Skip to Main content Skip to Navigation

Laser based sensor fusion and control for the tele-operation of minidrones

Abstract : Aerial robotics is a prominent field of research that has seen great commercial success during the last years due to the development of highly efficient and affordable small-sized airborne platforms, commonly referred to as mini-drones. This has opened the way to promising new applications in surveillance and inspection tasks. In recent years, this has been a key subject of research in the power industry, where power utilities are subject to deterioration due to atmospheric conditions and require extensive monitoring programs. Mini-drones have the potential of fully automating the inspection process, further reducing costs and inspection times. In this context, this thesis addresses autonomous electric tower inspections with mini-drones. Namely, self-localization, the first step in a long series of tasks towards achieving fully autonomous capabilities, is the main focus of this work. We explore how 2D laser scanners can be coupled with commonly available sensors to estimate a mini-drone's 6 degree of freedom pose in real-time, using uniquely on-board sensing and processing capabilities. This thesis develops topics from classic scan matching algorithms, such as the iterative closest point (ICP) algorithm and proposed adaptations to the electric tower scene, to sensor fusion and feed-back control. Validations based on experimental flights and extensive simulations are presented.
Document type :
Complete list of metadatas

Cited literature [122 references]  Display  Hide  Download
Contributor : Abes Star :  Contact
Submitted on : Thursday, September 27, 2018 - 12:04:08 PM
Last modification on : Sunday, October 25, 2020 - 7:05:25 PM
Long-term archiving on: : Friday, December 28, 2018 - 2:17:46 PM


Version validated by the jury (STAR)


  • HAL Id : tel-01882687, version 1


Carlos Viña. Laser based sensor fusion and control for the tele-operation of minidrones. Robotics [cs.RO]. Université Pierre et Marie Curie - Paris VI, 2017. English. ⟨NNT : 2017PA066525⟩. ⟨tel-01882687⟩



Record views


Files downloads