University of Twente Student Theses


Optimization of UAV flight plans in difficult landscapes

Wang, Ziyao (2020) Optimization of UAV flight plans in difficult landscapes.

Link to full-text:
(only accessible for UT students and staff)
Abstract:With the continuous progress and development of technology, Unmanned Aerial Vehicle (UAV) have been widely used in the field of surveying and mapping. UAV is able to collect high-resolution, close-range photogrammetry data at a low cost, and the flexibility of the UAV allows it to customize the carried sensors according to the mission requirements. However, due to the limitation of UAV flight height, there is often insufficient coverage when performing photogrammetry mission in undulating terrain as mountainous areas, which affects quality of data set. In order to avoid insufficient coverage, the commonly used method in actual flight is to set the image overlap much higher than required, which means that the distribution of viewpoints will be denser, further causing data redundancy. In addition, another limitation of UAV is endurance, a common battery-powered UAV can only support a flight for 15 to 30 minutes. Images acquired from different acquisition time may affect quality of data. Based on the reasons above, this research proposes a novel methodology to implement an algorithm for UAV flight planning in mountainous areas to filter out redundant data and design the path with shortest flight time under ensuring the quality of data. The Digital Surface Model (DSM) of the study area, user-defined GSD and overlap percentage will be used as the initial input of the algorithm. According to input parameters, the algorithm can calculate the position of all viewpoints in the flight mission. However, this initial dense viewpoint network does not show a significant improvement in accuracy or coverage. At the same time, intensive sampling based on this network will cause data redundancy and increase the time for both data acquisition and image processing. Therefore, one of the functions of the algorithm is to filter out those redundant images. For this purpose, principle for coverage check is proposed, that is all the points in study area need to be covered in at least “n” images. The filtering procedure works following this principle to remove those redundant images. Then, for all reserved viewpoints, using Simulated Annealing Algorithm (SAA) to design the optimal path with the shortest flight time. The final output of the algorithm is an optimal flight path composed of the position of all the points for acquiring images. The algorithm has been implemented in MATLAB. All the tests were performed on synthetic DSM with different shapes and terrain, in order to test the performance of the algorithm in different scenarios.
Item Type:Essay (Master)
Faculty:ITC: Faculty of Geo-information Science and Earth Observation
Programme:Geoinformation Science and Earth Observation MSc (75014)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page