Renesas

butcherano

Regular
Hi @butcherano ,

Just between you and me, I think there is a possibility that Akida could be used in these sensors, but keep it under your hat for now.

This Renesas patent has an early priority date of 20191323, and a second priority date of 20201215. That may indicate that some new tech was included in the second priority document. It would be nice to find out what new tech was disclosed in the second document.

The specification does discuss NNs in a general manner, but does not disclose the nitty-gritty of how to make a NN circuit, so it would be interesting to know where they got their NN from, n'est-ce pas? I suppose we could count all the available NN SoC IPs available at the time.

They have also thrown in a reference to analog NNs to cover themselves.

It's also interesting to compare the patent drawings with the one in your abstract.

View attachment 10494


US2021190750A1 System and Method to Avoid the Influence of Ozone for a Gas Sensor

Priority: 20191223; 20201215.
Note 2 priority dates indicates new material was added in filing the complete application US202017122761A.

View attachment 10492


View attachment 10493

As illustrated in FIG. 1C, AI processing may be a neural network 142 . The main components of neural network 142 are input nodes 148 that receives the data from pre-processing 140 , nodes in hidden layers 146 where data is processed by sequentially applying weighted functions to process data and pass to the next layer. Finally, output nodes 150 provided the results. The weighted parameters of each of the functions executed in each of the nodes can be set by a training algorithm that rely on a training data set. Memory 124 is large enough to store the weighting parameters for operation of neural network 142 . In some cases, neural network 142 may be, at least partially, trained in a supervised learning method where gas sensor 100 processes a set of known gas samples to set the weighting parameters. In either case, neural network 142 may provide much more reliable results for the gas dependent results output by gas sensor 100 .

[0041] Neural network 142 can be implemented digitally in processor 122 or may be implemented in analog circuits in processor 122 . In some embodiments the weighting factors to each node may be digitized and the nodes of neural network 142 may be implemented by analog circuitry. Whether processing in each node is performed digitally or by analog circuits, the weighting factors are stored in memory 124 and coupled to each of the nodes of input nodes 148 , hidden layers 146 , and output nodes
150 .


As I said, don't tell the others - we don't want to start a riot - at least til I've got some more at these prices.
Just one more piece of ammunition for Akida being embedded on the Renesas ZMOD gas sensors is the improvement that they’ve made to the power consumption over the last couple of years.

From the 1st generation back in 2020 to the 2nd generation available now, the power consumption has improved by 3 orders of magnitude from 1.5mW to only 160uw.

Akida would definitely be capable of achieving this. I can’t see any other changes that might account for this massive improvement.

1657343527304.jpeg




1657343280887.jpeg



June 2020 datasheet: https://docs.rs-online.com/e05c/A700000007499355.pdf

Dec 2021 datasheet: https://www.renesas.com/us/en/document/dst/zmod4410-datasheet
 
Last edited:
  • Like
  • Fire
  • Love
Reactions: 35 users

Diogenese

Top 20
Just one more piece of ammunition for Akida being embedded on the Renesas ZMOD gas sensors is the improvement that they’ve made to the power consumption over the last couple of years.

From the 1st generation back in 2020 to the 2nd generation available now, the power consumption has improved by 3 orders of magnitude from 1.5mW to only 160uw.

Akida would definitely be capable of achieving this. But I can’t see any other changes that might account for this massive improvement.

View attachment 10985



View attachment 10984


June 2020 datasheet: https://docs.rs-online.com/e05c/A700000007499355.pdf

Dec 2021 datasheet: https://www.renesas.com/us/en/document/dst/zmod4410-datasheet


2022 ................................................................................................................................. 2020

1657344038175.png
1657344085883.png


Alternate algorithms and firmware implementations ....

With the ultra-low operating current ...

... an excellent choice for low-voltage and ultra-low power ...

Not to mention the NVM configuration and calibration store.

Akida could be said to provide "alternate .... firmware configurations" by reconfiguration of the 8 NPUs in 2 nodes.

Akida's activation-only sparsity provides ultra-low operating current.
 
  • Like
  • Love
  • Fire
Reactions: 27 users

Makeme 2020

Regular


VOLUME NO. 0205/16

EdgeCortix Collaborates with Renesas to Deliver Enhanced Feature-Rich Compiler for the Renesas DRP-AI AI-Accelerator​

PR Newswire
12 Jul 2022, 23:05 GMT+10


TOKYO, July 12, 2022 /PRNewswire/ -- EdgeCortix Inc., an innovative fabless semiconductor design company with a software first approach, focused on delivering class-leading compute efficiency and latency for edge artificial intelligence (AI) inference, announced today a collaboration with Renesas Electronics Corporation (Renesas).




(PRNewsfoto/Edgecortix, Inc.)




Through this collaboration, EdgeCortix has taken its industry leading heterogeneous platform-based compiler framework MERA and developed a new compiler, DRP-AI* TVM for Renesas' DRP-AI1 accelerator. The new compiler is available with associated software and tools and works in combination with Renesas' DRP-AI tools.
"We are eager to apply the power of our MERA compiler across many heterogeneous environments, including leading FPGA boards, EdgeCortix's own custom AI-Inference ASIC and today, integration with Renesas' DRP-AI," said Sakyasingha Dasgupta, Founder and CEO of EdgeCortix. "We are very pleased that Renesas has realized the value, utility, and performance that our MERA solution offers in developing the compiler for DRP-AI. By applying EdgeCortix's MERA compiler technology to DRP-AI TVM, this combination will create significant business opportunities and value for both Renesas and their end customers in four key functional areas. Namely, Expanded Model Support, ML Framework Expansion, Support for Floating-point 16 and overall Performance Enhancements."
EdgeCortix Delivered Key Business Outcomes via MERA collaboration with Renesas DRP-AI Tool:

  • Expanded Model Support: More robust AI model support (20+ models explored) with significantly enhanced flexibility and end-user ease-of-use enhancement.
  • ML Framework expansion: Future proofing the DRP-AI product utility, by adding PyTorch support and making the ONNX support more robust. TensorFlow support to be added in ongoing future work.
  • Support for Floating-point 16 bit to MERA and extended OSS tool Apache TVM. Lower-precision support being added in ongoing future work.
  • Performance Enhancements: Improved performance especially for models with operators shared between host CPU and DRP-AI (a new feature added with this integration work).
"We recognized immediately the value of adding the MERA compiler and associated tool set to the RZ/V MPU series, as we expect many of our customers to implement application software including AI technology," said Shigeki Kato, Vice President, Enterprise Infrastructure Business Division at Renesas. "As we drive innovation to meet our customer's needs, we are collaborating with EdgeCortix to rapidly provide our customers with robust, high-performance and flexible AI-inference solutions. The EdgeCortix team has been terrific, and we are excited by the future opportunities and possibilities for this ongoing relationship."
About EdgeCortix Inc.
EdgeCortix is a fabless semiconductor design company focused on enabling energy-efficient edge intelligence. It was founded in 2019 with the radical idea of taking a software first approach, while designing an artificial intelligence specific runtime reconfigurable processor from the ground up using a technique called "hardware & software co-exploration". Targeting advanced computer vision applications first, using proprietary hardware and software IP on existing processors like FPGAs and custom designed ASIC, the company is geared towards positively disrupting the rapidly growing AI hardware space across defense, aerospace, smart cities, industry 4.0, autonomous vehicles and robotics.
For more details, contact: info@edgecortix.com
Copyright 2022 EdgeCortix, Inc. All rights reserved worldwide. EdgeCortix, the EdgeCortix logo and Dynamic Neural Accelerator are trademarks or registered trademarks of EdgeCortix, Inc. (or its group companies) in the US and other countries. EdgeCortix Inc. is a wholly owned subsidiary of EdgeCortix Pte. Ltd.
1 Dynamically Reconfigurable Processor for AI
 
  • Like
Reactions: 4 users

stuart888

Regular
A new Renesas Soc Board with a Spiking Voice Activation, that goes near zero power when not in use? More of the 1000 dots!

Darn cool video too from start to finish, as it shows Renesas comes prepared to show off its product, in a fully functional testing state, and ready to set up follow-up meetings. I am impressed with this Renesas demo. This partnership is going to be strong, and they are perfect for us, to be the sales lead, a marketing engine.


1657658921958.png


 
  • Like
  • Fire
Reactions: 18 users

stuart888

Regular
Our customer/partner Renesas has an Image Preprocessing for AI Inference session on Friday. There is that word Inference again! So good their marketing AI, and now on behalf of Brainchip. Plus, I listened to Megachips Episode 16 again today. Megachips is focused on consumer markets, the things that get to sales shelf quicker. Renesas and Megachips, both worldly, different markets, different product targets to a large degree. Great job to the Brainchip Team.

https://www.doulos.com/partner-webinars/image-preprocessing-for-ai-inference

1657660115440.png
 
  • Like
  • Fire
Reactions: 19 users

stuart888

Regular
Last thought on Renesas. It is great that they also implement more power-hungry AI cloud-based solutions too. When future customers work with them, they open up the menu and see a new section: there is AI on a Power Diet Section on the Menu! With Brianchip on the menu, they might want the ingredients of: Low Power, Low Cost, and Matches Patterns of Any Kind (voice, video, vibration, etc). PVDM, show Akida a pattern, we can match it!
 
  • Like
  • Love
  • Fire
Reactions: 8 users
A must watch imo



one interesting bit 100 Million plus mcu's to be sold. Not bad. Haven't seen the lot yet. Thought i would share.

SC
 
  • Like
  • Fire
  • Love
Reactions: 13 users
A must watch imo



one interesting bit 100 Million plus mcu's to be sold. Not bad. Haven't seen the lot yet. Thought i would share.

SC

In a place that shall not be named, the Valdemort of forums I stated or guessed really that imo the first Renesas application would be in factories. At the end of this video they mention just that as a main application of their tech in predictive maintenance in industrial.

SC
 
  • Like
  • Fire
Reactions: 6 users

Iseki

Regular
A new Renesas Soc Board with a Spiking Voice Activation, that goes near zero power when not in use? More of the 1000 dots!

Darn cool video too from start to finish, as it shows Renesas comes prepared to show off its product, in a fully functional testing state, and ready to set up follow-up meetings. I am impressed with this Renesas demo. This partnership is going to be strong, and they are perfect for us, to be the sales lead, a marketing engine.


View attachment 11172


Akida can do so much more than activate on "voice energy". It can activate on specific key words. This is much more powerful.

If you're in a car where there are many different types of sensors, each of these sensors can be monitored for set patterns eg vibrational noise, driver behavior, object detection in cameras etc, and all the responses can be managed by sending just the necessary data to classically trained AI models when required, saving vast amounts of power.

So in reality Akida will come to save its power hungry numerical processing AI competitors by allowing them to operate in a strategic, economical way. It's a complete win-win.

As soon as one of the major players eg NVIDIA or ARM can incorporate the Akida IP on it's AI boards/chips to do just this, the flood gates will open for our world changing technology.
 
  • Like
  • Love
  • Fire
Reactions: 16 users
Renesas have just acquired Reality AI


"Headquartered in Columbia, Maryland, U.S., Reality AI offers a wide range of embedded AI and Tiny Machine Learning (TinyML) solutions for advanced non-visual sensing in automotive, industrial and commercial products. Combining Reality AI’s best-in-class AI inference technologies with Renesas’ extensive MCU and MPU product portfolios will enable seamless implementation of machine learning and signal processing. The acquisition will allow Renesas to expand its tool suite and software offerings for AI applications and increase its in-house capability to provide highly optimized endpoint solutions that combine both hardware and software."

I think someone mentioned the other day Brainchip could be involved with Fujitsu. Fujitsu happens to be a partner of Reality AI, and I guess Brainchip could be working closely with Reality AI through Renesas, so the connections could be there.
 
  • Like
  • Thinking
  • Fire
Reactions: 16 users
D

Deleted member 118

Guest
Renesas have just acquired Reality AI


"Headquartered in Columbia, Maryland, U.S., Reality AI offers a wide range of embedded AI and Tiny Machine Learning (TinyML) solutions for advanced non-visual sensing in automotive, industrial and commercial products. Combining Reality AI’s best-in-class AI inference technologies with Renesas’ extensive MCU and MPU product portfolios will enable seamless implementation of machine learning and signal processing. The acquisition will allow Renesas to expand its tool suite and software offerings for AI applications and increase its in-house capability to provide highly optimized endpoint solutions that combine both hardware and software."

I think someone mentioned the other day Brainchip could be involved with Fujitsu. Fujitsu happens to be a partner of Reality AI, and I guess Brainchip could be working closely with Reality AI through Renesas, so the connections could be there.

Cool, as with them adding a software company for other devices should then make brainchip more $$$
 
  • Like
Reactions: 7 users
Cool, as with them adding a software company for other devices should then make brainchip more $$$
The Ecosystem has grown just that little bit more!
 
  • Like
  • Love
Reactions: 6 users
D

Deleted member 118

Guest
 
  • Like
  • Fire
Reactions: 5 users
Not sure if this has been covered before and I’m keen to here @Diogenese's thoughts on this…but is it possible that the 2nd generation gas sensors which are available now from Renesas have Akida on board? And is this where our first revenue from Renesas will come from?

These are the ZMOD4410 indoor air quality sensor and ZMOD4510 outdoor air quality sensors. With the ZMOD4450 refrigeration air quality sensor to follow shortly once they release the 2nd generation.

Key catch phrases:
  • “Embedded” AI (ie. on the sensor and not on the user’s MCU) – Anil has been saying that this is where Akida belongs all along
  • “Leverages neural network training”
  • The sensors have a CMOS signal conditioning IC (and Akida uses pure CMOS logic process)
  • Ultra-low power (only 160 micro watts, which rules out the Renesas in-house DRP-AI and also rules out software AI)
  • “This sensor can be run for months or years on AA batteries”…where have we heard that before?!
The global gas sensor market was US$2.5B in 2021. These sensors sell for around A$5 each. So even if Renesas only have 10% of this market that equates to 70mil sensors per year. That alone justifies our current market cap with plenty of blue sky left.

A few links and snippets of info below…


Environmental Sensing Solutions brochure

https://www.renesas.com/us/en/document/bro/environmental-sensing-solutions

View attachment 10471
View attachment 10472


Blog post from January 2022

https://www.renesas.com/us/en/blogs/breath-fresh-air-battery-powered-air-quality-monitoring


View attachment 10473


ZMOD4410 Indoor Air Quality Sensor Application Note

https://www.renesas.com/us/en/document/apn/zmod4410-application-note-tvoc-sensing

View attachment 10474


ZMOD4510 Outdoor Air Quality Sensor Overview


https://www.renesas.com/us/en/produ...-outdoor-air-quality-sensor-platform#overview

View attachment 10475


ZMOD4510 Outdoor Air Quality Sensor Datasheet

https://www.renesas.com/us/en/document/dst/zmod4510-datasheet


View attachment 10476
Love your work @butcherano, if this isn't AKIDA, I'd like to know who the hell it is, cause it looks like good sheet! 😛
 
  • Like
  • Fire
  • Love
Reactions: 11 users

uiux

Regular
Not sure if this has been covered before and I’m keen to here @Diogenese's thoughts on this…but is it possible that the 2nd generation gas sensors which are available now from Renesas have Akida on board? And is this where our first revenue from Renesas will come from?

These are the ZMOD4410 indoor air quality sensor and ZMOD4510 outdoor air quality sensors. With the ZMOD4450 refrigeration air quality sensor to follow shortly once they release the 2nd generation.

Key catch phrases:
  • “Embedded” AI (ie. on the sensor and not on the user’s MCU) – Anil has been saying that this is where Akida belongs all along
  • “Leverages neural network training”
  • The sensors have a CMOS signal conditioning IC (and Akida uses pure CMOS logic process)
  • Ultra-low power (only 160 micro watts, which rules out the Renesas in-house DRP-AI and also rules out software AI)
  • “This sensor can be run for months or years on AA batteries”…where have we heard that before?!
The global gas sensor market was US$2.5B in 2021. These sensors sell for around A$5 each. So even if Renesas only have 10% of this market that equates to 70mil sensors per year. That alone justifies our current market cap with plenty of blue sky left.

A few links and snippets of info below…


Environmental Sensing Solutions brochure

https://www.renesas.com/us/en/document/bro/environmental-sensing-solutions

View attachment 10471
View attachment 10472


Blog post from January 2022

https://www.renesas.com/us/en/blogs/breath-fresh-air-battery-powered-air-quality-monitoring


View attachment 10473


ZMOD4410 Indoor Air Quality Sensor Application Note

https://www.renesas.com/us/en/document/apn/zmod4410-application-note-tvoc-sensing

View attachment 10474


ZMOD4510 Outdoor Air Quality Sensor Overview


https://www.renesas.com/us/en/produ...-outdoor-air-quality-sensor-platform#overview

View attachment 10475


ZMOD4510 Outdoor Air Quality Sensor Datasheet

https://www.renesas.com/us/en/document/dst/zmod4510-datasheet








View attachment 10476

@butcherano @Diogenese

I found this in the datasheet from digikey:


Screenshot_20220814-195137.png


Check out the date of the revision ;)
 
  • Like
  • Fire
Reactions: 16 users

stuart888

Regular
  • Like
  • Love
  • Fire
Reactions: 17 users

uiux

Regular

Renesas Said to Explore Sale of Military ‘Rad Hard’ Business
  • Radiation-resistant chip business sale could fetch $1 billion
  • Tokyo-based Renesas has been focusing on supplying automakers
 
  • Like
  • Wow
  • Fire
Reactions: 12 users

Cardpro

Regular
Although I posted on the main one, below is an extract from the recent Job Ad from Renesas (AI Application Engineer).

【Role and Responsibility】

 ・research latest AI technologies (Framework, inference tool methodology, IP)
 ・benchmark/optimize open network, customer network
 ・consult and support Tier1 and OEM to utilize AI tool and IP in convination with other Computer Vision IPs in ADAS/AD SoC
 ・Create "easy to develop" application note



【Background】

ADAS/AD is one of Renesas focus growing area. AI is the most important technology in this application for OEM/Tier1.

Renesas already acquired >1B$ business and high quality support is required for OEM/Tier1 for their smooth development and mass production launch.

Full Description can be found on below link:
 
Last edited:
  • Like
Reactions: 7 users

Learning

Learning to the Top 🕵‍♂️
I think we just put Akida in here!
It's great to be a shareholder
 
  • Like
  • Fire
Reactions: 13 users
Top Bottom