Ritika Shrivastava

EECS Department, University of California, Berkeley

Technical Report No. UCB/EECS-2022-154

May 20, 2022

http://www2.eecs.berkeley.edu/Pubs/TechRpts/2022/EECS-2022-154.pdf

Drones are becoming increasingly prevalent due to their affordability, agility, and size. With this increased usage it is important to account navigability in a variety of terrains. This required detailed understanding of the environment. Most 3D environment reconstruction techniques use LiDAR or RGB-D cameras. However, LiDAR is too expensive and heavy for drones and RGB-D cameras have limited a field of view. We worked on a hardware design with a fisheye camera and RGB-D camera for 3D environment construction. This paper also explores improvements to SLAM module results through trajectory alignment. The use of least-squares estimations of transformation parameters given two points is shown it improve the rotational and translation error.

Advisors: S. Shankar Sastry

---

BibTeX citation:

@mastersthesis{Shrivastava:EECS-2022-154,
    Author= {Shrivastava, Ritika},
    Title= {Environment Reconstruction from an Aerial Perspective with RGB-D and Fisheye Cameras},
    School= {EECS Department, University of California, Berkeley},
    Year= {2022},
    Month= {May},
    Url= {http://www2.eecs.berkeley.edu/Pubs/TechRpts/2022/EECS-2022-154.html},
    Number= {UCB/EECS-2022-154},
    Abstract= {Drones are becoming increasingly prevalent due to their affordability, agility, and size. With this increased usage it is important to account navigability in a variety of terrains. This required detailed understanding of the environment. Most 3D environment reconstruction techniques use LiDAR or RGB-D cameras. However, LiDAR is too expensive and heavy for drones and RGB-D cameras have limited a field of view. We worked on a hardware design with a fisheye camera and RGB-D camera for 3D environment construction. This paper also explores improvements to SLAM module results through trajectory alignment. The use of least-squares estimations of transformation parameters given two points is shown it improve the rotational and translation error.},
}

EndNote citation:

%0 Thesis
%A Shrivastava, Ritika 
%T Environment Reconstruction from an Aerial Perspective with RGB-D and Fisheye Cameras
%I EECS Department, University of California, Berkeley
%D 2022
%8 May 20
%@ UCB/EECS-2022-154
%U http://www2.eecs.berkeley.edu/Pubs/TechRpts/2022/EECS-2022-154.html
%F Shrivastava:EECS-2022-154