Development of navigation software system for obstacle avoidance in a dynamic environment
Nathan Carmack, Maged B. Mikhail · 2016
The goal of this paper is to improve mobile robot navigation in a dynamic environment. In this paper a navigation software system was developed that allows a mobile robot to autonomously traverse a dynamic environment. The developed system accomplishes this by modeling an obstacle's motion and then predicting that object's trajectory. The navigation software system developed uses the potential field method to navigate the mobile robot. Any objects within the robots vicinity are tracked using a simulated laser scanner. These detected objects are avoided by combining different behaviors, all of which have been developed to navigate the mobile robot. The navigation software system was tested and implemented in a virtual environment that required the mobile robot reach a designated target while avoiding moving obstacles. The navigation software system showed an evident characteristic of not moving the mobile robot into a moving object's trajectory. This characteristic was demonstrated by either having the robot navigating behind the moving object or pausing and waiting for the moving object to pass. The demonstrated characteristic of the developed navigation software system will improve the safety of mobile robots navigating dynamic environments.