Top 10 Real-World Machine Learning Projects for Beginners to Try in 2022
Emerging technologies are being incorporated into machine learning and deep learning techniques along with the advancement of artificial intelligence(AI). More and more people are running machine learning models on small, low-power microcontrollers and incorporating them into projects. This all makes deploying machine learning projects more accessible and affordable.
Today I’m going to introduce you to the best real-world machine learning projects for beginners to try in 2022. Before starting the project introduction, I would like to recommend to you the best microcontrollers and compact devices available for machine learning at ultra-low costs as low as $5.
XIAO Series
- ARM Cortex-M0+ 32bit 48MHz MCU
- Compatible with Arduino IDE
- Breadboard-friendly
- As small as a thumb(20×17.5mm) for wearable devices and small projects
- Multiple development interface
- Powerful Raspberry PI RP2040 chip
- Dual-core ARM Cortex M0+ processor
- Rich on-chip resources
- Supports Arduino, MicroPython and CircuitPython
XIAO SERIES
- Nordic nRF52840, ARM® Cortex™-M4 32-bit processor
- Bluetooth 5.0, NFC, and ZigBee module with onboard antenna
- Ultra-low sleep power: 5 μA
- Onboard 2 MB flash
- Rich interfaces to expand more possibilities
- Advanced version of the board XIAO BLE
- Extra Onboard PDM microphone and 6-axis IMU
- Corresponded to TinyML AI project: Sensing, Processing, Communication 3 in 1 Node
Wio Terminal
Wio Terminal-$36.90
Powerful MCU: Microchip ATSAMD51P19 with ARM Cortex-M4F core
Reliable Wireless Connectivity:Realtek RTL8720DN, dual-band 2.4Ghz / 5Ghz Wi-Fi
Complete system equipped with Screen + Development Board + I/O Interface + Enclosure
Highly Integrated Design: Screen, MCU, IMU, WIFI, more add-ons housed in a compact enclosure
Raspberry Pi 40-pin Compatible GPIO enables installation as a peripheral to the Raspberry Pi
External onboard multi-functional Grove ports: Compatible with 300+ Grove to explore with IoT
USB OTG Support：can act as a USB host or USB client
Support Arduino, CircuitPython, Micropython, ArduPy, AT Firmware, Visual Studio Code
Azure Certified Device: Sense and tag real-world data and visualize through Azure IoT Central
Without further ado, let’s start learning about embedded machine learning projects for beginners that can be used in a variety of scenarios! If you already have XIAO series products or Wio Terminal, I believe you can use them to deploy and develop these projects more easily.
Seeed Wio Terminal programmed with Codecraft/Edge Impulse is a great tool for beginners to get started with tinyML (embedded machine learning).
This project mixes machine learning with Wio Terminal to bring an easy introduction to machine learning gesture recognition. Using gesture recognition with built-in light sensors, the Wio terminal will be able to recognize rock, paper and scissors gestures and display the corresponding images on the screen.
In this project will be used:
The project collects machine learning data on correct and incorrect postures through Edge Impulse training. In this project, you can choose to customize your own PCB board, or you can use XIAO and its expansion board, and match it with a GroveLIS3DH accelerometer module. Now deploy this project to your back to monitor correct posture!
In this project will be used:
Are you aware of sun damage risks? While many of us enjoy sunny days in our spare time, harsh sunlight can also have a detrimental effect on our health. This BLE smartwatch can be used to predict sun damage risk and inform users. The project trained and tested the artificial neural network model (ANN) using Edge Impulse, and uploaded the model deployment to XIAO BLE, which can easily collect data and run the neural network model.
In this project will be used:
Our lovely pets should be more active to maintain good mood and health. This pet activity tracker uses XIAO BLE Sense and Edge Impulse to track various activity data that identifies dogs. The tinyML model predicts activity based on data from the 3 Axis IMU on the XIAO BLE Sense.
This project can be implemented with just a microcontroller XIAO BLE Sense, doesn’t it sound great? XIAO BLE Sense is equipped with a powerful Nordic nRF52840 MCU, designed with Bluetooth 5.0 module. It features an onboard antenna, 6 Dof IMU, microphone, all of which make it an ideal board to run AI using TinyML and TensorFlow Lite.
Come and get a XIAO BLE Sense to try out this project.
In this project will be used:
Backcountry camping is a great experience, as long as the activity is safe, especially on dark nights. This project runs the TinyML model on the Wio Terminal, using the MLX90640 thermal imaging camera that works well without light. It can identify if an animal or human is approaching even in the dark, sending data to the Helium LoRa network to AWS to alert campers.
If you are a camping enthusiast, or any other place you want to be safe, deploy this solution around you.
In this project will be used:
When realizing the safety problem of the leaking fuel tank, Mithun Das’ daughter Sashrika designed a “gas leak detector” based on the Wio Terminal. It is connected to a multi-channel gas sensor and fan. This project runs a tinyML model on a microcontroller to detect the smell of diesel in the air. Push notifications are sent via the Blynk app when diesel is detected.
In this project will be used:
It has become very common for most people to spend their studies and work on the computer. Using a mouse for a long time can cause soreness in our wrists, forearms and other parts.
The project utilizes GSR (galvanic skin response) and EMG (electromyography) measurements to indicate forearm muscle soreness, and creates a budget-friendly device to predict muscle soreness levels in hopes of avoiding permanent mouse-related injuries. Wio Terminal is used for this project as it makes it easy to collect and display muscle soreness data.
In this project will be used:
Water resources are becoming more and more precious due to rising global temperatures and the limited availability of water. The imbalance between irrigation demand and water resources is urgently needed to improve water efficiency and agricultural water management.
The project collects irrigation water level data through thermal imaging, uses TensorFlow to build and train a neural network model, and runs the model directly on Wio Terminal.
In this project will be used:
Do you think it’s possible to perform handwriting recognition with just a single distance sensor? The answer to that is, well, sort of! This project uses machine learning on time series data from just one ToF sensor to recognise handwriting gesture patterns! While it’s very much a proof of concept project and far from actual implementation, I hope this inspires you to think of crazy ideas for your own project!
In this project will be used:
Weather stations are a popular project amongst the maker community. Why not take it one step further and add Edge AI capabilities to enable local weather predictions? This project by Dimitry Maslov does exactly that.
In this project will be used:
This project is also part of our Learn TinyML using Wio Terminal and Arduino IDE series. Be sure to check each of them out!
- Learn TinyML using Wio Terminal and Arduino IDE #1 Intro
- Learn TinyML using Wio Terminal and Arduino IDE #2 Audio Scene Recognition and Mobile Notifications
- TinyML using Wio Terminal and Arduino IDE #3 People Counting and Azure IoT Central Integration
- Learn TinyML using Wio Terminal and Arduino IDE #4 Weather prediction with Tensorflow Lite for Microcontrollers
- Learn TinyML using Wio Terminal and Arduino IDE #5 Anomaly detection for Predictive maintenance with new Blynk IoT Platform release
- Learn TinyML using Wio Terminal and Arduino IDE #6 Speech recognition on MCU – Speech-to-Intent
- Learn TinyML using Wio Terminal and Arduino IDE #7 Machine Learning on ARM Cortex M0+ MCU Seeeduino XIAO and XIAO RP2040
Summary
Summary
The above are the machine learning projects for beginners introduced today. There is no doubt that the XIAO series or Wio Terminal is the best choice for you to start your own machine learning project. Has this article sparked your desire to explore machine learning projects? What is your next embedded machine learning idea? Let us know in the comments, and if you need any help, we’ll do our best to help you make your shiny project idea a reality.
About Author
