Home

About

Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network
FAQHELPSITE MAPCONTACT US


  Advanced Search  

 
Multi-sensor Registration and Integration for Inspection and Navigation Faysal Boughorbel, Andreas Koschan, Mongi Abidi
 

Summary: Multi-sensor Registration and Integration for Inspection and Navigation
Faysal Boughorbel, Andreas Koschan, Mongi Abidi
Department of Electrical and Computer Engineering, University of Tennessee, Knoxville, TN, 37996, fboughor@utk.edu
Abstract This paper describes a unified framework for addressing the problem of multi-sensor registration and
integration in the context of inspection by autonomous platforms. The task is approached from an optimization angle by
designing an effective general- purpose registration criterion that can be employed for both aligning 3D and 2D datasets.
The resulting pipeline reconstructs full 3D representations of the scenes and objects of interest, including multi-spectral
texture overlay. The obtained information rich representation is very useful in automating decision and reducing the
cognitive load on human operators
I. INTRODUCTION
Mobile and fixed robotic systems for inspecting
hazardous environments that are equipped with multiple
imaging devices are deployed in many facilities around
the world. These systems rely mostly on human operators
to interpret the sensory input and decide the appropriate
action. To achieve further autonomy for these systems, in
labor-intensive tasks such as nuclear waste sorting and
critical components control, the robots need to acquire a
more complete representation of the objects that are
manipulated, or the environments in which they will

  

Source: Abidi, Mongi A. - Department of Electrical and Computer Engineering, University of Tennessee
Koschan, Andreas - Imaging, Robotics, and Intelligent Systems, University of Tennessee

 

Collections: Computer Technologies and Information Sciences; Engineering