6DOF Object Recognition and Positioning for Robotics Using Next Best View Heuristics
(2018) In LU-CS-EX 2018-12 EDA920 20172Department of Computer Science
- Abstract
- The accuracy and portability of depth cameras have increased by a lot in recent years, which allows for advanced 3D scanning of the environment for robotic applications. In this thesis we have developed a system that uses a depth camera mounted on a robot arm to identify and localize arbitrary objects, and give hints on how to move the camera to get better localization results. The system works by generating virtual views of objects to identify them in a point cloud generated by the depth camera. This data is then used to estimate a pose of the object, and generate a hint on where to move the camera next. After a new point cloud is taken, it is merged with the previous cloud which allows the system to iteratively get more confident in the... (More)
- The accuracy and portability of depth cameras have increased by a lot in recent years, which allows for advanced 3D scanning of the environment for robotic applications. In this thesis we have developed a system that uses a depth camera mounted on a robot arm to identify and localize arbitrary objects, and give hints on how to move the camera to get better localization results. The system works by generating virtual views of objects to identify them in a point cloud generated by the depth camera. This data is then used to estimate a pose of the object, and generate a hint on where to move the camera next. After a new point cloud is taken, it is merged with the previous cloud which allows the system to iteratively get more confident in the identification and pose of the object. (Less)
- Popular Abstract (Swedish)
- Robotar som anpassar sig efter omgivningen utan hjälp från människor blir vanligare och vanligare. Detta arbete skapar ett system för seende anpassat för robotar, som möjliggör identifiering och lokalisering av objekt.
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/8959961
- author
- Svensson, Andreas LU and Ganslandt, Alexander LU
- supervisor
-
- Mathias Haage LU
- Elin Anna Topp LU
- organization
- course
- EDA920 20172
- year
- 2018
- type
- H3 - Professional qualifications (4 Years - )
- subject
- keywords
- Robotics, point clouds, object identification, pose estimation, localization, point cloud merging, hint system, next best view
- publication/series
- LU-CS-EX 2018-12
- report number
- LU-CS-EX 2018-12
- ISSN
- 1650-2884
- language
- English
- id
- 8959961
- date added to LUP
- 2018-10-15 10:09:29
- date last changed
- 2018-10-15 10:09:29
@misc{8959961, abstract = {{The accuracy and portability of depth cameras have increased by a lot in recent years, which allows for advanced 3D scanning of the environment for robotic applications. In this thesis we have developed a system that uses a depth camera mounted on a robot arm to identify and localize arbitrary objects, and give hints on how to move the camera to get better localization results. The system works by generating virtual views of objects to identify them in a point cloud generated by the depth camera. This data is then used to estimate a pose of the object, and generate a hint on where to move the camera next. After a new point cloud is taken, it is merged with the previous cloud which allows the system to iteratively get more confident in the identification and pose of the object.}}, author = {{Svensson, Andreas and Ganslandt, Alexander}}, issn = {{1650-2884}}, language = {{eng}}, note = {{Student Paper}}, series = {{LU-CS-EX 2018-12}}, title = {{6DOF Object Recognition and Positioning for Robotics Using Next Best View Heuristics}}, year = {{2018}}, }