar. The car world is largely in agreement: Radar sensors are necessary for safe automated and autonomous driving. Only one did not want to admit this: Tesla founder Elon Musk. The CEO of the electric car pioneer has so far relied entirely on cameras.
Display
Now, according to experts, a spectacular U-turn is looming. The experts expect that Musk will soon start installing a self-developed, high-resolution radar system in his cars. This is indicated by documents that Tesla has submitted to the US supervision Federal Communications Commission (FCC), which monitors all radio wave-emitting devices in America, including radars.
The pivot is a sensitive topic for the most valuable car company in the world. Because as early as 2016, Musk promised buyers to develop a practical autopilot that will "enable your car to earn money for you if you don't use it." So a private Robotaxi. But he has not succeeded to do so to this day.
Now Tesla is taking a path that other manufacturers have long since taken with less full-bodied announcements. "All leading manufacturers use radar - only Tesla at the moment," says Klaus Schmitz, semiconductor specialist at the management consultancy Arthur D. Little. But that should change soon. The US company is in the process of bringing the so-called hardware 4 generation into its vehicles.
The radar system developed by Tesla itself is a modern, so-called 4D system. In addition to distance, speed and direction of movement, such solutions also recognize the height of objects. "In addition, such sensors also offer a significantly improved resolution," says Peter Fintl, chip expert at the consulting company Capgemini.
Guidehouse analyst Sam Abuelsamid, who has been observing Tesla for many years, has analyzed the documents submitted to the FCC. "The new solution is significantly more powerful than previous systems," he tells the Handelsblatt. Thus, there are more radar antennas, which allows a significantly higher resolution using special software.
Among other things, this could solve Tesla's problem with so-called phantom braking due to misinterpreted camera data. For example, the old system sometimes interprets photos of people on truck tarpaulins as real pedestrians. "This does not happen to a radar," says Abuelsamid. Tesla is thus closing out to other manufacturers.
Tesla initially did not respond to a Handelsblatt request for the new sensor.
The camera recordings that Tesla has used so far are evaluated and interpreted by artificial intelligence (AI) - analogous to a human driver who can only rely on his eyes. Musk has so far rejected the installation of additional systems: "In my opinion, this is a crutch," he had said in the past.
Tesla in court for fatal accidents
Tesla's previous autopilot is most likely to correspond to a so-called Level 2 system, i.e. a solution that only supports the driver (Level 5 corresponds to fully autonomous driving). Which does not prevent Tesla from marketing the autopilot as a "Full Self-Driving Beta" (FSD Beta).
This is a problem for supervision: In August 2021, the US Highway Traffic Safety Administration (NHTSA) initiated an investigation of the autopilot in 765,000 Tesla vehicles after about a dozen accidents. This spring, the company is on trial in the first jury trials for fatal accidents for which the autopilot is blamed.
Experts consider the installation of radar sensors for autopilot applications to be indispensable. "In the industry, it is undisputed that due to the technical limitations of purely camera-based systems, a supplement by further senses - i.e. sensors - makes sense," emphasizes consultant Fintl. Especially at higher speeds, for example on the highway, "radars are practically indispensable due to their long range".
Tesla has a zigzag course behind it in autonomous driving: At the introduction of the autopilot in 2015, the hardware of the supplier Mobileye was used, which included cameras, radar and ultrasonic sensors. After a fatal accident in 2016, the Israeli company withdrew from the partnership on the grounds that Tesla had abused the system.
With the autopilot version 2 from October 2016, Tesla introduced its own system, which included eight cameras, a long-range radar and twelve low-resolution ultrasonic sensors, which are mainly used for parking. At the beginning of 2021, the company removed the radar sensor, then also the ultrasonic devices in 2022. To date, some parking functions no longer work.
Tesla was able to save money through various technologies
Tesla officially justified the removal of the radar by the fact that it did not really help. However, it can be assumed that the chip shortage played a significant role," says analyst Abuelsamid. By cancelling, Tesla saved around 40 dollars per vehicle - a lot of money in the narrowly calculated car industry. The resolution of the radar was also only low. By abolishing the ultrasonic devices, Tesla has saved another around 100 dollars per vehicle, according to the analyst's bill.
However, the new technology should not help existing customers who have paid up to $15,000 for the FSD beta package. A retrofitting of the radar sensor and the necessary on-board computer makes "neither technical nor financial sense" for Tesla, says Abuelsamid.
For many owners, this should be annoying in view of the full-bodied announcements. "The vehicles sold so far will not perform nearly as well in autopilot as those with the new hardware. However, Musk has told all buyers from 2016 that they have all the necessary hardware for autonomous driving at level 5," says the analyst. Now Musk can't deliver what he promised. "Tesla has maneuvered herself into a real dead end."
First publication: 03/24/2023, 04:00 a.m.
Sirod69
bavarian girl ;-)
AUTONOMOUS DRIVINGNow this U-turn by Elon Musk sounds rather intriguing…
https://www.handelsblatt.com/techni...en-tesla-setzt-wieder-auf-radar/29052020.html
I was just about to copy the article when this paywall thingy came up.
Any Handelsblatt subscribers that can help out?
Tesla is back on radar
The US company is the only car manufacturer to have relied entirely on cameras for autonomous driving. A mistake, as Elon Musk now had to see.
New York, Munich It doesn't work without radar. The automotive world is largely in agreement: radar sensors are necessary for safe, automated and autonomous driving. Only one person didn't want to believe that: Tesla founder Elon Musk. The CEO of the electric car pioneer has so far relied entirely on cameras.
According to experts, a spectacular turnaround is now in the offing. Experts expect that Musk will soon start installing a high-resolution radar system he developed himself in his cars. This is indicated by documents that Tesla has submitted to the US Federal Communications Commission (FCC), which monitors all radio wave-emitting devices in America, including radar devices.
The swing is a sensitive issue for the most valuable car company in the world. Back in 2016, Musk promised buyers he would develop a viable autopilot that will "allow your car to make money for you when you're not using it." A private robotaxi, then. But to this day he has not succeeded.
Autonomous driving: Tesla follows the competition
Now Tesla is taking a path that other manufacturers have long been taking with less full-bodied announcements. "All leading manufacturers use radar - only Tesla doesn't currently," says Klaus Schmitz, semiconductor specialist at the management consultancy Arthur D. Little. But that is likely to change soon. The US group is in the process of bringing the so-called Hardware 4 generation into its vehicles.
TOPICS OF THE ARTICLE
Tesla Autonomous Driving Electromobility Automotive Industry Robot Elon Musk
The radar system developed by Tesla itself is a modern, so-called 4D system. In addition to distance, speed and direction of movement, such solutions also recognize the height of objects. "In addition, such sensors also offer a significantly improved resolution," says Peter Fintl, chip expert at the consulting firm Capgemini.
Guidehouse analyst Sam Abuelsamid, who has been watching Tesla for many years, analyzed the filings filed with the FCC. "The new solution is significantly more powerful than previous systems," he tells Handelsblatt. There are more radar antennas, which allows a significantly higher resolution using special software.
Among other things, this could solve Tesla's problem with so-called phantom braking due to incorrectly interpreted camera data. For example, the old system sometimes interprets photos of people on truck tarpaulins as real pedestrians. "That doesn't happen with a radar," says Abuelsamid. "Tesla is catching up with other manufacturers."
Tesla initially did not respond to a Handelsblatt inquiry about the new sensor.
The camera recordings that Tesla has been using so far are evaluated and interpreted by artificial intelligence (AI) - analogous to a human driver who can only rely on his eyes. Musk has so far refused to install additional systems: "In my opinion, that's a crutch," he had said in the past.
Tesla in court over fatal accidents
Tesla's previous autopilot most closely corresponds to a so-called Level 2 system, i.e. a solution that only supports the driver (Level 5 corresponds to fully autonomous driving). Which doesn't stop Tesla from marketing the autopilot as a "Full Self-Driving Beta" (FSD Beta).
This is a problem for supervisors: In August 2021, the US National Highway Traffic Safety Administration (NHTSA) launched an investigation into the autopilot in 765,000 Tesla vehicles after around a dozen accidents. This spring, the company faces its first jury trial in fatal accidents blamed on autopilot.
Experts consider the installation of radar sensors essential for autopilot applications. "In the industry, it is undisputed that due to the technical limitations of purely camera-based systems, it makes sense to add other senses - i.e. sensors," emphasizes consultant Fintl. Especially at higher speeds, such as on the freeway, "radars are practically indispensable due to their long range".
Tesla has taken a zigzag course in autonomous driving: When it introduced Autopilot in 2015, it used hardware from supplier Mobileye, which included cameras, radar and ultrasonic sensors. After a fatal accident in 2016, the Israeli company withdrew from the partnership on the grounds that Tesla had abused the system.
With Autopilot version 2 from October 2016, Tesla introduced its own system that included eight cameras, a long-range radar and twelve low-resolution ultrasonic sensors, which are mainly used when parking. The company removed the radar sensor in early 2021, followed by the ultrasonic devices in 2022. To this day, some parking functions no longer work.
Tesla was able to save money through various technologies
"Tesla officially justified the removal of the radar with the fact that it doesn't really help. However, it can be assumed that the lack of chips played a significant role,” says analyst Abuelsamid. The deletion saved Tesla around $40 per vehicle - a lot of money in the tightly calculating auto industry. The resolution of the radar was also only low. By getting rid of the ultrasound devices, Tesla saved another roughly $100 per vehicle, according to the analyst's calculation.
However, existing customers who have paid up to $15,000 for the FSD beta package are unlikely to benefit from the new technology. Retrofitting the radar sensor and the necessary on-board computer makes "neither technical nor financial sense" for Tesla, says Abuelsamid.
For many owners, this should be annoying given the full-bodied announcements. “The vehicles sold to date will not perform nearly as well on autopilot as those with the new hardware. But Musk has told all buyers since 2016 that they have all the necessary hardware for Level 5 autonomous driving,” the analyst said. Now Musk cannot deliver what he promised. "Tesla has maneuvered itself into a real dead end."
Similar threads
