Search

Unmanned aerial vehicles (UAV), also known as drones have been used for a variety of reasons and the commercial drone market growth is expected to reach remarkable levels in the near future. However, some drone users can mistakenly or intentionally fly into flight paths at major airports, flying too close to commercial aircraft or invading people’s privacy. In order to prevent these unwanted events, counter-drone technology is needed to eliminate threats from drones and hopefully they can be integrated into the skies safely. There are various counter-drone methods available in the industry. However, a counter-drone system supported by an artificial intelligence (AI) method can be an efficient way to fight against drones instead of human intervention. In this paper, a deep reinforcement learning (DRL) method has been proposed to counter a drone in a 3D space by using another drone. In a 2D space it is already shown that the deep reinforcement learning method is an effective way to counter a drone. However, countering a drone in a 3D space with another drone is a very challenging task considering the time required to train and avoid obstacles at the same time. A Deep Q-Network (DQN) algorithm with dueling network architecture and prioritized experience replay is presented to catch another drone in the environment provided by an Airsim simulator. The models have been trained and tested with different scenarios to analyze the learning progress of the drone. Experiences from previous training are also transferred before starting a new training by pre-processing the previous experiences and eliminating those considered as bad experiences. The results show that the best models are obtained with transfer learning and the drone learning progress has been increased dramatically. Additionally, an algorithm which combines imitation learning and reinforcement learning is implemented to catch the target drone. In this algorithm, called deep q-learning from demonstrations (DQfD), expert demonstrations data and self-generated data by the agent are sampled and the agent continues learning without overwriting the demonstration data. The main advantage of this algorithm is to accelerate the learning process even if there is a small amount of demonstration data. This work was funded partially by the AGAUR under grant 2020PANDE00141, the Ministry of Science and Innovation of Spain under grant PID2020-116377RB-C21 and the SESAR Joint Undertaking (JU) project CORUS-XUAM, under grant SESAR-VLD2 101017682. The JU receives support from the European Union’s Horizon 2020 research and innovation program and SESAR JU members other than the Union.

In the next years, the unmanned air business is expected to have an average annual growth rate of 14.5 per cent. Last-mile delivery, inspection works and security tasks are the most expected missions that those unmanned aircraft (UA) will execute. Most of these missions are well suited for multi-copters: small airframes with vertical take-off and landing capabilities. Large fleets of UA will be managed by new aerial logistic centers where flight plans will be created and monitored, the payload will be prepared, and fast battery replacement will allow continuous flights to obtain maximum benefit. Beyond visual line-of-sight capabilities is a must for those logistic center businesses. Anticipating the scalability of unmanned aircraft growth is the aim of this paper. For this, a simulation tool has been developed which generates unmanned traffic flights from completely parameterized inputs: the geographic area and the type and number of operations, aircraft and operators. For this paper, the tested scenario is a logistics industrial polygon with increasing delivery traffic, the Martorell industrial area (5.8 km2). Flights have a random altitude from 80 m to 120 m. En- route phases have some slight turns to make them more realistic. The time of departure follows a Box-Muller algorithm during the declared business hours, centered in the peak declared hour. This work was funded by the Ministry of Economy, Industry, and Competitiveness of Spain under GrantNumber TRA2016-77012-R

In recent years, prior to COVID-19, capacity shortfalls in airspace and airports inevitably caused an increase in aircraft delays. Therefore, when it returns to normal conditions, the airspace will exhibit the same capacity limits, even under normal weather conditions. To ensure that air traffic remains safe, reliable, and efficient in adverse weather conditions, planning and coordination activities through a Collaborative Decision Making process are required to deliver the most effective Air Traffic Flow and Capacity Management services to Air Traffic Control and Aircraft Operators. Nowadays, this task is based on air traffic controllers’ experience and historical data. That means that the Flow Manager Positions and the Network Manager operators have to process a huge amount of information, and the detection of future overloads is based on past experiences. Moreover, due to the inherent uncertainty of weather information, a reliable decision support framework is required to handle these situations as efficiently as possible. We propose a Deep Learning model able to extract the relationship between both the historical data and the implemented actions, accurately identifying the intervals of time that must be regulated. The proposed model achieves an accuracy between 80% and 90% across six traffic volumes belonging to both the MUAC and REIMS regions, a recall higher than 85%, and an F1-score higher than 0.8 in all the cases. Furthermore, the confidence-level analysis shows a really high activation when making a prediction. Finally, the SHapley Additive exPlanations method is applied to identify the most relevant input features. This work was funded EUROCONTROL under Ph.D. Research Contract No. 18-220569-C2 and by the Ministry of Economy, Industry, and Competitiveness of Spain under GrantNumber TRA2016-77012-R.

Artificial Intelligence (AI) has been used in different research areas in aerospace to create an intelligent system. Especially, an unmanned aerial vehicle (UAV), known as a drone, can be controlled by AI methods such as deep reinforcement learning (DRL) in different purposes. Drones with DRL become more intelligent and eventually they can be fully autonomous. In this paper, DRL method supported by real time object detection model is proposed to detect and catch a drone. Additionally, the results are analyzed by comparing the time to catch the target drone in seconds between DRL method, human pilot and an algorithm which directs the drone towards the target position without using any AI method or navigation and guidance method. The main idea is to catch a drone in an environment as fast as possible without crashing any obstacles inside the environment. In DRL method, the agent is a quadcopter drone and it is rewarded in each time step by the environment provided by Airsim flight simulator. Drone is trained to catch the target drone by using DRL model which is based on deep Q-Network algorithm. After training, the tests have been made by the agent drone with DRL model and human pilots to catch stationary and non-stationary target drone. The training and test results show that the agent drone learns to catch target drone which can be a stationary and a non-stationary. In addition. the agent avoids crashing any obstacles in the environment with a minimum success rate of 94%. Also, DRL model performance is compared with the human pilot performances and the agent with DRL model shows better time to catch the target drone. Human pilots struggle to control the drone by using remote controller when catching the target in simulation. However, the agent with DRL model is rarely missing the target when trying to catch the target. This work was funded partially by the AGAUR under grant 2020PANDE00141, the Ministry of Science and Innovation of Spain under grant PID2020-116377RB-C21 and the SESAR Joint Undertaking (JU) project CORUS-XUAM, under grant SESAR-VLD2 101017682. The JU receives support from the European Union’s Horizon 2020 research and innovation programme and the SESAR JU members other than the Union. Special thanks to Dr. Pablo Royo Chic for his assistance to our paper by piloting a drone to catch the target drone.

Remotely Piloted Aircraft Systems (RPAS) are increasingly becoming a part of our day-to-day life, and the wide range of their possible applications is creating a new industry with a large economic potential that is pushing the technological developments at a much faster pace than that for manned aviation. However, due to constraints arising from safety and operational considerations, RPAS can currently only fly in segregated airspace, making their integration in the civil airspace an unsolved challenge. RPAS have to guarantee a level of safety at least equal to that of manned aircraft. Manned aviation follows the rules of the air. The notion “well-clear” is used in ICAO rules without giving exact definition of the term. Here, rules rely on the perception and judgement of the pilot. The capability to perform RWC and DAA is required for the full integration of RPAS with General Air Traffic (GAT). To this end, several rule-making bodies are working to define the required safety and performance objectives. An RPAS needs proper mathematical definitions of Remain-Well-Clear (RWC) in order to operate safely. When implementing RPAS into civil airspace, exact well-clear parameters have to be determined. In 2013, the second FAA workshop on Unmanned Aerial Systems (UAS) concluded that ‘there is a need for establishing an unambiguous and quantitative definition for well clear that can be used as a separation performance standard for an aircraft system’. In 2014, the SAA Science and Research Panel (SARP) provided a Well Clear Recommendation to RTCA SC-228, which was subsequently used in the first Minimum Operational Performance Standards for Detect and Avoid (DAA) systems, DO-365. The URClearED project intends to define the requirements and capabilities for the RWC function of a DAA system of an RPAS operating in Class D-G airspace also interacting with VFR flights. Particularly in airspace classes F-G IFR aircraft will not receive ATC-provided separation. Different European countries may define airspace classes and rules differently. Further, the congestion of the airspace and the characteristics of the VFR traffic may complicate the situation. When crossing state border airspace class and rules may change. This paper will report the definition of scenarios and use cases selected to evaluate the RWC function under the aforementioned conditions. RWC volume and threshold selection will be evaluated in a subsequent simulation and assessment phase. Scenarios will encompass all the relevant elements under analysis. The type of encounter are a primary factor, not only in terms of its geometry but also in terms of the rules under which the conflicting aircraft are operating and the specific airspace classes involved. This will impact the roles and responsibilities of involved actors. Transitions between different classes of airspace will also be considered, including states cross-border situations. © 2021, American Institute of Aeronautics and Astronautics Inc.. All rights reserved. This project has received funding from the SESAR Joint Undertaking (JU) under grant agreement No 892440. The JU receives support from the European Union’s Horizon 2020 research and innovation programme and the SESAR JU members other than the Union. This paper only reflects the author’s views and the SESAR JU is not responsible for any use that may be made of the information it contains.

Counter-drone technology by using artificial intelligence (AI) is an emerging technology and it is rapidly developing. Considering the recent advances in AI, counter-drone systems with AI can be very accurate and efficient to fight against drones. The time required to engage with the target can be less than other methods based on human intervention, such as bringing down a malicious drone by a machine-gun. Also, AI can identify and classify the target with a high precision in order to prevent a false interdiction with the targeted object. We believe that counter-drone technology with AI will bring important advantages to the threats coming from some drones and will help the skies to become safer and more secure. In this study, a deep reinforcement learning (DRL) architecture is proposed to counter a drone with another drone, the learning drone, which will autonomously avoid all kind of obstacles inside a suburban neighborhood environment. The environment in a simulator that has stationary obstacles such as trees, cables, parked cars, and houses. In addition, another non-malicious third drone, acting as moving obstacle inside the environment was also included. In this way, the learning drone is trained to detect stationary and moving obstacles, and to counter and catch the target drone without crashing with any other obstacle inside the neighborhood. The learning drone has a front camera and it can capture continuously depth images. Every depth image is part of the state used in DRL architecture. There are also scalar state parameters such as velocities, distances to the target, distances to some defined geofences and track, and elevation angles. The state image and scalars are processed by a neural network that joints the two state parts into a unique flow. Moreover, transfer learning is tested by using the weights of the first full-trained model. With transfer learning, one of the best jump-starts achieved higher mean rewards (close to 35 more) at the beginning of training. Transfer learning also shows that the number of crashes during training can be reduced, with a total number of crashed episodes reduced by 65%, when all ground obstacles are included. This work was funded by the Ministry of Science, Innovation and Universities of Spain under Grant No. TRA2016-77012-R, TRA2019 This work was funded by the Ministry of Science, Innovation and Universities of Spain under Grant No. TRA2016-77012-R, TRA2019

Opening the sky to new classes of airspace user is a political and economic imperative for the European Union. Drone industries have a significant potential for economical growth according to the latest estimations. To enable this growth safely and efficiently, the CORUS project has developed a concept of operations for drones flying in Europe in very low-level airspace, which they have to share that space with manned aviation, and quite soon with urban air mobility aircraft as well. U-space services and the development of smart, automated, interoperable, and sustainable traffic management solutions are presented as the key enabler for achieving this high level of integration. In this paper, we present the U-space concept of operations (ConOps), produced around three new types of airspace volume, called X, Y, and Z, and the relevant U-space services that will need to be supplied in each of these. The paper also describes the reference high-level U-space architecture using the European air traffic management architecture methodology. Finally, the paper proposes the basis for the aircraft separation standards applicable by each volume, to be used by the conflict detection and resolution services of U-space. This work has been partially funded by the SESAR Joint Undertaking, a body of the European Commission, under grant H2020 RIA-763551 and by the Ministry of Economy and Enterprise of Spain under contract TRA2016-77012-R.

© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. Any emergent technology in history has raised an initial rejection by part of the society. Added to the several problems that the non-mature technology may have, the lack of any previous experience about side effects and the human’s psychological fear to the unknown play an important influence in its acceptance. As drones bring up high social and economical expectations due to their capabilities and bussiness applications, the social acceptance is key to the complete development of drone technology's potential. Experts believe that social acceptance is ruled by a balance between beneficial usages and inconvenient issues regarding the technology. This balance in the aeronautical sector is also conditioned by the strict safety policies and regulations of the airspace and the current airspace users. To analyse this balance situation in actual and future environments, regarding drone technology, different use cases will be presented. These use cases have been proposed and analysed by different stakeholders from the U-space community network, a network of airspace and drone stakeholders who participated in the context of the SESAR CORUS project.\par The purpose of this paper is to analyse some of these use cases by obtaining responses from different stakeholders point of view using a survey in order to see how economical, safety and political aspects are balanced in each one of the cases. From the survey responses we will perform an analysis by means of three different acceptance indicators, one for each aspect commented. Main results and conclusions point out that the economical indicator is, in general, positive, especially for the low cost payload use cases. In contrast the economical indicator is close to neutral for city transport and airports use cases, which leads to propose some economical promotion action may be needed to make them a reality. For the safety indicator we observe that they are close to negative values as use case complexity increases. Thus we can conclude that some of the proposed missions start affecting the current levels of safety. Finally, the political indicator is mostly neutral, with some positive trends for scenarios related with inspection tasks or done in non-populated areas. We thank our colleagues from the SESAR CORUS initia-tive: the project partners developping the use cases and theconcept of operation; and the U-space community network,who provided insight and expertise that greatly assisted theresearch. We also thank all our colleagues from the ICARUSResearch group (UPC) for the assistance with the surveys andfor the comments that greatly improved the manuscript.This work has been partially funded by the Ministry ofScience, Innovation and Universities of Spain under grantnumber TRA2016-77012-R and by the SESAR Joint Under-taking under the European Unions Horizon 2020 research andinnovation programme under grant agreement RIA-763551.

This document describes a key aspect of NtoM, a concept of operations (ConOps) currently under development, which focuses on the awareness, productivity and safety of Remotely Piloted Aircraft System (RPAS) pilots controlling several flights at once in non-segregated airspace. An explanation will be given of how the ConOps suggests capturing, representing, managing and predicting the workload of the pilots. To illustrate some of the features of the concept, it was necessary to define a representation of the workload associated to the tasks. A synthetic task environment that used the NtoM prototype was built and used to evaluate the requirements of time and attention of pseudo-pilots based on their performance while executing the tasks and task overlaps, determine the top threshold of workload allowed for a pilot and detect incompatibilities among tasks. These values served as a reference to design demanding test scenarios, which helped to reveal weaknesses and inspire improvements that were addressed in the following stage of development.