Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor | Faculty of Engineering | en_US |
| dc.creator | Zhang, Guohao | - |
| dc.identifier.uri | https://theses.lib.polyu.edu.hk/handle/200/9066 | - |
| dc.language | English | en_US |
| dc.publisher | Hong Kong Polytechnic University | - |
| dc.rights | All rights reserved | en_US |
| dc.title | Adaptive kalman filter for GNSS/INS integration based on PIXHAWK for UAV applications | en_US |
| dcterms.abstract | Since the Unmanned Aerial Vehicle (UAV) is developed with higher accuracy and mobility, its usage is increased for civil applications. However, those civil applications always require UAV to operate near or inside the urban area. The GNSS signal will suffer from an extra traveling distance by signal reflection from building surface, and further introduce large localization error, namely the multipath effect. Since the multipath effect is unable to be solved but to mitigate its influence, the multi-sensor integrated localization method is the common method to decrease the multipath error. The Kalman Filter (KF) is widely used to integrate GNSS with INS measurements with a confident coefficient in between. As for better localization estimation from KF in complex urban environment, the target is to tune an appropriate weighting between GNSS and INS for different situation during operation. In this study, an adaptive Kalman filter is developed to adjust the measurement noise covariance matrix with different GNSS measurement conditions. The Allan Variance Analysis technique is used to study the characteristics of the IMU sensors and determines the process noise covariance matrix. Then the principle component analysis is employed to study the relationship between positioning error and GNSS features and selecting the key features and classification range. The supervised machine learning model is trained with real operation data covered most situations, and further used to predict the current GNSS measurement condition during operations. To relieve the mis-classification error from machine learning, the fuzzy logic algorithm is designed to avoid the sudden change of classification result. Finally, the localization performance of the proposed adaptive Kalman filter is compared with different classification method, including decision tree, random forest and random forest with fuzzy logic. The result proving that, the presented adaptive Kalman filter using random forest with fuzzy logic can achieve better GNSS measurement condition classification and further obtain more accurate localization result, ensure the safety of UAV operating in urban area. | en_US |
| dcterms.extent | x, 99 pages : color illustrations | en_US |
| dcterms.isPartOf | PolyU Electronic Theses | en_US |
| dcterms.issued | 2017 | en_US |
| dcterms.educationalLevel | M.Sc. | en_US |
| dcterms.educationalLevel | All Master | en_US |
| dcterms.LCSH | Hong Kong Polytechnic University -- Dissertations | en_US |
| dcterms.LCSH | Drone aircraft | en_US |
| dcterms.LCSH | Vehicles, Remotely piloted | en_US |
| dcterms.LCSH | Drone aircraft -- Control systems | en_US |
| dcterms.accessRights | restricted access | en_US |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| 991021962224803411.pdf | For All Users (off-campus access for PolyU Staff & Students only) | 5.37 MB | Adobe PDF | View/Open |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/9066