This mono-repository contains the entire codebase for MDP.
It is divided into the following components:
- RPi - Contains the Raspberry Pi codebase that is responsible for communicating between the Android tablet, STM, image recongition service & pathfinding service.
- Image Rec - Contains codes responsible for AI image recognition training. Model used is Yolov8, a You Only Look Once Image recognition algorithm. Detection method is instance segmentation. Includes archive of AI models trained. Dataset for this model can be found here.
- Android - The Android tablet codebase.
- Openapi Simulator Client - An outdated simulator for the pathfinding service. See
Simulator Clientinstead. It is built using NextJS and uses a generated OpenAPI client to communicate with the pathfinding service. (It might no longer work as it was abandoned after the completion of the MDP checklist). - Robot - The STM codebase.
- Service - The pathfinding algorithm. It is built using Flask & exposes REST endpoints. These REST endpoints are documented using Swagger/OpenAPI 3.
- Simulator Client - A simulator for the pathfinding service. It is built using NextJS and uses a generated OpenAPI client to communicate with the pathfinding service. (It might no longer work as it was abandoned after the completion of the MDP checklist).
Other misc components:
- music - Unused Pokemon theme song that we originally wanted to blast out of a speaker attached to the robot. We unforunately did not have the time to achieve this.
Task 1:
Task 2:
