This thesis proposes a machine learning-based approach for detecting falls from bikes that lead to the activation of an airbag jacket. The system is designed to utilize sensors integrated within the jacket to detect anomalies in riding patterns and identify potential falls. Anomaly detection algorithms are used to analyze the data collected from the sensors, and various machine learning techniques are employed to differentiate normal from abnormal riding patterns. The proposed system is evaluated through the use of real-world sensor data collected during bike rides by people wearing the airbag jacket. The results demonstrate the effectiveness of the proposed approach in accurately identifying falls and triggering the airbag jacket. The thesis provides insight into the development of a technology that has the potential to improve rider safety and reduce the risk of serious injury in the event of a fall from a bike.
An Anomaly Detection-based approach to Crash Detection in dynamic sports.
BERTIPAGLIA, BEATRICE SOFIA
2022/2023
Abstract
This thesis proposes a machine learning-based approach for detecting falls from bikes that lead to the activation of an airbag jacket. The system is designed to utilize sensors integrated within the jacket to detect anomalies in riding patterns and identify potential falls. Anomaly detection algorithms are used to analyze the data collected from the sensors, and various machine learning techniques are employed to differentiate normal from abnormal riding patterns. The proposed system is evaluated through the use of real-world sensor data collected during bike rides by people wearing the airbag jacket. The results demonstrate the effectiveness of the proposed approach in accurately identifying falls and triggering the airbag jacket. The thesis provides insight into the development of a technology that has the potential to improve rider safety and reduce the risk of serious injury in the event of a fall from a bike.File | Dimensione | Formato | |
---|---|---|---|
Bertipaglia_BeatriceSofia.pdf
accesso riservato
Dimensione
7.91 MB
Formato
Adobe PDF
|
7.91 MB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/52261