ORCiD

Dongbin Lee: 0000-0002-5307-0374

Department

Electrical and Computer Engineering

Document Type

Article

Publication Title

Online Journal of Robotics & Automation Technology

ISSN

2832-790X

Volume

2

Issue

2

DOI

10.33552/OJRAT.2023.02.000534

First Page

1

Last Page

16

Publication Date

2023

Abstract

This paper examines not only the main unresolved theoretical question, centered on the responsibility gap issue in autonomous robots, but also examines whether the robot can be responsible via moral conditioning and our proposed method bumper theory. Our scientific inquiry aims to discuss what is required to meet the goals and objectives of moral robotics, as well as promote dialogues and acquire quantitative results that hope to make robots responsible moral agents. Robots are morally conditioned with bumper theory by first detecting human-beings, dogs, or cars to avoid collision and reinforce prosocial morality. We adopted a scientific method such as deep machine learning and computer vision incorporated into a single board computer (Jetson Nano) to start detection and recognition processes of identifying objects including humans. This paper explores different models of computer vision and finds that Yolov4-tiny is the best use case on the constrained environment of the Jetson Nano to work towards building a system to solve this responsibility gap.

Creative Commons License

Creative Commons Attribution-NonCommercial 4.0 International License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License

Download

Share

COinS