High-definition mapping is fundamental for self-driving vehicles. In this thesis we describe different approaches for online map validation whose goal is to verify if reality and map data are inconsistent. A probabilistic framework to perform the sensor fusion is defined and a spatial correlation is introduced to interpolate the information. The result is a probabilistic representation of the map whose assumed values represent the probability with which the map is valid in every point.

Map Validation for Autonomous Driving Systems

Fabris, Andrea
2020/2021

Abstract

High-definition mapping is fundamental for self-driving vehicles. In this thesis we describe different approaches for online map validation whose goal is to verify if reality and map data are inconsistent. A probabilistic framework to perform the sensor fusion is defined and a spatial correlation is introduced to interpolate the information. The result is a probabilistic representation of the map whose assumed values represent the probability with which the map is valid in every point.
2020-01-07
autonomous, driving
File in questo prodotto:
File Dimensione Formato  
andrea_fabris.pdf

accesso aperto

Dimensione 8.19 MB
Formato Adobe PDF
8.19 MB Adobe PDF Visualizza/Apri

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

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12608/22888