FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2964976771> ?p ?o ?g. }

• W2964976771 abstract "This thesis presents the design, implementation, and validation of a novel nonlinearfilteringbased Visual Inertial Odometry (VIO) framework for robotic navigation in GPSdeniedenvironments. The system attempts to track the vehicle’s ego-motion at each timeinstant while capturing the benefits of both the camera information and the Inertial MeasurementUnit (IMU). VIO demands considerable computational resources and processingtime, and this makes the hardware implementation quite challenging for micro- and nanoroboticsystems. In many cases, the VIO process selects a small subset of tracked featuresto reduce the computational cost. VIO estimation also suffers from the inevitable accumulationof error. This limitation makes the estimation gradually diverge and even fail totrack the vehicle trajectory over long-term operation. Deploying optimization for the entiretrajectory helps to minimize the accumulative errors, but increases the computational costsignificantly. The VIO hardware implementation can utilize a more powerful processorand specialized hardware computing platforms, such as Field Programmable Gate Arrays,Graphics Processing Units and Application-Specific Integrated Circuits, to accelerate theexecution. However, the computation still needs to perform identical computational stepswith similar complexity. Processing data at a higher frequency increases energy consumptionsignificantly. The development of advanced hardware systems is also expensive andtime-consuming. Consequently, the approach of developing an efficient algorithm will bebeneficial with or without hardware acceleration. The research described in this thesisproposes multiple solutions to accelerate the visual inertial odometry computation whilemaintaining a comparative estimation accuracy over long-term operation among state-ofthe-art algorithms.This research has resulted in three significant contributions. First, this research involvedthe design and validation of a novel nonlinear filtering sensor-fusion algorithm using trifocaltensor geometry and a cubature Kalman filter. The combination has handled the systemnonlinearity effectively, while reducing the computational cost and system complexity significantly.Second, this research develops two solutions to address the error accumulationissue. For standalone self-localization projects, the first solution applies a local optimizationprocedure for the measurement update, which performs multiple corrections on a singlemeasurement to optimize the latest filter state and covariance. For larger navigationprojects, the second solution integrates VIO with additional pseudo-ranging measurementsbetween the vehicle and multiple beacons in order to bound the accumulative errors. Third,this research develops a novel parallel-processing VIO algorithm to speed up the executionusing a multi-core CPU. This allows the distribution of the filtering computation on eachcore to process and optimize each feature measurement update independently.The performance of the proposed visual inertial odometry framework is evaluated usingpublicly-available self-localization datasets, for comparison with some other open-sourcealgorithms. The results illustrate that a proposed VIO framework is able to improve theVIO’s computational efficiency without the installation of specialized hardware computingplatforms and advanced software libraries." @default.
• W2964976771 created "2019-08-13" @default.
• W2964976771 date "2019-05-01" @default.
• W2964976771 modified "2023-09-27" @default.
• W2964976771 title "Computationally-efficient visual inertial odometry for autonomous vehicle" @default.
• W2964976771 hasPublicationYear "2019" @default.
• W2964976771 type Work @default.
• W2964976771 sameAs 2964976771 @default.
• W2964976771 citedByCount "0" @default.
• W2964976771 crossrefType "dissertation" @default.
• W2964976771 hasConcept C111919701 @default.
• W2964976771 hasConcept C11413529 @default.
• W2964976771 hasConcept C121332964 @default.
• W2964976771 hasConcept C1276947 @default.
• W2964976771 hasConcept C13662910 @default.
• W2964976771 hasConcept C154945302 @default.
• W2964976771 hasConcept C19966478 @default.
• W2964976771 hasConcept C2779851693 @default.
• W2964976771 hasConcept C31972630 @default.
• W2964976771 hasConcept C41008148 @default.
• W2964976771 hasConcept C45374587 @default.
• W2964976771 hasConcept C49441653 @default.
• W2964976771 hasConcept C5799516 @default.
• W2964976771 hasConcept C79061980 @default.
• W2964976771 hasConcept C79403827 @default.
• W2964976771 hasConcept C90509273 @default.
• W2964976771 hasConceptScore W2964976771C111919701 @default.
• W2964976771 hasConceptScore W2964976771C11413529 @default.
• W2964976771 hasConceptScore W2964976771C121332964 @default.
• W2964976771 hasConceptScore W2964976771C1276947 @default.
• W2964976771 hasConceptScore W2964976771C13662910 @default.
• W2964976771 hasConceptScore W2964976771C154945302 @default.
• W2964976771 hasConceptScore W2964976771C19966478 @default.
• W2964976771 hasConceptScore W2964976771C2779851693 @default.
• W2964976771 hasConceptScore W2964976771C31972630 @default.
• W2964976771 hasConceptScore W2964976771C41008148 @default.
• W2964976771 hasConceptScore W2964976771C45374587 @default.
• W2964976771 hasConceptScore W2964976771C49441653 @default.
• W2964976771 hasConceptScore W2964976771C5799516 @default.
• W2964976771 hasConceptScore W2964976771C79061980 @default.
• W2964976771 hasConceptScore W2964976771C79403827 @default.
• W2964976771 hasConceptScore W2964976771C90509273 @default.
• W2964976771 hasLocation W29649767711 @default.
• W2964976771 hasOpenAccess W2964976771 @default.
• W2964976771 hasPrimaryLocation W29649767711 @default.
• W2964976771 hasRelatedWork W1541157786 @default.
• W2964976771 hasRelatedWork W1982970530 @default.
• W2964976771 hasRelatedWork W2012850966 @default.
• W2964976771 hasRelatedWork W2062382783 @default.
• W2964976771 hasRelatedWork W2078680214 @default.
• W2964976771 hasRelatedWork W2102074742 @default.
• W2964976771 hasRelatedWork W2346926652 @default.
• W2964976771 hasRelatedWork W2575448526 @default.
• W2964976771 hasRelatedWork W2768206885 @default.
• W2964976771 hasRelatedWork W2772319560 @default.
• W2964976771 hasRelatedWork W28109220 @default.
• W2964976771 hasRelatedWork W2902084102 @default.
• W2964976771 hasRelatedWork W3034588296 @default.
• W2964976771 hasRelatedWork W3038053746 @default.
• W2964976771 hasRelatedWork W3112137689 @default.
• W2964976771 hasRelatedWork W3129050860 @default.
• W2964976771 hasRelatedWork W3163600632 @default.
• W2964976771 hasRelatedWork W46039910 @default.
• W2964976771 hasRelatedWork W563082167 @default.
• W2964976771 hasRelatedWork W1561472694 @default.
• W2964976771 isParatext "false" @default.
• W2964976771 isRetracted "false" @default.
• W2964976771 magId "2964976771" @default.
• W2964976771 workType "dissertation" @default.