The necessity of a reliable indoor navigation system in recent year is trending up with the increased time spent in large buildings such as big malls, hospitals or museums. GPS is a well-known technology already utilized for outdoor environment but the satellite signal cannot reliably pass into thick constructions and therefore researchers are looking for the best strategy to localize people in this context as well. Such problem is called simultaneous localization and mapping (SLAM) and it is the first step in order to build an indoor positioning system (IPS). Many strategies were adopted, mostly based on beacons which are required to be installed in the environment; however this thesis aims to propose an approach that does not involve external sensors or complex setups to run. Project Tango is an early attempt by Google to build a platform able to recognize what lies around a mobile device and our research wants to exploit the feature in order to bring a navigation system for indoors. This thesis will attempt first to solve the SLAM challenge using this technology without an external structural help and then subsequently we will try to build a navigation system on top of it. The final result is then moved on Tango Navigator, a mobile application developed for indoor navigation. The software consists of two parts: a mapping phase and a subsequently navigation phase. The application is tested in a real environment and early experiments show good promises for further developments going forward.
La necessità di un sistema di navigazione per interni affidabile è in salita negli ultimi anni con l’aumentare del tempo trascorso in ampi complessi come grandi centri commerciali, ospedali o musei. Il GPS è una tecnologia ben conosciuta già utilizzata per ambienti esterni, tuttavia il segnale satellitare non è in grado di oltrepassare in maniera affidabile le costruzioni dense e per tale ragione i ricercatori stanno cercando la strategia migliore per localizzare le persone anche in tale contesto. Questo problema è chiamato simultaneous localization and mapping (SLAM) ed è il primo passo per costruire un indoor positioning system (IPS). Molte strategie sono state adottate, prevalentemente basate su beacon che richiedono di essere installati nell’ambiente; tuttavia il nostro scopo è di proporre un approcio che non coinvolga sensori esterni o configurazioni complicate per funzionare. Project Tango è un primo tentativo di Google di costruire una piattaforma in grado di riconoscere ciò che circonda un dispositivo mobile e la nostra ricerca vuole sfruttare questa capacità per portare un sistema di navigazione per interni. La nostra tesi proverà innanzitutto a risolvere la sfida della SLAM usando questa tecnologia senza aiuti strutturali esterni e conseguentemente proveremo a costruire un sistema di navigazione sopra di esso. Il risultato finale è poi trasferito su Tango Navigator, un’applicazione mobile per la navigazione indoor. Il software consiste di due parti: una fase di mappatura ed una successiva fase di navigazione. L’applicazione è testata in un ambiente reale e i primi esperimenti mostrano buone promesse per gli ulteriori sviluppi nel futuro.
Tango Navigator : an indoor navigation system
TRAN, KHANH HUY PAOLO
2016/2017
Abstract
The necessity of a reliable indoor navigation system in recent year is trending up with the increased time spent in large buildings such as big malls, hospitals or museums. GPS is a well-known technology already utilized for outdoor environment but the satellite signal cannot reliably pass into thick constructions and therefore researchers are looking for the best strategy to localize people in this context as well. Such problem is called simultaneous localization and mapping (SLAM) and it is the first step in order to build an indoor positioning system (IPS). Many strategies were adopted, mostly based on beacons which are required to be installed in the environment; however this thesis aims to propose an approach that does not involve external sensors or complex setups to run. Project Tango is an early attempt by Google to build a platform able to recognize what lies around a mobile device and our research wants to exploit the feature in order to bring a navigation system for indoors. This thesis will attempt first to solve the SLAM challenge using this technology without an external structural help and then subsequently we will try to build a navigation system on top of it. The final result is then moved on Tango Navigator, a mobile application developed for indoor navigation. The software consists of two parts: a mapping phase and a subsequently navigation phase. The application is tested in a real environment and early experiments show good promises for further developments going forward.File | Dimensione | Formato | |
---|---|---|---|
2018_04_Tran.pdf
accessibile in internet per tutti
Descrizione: Testo della tesi
Dimensione
16.97 MB
Formato
Adobe PDF
|
16.97 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/10589/140203