Abstract:

Abstract Mapping is an essential and basic requirement for a mobile robot in order to be able to navigate autonomously. This paper proposes a solution for autonomous navigation and real-time mapping using the virtual humanoid robot called NAO. The robot navigates through its environment using ultrasonic sensors only and develops a 2-D map of the environment. For implementation and testing, the Webots simulator was used. It provides real-time values, modification and designing of the 3-D world arena, and plugins for other parameters control. We test autonomous navigation in differently shaped environments. The proposed mathematical algorithm allows the autonomous navigation of the robot and calculates the position of the robot and the obstacles (if any). The results indicate that the algorithm can localize the robot within the environment whereas the accuracy in localization can be increased by adding a control constant to the orientation of the robot. The results demonstrate that the algorithm is more effective in the rectangular arena than in the triangular and pentagon arenas.

Keywords: