Smart City

Smart Trashcan

AS034

Marco Scipioni (Queens University of Charlotte)

Sep 30, 2021 2388 views

Smart Trashcan

The practice of recycling helps reduce pollution, greenhouse gas emissions, and the amount of waste that is disposed in landfills. The majority of Americans unfortunately does not embrace the three Rs (reduce, reuse, recycle). The lack of adequate knowledge for sorting and recycling materials is one of the biggest barriers to being green. Recycling is a behavior that can be improved through technology. This proposed project is centered around the creation of a smart trash can prototype designed to create awareness among students at Queens University and in its neighboring community on the importance of correctly sorting waste items. The smart-trash can has both a hardware and a software component. The project will specifically focused on developing a working prototype and deep learning (DL) model using the Intel FPGA Cloud Connectivity kit in combination with Microsoft Azure IOT. The model is able to correctly classify different types of disposable and recyclable food service items (paper cups, paper boxes, paper trays, food containers, etc.) commonly found in the Queens University’s cafeteria and around campus. The classification is used by the hardware to provide a visual prompt to indicate the bin for a particular waste item. This can lead to improving the process of pre-sorting recyclable materials once the smart trash can is fully deployed on campus.

Project Proposal


1. High-level project introduction and performance expectation

The system we propose is a smart trash can that provides awareness to the user on what should be recycled versus what should not be recycled and can correctly classify the objects presented to it. To achieve this, FPGA Cloud Connectivity Kit enabled with Azure IoT, a USB Camera, PIR motion sensor, differently colored LEDs for each different class of recyclables and their corresponding bins. The FPGA will ensure the appropriate processing speed to correctly capture pictures and correctly classify them. The camera would be positioned above the PIR motion sensor which would be triggered once motion is sensed. Lastly, a different color LED will lit to indicate in which recycling bin the object needs to be disposed. The user would then place the object in the bin thus correctly sorting the material.

2. Block Diagram

3. Expected sustainability results, projected resource savings

Once implemented using FPGA, the smart trash can system would have a very fast processing speed to be able to capture pictures of the disposable objects and classify them in about 1 seconds without suffering significant lags. Such speeds would be hard to achieve using microcontrollers. The smart trash can beta prototype would be positioned in different areas of the Queens University campus to interact with students to discover their recycling habits to  The smart trash can would become spark interest in the important topic of recycling and encourage students to sort recyclables correctly.

4. Design Introduction

5. Functional description and implementation

6. Performance metrics, performance to expectation

7. Sustainability results, resource savings achieved

8. Conclusion

0 Comments



Please login to post a comment.