Evaluation on a mobile robotic software platform

• Main Author: Michael Sanjaya Wira.
• Other Authors: Lau Wai Shing, Michael
• Format: Final Year Project
• Language: English
• Published: 2010
• Subjects: DRNTU::Engineering::Mechanical engineering::Robots
• Online Access: http://hdl.handle.net/10356/40452
• Institution: Nanyang Technological University

Simultaenous Localization and Mapping (SLAM) is a technique used for an autonomous robot to create a map from an unknown environment by using landmarks and at the same time using the map to locate where the robot is located on the map. SLAM offers advantage in providing robot's position and orientation (pose) and where landmarks are located better than a GPS for indoor tasks or in constrained environment. SLAM -unlike GPS- does not rely on external equipments to do its localization rather it is using sensors placed on board. The development of SLAM performance will bring more opportunities to use robot for exploration or for search and rescue. Nowadays, numerous techniques of SLAM has been developed and used, and one of the techniques that has been developed is vSLAM. vSLAM uses camera and object recognition algorithm to help its SLAM capability. This technique offers a SLAM performance with capability to recognize and associate the newly detected landmarks with landmark that has been detected previously. This capability has brought a new solution for handling kidnapping problem, whereby a robot is being carried to arbitrary position.