Washing Machine Load Weighing

Table of Contents

• Introduction
• Supported Combinations
• Dataset and Model Details
  • Dataset
  • Model Architecture
  • Input Features
  • Output
• Feature Extraction Configuration
• Steps to Run the Example
• See Also
• Sample Output

Introduction

Prediction of weight of clothes in a washing machine eliminates the need for a physical weight sensor, which is typically required for automatic water level adjustment. This reduces manufacturing costs and provides a reliable sensor-less solution.

This project utilizes machine learning models that take into account various features such as voltage, current, and speed of the washing machine motor to predict the weight of the clothes. This ensures that the water level can be adjusted accurately without relying on a physical weight sensor, which can be prone to mechanical failure.

Supported Combinations

Parameter	Value
CPU + OS	r5fss0-0 nortos
Toolchain	ti-arm-clang
Board	am261x-lp
Example folder	examples/ai/washing_machine_load_weighing/

Dataset and Model Details

Dataset

The dataset consists of 6 current, voltage, and speed-based features obtained from TI Lab. The dataset has 100g precision from 0g to 900g.

Variable	Description
Vd	Voltage component along d-axis
Id	Current component along d-axis
Vq	Voltage component along q-axis
Iq	Current component along q-axis
Iqref	Reference Current
Speed	Speed of the washing machine motor

Model Architecture

The model is a CNN-based architecture with approximately 13,000 parameters consisting of 3 convolution layers and 2 fully connected layers.

Input Features

The model takes 4D input (N,C,H,W):

• N (1): batch size which is restricted to 1
• C (6): channels which is 6 for the features
• H (512): samples of timeseries signals which is 512 in this example
• W (1): width of samples is restricted to 1 for timeseries applications

Output

This model produces a 1D output representing the weight of the clothes in the washing machine (in grams).

Feature Extraction Configuration

For this regression task, a SimpleWindow transformation is used, which makes use of the previous 512 datapoints to predict the current instance of the target variable. No FFT or frequency-domain feature extraction is applied.

Steps to Run the Example

• When using CCS projects to build, import the CCS project for the required combination and build it using the CCS project menu (see Using SDK with CCS Projects).
• When using makefiles to build, note the required combination and build using make command (see Using SDK with Makefiles)
• Launch a CCS debug session and run the executable, see CCS Launch, Load and Run
• The application will predict load weight and compare against expected values

See Also

AI Examples

Sample Output

Washing Machine Load Weighing Example Started ...
All tests have passed!!
Golden vectors matched: 3 not matched: 0