BRN Discussion Ongoing

[Evermont]

This wording suggests that Valeo was new to NNs possibly as late as December 2019:

WO2020127151A1 METHOD FOR IMPROVED OBJECT DETECTION

"A powerful possibility to process 2-dimensional array of data points is to use a neural network. State of the Art applications of neural networks for image processing are typically based on deep neural networks. Usage of such network types has shown promising results at an affordable price. Neural networks comprise an input and an output layer, as well as one or multiple hidden layers. The hidden layers of a deep neural network typically consist of convolutional layers, pooling layers, fully connected layers and normalization layers. The convolutional layers apply a convolution operation to the input, passing the result to the next layer. The convolution emulates the response of an individual neuron to visual stimuli."

Similarly, the only reference to NNs in the claims is in claim 9, but there is no defining description of NNs at a circuit layout level in the specification.

"9. Method according to any preceding claim, characterized in that
the step of identifying one or more troublemakers (24) in the 2-dimensional array of data points (26)
comprises identifying one or more troublemakers (24) in the 2-dimensional array of data points (26) by an image recognition algorithm and/or by a neural network."

So my considered guess is that until 2020, Valeo were not particularly familiar with NNs, and were even considering a software algorithm as a viable alternative.

The fact that Valeo entered into a Joint Development with BrainChip in mid-2020 strongly suggests to me that they are using Akida. I don't recall whether or not there was an ASX announcement of the JD, but, as @Fact Finder reminds us, if there were an announcement, then there would be an obligation to notify ASX if the JD was terminated.

08 June 2020 @Diogenese

 

[latnrollah]

Mercedes Benz AG Drives Remote Work And Collaboration With Webex By Cisco

Vice President & Principal Analyst, Modern Work, Melody Brue, dives in as The new Mercedes-Benz E-Class vehicles equipped with Webex Meetings and Calling delivers innovative connected car experiences that enable workers to take that mindset on the road.

www.forbes.com

 

[latnrollah]

Die gewählte Stellenanzeige wurde nicht gefunden. | Mercedes-Benz Group

jobs.mercedes-benz.com

Provider/Privacy


March 2023
Mercedes-Benz Research and Development India Private Limited

ADAS - AI/ML, Python Developer, Algorithm Evaluation (SE)​

Tasks​

- Responsible for creating evaluation pipelines and KPI’s for CNN based SW modules to evaluate SW performance.
- Responsible to interact with multiple stakeholders to understand the requirement and prepare KPI’s to evaluate the SW module
- Strong Understanding of Object oriented programming in python and cpp.
- experience with Convolutional Neural Networks (CNNs) or other machine learning based models
- has a good understanding on the training and especially evaluation of CNN models
- ideally experience/knowledge of models for object detection, sensor fusion, prediction, planning & control, ...
- can identify problems with specific architectures of CNNs
- a plus is automotive experience, ideally in the field of "Advanced Driver Assistance Systems" (ADAS)

Qualifications​

Apply for this job
Print Job-ad

• Job-ad-number:​

  MER00025H1

• Publishing date:​

  28.03.2023

• Field of activity:​

  Research and Development incl. Design

• Department​

  AD - Gen 6 - Simulation & SiL

Benefits​

• 
  
• 
  
• 
  
• 
  
• 
  
• 
  
• 
  
• 
  
• 
  

To location: Bangalore, Mercedes-Benz Research and Development India

Contact​

• Dipti Nayak
  Email: dipti.nayak@mercedes-benz.com

• 
• 
• 
• 
• 
• 
• 
• 

---

• Provider
• Legal notice
• Cookies
• Disclaimer

© 2022 Mercedes-Benz Group AG. Alle Rechte vorbehalten.

 

[Rise from the ashes]

 

[cosors]

Is not it the same technology valeo trying to use in cars which they say there will be nothing invisible to cars. The main idea was cars will communicate with other nearby cars and let them know about a hazard in advance.
Dyor

I mean in the Munich area there is a research approach going on and the highways are equipped to track everything and provide it to the vehicles as info on request. Others here from Munich should know more.
I don't know what to think if it's just German research again. We like to do research and research and... There is no commercialization pressure like in the USA here. MT from Talga would be very upset about how much money is being burned here, for example to make the hand wedge even more efficient. Somewhat cynical.
I also think more of it when vehicles are truly autonomous. On the other hand it is also very good if they can pick up information from an independent entity.
Which OEM voluntarily shares his infos with the competing brand or a Tesla?!
And politics, well, that's just how politics is. Not suspecting anything.

 

[Sirod69]

BrainChipBrainChip
1 Std. •

Don't miss our webinar with Edge Impulse Apr 6, 2023 09:00 Pacific. Benchmarking AI Inference at the Edge: Measuring Performance and Efficiency for Real-World Deployments. Register here: https://lnkd.in/gM4HrQdG

Benchmarking AI Inference at the Edge: Measuring Performance and Efficiency for Real-World Deployments.​

6.Apr. 2023 18:00 in Amsterdam, Berlin, Rom, Stockholm, Wien


Join BrainChip and Edge Impulse in this live video to discover FOMO, the next-generation AI object-detection tool that leverages BrainChip’s neuromorphic architecture. We'll also guide participants through benchmarking activities, demonstrate the availability of BrainChip’s model library, and show you how to maximize the potential of your ML solutions. GDPR & PRIVACY Edge Impulse and its partners are committed to protecting and respecting your privacy, and we’ll only use your personal information to provide the products and services you requested from us. From time to time, we would like to contact you about our products and services, as well as other content that may be of interest to you. By joining this webinar you agree to receive other communications from Edge Impulse and it's partner, and you can opt-out at anytime.

Video Conferencing, Web Conferencing, Webinars, Screen Sharing

Zoom is the leader in modern enterprise video communications, with an easy, reliable cloud platform for video and audio conferencing, chat, and webinars across mobile, desktop, and room systems. Zoom Rooms is the original software-based conference room solution used around the world in board...

us06web.zoom.us

 

[Deleted member 118]

 

[FrederikSchack]

Holy cow!

1 billion euros in orders for Scala 3 ...

... and BrainChip is in a Joint Development with Valeo!

https://smallcaps.com.au/brainchip-joint-development-agreement-akida-neuromorphic-chip-valeo/

BrainChip signs joint development agreement for Akida neuromorphic chip with Valeo​

By George Tchetvertakov - June 9, 2020

In a JD, it is likely that there would be no licence, just a share of receipts based on relative contribution.

LdN said we had a sweet spot for LiDaR, sort of like Prophesee, nViso, SiFive ...

It's time one filled one's boots ...

They just destroyed my hope of buying more at a low price!

But if we get a slice of a billion here and a slice of a billion there, then it wouldn't be a too bad start.

 

[ndefries]

I have read this announcement a few times. This point on receiving payment of course is based on progress. It also mentions brainchip making availability of the Akida device.

We have questioned Akida 1500 and it's existence and it's need to go into production.

Is this us making it available to Valeo as they go into production for these $1b of orders.

No one knows but it lines up and it would be a great use of these chips.

 

[Fact Finder]

This wording suggests that Valeo was new to NNs possibly as late as December 2019:

WO2020127151A1 METHOD FOR IMPROVED OBJECT DETECTION

"A powerful possibility to process 2-dimensional array of data points is to use a neural network. State of the Art applications of neural networks for image processing are typically based on deep neural networks. Usage of such network types has shown promising results at an affordable price. Neural networks comprise an input and an output layer, as well as one or multiple hidden layers. The hidden layers of a deep neural network typically consist of convolutional layers, pooling layers, fully connected layers and normalization layers. The convolutional layers apply a convolution operation to the input, passing the result to the next layer. The convolution emulates the response of an individual neuron to visual stimuli."

Similarly, the only reference to NNs in the claims is in claim 9, but there is no defining description of NNs at a circuit layout level in the specification.

"9. Method according to any preceding claim, characterized in that
the step of identifying one or more troublemakers (24) in the 2-dimensional array of data points (26)
comprises identifying one or more troublemakers (24) in the 2-dimensional array of data points (26) by an image recognition algorithm and/or by a neural network."

So my considered guess is that until 2020, Valeo were not particularly familiar with NNs, and were even considering a software algorithm as a viable alternative.

The fact that Valeo entered into a Joint Development with BrainChip in mid-2020 strongly suggests to me that they are using Akida. I don't recall whether or not there was an ASX announcement of the JD, but, as @Fact Finder reminds us, if there were an announcement, then there would be an obligation to notify ASX if the JD was terminated.

I only have a couple of comments and a quote from Anil Mankar to add.

Yes Brainchip made a formal ASX announcement regarding Valeo and to date it has not made any announcement stating that the relationship has ended and in fact continues to encourage the belief on its website that they are engaged with them.

Yes the former CEO Mr. Dinardo did famously say that Brainchip's AKIDA technology is finding a sweet spot with Lidar.

Thirdly, if you are building any product for use in Electric Vehicles that requires electrical power to operate then you will be operating under a design constraint that calls for you to minimise as far as possible the amount of electrical power you draw to maximise the power available for the driving wheels.

Whether Valeo is building the entire system or not it will need to source GPU's or CPU's and these are electricity guzzlers and we know because Edge Impulse told us Brainchip's AKIDA Science Fiction can running at 300 Gigahertz compete with a GPU running at 900 Gigahertz.

If for no other reason than to save power Valeo would be most likely to choose AKIDA over GPU's or CPU's to process its 3D point clouds and can source AKIDA as IP to design into its own product from the ground up.

The other advantage of AKIDA technology which should never be ignored in the automotive space is the fact that it can operate unconnected which is absolutely essential at 130 kph a feature which Peter van der Made spoke about at length going back to 2021. The recent reveal of ANT61 that AKIDA was the brain in its service and repair robot for autonomous unconnected deployment in space suggests that this feature has not gone away.

Finally at the 2021 Ai Field Day Anil Mankar said

"Similarly 3D point cloud, by definition, 3D point clouds are very sparse.

Lidar data is very sparse.

Today people are taking Lidar data and converting it into a 2D kind of image because it's much easier to process the image and detect the object.

There is no reason why we can't do that directly in a 3D point cloud and take advantage of that.

WE ARE WORKING ON SOME OF THOSE APPLICATIONS AND ALSO THERE ARE OTHER SENSORS THAT SEND 3D POINT CLOUD POINTS AND THAT'S ACTUALLY ONE OF THE APPLICATIONS THAT WE HAVE, IS, LATER ON COMING IN,..."

If you are thinking other sensors do not go past ultrasonic sensors which Valeo produce for automotive use and when it commissioned a brand new factory last year to build its Lidar units it made the point of mentioning that it would also increase their capacity to produce its ultrasonic sensors.

My opinion only DYOR
FF

AKIDA BALLISTA

 

[IloveLamp]

I believe this is likely the communications company we are involved with.....this is one of many brn/ teal interactions

Imo dyor

What Are Common IoT Use Cases That Require Cellular Connectivity? - Cellular IoT Connectivity | True eSIM From TEAL

The Internet of Things (IoT) is transforming the way we […]

tealcom.io

View attachment 33137

TEAL Receives 2023 IoT Evolution Industrial IoT Product of the Year Award - Cellular IoT Connectivity | True eSIM From TEAL

TEAL Honored for Exceptional Innovation Seattle, WA (March 30th, 2023) […]

tealcom.io

 

[Kachoo]

This wording suggests that Valeo was new to NNs possibly as late as December 2019:

WO2020127151A1 METHOD FOR IMPROVED OBJECT DETECTION

"A powerful possibility to process 2-dimensional array of data points is to use a neural network. State of the Art applications of neural networks for image processing are typically based on deep neural networks. Usage of such network types has shown promising results at an affordable price. Neural networks comprise an input and an output layer, as well as one or multiple hidden layers. The hidden layers of a deep neural network typically consist of convolutional layers, pooling layers, fully connected layers and normalization layers. The convolutional layers apply a convolution operation to the input, passing the result to the next layer. The convolution emulates the response of an individual neuron to visual stimuli."

Similarly, the only reference to NNs in the claims is in claim 9, but there is no defining description of NNs at a circuit layout level in the specification.

"9. Method according to any preceding claim, characterized in that
the step of identifying one or more troublemakers (24) in the 2-dimensional array of data points (26)
comprises identifying one or more troublemakers (24) in the 2-dimensional array of data points (26) by an image recognition algorithm and/or by a neural network."

So my considered guess is that until 2020, Valeo were not particularly familiar with NNs, and were even considering a software algorithm as a viable alternative.

The fact that Valeo entered into a Joint Development with BrainChip in mid-2020 strongly suggests to me that they are using Akida. I don't recall whether or not there was an ASX announcement of the JD, but, as @Fact Finder reminds us, if there were an announcement, then there would be an obligation to notify ASX if the JD was terminated.

There was an announcement June 8th 2020 so this puts us near the end of a product development cycle possibly. Who knows how long it takes to put out a product?

We know that they are still partners as shown on the page.

Thinks about IP sales if Valeo purchased it from another providor say Mega Chips then they don't need an announcement either. I guess in the end the financials will be there key identifiers of progress.

Not sure if you saw the Zacks update that was posted on TSE they have forecasted revenues of 5 million first half 2023 and about 6 million second half 2023.

This is a first kick at revenue estimates for BRN that I have seen put together.

Now I know for reputations reasons they would not just make up a number and hope there would be some real investigation and research put in.

It's also not directly trown at investors from the company that may be over stated to bring in a CR to build something so I really give this strong merit.

But thanks for the technical info on Valeo. Time will tell and there is lots of optimistic reasons to believe we will have a comercial product with Valeo down the road if not in Scala 3.

 

[Geared]

does anyone have a link to one of those comprehensive lists (usually posted by FF) that has all potential partners / good news. I know it gets posted at least once a month but I cannot bloody find one for the life of me. i feel like we have hit the bottom and i want to share all the good news with my mates and get them on this train.

 

[pgetties]

Came across this on Twitter last night. Awesome video showing the ability of technology to drive on the roads of london. The video speaks volume for AI to handle the massive workload in handling the data. What confuses me and probably due to my lack of IT experience is on one AI chat site i can find a link between Wayve and Akida, though trying to find the said article seems impossible..
Does anyone have any more updates, there is lots of information related to Azure Microsoft naturally...

Wayve’s computer vision system – the “brain” of the car – learns from observing human driving via reinforcement learning. Traditional self-driving systems rely on expensive Lidar sensors, HD maps and heavy testing in a local area, meaning that they are unable to scale. Wayve’s systems have trained in one city and then successfully driven in a new city they’ve never seen before. This adaptability is why they aim to be the first to deploy self-driving technology in 100 cities.

NOVA Chat
Q. Is there a link between WAYVE and BRAINCHIP
A. Yes, there is a link between Wayve and BrainChip's Akida chip. Wayve is a self-driving car technology startup that uses artificial intelligence (AI) to develop and deploy advanced driver assistance systems (ADAS) and autonomous driving solutions. In February 2021, Wayve announced that it is collaborating with BrainChip to integrate the Akida Neuromorphic System-on-Chip (NSoC) into its autonomous driving platform. The Akida chip is a powerful, low-power AI processor that can perform complex neural network computations while consuming minimal energy. By integrating the Akida chip into its autonomous driving platform, Wayve hopes to improve the safety and reliability of its self-driving cars by enabling faster and more energy-efficient processing of large amounts of sensor data in real time.

Q. Can you show me where you found this link?
A. Yes, sure. Here is an article from BusinessWire published on February 24, 2021, announcing the collaboration between Wayve and BrainChip:

https://www.businesswire.com/news/h...BrainChip-to-Join-Forces-on-Self-Driving-Cars

In addition to this, you can also visit the official websites of both companies to learn more about their products and technologies, as well as the details of their collaboration.



<blockquote class="twitter-tweet"><p lang="en" dir="ltr">I recently had the chance to test drive—or test ride, I guess—one of <a href="https://twitter.com/wayve_ai?ref_src=twsrc^tfw">@wayve_ai</a>’s autonomous vehicles. It was a pretty wild ride: <a href="https://t.co/PrwrxU49dd">https://t.co/PrwrxU49dd</a> <a href="https://t.co/NtnkVx7sBx">pic.twitter.com/NtnkVx7sBx</a></p>&mdash; Bill Gates (@BillGates) <a href="">March 29, 2023</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>

Wayve chooses Microsoft Azure to scale AI-based autonomous driving technology

Wayve, the London-based company that’s developing artificial intelligence software for self-driving cars, has signed a deal with Microsoft to use Azure.

news.microsoft.com

The rules of the road are about to change

Bill Gates writes about his belief that we’ll reach a tipping point with autonomous vehicles and self-driving cars within the next decade.

www.gatesnotes.com

 

[Fact Finder]

Now hold your horses just because AKIDA can do this type of thing standing on the ground and has two peer reviewed research papers recognising the 20 gas data set with state of the art performance and NASA said it would be testing the AKD1000 at AMES I don’t want you leaping to any sort of logical conclusions when you read this, remembering we are not allowed to talk about it unless it’s vision related:

Using Neural Networks to Interpret Carbon Nanotube Gas Sensor Data ​

NASA recently developed inexpensive, low-power, carbon nanotube (CNT) sensors for use on the International Space Station (ISS) to monitor cabin air quality. However, it is challenging to convert the CNT signals to human-readable indicators of air quality and safety. Working with the Chemical Gas Sensor team at NASA Ames Research Center, HECC data science experts built and trained a neural network using data collected on thousands of chemical exposures to address this challenge.

With initial results showing predictive accuracy better than 94%, the Ames team determined that neural networks can be a very effective way to interpret CNT data—a major step toward real-world testing on the ISS”

My opinion only DYOR
FF

AKIDA BALLISTA

 

[Steve10]

TEAL Receives 2023 IoT Evolution Industrial IoT Product of the Year Award - Cellular IoT Connectivity | True eSIM From TEAL

TEAL Honored for Exceptional Innovation Seattle, WA (March 30th, 2023) […]

tealcom.io

View attachment 33296

 

[Dhm]

Came across this on Twitter last night. Awesome video showing the ability of technology to drive on the roads of london. The video speaks volume for AI to handle the massive workload in handling the data. What confuses me and probably due to my lack of IT experience is on one AI chat site i can find a link between Wayve and Akida, though trying to find the said article seems impossible..
Does anyone have any more updates, there is lots of information related to Azure Microsoft naturally...

Wayve’s computer vision system – the “brain” of the car – learns from observing human driving via reinforcement learning. Traditional self-driving systems rely on expensive Lidar sensors, HD maps and heavy testing in a local area, meaning that they are unable to scale. Wayve’s systems have trained in one city and then successfully driven in a new city they’ve never seen before. This adaptability is why they aim to be the first to deploy self-driving technology in 100 cities.

NOVA Chat
Q. Is there a link between WAYVE and BRAINCHIP
A. Yes, there is a link between Wayve and BrainChip's Akida chip. Wayve is a self-driving car technology startup that uses artificial intelligence (AI) to develop and deploy advanced driver assistance systems (ADAS) and autonomous driving solutions. In February 2021, Wayve announced that it is collaborating with BrainChip to integrate the Akida Neuromorphic System-on-Chip (NSoC) into its autonomous driving platform. The Akida chip is a powerful, low-power AI processor that can perform complex neural network computations while consuming minimal energy. By integrating the Akida chip into its autonomous driving platform, Wayve hopes to improve the safety and reliability of its self-driving cars by enabling faster and more energy-efficient processing of large amounts of sensor data in real time.

Q. Can you show me where you found this link?
A. Yes, sure. Here is an article from BusinessWire published on February 24, 2021, announcing the collaboration between Wayve and BrainChip:

https://www.businesswire.com/news/h...BrainChip-to-Join-Forces-on-Self-Driving-Cars

In addition to this, you can also visit the official websites of both companies to learn more about their products and technologies, as well as the details of their collaboration.



<blockquote class="twitter-tweet"><p lang="en" dir="ltr">I recently had the chance to test drive—or test ride, I guess—one of <a href="https://twitter.com/wayve_ai?ref_src=twsrc^tfw">@wayve_ai</a>’s autonomous vehicles. It was a pretty wild ride: <a href="https://t.co/PrwrxU49dd">https://t.co/PrwrxU49dd</a> <a href="https://t.co/NtnkVx7sBx">pic.twitter.com/NtnkVx7sBx</a></p>&mdash; Bill Gates (@BillGates) <a href="">March 29, 2023</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>

Wayve chooses Microsoft Azure to scale AI-based autonomous driving technology

Wayve, the London-based company that’s developing artificial intelligence software for self-driving cars, has signed a deal with Microsoft to use Azure.

news.microsoft.com

The rules of the road are about to change

Bill Gates writes about his belief that we’ll reach a tipping point with autonomous vehicles and self-driving cars within the next decade.

www.gatesnotes.com

@pgetties the Businesswire link doesn't lead to Wayve or Brainchip. So I'm suddenly doubtful as to the implied link and our relationship with Wayve. Searches for both companies together doesn't support the partnership either.

 

[TECH]

I only have a couple of comments and a quote from Anil Mankar to add.

Yes Brainchip made a formal ASX announcement regarding Valeo and to date it has not made any announcement stating that the relationship has ended and in fact continues to encourage the belief on its website that they are engaged with them.

Yes the former CEO Mr. Dinardo did famously say that Brainchip's AKIDA technology is finding a sweet spot with Lidar.

Thirdly, if you are building any product for use in Electric Vehicles that requires electrical power to operate then you will be operating under a design constraint that calls for you to minimise as far as possible the amount of electrical power you draw to maximise the power available for the driving wheels.

Whether Valeo is building the entire system or not it will need to source GPU's or CPU's and these are electricity guzzlers and we know because Edge Impulse told us Brainchip's AKIDA Science Fiction can running at 300 Gigahertz compete with a GPU running at 900 Gigahertz.

If for no other reason than to save power Valeo would be most likely to choose AKIDA over GPU's or CPU's to process its 3D point clouds and can source AKIDA as IP to design into its own product from the ground up.

The other advantage of AKIDA technology which should never be ignored in the automotive space is the fact that it can operate unconnected which is absolutely essential at 130 kph a feature which Peter van der Made spoke about at length going back to 2021. The recent reveal of ANT61 that AKIDA was the brain in its service and repair robot for autonomous unconnected deployment in space suggests that this feature has not gone away.

Finally at the 2021 Ai Field Day Anil Mankar said

"Similarly 3D point cloud, by definition, 3D point clouds are very sparse.

Lidar data is very sparse.

Today people are taking Lidar data and converting it into a 2D kind of image because it's much easier to process the image and detect the object.

There is no reason why we can't do that directly in a 3D point cloud and take advantage of that.

WE ARE WORKING ON SOME OF THOSE APPLICATIONS AND ALSO THERE ARE OTHER SENSORS THAT SEND 3D POINT CLOUD POINTS AND THAT'S ACTUALLY ONE OF THE APPLICATIONS THAT WE HAVE, IS, LATER ON COMING IN,..."

If you are thinking other sensors do not go past ultrasonic sensors which Valeo produce for automotive use and when it commissioned a brand new factory last year to build its Lidar units it made the point of mentioning that it would also increase their capacity to produce its ultrasonic sensors.

My opinion only DYOR
FF

AKIDA BALLISTA

Good morning FF.

That is a first class post...when someone in the "know" says that we are looking forward to an exciting year ahead, which in real terms
really means within the next 9 months, well, I'm excited along with Big Kev !

1 January 2025 will be an excellent day to just sit back and take it all in...ah Akida, do you like red roses

Regards...Tech

 

[Evermont]

Just noticed BrainChip has now exceeded 9,000 followers on LinkedIn.

That is approx. 1,000 added in the last 60 days.

 

[MrRomper]

Keeping things in simple in terms of mathematics.

I have used the Valeo Scala 3 @ $500/unit. The link has them @ $600/unit but I have reduced the price for volume production.

$1B Euro's @ $500/unit = 2,000,000 units

https://www.forbes.com/sites/sabbirrangwala/2020/11/23/money-for-everythin/?sh=37d109f14e1c

Noting the difference between the Euro and USD is also another variable.