| Titre : |
UAV photogrammetry |
| Type de document : |
Thèse/HDR |
| Auteurs : |
Henri Eisenbeiss, Auteur |
| Editeur : |
Zurich : Institut für Geodäsie und Photogrammetrie IGP - ETH |
| Année de publication : |
2009 |
| Collection : |
IGP Mitteilungen, ISSN 0252-9335 num. 105  |
| Importance : |
203 p. |
| Format : |
21 x 30 cm |
| ISBN/ISSN/EAN : |
978-3-906467-86-3 |
| Note générale : |
Bibliographie
A dissertation submitted to ETH ZURICH for the degree of Doctor of Sciences |
| Langues : |
Anglais (eng) |
| Descripteur : |
[Vedettes matières IGN] Photogrammétrie numérique
[Termes IGN] acquisition d'images
[Termes IGN] données lidar
[Termes IGN] drone
[Termes IGN] lasergrammétrie
[Termes IGN] modèle numérique de surface
[Termes IGN] montagne
[Termes IGN] orthophotographie
[Termes IGN] Pérou
[Termes IGN] photogrammétrie aérienne
[Termes IGN] photogrammétrie numérique
[Termes IGN] site archéologique
[Termes IGN] Suisse
[Termes IGN] télémétrie laser aéroporté
|
| Index. décimale : |
33.30 Photogrammétrie numérique |
| Résumé : |
(Auteur) UAVs are unmanned aerial vehicles. Hence, UAV photogrammetry can be understood as a new photogrammetric measurement tool. UAV photogrammetry opens various new applications in the close range domain, combining aerial and terrestrial photogrammetry, but also introduces low-cost alternatives to the classical manned aerial photogrammtery. This thesis deals with the challenging task: "The use of UAV systems as photogrammetric data acquisition platforms " and the work was conducted by the author at IGP at ETH Zurich from 2003 to 2009.
In this work, a new terminology, UAV Photogrammetry, was introduced. UAV Photogrammetry describes photogrammetric measurement platforms, which operate as either remotely controlled, semi-autonomously, or autonomously, all without a pilot sitting in the platform, and the photogrammetric processing of UAV images. The broad definition covers balloons, kites, gliders, airships, rotary and fixed wing UAVs with the capability for photogrammetric data acquisition in manual, semi-automated and automated flight mode. To more clearly define UAVs, a comprehensive literature review on UAVs used in photogrammetry will be given. Using the existing categorizations of UAVs, the outcome of the literature review and our experiments, a new classification for UAVs, with respect on low-cost and high-end systems, the real-time processing capability (based on the integrated sensors and the flight performance, as well as the influence of environmental conditions), was provided. For this work, UAV systems, based on the new classification, will be selected to demonstrate how UAVs can be applied for photogrammetric data acquisition and processing.
In addition to the revised definitions and classifications of UAVs, we have devised a new generic workflow for the photogrammetric UAV flight planning, image acquisition, quality control and data processing. This workflow can specifically be adapted to individual UAV-systems and applications. Thus, specific tools, such as flight planning, are developed. Based on our results, UAV flight control systems and the flight operations were improved and commercial and in-house developed software packages were additionally evaluated for the processing of UAV image data. The proposed workflow allows the combination of office and field work, enabling the first results to be available during the field work for preliminary analysis.
This dissertation also emphasizes the influence of the manual, assisted and autonomous control of the UAV system on the flight performance during the data acquisition, which in turn influences the results and the feasibility of the photogrammetric data processing. Therefore, an autonomous tachymeter tracking of the flight trajectory was performed firstly with an UAV system. A test field for UAVs was then established at the campus Honggerberg at ETH Zurich. This test field enables us to assess the flight performance of various UAV systems. Specifically, an UAV system combined with a light weight laser scanner acquired a DSM over our test field.
Complementary to the developed workflow the accomplished real world application, this work has shown the great potential of using UAVs in photogrammetry and upcoming applications. The focus of our applications was particularly on archaeology and environmental applications, which allowed us to prove our established workflow and to underline the high potential of UAV systems for specific photogrammetric tasks, specifically the use of autonomous operated and stabilized UAV systems. :
• In the frame of the Nasca/Palpa project, the pre-Inca settlement Pinchango Alto (Peru) was documented with our autonomous flying model helicopter. For the first time an accurate and dense elevation model (10cm resolution) of an archaeological site was generated automatically using our in-house developed software. A detailed analysis of the generated data was done using terrestrial laser scanning data. The comparison of both data sets showed that the mean difference between the elevation models was less than 1cm with a standard deviation of 6cm.
• The world heritage Maya site Copan (Honduras) was documented with our system in 2009. The preliminary results, achieved directly after the data acquisition, underlined the expectations and potentials for up-coming
archaeological analysis and investigations.
• The castle Landenberg was documented with 1cm resolution using terrestrial and UAV images.
• In the framework of this dissertation, the rockslide Randa (Switzerland) was documented with LiDAR and image data taken from a manned helicopter. The results from the manned system were compared to the observed data from a UAV for a small part of the rockslide. The elevation model generated from the UAV image data had a substantial higher resolution and showed less occlusions than the LiDAR data. These differences result from the possibility to acquire image data close to the rockslide using an autonomous operated UAV, without endangering human life, and using a specific flight planning tool for UAVs in mountainous areas. This particular application showed also the limitations of existing UAVs in the data acquisition in Alpine areas, due to the high altitudes. However recent developments of new UAV systems are now ready in future work to operate in Alpine areas.
• Using a quadrotor, equipped with a light weight compact camera, it was possible to generate elevation models of a gravel pit at different time periods (February and April). The image data was acquired in the semi-automated and automated mode, allowing an estimation of the volume dug from the pit over time.
• An UAV was used in a study for the evaluation of genetically modified maize (simulated) on conventional maize. High resolution elevation models and orthoimages were produced from two campaigns. The generated spatial data was integrated into a GIS and then used with measures for the out-crossing in maize, to analyse the inclination, height, distance and wind on the influence of cross-pollination in maize.
Finally, based on the experimental results, a new particular workflow for combination of image and LiDAR sensors on one UAV platform was provided. |
| Note de contenu : |
1 Introduction
1.1 Definition of UAVs
1.1.1 UAV photogrammetry
1.1.2 Advantages of UAVs
1.1.3 Limitations in the use of UAVs
1.1.4 Categorization of photogrammetric platforms
1.2 Research goals
1.2.1 Rationale
1.3 Outline
2 Literature review
2.1 Pre 2004
2.1.1 Early investigations
2.1.2 Balloons
2.1.3 Airship
2.1.4 Model helicopters
2.1.5 A7fev
2.1.6 Fixed wing aircraft
2.2 Developments from 2004 to 2007
2.3 2008 until present
2.3.1 VAV-borne LiDAR
2.3.2 Rotary wing UAVs
2.3.3 Balloons
2.3.4 Open source systems
2.3.5 Further applications and developments
2.3.6 IGP ETH Zurich
2.4 Conclusions
3 UAV-systems
3.1 General overview
3.1.1 Classification of UAVs
3.1.2 Regulations
3.1.3 UAV communities
3.2 Open Source and Low Cost UAV Systems: OM-Class
3.2.1 Mikrokopter
3.2.2 Manual controlled system ,,Susi"
3.3 Micro & Mini UAV Systems: M-Class
3.3.7 Quadrotors
3.3.2 Model helicopter
3.3.3 Fixed wing UA Vs
3.4 Large Payload UAV Systems: L-Class
3.4.1 Aeroscout
3.4.2 Geocopter
3.5 Conclusions
4 Project workflow and image data acquisition
4.1 Workflow
4.2 Flight Planning
4.2.1 Flight planning for UAVs
4.2.2 Integration into the mission planning software
4.3 Manual versus autonomous flight
4.3.1 Example model helicopter
4.3.2 Manually controlled Kites and Zeppelin
4.4 Analysis of the trajectory of autonomous UAV flights
4.4.1 Flight trajectory versus predefined flight path
4.4.2 Influence of the flight modus
4.4.3 Tracking tachymetry
4.5 Main achievements for the autonomous flights
5 Photogrammetric data processing
5.1 Image orientation
5.2 Photogrammetric products
5.2.1 Generation of digital surface models
5.2.2 Orthoimage and 3D Visualization
5.3 UAV-borne laser scanning
5.3.7 Test field Campus Honggerberg
5.3.2 First test flights 2009
5.4 Conclusions
6 Archaeological applications and cultural heritage documentation
6.1 Motivation
6.2 Pinchango Alto (Peru) a pre-Inca settlement
6.3 The Maya site Copan (Honduras)
6.4 Castle Landenberg (Switzerland)
6.5 Conclusions
7 Monitoring of hazards, environmental and agricultural applications
7.1 The rockslide Randa (Switzerland)
7.2 Estimation of the volumetric changes of gravel-pits
7.3 Agriculture: The maize field project
7.4 Main achievements
8 Conclusions and perspectives
8.1 Conclusions
8.2 Perspectives |
| Numéro de notice : |
15508 |
| Affiliation des auteurs : |
non IGN |
| Autre URL associée : |
http://dx.doi.org/10.3929/ethz-a-005939264 |
| Thématique : |
IMAGERIE |
| Nature : |
Thèse étrangère |
| DOI : |
10.3929/ethz-a-005939264 |
| En ligne : |
https://ethz.ch/content/dam/ethz/special-interest/baug/igp/igp-dam/documents/PhD [...] |
| Format de la ressource électronique : |
URL |
| Permalink : |
https://documentation.ensg.eu/index.php?lvl=notice_display&id=62741 |
|  |
UAV photogrammetry (2009)
