Happy Hedgehog - lawnmower robot with sensors to detect and protect wild animals

The underlying project - in cooperation with the FHNW - deals with the implementation of machine morality into a lawnmower robot. Based on a prototype it will be shown how the lawnmower robot detects a hedgehog and handles the situation based on machine ethics rules.

Beugger, Michel & Chang, Vay Lien & Graf, Emanuel & Bollier, Kevin, 2020

Type of Thesis Projektarbeit/Praxisprojekt
Client FHNW School of Business
Supervisor Schwaferts, Dino, Richards, Bradley
Views: 38 - Downloads: 7
Machine ethics currently mainly refers to humans, although there are many possible use cases in the animal world. A big issue in the field of self-driving lawnmowers are injured or dead hedgehogs due to missing protection mechanism. Ethical aspects in regard to animals are not implemented in today's lawnmowers - essential sensors are missing in order to execute a meaningful decison of action.
The process was divided into three main phases. The first phase dealt with the evaluation of a suitable lawnmower as foundation for additional sensors. Due to access restriction to the coding part of already existing lawnmowers, we decided to imitate such a machine with the help of a so-called PiCar. In the second step, various ideas were evaluated regarding sensors. Finally, the PiCar code required several adjustments for the interaction with the sensors. Furthermore, a model had to be trained to recognize hedgehogs where the part of machine learning came into the project.
The project ensures appropriate actions taken to prevent animal (hedgehog) harm. The prototype can be used for presentations to show the implemented solution. In addition, the project provides a solid foundation for further research in the field of machine ethics with focus on animals. The project is easy expandable for other animals and adaptable by additional sensors.
Studyprogram: Business Information Technology (Bachelor)
Keywords hedgehog, lawnmower, machine ethics, machine learning
Confidentiality: öffentlich
Type of Thesis
Projektarbeit/Praxisprojekt
Client
FHNW School of Business, Olten
Authors
Beugger, Michel & Chang, Vay Lien & Graf, Emanuel & Bollier, Kevin
Supervisor
Schwaferts, Dino, Richards, Bradley
Publication Year
2020
Thesis Language
English
Confidentiality
Public
Studyprogram
Business Information Technology (Bachelor)
Location
Basel
Keywords
hedgehog, lawnmower, machine ethics, machine learning