Currently, the safety of operators working on railway construction site depends to a large extent on the ability to correctly follow a series of requirements and procedures which, however, very often do not prove effective in avoiding even fatal accidents. The thesis is aimed at the development of algorithms, based on AI, integrating information made available from cameras, thermocameras, lidar and radars. This will allow to reach a new safety paradigm in which machines will be able to autonomously recognize potentially dangerous conditions and intervene to guarantee the safety of the operators
La sicurezza degli operatori che lavorano nei cantieri ferroviari dipende attualmente in larga misura dalla capacità di seguire correttamente una serie di requisiti e procedure che, tuttavia, molto spesso non si rivelano efficaci nell’evitare anche incidenti mortali. La tesi è volta allo sviluppo di algoritmi basati sull’intelligenza artificiale, in grado di integrare le informazioni fornite da telecamere, termocamere, lidar e radar. Questo consentirà di raggiungere un nuovo paradigma di sicurezza, in cui le macchine saranno in grado di riconoscere autonomamente condizioni potenzialmente pericolose e intervenire per garantire la sicurezza degli operatori.
Improving safety of railway operating machines through Artificial Inteligence
HORVAT, BORIS
2024/2025
Abstract
Currently, the safety of operators working on railway construction site depends to a large extent on the ability to correctly follow a series of requirements and procedures which, however, very often do not prove effective in avoiding even fatal accidents. The thesis is aimed at the development of algorithms, based on AI, integrating information made available from cameras, thermocameras, lidar and radars. This will allow to reach a new safety paradigm in which machines will be able to autonomously recognize potentially dangerous conditions and intervene to guarantee the safety of the operators| File | Dimensione | Formato | |
|---|---|---|---|
|
2025_10_Horvat.pdf
accessibile in internet per tutti a partire dal 22/09/2026
Descrizione: Thesis File
Dimensione
85.19 MB
Formato
Adobe PDF
|
85.19 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/243436