NEW Releasing: Grove AI HAT for Edge Computing and Look forward to your ideas!

[yokonav]

@jeremy882 post_id=52729 time=1552986862 user_id=22261

[quote=yokonav post_id=52675 time=1552605406 user_id=13948] This looks amazing with a lot of possibilities! If we are using it with Raspberry Pi, we do not need wifi, bluetooth, sd card module but of course as a standalone unit we might need it. I was working on the "Recognition of Activities of Daily Lives (ADL)" and there was requirement for a 3-axis accelerometer. The problem is: if we use 3-axis accelerometer grove module, for better result it should be mounted parallel to the board. It would be very helpful to add a 3-axis accelerometer on the board itself which I think would not take much space. It will be beneficial for other use cases like auto-rotate lcd screen, free fall detection (to save data before it crashes!), almost all robotics projects (auto driving) needs it. I am eager to try it out soon! [/quote] Thanks for the suggestion, a 3-axis accelerometer sensor integrated on the board is surely more convenient than using Grove. Will try to find a cost-effective one so that not adding too much cost for our users, do you have any recommendation for the 3-Axis accelerometer sensor?

I think ST LIS2DH12 is low-cost and ultra-low-power and also useful for gesture recognition, free fall and wake up detection and many other features.

[jeremy882]

@marioalemn post_id=52668 time=1552581110 user_id=18827

A connector for a battery pack would be great! So the battery pack could power both the HAT and the Pi itself.

For Raspbeery Pi + the AI HAT, the lithum battery would run out very rapidly :mrgreen:

[elaine.wu]

UPDATE: We just got the first look of this Grove AI HAT for Edge Computing and Raspberry Pi based on the Sipeed MAIX-I, including LCD and Camera Interfaces, connectors for Digital, I2C, UART and Analog.

Don’t hesitate to share your thoughts and ideas with us in this forum! Let us know what features you would like to see and have the chance of making it to the final product.

[yokonav]

Wow, looking cool! Hoping to see an accelerometer in the next revision and want to try it out soon.

[julie.weber]

@jeremy882 post_id=52740 time=1553073875 user_id=22261

[quote=marioalemn post_id=52668 time=1552581110 user_id=18827] A connector for a battery pack would be great! So the battery pack could power both the HAT and the Pi itself. [/quote] For Raspbeery Pi + the AI HAT, the lithum battery would run out very rapidly :mrgreen:

I think it’s a good idea to add a connector for a battery pack, not to power the board with the battery only but to have an UPS for short power failure. It could be very useful to prevent data corruption.

[baqwas]

Hello,



Is there a release schedule for the revisions posted (for easy lookup) and order placement?



Thanks.

[jeremy882]

Hi Guys,



Thanks for all the wonderful ideas! Here is an update about things we are going to add to the next version.

1. There AI HAT can power Raspberry Pi directly. Just use the onboard switch to switch power mode.
   
2. There will be a microphone on board, the same microphone as the Sipeed 6+1 micarray.
   
3. There will be an ADXL345, 3-Axis Accelerometer on board. We choose this one because there is already a Grove version of the sensor, the software is ready, easier to get a sample, and also lower cost.
   
4. There will be a user button, and the boot LED can be used as user LED.
   
5. The 40-Pin interface will be changed to SMT female header.
   
6. The JTAG package will be smaller.
   
   
   
   We will finish the layout in a week, and get samples of the second version in the week of 22-Apr. If everything goes well, we will do pre-order in early May.
   
   
   
   Welcome to more ideas about how to using Grove with this HAT~

[jeremy882]

@baqwas post_id=53098 time=1554383331 user_id=14923

Hello,

Is there a release schedule for the revisions posted (for easy lookup) and order placement?

Thanks.

Hi baqwas,



We will finish the layout in a week, and get samples of the second version in the week of 22-Apr. If everything goes well, we will do pre-order in early May. Thank you~

[seeed622]

This processor has an amazing APU ( audio processing unit ) that is capable of handling ( 8 ) 192 khz audio streams with hardware accelerated FWT processing. It would be a shame not to have a way to access those features via I2S. It would make an amazing engine for processing high-end audio, especially in combination with the Pi.

[hitmanarky]

Please consider adding a camera that is ideal for future machine vision applications. At the moment Pregius Gen 3 Global Shutter CMOS is the my favorite.



<LINK_TEXT text=“https://www.sony-semicon.co.jp/products … egius.html”>https://www.sony-semicon.co.jp/products_en/IS/sensor0/industry/technology/pregius.html</LINK_TEXT>



What do you expect from a good machine vision camera?

• Accurate color reproduction
  
• No chromatic aberration
  
• Good low light performance
  
• Interchangeable lens with mount system
  
• High frame rate upto 120 fps
  
• Ability to access and change the parameters of the camera settings (white balance etc)
  
  
  
  For conservation projects, the embedded vision camera need specific requirements.
  
  
  
  Camera with IR cut filter / IR-white-red flash /star light camera are ideal.

[jeremy882]

@seeed622 post_id=53329 time=1555607705 user_id=36679

This processor has an amazing APU ( audio processing unit ) that is capable of handling ( 8 ) 192 khz audio streams with hardware accelerated FWT processing. It would be a shame not to have a way to access those features via I2S. It would make an amazing engine for processing high-end audio, especially in combination with the Pi.

Indeed. And thanks for precvious feedback in this forum, we added an on-board microphone, and we have also breaked out I2S of the processpr to Raspberry Pi 40-Pin connector.

[jeremy882]

@hitmanarky post_id=53385 time=1556038755 user_id=111099

Please consider adding a camera that is ideal for future machine vision applications. At the moment Pregius Gen 3 Global Shutter CMOS is the my favorite. :)

https://www.sony-semicon.co.jp/products_en/IS/sensor0/industry/technology/pregius.html

What do you expect from a good machine vision camera?

* Accurate color reproduction
* No chromatic aberration
* Good low light performance
* Interchangeable lens with mount system
* High frame rate upto 120 fps
* Ability to access and change the parameters of the camera settings (white balance etc)

For conservation projects, the embedded vision camera need specific requirements.

Camera with IR cut filter / IR-white-red flash /star light camera are ideal.

We will have a beginner kit for this Grove AI HAT soon and will include a camera in the kit. The idea is to include the camera that is being used by those Sipeed boards because it is cost-effective and can handle basic tasks. The Sony cameras can be great accessories to be purchased separately for higher level usage.

Top

[andriyadi]

Good progress.



What about connector for external 6 Mic Array that’s already available for Maix, is it already there? Would be great to have it.

[jeremy882]

An update: Face Recognition Demo using Grove AI HAT for Edge Computing and Raspberry Pi



https://youtu.be/5BF3ExL1HOQ



<LINK_TEXT text=“https://project.seeedstudio.com/SeeedSt … -pi-3e100f”>https://project.seeedstudio.com/SeeedStudio/face-count-and-display-using-grove-ai-hat-and-pi-3e100f</LINK_TEXT>

[ronwagner]

Artificial intelligence technology has been applied in various fields. Grove AI HAT for Edge Computing is a surprising news .And l want to share some discussions about AI : <LINK_TEXT text=“https://docsbay.net/ai-at-the-edge-the- … -of-things”>https://docsbay.net/ai-at-the-edge-the-next-frontier-of-the-internet-of-things</LINK_TEXT>

[tristan.mumford]

I don’t know if it will work for anyone else. I put this together yesterday.



https://github.com/experimentech/kendryte-toolchain-arm



It’s based on the native toolchain, so it may need some libraries installed as per the Kendryte-gnu-toolchain installation instructions. Or it may not work at all for others.

I have used it on my RPi3 and Jetson nano so far to build, upload and run some example sketches.

[niketsurve12345]

Well explained…

[jaikrishan5591]

The idea is to distribute the data storage and intelligence in … Disclaimer: SeeedStudio has contacted me to review the Grove AI Hat for Edge Computing. … Hardware is actually evolving so quickly that we can see small .

[zamir138]

@bernard.vanhaecke671 post_id=52673 time=1552589206 user_id=32725

Awesome stuff! I suggest to put at least a couple more Grove connector sockets on the board, there seems to be enough room left. This would allow to connect several additional Grove sensors (there are many for about anything), which could be used as additional input to ML models and allow to do sensor fusion projects (because using AI for camera only is somewhat limiting). Grove actuators can be hooked up as well to sockets (relays, motor controller, etc.. there are a lot of Grove accessories) and can for example be controlled from the output of a ML model; this is easy to do. I suggest to put the additional Grove sockets vertically, just like on the Seeeduino Lotus board instead of lining them up at the edge of the PCB. You could probably have 4 analog and 4 digital ports, instead of 2.

We are going to release two versions, one based on the Sipeed MAIX-I with WiFi and another without. Both can work as a Raspberry Pi HAT, or work by themselves. The basic function and features are confirmed and listed here. But we are open to suggestions from the community about how to make this product.

[deous]

You could add LAN controller f.ex.ENC28J60 or W5500 instead of wifi.

Also fpc for 16bit and 24bit LCDs