2023 Google Solution Challenge

🌿Garbage Collection

Member

• Sangwon Choi (Department of AI Convergence, Soongsil University)
• Sanghyeon Sim (Department of Computer Science, Soongsil University)
• Jungmin Lee (Department of Computer Science, Soongsil University)
• Hwangon Jang (Department of Software, Soongsil University)

Related Repository

• Garbage Collection Client
• Garbage Collection Server
• Garbage Collection AI

Introduction Our Solution

• By using our app, users can identify the type of trash they need to dispose of through AI, as well as locate nearby garbage cans
• This will enable them to properly dispose of their waste, contributing to the United Nations' Sustainable Development Goals of Sustainable Cities and Communities, Clean Water and Sanitation, and ultimately creating a cleaner environment for everyone.

Architecture

• Our app was developed with a flutter on the front end, and we used firebase for authentication/authorization, and google map to locate the trash can. The backend developed a server with a spring boot and distributed it through CloudRun, and the AI model was made into tensorflow and distributed through the compute engine.

UI

1. When enter the app, the locations of nearby garbage cans are displayed on the map with green markers.

2. When press the camera button to see what trash I'm trying to throw away, then AI tells me the type of trash and recommends a nearby garbage can.

3. When you click on the recommended garbage can, its location will be marked on the map with a red marker.

4. When the user arrives at the recommended garbage can and disposes of the plastic trash, they will receive a plastic item card as a reward.

5. If the garbage can is full, users can report its status to the local government through the "Report" button.

6. Users can create their own collection of item cards by earning them, and compete with other app users to climb the rankings.