A mission planning framework for fleets of connected drones

• Autor(a) principal: Silva, Ana Margarida Oliveira da Costa e
• Data de Publicação: 2021
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10773/31385
• topic: Autonomous aerial vehicles; Drones; Control platform; User-friendly description language; Missions

The usage of aerial drones has become more popular as they also become more accessible, both in economic and usability terms. Nowadays, these vehicles can present reduced dimensions and a good cost-benefit ratio, which makes it possible for several services and applications supported by aerial drone networks to emerge. Some scenarios that benefit from the use of aerial drones are the monitoring of emergency situations and natural disasters, the patrolling of urban areas and support to police forces, and tourist applications such as the real-time video transmission of points of interest. It is common for the control of the drone to be dependent on human intervention in these situations, which requires professionals specialized in its control. However, in recent years, several solutions have emerged that enable the autonomous flight of these vehicles, minimizing manual interference. Taking into account the enormous diversity of use cases, many of the existing solutions for autonomous control focus on specific scenarios. Generic mission planning platforms also exist, but most of them only allow missions consisting of linear waypoints to be traversed. These situations translate into a mission support that is not very flexible. In this dissertation, we propose a modular infrastructure that can be used in various scenarios, enabling the autonomous control and monitoring of a fleet of aerial drones in a mission context. This platform has two main components, one integrated into the onboard computer of the vehicle, and the other one in the ground control. The former allows the communication with the flight controller so that it can collect telemetry data and send movement instructions to the drone. The latter allows to monitor this data and send the commands remotely, also enabling robust mission planning with multiple drones. A mission can be described in a script that the ground module interprets, sending the commands to the assigned vehicles. These missions can describe different paths, modifying the behaviour of the drones according to external factors, such as a sensor reading. It is also possible to define plugins to be reused in various missions, for example, by integrating an algorithm that ensures that all drones maintain connectivity. The solution was evaluated in scenarios with a single drone and with the collaboration of multiple drones. The tests were performed in a simulated environment and also in an environment with real drones. The observed behaviour is similar in both scenarios.