Target Detection, Indoor Scene Classification, Visual and three-dimensional mapping for service robots in Healthcare

Loading...
Thumbnail Image

URL

Journal Title

Journal ISSN

Volume Title

Sähkötekniikan korkeakoulu | Master's thesis

Date

2021-06-14

Department

Major/Subject

Autonomous Systems

Mcode

ELEC3055

Degree programme

Master's Programme in ICT Innovation

Language

en

Pages

63

Series

Abstract

This thesis details work on different techniques used to implement service robots for indoor environments. This included two machine learning techniques: Target Detection and Indoor scene classification as well as two computer vision techniques: Visual mapping and three-dimensional mapping. Using these techniques, we tried to make service robots better in environments like hospitals. Assistance provided by service robots will help staff in managing tedious tasks without any problem. We used different techniques for mapping and localization so service robots can autonomously navigate from floor to floor. Depth cameras were used to make recognition and mapping better for indoor environments.

Description

Supervisor

Fontanelli, Daniele

Thesis advisor

Zhou, Quan

Keywords

localization, mapping, object recognition, scene recognition, visual mapping, three dimensional mapping

Other note

Citation