Advanced

Dynamic path planning of initially unknown environments using an RGB-D camera

Gustafzelius, Sara (2015)
Department of Automatic Control
Abstract
In this thesis an RGB-D camera was used with the goal to perform dynamic path planning in an initially unknown environment. Depth data from an RGB-D camera together with a discretizising algorithm is continuously used for maintaining an obstacle map of the environment which within the path planning algorithm D* Lite [S. Koening, 2005] is performed on the flight.

Experiments were conducted on two different systems, on Combine’s hexacopter and on a Gantry Tau robot at the Robot Lab of the Department of Automatic Control, LTH. On Combine’s hexacopter different tracking algorithms such as ICP, Translation Approximation and SDF where evaluated for 3D positioning while the robots internal positioning where used on the Gantry Tau robot.

... (More)
In this thesis an RGB-D camera was used with the goal to perform dynamic path planning in an initially unknown environment. Depth data from an RGB-D camera together with a discretizising algorithm is continuously used for maintaining an obstacle map of the environment which within the path planning algorithm D* Lite [S. Koening, 2005] is performed on the flight.

Experiments were conducted on two different systems, on Combine’s hexacopter and on a Gantry Tau robot at the Robot Lab of the Department of Automatic Control, LTH. On Combine’s hexacopter different tracking algorithms such as ICP, Translation Approximation and SDF where evaluated for 3D positioning while the robots internal positioning where used on the Gantry Tau robot.

For discretization purposes we compare the use of Box Approximation and Signed Distance Function (SDF) for creating the obstacle map. (Less)
Please use this url to cite or link to this publication:
author
Gustafzelius, Sara
supervisor
organization
year
type
H3 - Professional qualifications (4 Years - )
subject
ISSN
0280-5316
other publication id
ISRN LUTFD2/TFRT--5980--SE
language
English
id
7442930
date added to LUP
2015-07-03 09:34:22
date last changed
2015-07-03 09:34:22
@misc{7442930,
  abstract     = {In this thesis an RGB-D camera was used with the goal to perform dynamic path planning in an initially unknown environment. Depth data from an RGB-D camera together with a discretizising algorithm is continuously used for maintaining an obstacle map of the environment which within the path planning algorithm D* Lite [S. Koening, 2005] is performed on the flight.

 Experiments were conducted on two different systems, on Combine’s hexacopter and on a Gantry Tau robot at the Robot Lab of the Department of Automatic Control, LTH. On Combine’s hexacopter different tracking algorithms such as ICP, Translation Approximation and SDF where evaluated for 3D positioning while the robots internal positioning where used on the Gantry Tau robot.

 For discretization purposes we compare the use of Box Approximation and Signed Distance Function (SDF) for creating the obstacle map.},
  author       = {Gustafzelius, Sara},
  issn         = {0280-5316},
  language     = {eng},
  note         = {Student Paper},
  title        = {Dynamic path planning of initially unknown environments using an RGB-D camera},
  year         = {2015},
}