Javascript must be enabled to continue!
Information metrics for localization and mapping
View through CrossRef
Decades of research have made possible the existence of several autonomous systems that successfully and efficiently navigate within a variety of environments under certain conditions. One core technology that has allowed this is simultaneous localization and mapping (SLAM), the process of building a representation of the environment while localizing the robot in it.
State-of-the-art solutions to the SLAM problem still rely, however, on heuristic decisions and options set by the user. In this thesis we search for principled solutions to various aspects of the localization and mapping problem with the help of information metrics.
One such aspect is the issue of scalability. In SLAM, the problem size grows indefinitely as the experiment goes by, increasing computational resource demands. To maintain the problem tractable, we develop methods to build an approximation to the original network of constraints of the SLAM problem by reducing its size while maintaining its sparsity. In this thesis we propose three methods to build the topology of such approximated network, and two methods to perform the approximation itself.
In addition, SLAM is a passive application. It means, it does not drive the robot. The problem of driving the robot with the aim of both accurately localizing the robot and mapping the environment is called active SLAM. In this problem two normally opposite forces drive the robot, one to new places discovering unknown regions and another to revisit previous configurations to improve localization. As opposed to heuristics, in this thesis we pose the problem as the joint minimization of both map and trajectory estimation uncertainties, and present four different active SLAM approaches based on entropy-reduction formulation.
All methods presented in this thesis have been rigorously validated in both synthetic and real datasets.
Dècades de recerca han fet possible l’existència de nombrosos sistemes autònoms que naveguen eficaçment i eficient per varietat d’entorns sota certes condicions. Una de les principals tecnologies que ho han fet possible és la localització i mapeig simultanis (SLAM), el procés de crear una representació de l’entorn mentre es localitza el robot en aquesta.
De tota manera, els algoritmes d’SLAM de l’estat de l’art encara basen moltes decisions en heurístiques i opcions a escollir per l’usuari final. Aquesta tesi persegueix solucions fonamentades per a varietat d’aspectes del problema de localització i mappeig amb l’ajuda de mesures d’informació.
Un d’aquests aspectes és l’escalabilitat. En SLAM, el problema creix indefinidament a mesura que l’experiment avança fent créixer la demanda de recursos computacionals. Per mantenir el problema tractable, desenvolupem mètodes per construir una aproximació de la xarxa de restriccions original del problema d’SLAM, reduint així el seu tamany a l’hora que es manté la seva naturalesa dispersa. En aquesta tesi, proposem tres métodes per confeccionar la topologia de l’approximació i dos mètodes per calcular l’aproximació pròpiament.
A més, l’SLAM és una aplicació passiva. És a dir que no dirigeix el robot. El problema de guiar el robot amb els objectius de localitzar el robot i mapejar l’entorn amb precisió es diu SLAM actiu. En aquest problema, dues forces normalment oposades guien el robot, una cap a llocs nous descobrint regions desconegudes i l’altra a revisitar prèvies configuracions per millorar la localització. En contraposició amb mètodes heurístics, en aquesta tesi plantegem el problema com una minimització de l’incertesa tant en el mapa com en l’estimació de la trajectòria feta i presentem quatre mètodes d’SLAM actiu basats en la reducció de l’entropia.
Tots els mètodes presentats en aquesta tesi han estat rigurosament validats tant en sèries de dades sintètiques com en reals.
Title: Information metrics for localization and mapping
Description:
Decades of research have made possible the existence of several autonomous systems that successfully and efficiently navigate within a variety of environments under certain conditions.
One core technology that has allowed this is simultaneous localization and mapping (SLAM), the process of building a representation of the environment while localizing the robot in it.
State-of-the-art solutions to the SLAM problem still rely, however, on heuristic decisions and options set by the user.
In this thesis we search for principled solutions to various aspects of the localization and mapping problem with the help of information metrics.
One such aspect is the issue of scalability.
In SLAM, the problem size grows indefinitely as the experiment goes by, increasing computational resource demands.
To maintain the problem tractable, we develop methods to build an approximation to the original network of constraints of the SLAM problem by reducing its size while maintaining its sparsity.
In this thesis we propose three methods to build the topology of such approximated network, and two methods to perform the approximation itself.
In addition, SLAM is a passive application.
It means, it does not drive the robot.
The problem of driving the robot with the aim of both accurately localizing the robot and mapping the environment is called active SLAM.
In this problem two normally opposite forces drive the robot, one to new places discovering unknown regions and another to revisit previous configurations to improve localization.
As opposed to heuristics, in this thesis we pose the problem as the joint minimization of both map and trajectory estimation uncertainties, and present four different active SLAM approaches based on entropy-reduction formulation.
All methods presented in this thesis have been rigorously validated in both synthetic and real datasets.
Dècades de recerca han fet possible l’existència de nombrosos sistemes autònoms que naveguen eficaçment i eficient per varietat d’entorns sota certes condicions.
Una de les principals tecnologies que ho han fet possible és la localització i mapeig simultanis (SLAM), el procés de crear una representació de l’entorn mentre es localitza el robot en aquesta.
De tota manera, els algoritmes d’SLAM de l’estat de l’art encara basen moltes decisions en heurístiques i opcions a escollir per l’usuari final.
Aquesta tesi persegueix solucions fonamentades per a varietat d’aspectes del problema de localització i mappeig amb l’ajuda de mesures d’informació.
Un d’aquests aspectes és l’escalabilitat.
En SLAM, el problema creix indefinidament a mesura que l’experiment avança fent créixer la demanda de recursos computacionals.
Per mantenir el problema tractable, desenvolupem mètodes per construir una aproximació de la xarxa de restriccions original del problema d’SLAM, reduint així el seu tamany a l’hora que es manté la seva naturalesa dispersa.
En aquesta tesi, proposem tres métodes per confeccionar la topologia de l’approximació i dos mètodes per calcular l’aproximació pròpiament.
A més, l’SLAM és una aplicació passiva.
És a dir que no dirigeix el robot.
El problema de guiar el robot amb els objectius de localitzar el robot i mapejar l’entorn amb precisió es diu SLAM actiu.
En aquest problema, dues forces normalment oposades guien el robot, una cap a llocs nous descobrint regions desconegudes i l’altra a revisitar prèvies configuracions per millorar la localització.
En contraposició amb mètodes heurístics, en aquesta tesi plantegem el problema com una minimització de l’incertesa tant en el mapa com en l’estimació de la trajectòria feta i presentem quatre mètodes d’SLAM actiu basats en la reducció de l’entropia.
Tots els mètodes presentats en aquesta tesi han estat rigurosament validats tant en sèries de dades sintètiques com en reals.
Related Results
Indoor Localization System Based on RSSI-APIT Algorithm
Indoor Localization System Based on RSSI-APIT Algorithm
An indoor localization system based on the RSSI-APIT algorithm is designed in this study. Integrated RSSI (received signal strength indication) and non-ranging APIT (approximate pe...
Comparison of LA and PVC mapping using OCTARAY and OPTRELL catheters
Comparison of LA and PVC mapping using OCTARAY and OPTRELL catheters
AbstractBackgroundMultielectrode mapping catheters, such as the OCTARAY and OPTRELL, are essential in creating myocardial electroanatomical mapping in arrhythmias. The OCTARAY is a...
THE SECURITY AND PRIVACY MEASURING SYSTEM FOR THE INTERNET OF THINGS DEVICES
THE SECURITY AND PRIVACY MEASURING SYSTEM FOR THE INTERNET OF THINGS DEVICES
The purpose of the article: elimination of the gap in existing need in the set of clear and objective security and privacy metrics for the IoT devices users and manufacturers and a...
Comparison of Phase-Based 3D Near-Field Source Localization Techniques for UHF RFID
Comparison of Phase-Based 3D Near-Field Source Localization Techniques for UHF RFID
In this paper, we present multiple techniques for phase-based narrowband backscatter tag localization in three-dimensional space with planar antenna arrays or synthetic apertures. ...
Anchor Nodes Placement for Effective Passive Localization
Anchor Nodes Placement for Effective Passive Localization
Wireless sensor networks are composed of sensor nodes, which can monitor an environment and observe events of interest. These networks are applied in various fields including but n...
Corporate environmental reporting: what's in a metric?
Corporate environmental reporting: what's in a metric?
AbstractAlthough there has been increased attention to corporate environmental reports (CERs), there has yet to be a close examination of the metrics used in these reports. Metrics...
Impact of Model Knowledge on Acoustic Emission Source Localization Accuracy
Impact of Model Knowledge on Acoustic Emission Source Localization Accuracy
Reliable and precise damage localization in mechanical structures is of high importance in the context of structural health monitoring (SHM). The acoustic emission (AE) method has ...
Impact of Model Knowledge on Acoustic Emission Source Localization Accuracy
Impact of Model Knowledge on Acoustic Emission Source Localization Accuracy
Reliable and precise damage localization in mechanical structures is of high importance in the context of structural health monitoring (SHM). The acoustic emission (AE) method has ...

