3D model generation using an airborne swarm

Ruaridh Clark, Giuliano Punzo, Gordon Dobie, Rahul Summan, Charles Norman MacLeod, Stephen Pierce, Malcolm Macdonald, Gregour Bolton

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

82 Downloads (Pure)

Abstract

Using an artificial kinematic field to provide co-ordination between multiple inspection UAVs, the authors herein demonstrate full 3D modelling capability based on a photogrammetric system. The operation of the system is demonstrated by generating a full 3D surface model of an intermediate level nuclear waste storage drum. Such drums require periodic inspection to ensure that drum distortion or corrosion is carefully monitored. Inspection is performed by multiple airborne platforms, which would enable rapid inspection of structures that are inaccessible to on-surface remote vehicles and are in human-hazardous environments. A three-dimensional surface-meshed model of the target is then constructed in post-processing through photogrammetry analysis of the visual inspection data. The inspection environment uses a tracking system to precisely monitor the position of each aerial vehicle within the enclosure. The vehicles used are commercially available Parrot AR.Drone quadcopters, controlled through a computer interface connected over an IEEE 802.11n (WiFi) network, implementing a distributed controller for each vehicle. This enables the autonomous and distributed elements of the control scheme to be retained, while alleviating the vehicles of the control algorithm’s computational load. The control scheme relies on a kinematic field defined with the target at its centre. This field defines the trajectory for all the drones in the volume relative to the central target, enabling the drones to circle the target at a set radius while avoiding drone collisions. This function enables complete coverage along the height of the object, which is assured by transitioning to another inspection band only after completing circumferential coverage. Using a swarm of vehicles, the time until complete coverage can be significantly reduced.
Original languageEnglish
Title of host publicationE-Book of Abstracts 41st Annual Review of Progress in Quantitative Nondestructive Evaluation Conference
Place of PublicationAmes, Iowa
Number of pages8
Publication statusPublished - 1 Oct 2014
EventQuantitative Nondestructive Evaluation Conference, QNDE 2014 - Idaho, Boise, United States
Duration: 20 Jul 201425 Jul 2014

Conference

ConferenceQuantitative Nondestructive Evaluation Conference, QNDE 2014
CountryUnited States
CityBoise
Period20/07/1425/07/14

Keywords

  • autonomous
  • inspection photogrammetry
  • quadcopter
  • swarm

Fingerprint Dive into the research topics of '3D model generation using an airborne swarm'. Together they form a unique fingerprint.

  • Cite this

    Clark, R., Punzo, G., Dobie, G., Summan, R., MacLeod, C. N., Pierce, S., Macdonald, M., & Bolton, G. (2014). 3D model generation using an airborne swarm. In E-Book of Abstracts 41st Annual Review of Progress in Quantitative Nondestructive Evaluation Conference