Savjetovanje pravilnog držanja tijela tijekom bicikliranja primjenom strojnog učenja

undergraduate thesis

• Title (english): Advising on correct body posture in cycling by machine learning
• Author: Lovro Švenda
• Mentor: Vinko Lešić (mentor)
• Committee member: Vinko Lešić (član povjerenstva)
• Committee member: Filip Vrbanc (član povjerenstva)
• Committee member: Mihael Jakšić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2024-07-04, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Bike fitting je ključan za bicikliste, bilo profesionalce ili rekreativce, jer osigurava udobnost i smanjuje rizik od ozljeda. Kroz prilagodbu bicikla prema individualnim karakteristikama vozača, bike fitting omogućava efikasniju i ugodniju vožnju. Primjena strojnog učenja, posebno kroz algoritme kao što su linearna regresija, stabla odluka i neuronske mreže, omogućila je razvijanje aplikacije koja može pomoći biciklistima da sami prilagode svoj bicikl. Analiza videozapisa biciklista pružila je podatke potrebne za treniranje modela, koji su kasnije korišteni za predikciju optimalnih postavki bicikla. Uvođenjem šuma u podatke, modeli su postali robusniji i otporniji na varijabilnosti u stvarnim uvjetima. Rezultati pokazuju da modeli strojnog učenja, poput slučajne šume, pružaju točne predikcije i mogu pomoći u optimizaciji postavki bicikla. Ovi modeli mogu identificirati nepravilnosti u položaju vozača i preporučiti prilagodbe koje poboljšavaju udobnost i performanse. U budućnosti, daljnje istraživanje može uključivati proširenje skupa podataka kako bi se obuhvatili različiti tipovi bicikala, brzina vožnje, otpor zraka, tjelesne karakteristike vozača kao što su visina, težina, duljina nogu… Također, integracija senzora za praćenje stvarnog vremena može omogućiti dinamičko prilagođavanje postavki bicikla tijekom vožnje, pružajući još veću razinu personalizacije i udobnosti. Kroz kontinuirano unapređenje algoritama i povećanje baze podataka, bike fitting
može postati pristupačniji i učinkovitiji za sve bicikliste.

Abstract (english)

Bike fitting is crucial for cyclists, whether professionals or recreational riders, as it ensures comfort and reduces the risk of injuries. By adjusting the bike according to the individual characteristics of the rider, bike fitting enables a more efficient and enjoyable ride. The application of machine learning, particularly through algorithms such as linear regression, decision trees, and neural networks, has enabled the development of an application that can help cyclists adjust their bikes themselves. Video analysis of cyclists provided the necessary data to train the models, which were later used to predict optimal bike settings. By introducing noise into the data, the models became more robust and resistant to variability in real-world conditions. The results show that machine learning models, such as random forests, provide accurate predictions and can help optimize bike settings. These models can identify irregularities in the rider's position and recommend adjustments that improve comfort and performance. In the future, further research may include expanding the dataset to cover different types of bikes, riding speeds, air resistance, and physical characteristics of riders such as height, weight, and leg length. Additionally, the integration of real-time monitoring sensors can enable dynamic adjustment of bike settings during the ride, providing an even greater level of personalization and comfort. Through continuous improvement of algorithms and expansion of the database, bike fitting can become more accessible and effective for all cyclists.

• Keywords: Bike fitting; strojno učenje; slučajna šuma
• Keywords (english): Bike fitting; machine learning; random forest
• Language: croatian
• URN:NBN: urn:nbn:hr:168:279106
• 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: 2024-10-12 22:19:29