The thesis presents the development of a MATLAB-based framework for estimating the trajectory of a mobile platform operating in an agricultural environment and generating three-dimensional crop maps. The system is based on the ICP (Iterative Closest Point) algorithm integrated with sensor fusion techniques combining LiDAR, RGB, NIR, and IMU data, with the aim of improving localization accuracy. Performance is evaluated by comparing the results with GPS ground-truth measurements and by optimizing the main algorithm parameters. From the resulting maps, crops are automatically segmented and their corresponding agronomic parameters are extracted. Finally, the work includes an interactive interface that enables simple and intuitive access to the data.
La tesi presenta lo sviluppo di un framework in MATLAB per la stima della traiettoria di una piattaforma mobile in ambito agricolo e la generazione di mappe tridimensionali delle colture. Il sistema si basa sull’algoritmo ICP integrato con tecniche di sensor fusion tra LiDAR, RGB, NIR e IMU, con l’obiettivo di migliorare l’accuratezza della localizzazione. Le prestazioni vengono valutate confrontando i risultati con dati GPS di riferimento e ottimizzando i principali parametri dell’algoritmo. A partire dalle mappe ottenute, vengono segmentate automaticamente le colture e vengono estratti i relativi parametri agronomici. Infine, il lavoro include un’interfaccia interattiva che consente una consultazione semplice e immediata dei dati.
3D MAPPING FOR AUTONOMOUS ROBOTS IN AGRICULTURAL APPLICATIONS
GIGOTTI, ROSSANA
2025/2026
Abstract
The thesis presents the development of a MATLAB-based framework for estimating the trajectory of a mobile platform operating in an agricultural environment and generating three-dimensional crop maps. The system is based on the ICP (Iterative Closest Point) algorithm integrated with sensor fusion techniques combining LiDAR, RGB, NIR, and IMU data, with the aim of improving localization accuracy. Performance is evaluated by comparing the results with GPS ground-truth measurements and by optimizing the main algorithm parameters. From the resulting maps, crops are automatically segmented and their corresponding agronomic parameters are extracted. Finally, the work includes an interactive interface that enables simple and intuitive access to the data.| File | Dimensione | Formato | |
|---|---|---|---|
|
TESI_ROSSANA_GIGOTTI.pdf
non disponibili
Dimensione
28.72 MB
Formato
Adobe PDF
|
28.72 MB | Adobe PDF |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14253/67883