We were inspired to create this project by the desire to help people stay protected and not limit themselves in shopping. This project will solve the problem of distancing, healthcare, transport and goods delivery The issue of distancing, it is necessary to create a set of actions to slow the spread of infections, based on the reduction of close contact with people. Humanity is under the threat of epidemics and various viral attacks, it is important for us to maintain the availability and safety of delivery of various goods. For many people, the most important thing is that they want to preserve for this purpose it is necessary to preserve and show the buyer what is missing in his grocery basket to help stay healthy for many years of life.
What it does
We created Cleanbox to clean viruses from things, which is controlled from a web application. Our system consists of a box equipped with a UV lamp, microcontroller, and a lock. With the help of that box, people will sterilize the goods without fear of viruses brought into the house.
The other facet of our solution is a Myhealth for users who wonders about their healthy diet. Users can upload a shop receipt. We will show them the miss items in their shopping receipt and give the recommendations of the grocery. This recommendation is based on predictive NLP and ML models. This will help users understand about missing groceries.
How we built it
We created a box model and equipped it with an esp32 microcontroller that controls the lock and lamp. The microcontroller controls the electrical signals in the case of a lock, it sends a short-term electrical signal. For lamp
the electrical signal passes through the relay and turns on sterilization.
Health inside uses 2 different models fasttext and logistic regressions to predict vitamins based on product characteristics. We preprocess the input data and computed embeddings based on the fasttext model. Then we trained the logistic regression model for multilabel classification. These models were built in Python.
The website for users was built using Flask, Python, and Ajax and relation database PostgreSQL.
Challenges we ran into
We created our first box from an ordinary cardboard box, but after the organizers added time for work, we created a full-fledged box.
Accomplishments that we're proud of
We are proud that we have created a product that can be developed further and will be useful for distancing, healthcare, transport and goods delivery
What we learned
For the first time, we made an engineering solution together with a software
What's next for Here as Here
Part 1. Cleanbox+
Add shared boxes to apartments.
Add flexible cleaning time that users can select.
Add virtual assistant capability.
Optional sterilization lamp replacement function.
Creation of different sizes of boxes to improve the demand of different buyers.
Modification of the box for the use of sterilization in cars to prevent the risk of infection in carsharing.
Part 2. MyHealth.
Suggest items by vitamins based on the user's diet.
Compute percent of every vitamin in the grocery basket based on the daily norm for humans.
Push notification until when the product must be consumed.
Diet recommendation and selection of foods for people with diabetes or allergies.
Finding cheaper similar goods.