A hardware accelerator for ORB-SLAM

Simultaneous Localization And Mapping (SLAM) is a key component of self-driving cars. We study ORB-SLAM, a SLAM state-of-the-art solution, and develop a hardware accelerator for a critical part of it: ORB feature extraction. The accelerator achieve 8x speedup and 2000x energy consumption reduction.

Detalles Bibliográficos
Autor: Taranco, Raúl
Tipo de recurso: tesis de maestría
Fecha de publicación:2019
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/176803
Acceso en línea:https://hdl.handle.net/2117/176803
Access Level:acceso abierto
Palabra clave:Computer architecture
Software engineering
orb
slam
orb-slam
accelerador hardware
hardware accelerator
Arquitectura d'ordinadors
Enginyeria de programari
Àrees temàtiques de la UPC::Informàtica
Descripción
Sumario:Simultaneous Localization And Mapping (SLAM) is a key component of self-driving cars. We study ORB-SLAM, a SLAM state-of-the-art solution, and develop a hardware accelerator for a critical part of it: ORB feature extraction. The accelerator achieve 8x speedup and 2000x energy consumption reduction.