Flight coordination solutions for multirotor unmanned aerial vehicles

Abstract

[EN] As the popularity and the number of Unmanned Aerial Vehicles (UAVs) increases, new protocols are needed to coordinate them when flying without direct human control, and to avoid that these UAVs collide with each other. Testing such novel protocols on real UAVs is a complex procedure that requires investing much time, money and research efforts. Hence, it becomes necessary to first test the developed solutions using simulation. Unfortunately, existing tools present significant limitations: some of them only simulate accurately the flight behavior of a single UAV, while some other simulators can manage several UAVs simultaneously, but not in real time, thus losing accuracy regarding the mobility pattern of the UAV. In this work we address such problem by introducing Arducopter Simulator (ArduSim), a novel simulation platform that allows controlling in soft real-time the flight and communications of multiple UAVs, being the developed protocols directly portable to real devices. Moreover, ArduSim includes a realistic model for the WiFi communications link between UAVs, which was proposed based on real experiments. The chances that two UAVs get close to each other during their flights is increasing as more and more of them populate our skies, causing concerns regarding potential collisions. Therefore, this thesis also proposes the Mission Based Collision Avoidance Protocol (MBCAP), a novel UAV collision avoidance protocol applicable to all types of multicopters flying autonomously. It relies on wireless communications in order to detect nearby UAVs, and to negotiate the procedure to avoid any potential collision. Experimental and simulation results demonstrate the validity and effectiveness of the proposed solution, which typically introduces a small overhead in the range of 15 to 42 seconds for each risky situation successfully handled. The previous solution aims at UAVs performing independent flights, but they can also form a swarm, where more constraints have to be met to avoid collisions among them, and to allow them to complete their task efficiently. Deploying an UAV swarm instead of a single UAV can provide additional benefits when, for example, cargo carrying requirements exceed the lifting power of a single UAV, or when the deployment of several UAVs simultaneously can accelerate the accomplishment of the mission, and broaden the covered area. To this aim, in this work we present the Mission-based UAV Swarm Coordination Protocol (MUSCOP), a solution that allows multiple UAVs to perfectly coordinate their flight when performing planned missions. Experimental results show that the proposed protocol is able to achieve a high degree of swarm cohesion independently of the flight formation adopted, and even in the presence of very lossy channels, achieving minimal synchronization delays and very low position offsets with regard to the ideal case. Currently, there are some other scenarios, such as search and rescue operations, where the deployment of manually guided swarms of UAVs can be necessary. In such…