Dirt Classification Scheme on a built-up Robot Platform
Project Description

This project aims to build a dirt classification scheme on the embedded system of a locomotion robot. A dataset of various targeted objects need to be built. Neural network(e.g. Tiny YOLOv2) may be utilized for training and developping on a robot’s embedded system with a KPU chip.

Supervisor
SHI Ling
Quota
2
Course type
UROP1100
UROP2100
UROP3100
UROP4100
Applicant's Roles

Build a dataset of various targeted objects and train a neural network for object detection.

Applicant's Learning Objectives

1. Building a dataset that meets the targeted conditions.
2. Developing a neural network architecture for dirt classifications.

Complexity of the project
Challenging