University of Twente Student Theses


Texture Based Autonomous Reconstruction of 3D Point Cloud Models using Fully Actuated UAVs

Mathivanan, Rajavarman (2020) Texture Based Autonomous Reconstruction of 3D Point Cloud Models using Fully Actuated UAVs.

[img] PDF
Abstract:This master thesis proposes and validates an approach for ariel robotic system to autonomously build a 3D point cloud model of a 3D object using an unmanned aerial vehicle (UAV) in a static environment, without having prior information about the object. The proposed method relies on the textures on the surface of the object for UAV coverage path planning. The reconstruction process comprises of two flight modes, initial arbitrary flight and exploration flight.The 3D point cloud generated from initial arbitrary flight undergoes spatial segmentation to determine the uncovered regions to explore and improve the point cloud model better. The identified low dense regions during the initial flight are converted into the waypoints for the UAV to navigate. These waypoints are positioned in such a way that the trajectories are obstacle free. Polar coordinate system used around the object is used to achieve the obstacle free path. The simulation results showed that the proposed method, unlike trivial helical trajectory method, completes the point cloud model construction process autonomously without a priori knowledge about the object of interest. In addition to that, the proposed method completes the point cloud reconstruction in less time and high quality than the aforementioned helical solution.
Item Type:Essay (Master)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:50 technical science in general, 52 mechanical engineering, 54 computer science
Programme:Electrical Engineering MSc (60353)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page