Close
0%
0%

MasterWaste : A smart waste management system

A trash can able to classify and sort to facilitate recycling!

Similar projects worth following
233 views
0 followers
Introducing the MasterWaste: Revolutionizing Recycling with AI!
We are two passionate robotics students who have embarked on a mission to make recycling easier and more efficient for people everywhere. Our innovative solution, the MasterWaste, is a state-of-the-art smart waste container that utilizes cutting-edge technology to automatically sort various types of waste.
The system operates using a proximity capacitive sensor that acquires a specific value corresponding to each material.
This value is then sent to the computer where a previously trained AI model attributes it a category between Glass, Paper, Metal, or Organic.
The system then uses this label to redirect the trash into the right container that can be harvested later.
With this project we hope to provide a proper solution to the Green Hack challenge round of the 2023 Hackaday Prize

An overview of the situation :

    According to the United Nations Environment Program (UNEP), 11 billion tonnes of waste are generated and collected yearly around the world. The main solution implemented is, of course, the minimization of waste around the world but in situations where waste cannot be avoided, physical waste management in order to recycle becomes the next best thing.

According to the 2020 annual report of the International Environmental Technology Centre (IETP) however, only 11% of municipal trash undergoes treatment and this value greatly varies depending on the city. 

   This issue creates an incentive to create a simple household system designed that allows citizens to easily categorize their waste without having to trouble themselves. 

Our system will have to be cheap and non-invasive in order to be installable in a maximum of households. It also will have to be comprehensive so that users don't find themselves confused by it.

The idea : 

Classification : 

The original idea was to use the electric properties that vary between the different categories of waste to properly classify the trash using a decision tree algorithm.

To acquire those values we decide on using a capacitive sensor to detect the "degree" of conductivity of the material thrown. From there the idea is to make software that categorizes and operates the motors to put the trash in the correct bin.

The sensor and microprocessor must be cheap so that the overall system remains affordable

  • 1 × Heschen M18 Capacitive Proximity Sensor Switch LJC18A3-B-Z/BX Mounting hole diameter: 18mm Detecting distance: 1-10mm Working voltage: 10-30 VDC Current output: 200mA Output type: NPN NO(Normally Open) Cable length: 120cm/49" Detection Object: Can detect any dielectric

  • Update on classification

    Elijah Ki-Zerbo06/20/2023 at 10:58 0 comments

    After some testing, the decision tree algorithm using the scikit-learn algorithm is able to assign labels to dielectric constants (see video below)

    In this video, I input an arbitrary value for the dielectric constant received by the script to test the classification. While the code is not complete, it manages to assign a label of 0 (glass/plastic) when the constant is 0, For a constant of  40 the algorithm correctly guesses organic matter (1), and finally for a constant of 200, the algorithm classifies it as a metal (2)

  • Update on dielectric sensing

    Elijah Ki-Zerbo06/20/2023 at 10:17 0 comments

    We intend to add the results of the project in this section 

    Once received, we proceeded to test the capacitive sensor with the Arduino operations.

View all 2 project logs

Enjoy this project?

Share

Discussions

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates