Research Information System
27th Annual Symposium of the Pattern-Recognition-Association-of-South-Africa / 9th Robotics and Mechatronics Conference of South Africa (Robmech), Stellenbosch, South Africa, 30 November - 02 December 2016
This paper propose a new position configuration for 2D laser range finders (LRF) on mobile robots for simultaneous localization and mapping (SLAM) algorithms. In this study, a pair of 2D LRFs with 270 degrees scanning range are replaced with exterior corresponding angles to the front left/right and rear right/left of the vehicle. This configuration is named as TRAnsversal Positioned (TRAP) and offers a 360 degrees angle of field of view for the mobile robot without any blind region. This configuration ensures the AGV to identify anything inside the scanning area instantaneously. This study compares the TRAP configuration with the traditional configuration that the 2D LRF is positioned centred on the front and the rear for "G-Mapping" SLAM method. Comparison tests are achieved under ROS (Robot Operating System) environment and results shows that TRAP configuration achieves better performance over the traditional configuration and these results are given for benchmarking.