Prepoznavanje događaja u okruženju Interneta stvari primjenom strojnog učenja

undergraduate thesis

• Title (english): Events identification in the Internet of Things environment by applying machine learning
• Author: Matko Nikolić
• Mentor: Marko Jurčević (mentor)
• Committee member: Marko Jurčević (predsjednik povjerenstva)
• Committee member: Ivan Leniček (član povjerenstva)
• Committee member: Jure Konjevod (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2023-07-07, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

TinyML je novi i rastući način primjene strojnog učenja koji kombinira ugradbene uređaje s modelima strojnog učenja. Prednosti su: decentralizacija, brža obrada podataka, privatnost, rasterećenje središnjeg sustava. U teoretskom dijelu ovoga rada obrađeni su najpopularniji ugradbeni uređaji koji podržavaju TinyML kao i najpopularniji razvojni okviri. U praktičnome dijelu zadatka je razvijen model strojnog učenje za dva uređaja koja emuliraju palice korištene za signaliziranje za parking zrakoplova te na temelju akceleracije detektiraju o kojem signalu se radi. Korišteni uređaji su dva FiPy modula i dvije Pysense 2.0x senzorske pločice.

Abstract (english)

TinyML is a new and growing way of applying machine learning. TinyML is a combination of embedded devices with machine learning and the main advantages are: decentralization, faster data processing, enhanced privacy and relieving the servers. In the theoretical part of this thesis are described most popular embedded devices which support TinyML and the most popular frameworks. In the practical part, a model of machine learning is trained for two devices which represent wands used for aircraft marshalling signals and different signals are detected based on acceleration. For hardware, two FiPy development modules were used and two Pysense 2.0x sensor shields.

• Keywords: internet stvari; klasifikacija; strojno učenje; prepoznavanje pokreta; ugradbeni uređaji; TinyML; slučajne šume
• Keywords (english): internet of things; classification; machine learning; movement classification; embedded; TinyML; random forrest
• Language: croatian
• URN:NBN: urn:nbn:hr:168:541818
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-07-13 10:24:31