Abstract

Drone navigation is one of the hottest topics in Unmanned Aerial Vehicles. Simulating UAVs is a way to visualize how a drone can move without the risk of crashing a real drone. One way to navigate a drone in either a simulated or real-life environment is for the drone to have a path planning algorithm built into its programming. A path planning algorithm is a path detection system that can map out the most efficient way to get to a given target, before the drone actually begins moving. Up to this point, these path planning algorithms have only been created for a 2D environment. This report is attempting to create a 3D path planning algorithm using a drone in a pre-made obstacle course simulation. The main algorithms to be implemented in the simulation are A* and Dijkstra. Using a robotics framework, ROS , a simulation environment, Gazebo , and a sensor data visualizer, RViz, a simulated environment will be created which can run a 3D path planning script. The goal is to use the simulation to find the most efficient algorithm for the drone to find its way to its set destination in a 3D environment.

Full Paper