Autonomous navigation for multiple unmanned aerial vehicles (UAVs) in urban environments

Abstract

This paper presents an autonomous navigation system for multiple unmanned aerial vehicles (UAVs) in urban environments. The navigation system comprises a long-term route planner and a short-term cooperative collision avoidance function. A 3D model of a real urban terrain is used to represent the environment within which the UAVs operate, and the system takes the motion constraints of the UAVs into account when performing planning. The navigation system enables multiple independent UAVs to travel from their departure points to their destinations while predicting and avoiding collisions with one another and with static terrain. Monte Carlo simulations on the system indicate a worst-case route planner replan rate of 11.65 %, route planner failure rate of 9.82×10-4 %, and collision resolution failure rate of 0.12 % for the collision avoidance function.