| dc.contributor.author | Zhang, Peter Pifu. | en_US |
| dc.date.accessioned | 2014-10-21T12:35:34Z | |
| dc.date.available | 2007 | |
| dc.date.issued | 2007 | en_US |
| dc.identifier.other | AAINR31508 | en_US |
| dc.identifier.uri | http://hdl.handle.net/10222/54972 | |
| dc.description | This thesis addresses the problem of globally-consistent localization and mapping simultaneously for autonomous mobile robots in an unknown and unstructured 3D environment by sensor fusion. It belongs to the research area of simultaneous localization and mapping (SLAM) in a mobile robot community. The main contribution presented in this thesis is the development of a set of new algorithms for an autonomous mobile robot with full 3D SLAM ability by multi-sensor fusion. It can be classified in the following aspects: (1) A measurement system architecture designed for mobile robot localization and mapping in a large and unknown environment. Based on the general SLAM method, a simple structure is designed by using a stereo camera and a set of range sensors to solve the 3D SLAM problem in an unknown environment. (2) Registration uncertainty for robot self-localization in 3D . This is an approach to estimate registration uncertainty where the feature correspondence-based method is used during the process of robot pose estimation. (3) Algorithms for efficient map building in large area. By using set theory, a set of new algorithms for efficient mapping building is designed during the SLAM solution processing. (4) An algorithm design of a globally-consistent 3D SLAM by sensor fusion. One sensor such as stereo camera will be used for local SLAM and another sensor such as buoys or GPS will be used for global path estimation. The estimation results from both sensors will be fused in a global coordinate system to form a globally-consistent map and path for the mobile robot. (5) A mobile robot simulation system designed for full 3D SLAM application. In this 3D animation system, a robot would navigate in a virtual space and measure the features with all equipped sensors in its view field at every time. These measurements are used to evaluate algorithms for mobile robot in any environment. | en_US |
| dc.description | The strategies and methods provided in this dissertation can be used for any autonomous mobile robot. There are a number of directions in which the work presented here could be extended to many other challenging areas and the same strategies could be applied. | en_US |
| dc.description | Thesis (Ph.D.)--Dalhousie University (Canada), 2007. | en_US |
| dc.language | eng | en_US |
| dc.publisher | Dalhousie University | en_US |
| dc.publisher | | en_US |
| dc.subject | Engineering, Robotics. | en_US |
| dc.subject | Computer Science. | en_US |
| dc.title | Globally-consistent three-dimensional simultaneous localization and mapping with multi-sensor fusion. | en_US |
| dc.type | text | en_US |
| dc.contributor.degree | Ph.D. | en_US |
