Uporaba Kalmanovega filtra pri povezavi različnih senzorjev za določanje položaja v cestnem mobilnem kartirnem sistemu

Povzetek

Ključne besede

Podatki

Jezik:	Slovenski jezik
Leto izida:	2009
Izvor:	Ljubljana
Tipologija:	2.09 - Magistrsko delo
Organizacija:	UL FGG - Fakulteta za gradbeništvo in geodezijo
Založnik:	[U. Ranfl]
UDK:	528.51:528.8+625.7(043.3)
COBISS:	4553057
Št. ogledov:	1564
Št. prenosov:	522
Ocena:	0 (0 glasov)
Metapodatki:

Ostali podatki

Sekundarni jezik:	Angleški jezik
Sekundarni naslov:	Application of Kalman filter in multi-sensor position determination for road mobile mapping system
Sekundarni povzetek:	Technology of mobile mapping systems is used to provide metric and georeferenced spatial data. We can achieve geolocation of spatial data by the means of measurement techniques that can measure data while moving. These techniques should be independent from external disturbances. State of the art systems that integrate kinematic GPS method and inertial navigation are proved as appropriate to fulfill given demands. We can measure accurate coordinates for almost all time epochs with GPS system, but GPS measurements can be influenced by the external impacts so we can not achieve demanded accuracy. In such cases we have to use complementary inertial navigational system which is independent from external impacts. We can bridge only short GPS gaps with the inertial system because accuracy of the measured coordinates with inertial system rapidly decreases. We can successfully merge accurate GPS measurements with, form external impacts independent, inertial measurements, to provide continuous coordinates. Measurements from both systems are integrated by the means of Kalman filtering. Kalman filtering is widely used for calculating navigational data. Recursive algorithm is implemented in two steps: prediction and update of estimated parameters. In the first step is used dynamical model to describe time transition of state vector. In he second step are combined data form the previous step with current measurement update.
Sekundarne ključne besede:	geodesy;master of science thesis;GPS;INS;integrated navigation;Kalman filtering;mobile mapping system;
URN:	URN:NBN:SI
Vrsta datoteke:	application/pdf
Vrsta dela (COBISS):	Magistrsko delo
Komentar na gradivo:	Univ. v Ljubljani, Fak. za gradbeništvo in geodezijo
Strani:	XIV, 123 str.
Vrsta dela (ePrints):	thesis
Naslov (ePrints):	Application of Kalman filter in multi-sensor position determination for road mobile mapping system
Ključne besede (ePrints):	mobilni kartirni sistem;GPS;INS;integrirana navigacija;Kalmanov filter
Ključne besede (ePrints, sekundarni jezik):	GPS;INS;integrated navigation;Kalman filtering;mobile mapping system
Povzetek (ePrints):	Tehnologija mobilnih kartirnih sistemov je namenjena pridobivanju metričnih in geolociranih podatkov o prostoru. Za zagotovitev geolokacije prostorskih podatkov moramo uporabiti merske tehnike, ki omogočajo določevanje koordinat med premikanjem in so čimbolj neodvisne od zunanjih vplivov. V praksi se kot za zelo primerne izkažejo integrirani sistemi, ki združujejo kinematično GPS-izmero in inercialno navigacijo. S sistemom GPS lahko zagotavljamo natančne koordinate za večino primerov. GPS-meritve so lahko zelo obremenjene z zunanjimi vplivi in zato v določenih okoliščinah s sistemom ne moremo določiti koordinat z ustrezno natančnostjo. V takih premerih uporabimo komplementaren inercialni sistem, ki je neodvisen od zunanjih dejavnikov. Natančnost določevanja koordinat z inercialnimi sistemi s časom zelo hitro pada, zato lahko s takim sistemom premoščamo le kratkotrajne vrzeli v GPS-meritvah. Natančnost GPS-meritev in neodvisnost inercialnih meritev od zunanjih vplivov lahko uspešno uporabimo za neprekinjeno določevanje položaja. Meritve obeh sistemov združujemo s pomočjo Kalmanovega filtra. Kalmanov filter na široko uporabljamo pri preračunavanju podatkov za navigacijske potrebe. Rekurzivni postopek izvedemo v dveh stopnjah: napoved in obnovitev ocenjenih parametrov. V prvem koraku uporabimo dinamični model, ki opisuje povezave med neznankami v času. V drugem koraku kombiniramo podatke meritev, ki smo jih pridobili v predhodnem koraku (v napovedi) z novimi meritvami. V nalogi so predstavljene teoretične osnove navigacijske komponente mobilnega kartirnega sistema WideoCar 3 (senzorji, navigacijske enačbe, združevanje podatkov) in prikazan praktični primer uporabe združevanja GPS/INS-meritev s pomočjo Kalmanovega filtra. Praktični primer je izveden na podlagi testnih meritev in obdelan s programskim paketom Inertial Explorer.
Povzetek (ePrints, sekundarni jezik):	Technology of mobile mapping systems is used to provide metric and georeferenced spatial data. We can achieve geolocation of spatial data by the means of measurement techniques that can measure data while moving. These techniques should be independent from external disturbances. State of the art systems that integrate kinematic GPS method and inertial navigation are proved as appropriate to fulfill given demands. We can measure accurate coordinates for almost all time epochs with GPS system, but GPS measurements can be influenced by the external impacts so we can not achieve demanded accuracy. In such cases we have to use complementary inertial navigational system which is independent from external impacts. We can bridge only short GPS gaps with the inertial system because accuracy of the measured coordinates with inertial system rapidly decreases. We can successfully merge accurate GPS measurements with, form external impacts independent, inertial measurements, to provide continuous coordinates. Measurements from both systems are integrated by the means of Kalman filtering. Kalman filtering is widely used for calculating navigational data. Recursive algorithm is implemented in two steps: prediction and update of estimated parameters. In the first step is used dynamical model to describe time transition of state vector. In he second step are combined data form the previous step with current measurement update.
Ključne besede (ePrints, sekundarni jezik):	GPS;INS;integrated navigation;Kalman filtering;mobile mapping system
ID:	8311015