Why?
Only 30% of trash is reused or recycled, the remaining ends up in landfills polluting the environment. By tracking what trash we are producing based on zip code,
1). We can create maps of the most used products in a region, then suggest alternatives or greener products
2). We can collect the images people upload anonymously along with the weight of the product to the app to build models to train automatic trash sorting tools
3). We can track the trash supply chain and see how much of it is being reused or recycled, this can help local legislature create laws, as there are monopolies/companies like WM that control most waste management in the USA so they can report numbers that may not be true, but by moving things like this to the Cardano blockchain we can create a decentralized system which is free from the influence of these companies
4). By building a “Sorted at Source” system instead of “Single Stream Recycling” we can divert the trash to the right processing unit since, I found one of the major reasons that trash ends up in landfills is due to single-stream recycling although it is easier to execute, the people sorting are only a few when compared to the people producing the trash, so by rewarding the people for “Sorting at Source” by means of something like a smart contract we can incentives that reward them in Cardano for their clean and responsible acts.
5). We can help social programs and researchers see the blockchain data and use it to create clean up programs, educate students and the community about the effects of trash pollution, etc
6). We can support entrepreneurs who want to build businesses that specialize in the specific type of plastics like styrofoam or handling food waste that may not be available in waste management companies by showing information on the blockchain that is decentralized and can’t be corrupted by larger corporations
Project Description (What?)
An app and system to track what trash is being produced by product/category based on zip code, how much of it is being produced, and how the trash is being processed I.e is it reused, recycled, or just thrown in a landfill.
How?
I developed an app called “Track My Trash”, and have been testing it for a year, the use the app we just need the zip code, every time I dispose of something in the trash, I take a picture, measure the weight using a small weight machine and then dispose of it properly.
There is also a machine learning system I am training that will automatically classify the images so that in the future if it sees a 1-gallon milk container it can estimate the weight instead of the user entering
The user can also see the metrics of how much trash they produced based on weeks, months, or years
AWS Architecture
Explanation
The user first creates an account with basic information like email, password, name, zipcode using AWS Cognito. Before disposing any trash the user measures the weight using a scale sent to them and uploads the picture and some metadata like weight and description, the image is uploaded directly to S3 using a signed url using the AWS Cognito.
Once the image is uploaded to S3 a lambda function is invoked, it uses the Sagemaker API to get details about the image uploaded, it then creates a record in the project’s DynamoDB metadata table. This metadata will contain user provided information and the information returned by the classifier.
Different API’s can be developed using API Gateway to show the various metrics like consumption based on locality, based on trash type, the user can see their weekly, monthly, yearly consumption, we can based on consumption suggest alternative materials or products that are more sustainable
A commercial or government API can be developed to expose anonymous meta data to governments and other waste management companies to build new infrastructure to manage trash, develop sustainable materials, develop sustainability policies etc.
Mobile App Screenshots
Mobile App was developed using the Flutter Framework (https://flutter.dev/)
