3D Path Planning for Quadrotor Using Gazebo Simulator

Abstract

This research explores the 3D path planning for quadrotor, which is an unmannedaerial vehicle (UAV) with four rotors. The quadrotor is simulated using robot operating system(ROS) and Gazebo software, and is equipped with camera, GPS sensor, and inertialmeasurement unit (IMU) sensor to do the mapping of its environment. The packages used inROS are Hector Quadrotor package, joystick package, Octomap package, and MoveIt package.These packages were modified so that it could be integrated with each other and fulfill theobjective of this research. For the 3D path planning, a method called rapidly-exploring randomtree (RRT) is explored and implemented. Several experiments regarding the behavior of thequadrotor, the mapping, and the path planning were conducted to find out the performance andlimitations of the simulation. This simulation is set up so that it can be used to validate a newdesign of quadrotor before it is tested with a physical prototype.