Behavior Control and Navigation of Two-Wheeled Robot Using ROS-Gazebo

Abstract

Nowadays, robots serve a wide range of purposes, such as rescue and service. Due to the complexity of the environment, it is critical that the robot interacts with the environment and performs its task without losing context. This article uses a two-wheeled robot whose rotational speed is independently controlled by each wheel. This causes the robot to rotate in a horizontal plane without any steering. Our navigation algorithm, based on layered architecture of robot navigation, consists of two behaviors; Go to Goal and Avoid Obstacle. A hard-switch PID controller is then developed to track reference signals based on two behaviors. Using ROS-Gazebo and Python, all simulations and algorithms are developed and programmed for robot. Finally, the navigation algorithm is validated in two different environments based on their complexity.