The Use of Direction Matrix to Determine Autonomous Quad-Copter Drone Path when Facing Obstacle

Abstract

This paper presents a method to conduct path planning for Multi-Rotor Autonomous Aerial Vehicle such as quad-copter drone (Drone) in an urban environment. The priority of the path plan will be the safety of the drone from collisions with obstacles and efficiency of the path. In our proposed method, initially, the drone only knows its position and its target position. The agent studies the environment obstacles towards its target while conducting a movement. To perform the drone obstacle avoidance movement, we introduce the concept of 3 by 3 Direction Matrix usage to improve the A* algorithm. The experimental environment in this study will be dynamic, where various obstacles exist when flying at a relatively low altitude. The method is tested on a 3Dimensional environment, where the Drone could also face various dynamic obstacles. In addition, the target, which is the end destination in this experiment, is also dynamic. The target is able to move and the Drone must be able to plan a path and reach the target object. The experiment result shows that the proposed method is effective to be used as drone path planning in a dynamic environment.