Skip navigation
1 2 3 Previous

Single-Board Computers

156 posts
Edge Impulse is an online tool that makes training and deploying machine learning to the edge a breeze. Their built-in optimizations means that your models will be smaller and faster than most models of comparable accuracy, and far easier to train. Thankfully, the Edge Impulse Linux SDK makes it easy to get set up and start collecting data, training, and running inference on Avnet’s MaaXBoard. Previously, I showed how to get started with the Node.js SDK on MaaXBoard Mini with this hard h ...
Introduction I recently designed a frame and stand for the Avnet MaaXBoard MIPI LCD touch display   There were 3 main design goals . . . Provide some level of protection for the LCD display Frame can lay with the screen down and screen will not touch surface Provide stands for both portrait and landscape modes Frame can be lifted from stand Provide optional MaaXBoard MIPI CSI Camera mounts (one for each orientation)     If your are not already familiar with MaaXBoard RT, ...
I've been playing around with the eIQ tools on the Yocto image for MaaXBoard recently. Until now, I've mostly focused on Debian, but the eIQ layer for MaaXBoard Yocto was released earlier this year and it makes a number of machine learning tools available on the Yocto image. What is eIQ? It's NXP's custom collection of Neural Network compilers, libraries, and hardware abstraction layers. The following five inference engines are supported by eIQ: ArmNN TensorFlow Lite ONNX Runtime PyTorch O ...
A rendering of the ELSA-d spacecraft capturing a piece of space junk. (Image Credit: Astroscale)   On March 22nd, Astroscale, a Japan-based space startup, successfully launched ELSA-d (End-of-Life Services by Astroscale), a demonstration mission that intends to clean up orbital debris. The package contains two different payloads. One of them is a servicer satellite for future spacecraft, and the other is a client satellite, which de-orbits debris satellites for customers.   Astrosca ...
In the benchmarking results, we have seen that MaaXBoard Mini is a powerful board, with results comparable to Jetson Nano when benchmarked on the Mobilenet model: Despite it's name, the MaaXBoard Mini is mighty indeed! It boasts Quad Cortex-A53 at 1.8GHz and Cortex-M4F at 400MHz, a speedup compared to regular MaaXBoard's Quad Cortex-A53 at 1.5GHz and Cortex-M4F at 266MHz. At only $72.50, it's one of the most cost effective boards capable of running Machine Learning.   How to get set ...
Last spring I benchmarked machine learning inference on the MaaXBoard against similar single board computers. I looked at speed and power consumption while running inference using the object detection models MobileNet SSD V1 and V2. Recently, I benchmarked the slightly faster MaaXBoard Mini against the same boards. Here are the results in graphs.   Results for speed: Results for Power Consumption:   Results for Price: Thanks for reading! I hope to benchmark more of Avnet's b ...
Robotic dogs are being introduced in the law enforcement and national security units to support and protect them. (Image credit: NYPD Digidog & ABC7)   This year has been marked by a lot of violence towards both civilians and law enforcement officers, and the main question that is on the mind of most people is how to hold people accountable and always have the truth. Well, the solution might be in robotics. In Florida and New York, the law enforcement units are welcoming new members: ro ...
Introduction If you haven't noticed already, the MaaXBoard Mini supports Android 9 out of box! One thing that developers are probably wondering is how do I go through the process of actually developing my own application and deploying it on the MaaXBoard Mini? Well look no further, I have gone through the process of documenting everything required to get you started with your application development!   Version Date Version Revision 26 October 20 01 Initial Release   Required H ...
C++ is regularly revised. The current set of proposals are under review. They are called C++20. There are a few of the proposals that I'd like to try out. Some simplify object handling. Other move algorithms from run-time to compile time solve.   Microsoft and GCC (and others) released a compiler that supports many of the preview options. Check here for an overview of which compiler supports what.   Visual Studio 2019 Community Edition comes with the compiler built-in, if you sele ...
Avnet recently released the MaaXBoard Mini which is based on the NXP i.MX 8M Mini Processor. The MaaXBoard Mini is a low-cost, NXP i.MX 8M Mini processor-based, single-board computer ideal for embedded computing and smart edge IoT applications. The i.MX 8M Mini family of application processors are based on the Arm Cortex-A53 and Cortex-M4 cores, which provide industry-leading audio, voice and video processing for applications that scale from consumer home audio to industrial building automation ...
Introduction The Monarch Go Pi HAT is a cellular connectivity add-on card for SBC boards that have a Raspberry Pi HAT-compatible connector. Fitted with Sequans Monarch Go (or Monarch Go GPS) LTE Cat-M1 modem, this HAT board also facilitates expansion via Click boards using a compact onboard Click Shuttle connector. Noteworthy is that an integrated onboard USB-UART bridge interface allows this LTE HAT to also be used with other platforms.   Version Date Version Revision 18 June 20 01 Initi ...
Update: Want to see how well benchmarking performs on the MaaXBoard Mini? Updated results including the MaaXBoard Mini are HERE.   Introduction In the past two years, Machine Learning at the edge has proliferated. With more machine learning models capable of running on small, low power devices, and more devices capable of running them, it can be hard to know which device best suits your needs.As the ecosystem grows, benchmarking becomes more important. In order to choose a hardware platfo ...
1. Indroduction Python is a great general purpose high level programming language. It is beneficial for many reasons, one being that its simple syntax rules allows the language code to be readable and maintanable. It is fairly easy to develop also due to the wide array of available packages to support most types of development. When it comes to embedded development you may be asking why Python over C or C++. Each has their own advantages of course, mainly coming down to the debate of developmen ...
1. Introduction The MaaXboad by Avnet is based on the NXP i.MX 8M processor running quad Arm® Cortex®-A53s and a single Cortex-M4 core. Out of box the MaaXBoard supports both a Debian based linux image and a Android 9 image. In this blog lets discover what the Android enviroment has to offer.   To allow for a simple setup i'll be using a keyboard, mouse, and an external HDMI monitor to communicate with the MaaXBoard. If you wish you could connect to the board via TTL or use a SS ...
Thanks to Gianluca Filippini from EBV for creating the original tutorial. A slightly more complete version of this project is also available HERE   Intro These are the step-by-step instructions on how to set up a MaaXBoard to drive multiple servos with an accurate PWM external add-on card. The goal is to be able to control these motors with Python. We will reuse code originally written for RaspberryPi by Adafruit's Tony DiCola. Only few modifications are required to get everything working o ...

Filter Blog

By date: By tag: