Appliances

Accurately measure the weight of clothes inside a washing machine

Using AI to make your home appliances “smarter” and more energy efficient for a sustainable future.

Accurately measure the weight of clothes inside a washing machine Accurately measure the weight of clothes inside a washing machine Accurately measure the weight of clothes inside a washing machine
Accurately measure the weight of clothes inside a washing machine Accurately measure the weight of clothes inside a washing machine Accurately measure the weight of clothes inside a washing machine

Appliances

NanoEdge AI Studio

Asset tracking

Current sensor

Artificial intelligence can be used to achieve unprecedented levels of energy and water efficiency by more accurately measuring the weight of clothes inside a washing machine. The AI model generated by NanoEdge AI Studio significantly improves measurement accuracy compared to traditional algorithms by analyzing and learning the features of current signals.
This is only one example of what you can achieve using edge AI and NanoEdge AI Studio. This approach can easily be adapted to many other appliances or industrial machines.

Approach

This use case shows how to automatically optimize a washing machine's cycles by estimating the correct amount of water and energy to use based on the weight of the clothes placed inside the drum. Here is how we proceeded:
  • The current signals were measured directly from the motor control loop.
  • Magnets were used to simulate different weight conditions.
  • We used the Extrapolation algorithm provided by NanoEdge AI Studio to quickly build a regression machine learning model that accurately measures the weight of clothes inside the drum. The maximum estimation error reached was 100 g, which is three times better than the state-of-the-art solution on the market.
  • Then, we ported the machine learning library to the same microcontroller as the motor control algorithm. NanoEdge AI Studio optimized the library so it fits in the MCU's internal memory.

Additional sensors and network connectivity are no longer needed since the processing is done at the very deep edge, in the motor of the machine itself. Processing the signal directly on the MCU increases security, safety, energy, and cost since only meaningful data is output and processed.

Sensor

Current sensor inside a custom board (based on STM32G4).

Data

Data set Current signal with weight values as the label (available upon request)
Data format Time series

Results

R-Squared:
99.96% accuracy, 10.6 Kbytes of RAM, 9.5 Kbytes of Flash memory

Model created with

NanoEdge AI Studio

Model created with

Compatible with

Any STM32 MCU

Compatible with

Resources

Model created with NanoEdge AI Studio

A free AutoML software for adding AI to embedded projects, guiding users step by step to easily find the optimal AI model for their requirements.

Model created with NanoEdge AI Studio

Compatible with Any STM32 MCU

The STM32 family of 32-bit microcontrollers based on the Arm Cortex®-M processor is designed to offer new degrees of freedom to MCU users. It offers products combining very high performance, real-time capabilities, digital signal processing, low-power / low-voltage operation, and connectivity, while maintaining full integration and ease of development.

Compatible with Any STM32 MCU

You also might be interested by

Appliances | Agriculture

Coffee bean recognition

Classify coffee beans thanks to an image classification model from STM32 Model Zoo running on a STM32H7 microcontroller.

Smart homes | Appliances | Wearables | Entertainment

Handwriting recognition

Handwriting recognition on ultra-low-power MCU.

Smart buildings | Appliances | Entertainment

Hand posture recognition without camera module

Hand posture recognition running on STM32F401 based on ST multizone Time-of-Flight ranging sensor.