Skip navigation
> RoadTest Reviews

Brainium /SMARTEDGE AGILE - Review

Scoring

Product Performed to Expectations: 8
Specifications were sufficient to design with: 9
Demo Software was of good quality: 9
Product was easy to use: 9
Support materials were available: 10
The price to performance ratio was good: 8
TotalScore: 53 / 60
  • RoadTest: Brainium /SMARTEDGE AGILE
  • Buy Now
  • Evaluation Type: Test Equipment
  • Was everything in the box required?: Yes
  • Comparable Products/Other parts you considered: Severals if we take only sensors capabilities, none I know with the no coding maching learning capabilities
  • What were the biggest problems encountered?: None

  • Detailed Review:

    BRAINIUM WHAT HAPPENS ?

     

    The Agile SmartEdge, a low-energy, ready-to-use IoT device developed through a partnership between Avnet, Octonion, and ST Microelectronics. The first of a new category called "meta-detection". The SmartEdge Agile includes nine sensors and gives engineers and makers the ability to implement artificial intelligence (AI) with an easy-to-use interface and no coding.

     

    The Agile SmartEdge is a ready-to-use IoT product that ships an STMicroelectronics STM32 microcontroller. The solution is hosted in Microsoft's Azure Cloud and also includes a dashboard for easy operation and programming of feedback all without any lines of code. What bring speed and ease of development to the world of machine learning.

     

    INTRODUCTION

     

    The SmartEdge Agile is FCC, CE certified and equipped with Bluetooth Low Energy 5.0 technology, its internal battery can be charged via a USB C-type connection. The device is based on an 80MHz and 32-bit Brainium STM32 chip ARM cortex-M4. The set includes two 3D accelerometers, a 3D gyroscope, a 3D magnetometer, a pressure sensor, a temperature and humidity sensor, a microphone, an ambient light sensor and a proximity sensor.

     

    The whole Brainium system is focused on security. While the gateway application provides only BLE to TCP tunneling, data between device and cloud being sent over TLS 1.2 protocol to ensure the security, authentication, encryption and integrity of the data exchanged.

     

    The SmartEdge Agile is delivered with a USB-C cable and its instructions, the set is very compact with dimensions 32 x 17 x 64 mm, 27 grams weigh and fully recharged in 4 hours.

     

     

     

    We find on the top face the different sensors detailed above.

     

    WHERE TO START

     

    Brainiums portal for managing and using Agile devices is very easy to access and simple to use. It has lots of features that will help you set up your devices and create AI models for them. After charging your device, we will install the free mobile application on Android, iOS and Raspberry Pi. With the software, we will be able to communicate in BLE (Bluetooth Low Energy) with our SmartEdge Agile.

     

    Next ou need to create an account on Brainium portal. After signing up you will receive an email with your credentials, use it to connect to the portal on www.brainium.com/portal/

     

    Now we will pair your device with Brainium portal. To do so, your phone bluetooth and internet connection needs to be turned on. Turn on your device by pressing the button on it and holding it for more that 2 seconds. Once it will start blinking with blue light and should automatically connect to your phone.

     

     

     

    QUICK EXAMPLE

     

    In this first example, we will teach our SmartEdge Agile with machine learning and motion detection to identify the action of walking and jumping, for our first use, we are accompanied in your approach.

     

     

    To create a project, go to Brainium portal and open tab "Projects". Next click the "+" icon in the bottom right corner and modal will be displayed where you enter project name and project image.

     

     

    Now click on project icon and then you’ll be in the project page. Now you can click on "+" icon in the bottom right corner to assign the device to your project.

     

     

    To make your own AI model you need to go to the AI studio and  click the "+" icon to create a new workspace.

    There you will define the motions that you want to track and you will supply the model with examples that will be used for machine learning. To get that data you will have to record some data using your device.

    You will do that by creating a new motion data here "walk" and "jump", next we will record data on each action. The more data you have, the more accurate the result will be. Give the example number and click save data

     

     

    After the recording you can inspect the collected data on a graph to make sure that the the recording if good enought.

     

     

     

     

    When you are satisfied with the recorded data you can create the model who will represent the action. Select all recordings that will be used in model, from both motions and click "Generate Your Model". The model creation can take severals minutes to achieve.

     

     

    When model is done, you can assign it to your project, select Devices tab and again click on sensor icons. This time click on tab "AI Studio Rules" and select from dropdown menu your AI studio workspace. Then select your model that you’ve created for "walk" and "jump" movements.

     

     

    Next switch to "Data Tracking/Recording" tab where you can create widgets. Create one by clicking on the "Create Widget" button which opens up a modal for creating widgets. Just input widget name, type and visualisation type, click next, select your device you’ve added to the project and click "Finish". Now you’ve widget to monitor data to your project.

    For this example, I created 4 widgets. The first for tracking the battery and the other three for accelerating monitoring and responding machine learning.

     

     

    If we want to had an alert, we can go back to devices tab and click on sensor icons of your device. That will open a modal where you can click on "Rules" tab to add notifications. Click on "Add New Rule" button and fill requested fields with your conditions. Here I will be advise every minute if my devise have an acceleration above "2 m/s".

     

     

    CONCLUSION

     

    Here is a very quick example set up in less than 30 min with machine learning and all without coding. We can say that simplification is the key to this product, both in terms of getting started and starting a project. It is a product that will not fail to attract the curiosity of makers and industry.

    And it's not over, there is still much to do, only your imagination will be the limit. Personally, I'm going to dig a little more on the connectivity side with the REST API and MQTT.

     

    MANUALS/LINKS

     

    www.brainium.com/portal/

    https://brainium.blob.core.windows.net/public/docs/Brainium_User_Manual_1.8.pdf

    https://developer.brainium.com/


Comments

Also Enrolling

Enrollment Closes: Aug 10 
Enroll
Enrollment Closes: Sep 15 
Enroll
Enrollment Closes: Sep 8 
Enroll
Enrollment Closes: Aug 21 
Enroll
Enrollment Closes: Aug 28 
Enroll
Enrollment Closes: Aug 25 
Enroll
Enrollment Closes: Aug 18 
Enroll
Enrollment Closes: Aug 18 
Enroll
Enrollment Closes: Aug 17 
Enroll
Enrollment Closes: Aug 7 
Enroll