Pub Date : 2019-11-01DOI: 10.1109/REDUAS47371.2019.8999719
Hirad Goudarzi, D. Hine, Arthur G. Richards
This paper describes work in pursuit of pilot-only operation of a drone for inspection tasks in congested environments. In particular, the work focusses on the automation of mission-planning, monitoring and execution on the ground control station, reducing the need for additional personnel in support of the pilot. Geographic Information Systems are used for planning and behaviour trees for automated execution. Results of initial flight trials at the Clifton Suspension Bridge are reported.
{"title":"Mission Automation for Drone Inspection in Congested Environments","authors":"Hirad Goudarzi, D. Hine, Arthur G. Richards","doi":"10.1109/REDUAS47371.2019.8999719","DOIUrl":"https://doi.org/10.1109/REDUAS47371.2019.8999719","url":null,"abstract":"This paper describes work in pursuit of pilot-only operation of a drone for inspection tasks in congested environments. In particular, the work focusses on the automation of mission-planning, monitoring and execution on the ground control station, reducing the need for additional personnel in support of the pilot. Geographic Information Systems are used for planning and behaviour trees for automated execution. Results of initial flight trials at the Clifton Suspension Bridge are reported.","PeriodicalId":351115,"journal":{"name":"2019 Workshop on Research, Education and Development of Unmanned Aerial Systems (RED UAS)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121444608","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper presents an integration approach to deploy new mission planning methods for multi-vehicle systems within an open source ground control system (GCS), i.e. QGroundControl. As the autonomous mission planning features of QGroundControl consist in pre-planning waypoints within a geofence, there is no way to deal with scenarios where the agents of the swarm must counter moving targets of reduced size, or more generally any scenario where an algorithm of path planning is preferred instead of a manual planification. The proposed solution consists in extending the ”toolbox” of the application (QGCToolbox) with a new tool (QGCTool), in implementing a new controller and modifying views (FlightDisplayViewMap, PlanView) while enabling information to flow between components using Qt’s signals and slots. A test scenario is presented to verify new mission planning interface and path planning algorithm, that involves in this case eight UAVs operating in a zone to protect and an area to attack. Also a workaround is proposed to increase the default limitation of connections to QGroundControl. The proposed methodology is tested and validated with software in the loop and flight test.
{"title":"Implementation of Ground Control System for Autonomous Multi-agents using QGroundControl","authors":"Tristan Dardoize, Noé Ciochetto, Ju-Hyeon Hong, Hyo-Sang Shin","doi":"10.1109/REDUAS47371.2019.8999717","DOIUrl":"https://doi.org/10.1109/REDUAS47371.2019.8999717","url":null,"abstract":"This paper presents an integration approach to deploy new mission planning methods for multi-vehicle systems within an open source ground control system (GCS), i.e. QGroundControl. As the autonomous mission planning features of QGroundControl consist in pre-planning waypoints within a geofence, there is no way to deal with scenarios where the agents of the swarm must counter moving targets of reduced size, or more generally any scenario where an algorithm of path planning is preferred instead of a manual planification. The proposed solution consists in extending the ”toolbox” of the application (QGCToolbox) with a new tool (QGCTool), in implementing a new controller and modifying views (FlightDisplayViewMap, PlanView) while enabling information to flow between components using Qt’s signals and slots. A test scenario is presented to verify new mission planning interface and path planning algorithm, that involves in this case eight UAVs operating in a zone to protect and an area to attack. Also a workaround is proposed to increase the default limitation of connections to QGroundControl. The proposed methodology is tested and validated with software in the loop and flight test.","PeriodicalId":351115,"journal":{"name":"2019 Workshop on Research, Education and Development of Unmanned Aerial Systems (RED UAS)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116626656","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-11-01DOI: 10.1109/REDUAS47371.2019.8999685
Manuel J. Fernández, P. Sánchez-Cuevas, G. Heredia, A. Ollero
This paper is about an application of a method based on the ECIES (Elliptic Curve Integrated Encryption Scheme) to improve the security against malicious attacks of the UAVs (Unmanned Aerial Vehicles) communications system. This system is focused on improving the security conditions in extreme situations and preventing the aircraft for man-made incidents and cyber attacks. The paper briefly describes the different attacks that can affect to the operation of UAVs and the security methods that, nowadays, are used to guarantee the security during the operations. Moreover, it presents a solution to a strong vulnerability detected in the classical scheme used in UAV. This scheme uses ROS (Robot Operating System) as the core of the communication system to interconnect different devices and nodes in this paper, it is demonstrated that if an Intruder is able to enter in the local network of the UAV system, he/she is also able to impersonate the GCS (Ground Control Station) of the UAV and take control of it leading to an undesirable maneuver or even a dangerous crash against a building or a person. The security system proposed to avoid this consists of a simplified method based on ECIES sending packets, between UAV and GCS, which uses ECDSA (Elliptic Curve Digital Signature) and are ciphered in RSA (Rivest–Shamir–Adleman). Thus, it is possible to guarantee that the high level computer of the UAV is able to identify the identity of their GCS and prevent of being commanded by an unauthorized Intruder. Both, the vulnerability and the solution proposed have been experimentally tested and validated through software-in-the-loop simulations and in a outdoor scenario using a small UAV.
本文研究了一种基于椭圆曲线集成加密方案(ECIES)的方法在提高无人机通信系统抗恶意攻击安全性方面的应用。该系统的重点是改善极端情况下的安全条件，防止飞机发生人为事件和网络攻击。简要介绍了影响无人机运行的各种攻击，以及目前用于保证无人机运行安全的安全措施。此外，针对传统方案中检测到的强漏洞，提出了一种解决方案。该方案以ROS (Robot Operating System)作为通信系统的核心，实现了不同设备和节点之间的互联，本文论证了如果入侵者能够进入无人机系统的本地网络，他/她还可以冒充无人机的GCS (Ground Control Station)并控制无人机，从而导致无人机的不良机动甚至与建筑物或人员发生危险的碰撞。为了避免这种情况，提出的安全系统包括一种基于ECIES在无人机和GCS之间发送数据包的简化方法，该方法使用ECDSA(椭圆曲线数字签名)，并在RSA (Rivest-Shamir-Adleman)中进行加密。因此，有可能保证无人机的高级计算机能够识别其GCS的身份并防止被未经授权的入侵者指挥。该漏洞和提出的解决方案都已通过软件在环模拟和使用小型无人机的户外场景进行了实验测试和验证。
{"title":"Securing UAV communications using ROS with custom ECIES-based method","authors":"Manuel J. Fernández, P. Sánchez-Cuevas, G. Heredia, A. Ollero","doi":"10.1109/REDUAS47371.2019.8999685","DOIUrl":"https://doi.org/10.1109/REDUAS47371.2019.8999685","url":null,"abstract":"This paper is about an application of a method based on the ECIES (Elliptic Curve Integrated Encryption Scheme) to improve the security against malicious attacks of the UAVs (Unmanned Aerial Vehicles) communications system. This system is focused on improving the security conditions in extreme situations and preventing the aircraft for man-made incidents and cyber attacks. The paper briefly describes the different attacks that can affect to the operation of UAVs and the security methods that, nowadays, are used to guarantee the security during the operations. Moreover, it presents a solution to a strong vulnerability detected in the classical scheme used in UAV. This scheme uses ROS (Robot Operating System) as the core of the communication system to interconnect different devices and nodes in this paper, it is demonstrated that if an Intruder is able to enter in the local network of the UAV system, he/she is also able to impersonate the GCS (Ground Control Station) of the UAV and take control of it leading to an undesirable maneuver or even a dangerous crash against a building or a person. The security system proposed to avoid this consists of a simplified method based on ECIES sending packets, between UAV and GCS, which uses ECDSA (Elliptic Curve Digital Signature) and are ciphered in RSA (Rivest–Shamir–Adleman). Thus, it is possible to guarantee that the high level computer of the UAV is able to identify the identity of their GCS and prevent of being commanded by an unauthorized Intruder. Both, the vulnerability and the solution proposed have been experimentally tested and validated through software-in-the-loop simulations and in a outdoor scenario using a small UAV.","PeriodicalId":351115,"journal":{"name":"2019 Workshop on Research, Education and Development of Unmanned Aerial Systems (RED UAS)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126553573","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-11-01DOI: 10.1109/REDUAS47371.2019.8999673
C. Capitán, Á. Castaño, J. Capitán, A. Ollero
This paper presents a framework to handle threats for UAS operating in the U-space. During UAS operations, unexpected events or threats can happen and put the UAS in danger. In order to handle those situations, we propose a module for threats management that incorporates autonomous decision-making capabilities and is integrated with the U-space. First, we survey related works and compile a classification of event classes as well as possible contingency or emergency procedures. Then, we devise an approach to integrate our decision-making procedure within the U-space framework. This approach selects the proposed mitigation actions depending on the operational state of the UAS involved. Our framework takes a step forward in the integration of autonomous systems in the U-space, as there is no current solution for emergency management without the intervention of UAS operators.
{"title":"A framework to handle threats for UAS operating in the U-space","authors":"C. Capitán, Á. Castaño, J. Capitán, A. Ollero","doi":"10.1109/REDUAS47371.2019.8999673","DOIUrl":"https://doi.org/10.1109/REDUAS47371.2019.8999673","url":null,"abstract":"This paper presents a framework to handle threats for UAS operating in the U-space. During UAS operations, unexpected events or threats can happen and put the UAS in danger. In order to handle those situations, we propose a module for threats management that incorporates autonomous decision-making capabilities and is integrated with the U-space. First, we survey related works and compile a classification of event classes as well as possible contingency or emergency procedures. Then, we devise an approach to integrate our decision-making procedure within the U-space framework. This approach selects the proposed mitigation actions depending on the operational state of the UAS involved. Our framework takes a step forward in the integration of autonomous systems in the U-space, as there is no current solution for emergency management without the intervention of UAS operators.","PeriodicalId":351115,"journal":{"name":"2019 Workshop on Research, Education and Development of Unmanned Aerial Systems (RED UAS)","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125975105","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-11-01DOI: 10.1109/REDUAS47371.2019.8999711
E. Paiva, M. Gomez-Redondo, J. Rodas, Y. Kali, M. Saad, R. Gregor, H. Fretes
Unmanned aerial vehicles have become a disruptive technology, which has experienced exponential growth in several applications. The control of these vehicles is a fairly wide area and the cascade PID controller is the most used in practice. However, this latter structure doesn’t ensure high performances in the presence of unmodelled dynamics, uncertainties and external abrupt disturbances. To that end, this work proposes a new method that consists of a non-linear cascade configuration of the variable structure control between first order sliding mode based on exponential reaching law and modified super-twisting second order sliding mode algorithm. The developed method is tested on simulation on a quadrotor system, the results obtained demonstrate good performance for trajectory tracking and as well as other non-linear controller options, it is robust against unmodeled dynamics and disturbances.
{"title":"Cascade First and Second Order Sliding Mode Controller of a QuadRotor UAV based on Exponential Reaching Law and Modified Super-Twisting Algorithm","authors":"E. Paiva, M. Gomez-Redondo, J. Rodas, Y. Kali, M. Saad, R. Gregor, H. Fretes","doi":"10.1109/REDUAS47371.2019.8999711","DOIUrl":"https://doi.org/10.1109/REDUAS47371.2019.8999711","url":null,"abstract":"Unmanned aerial vehicles have become a disruptive technology, which has experienced exponential growth in several applications. The control of these vehicles is a fairly wide area and the cascade PID controller is the most used in practice. However, this latter structure doesn’t ensure high performances in the presence of unmodelled dynamics, uncertainties and external abrupt disturbances. To that end, this work proposes a new method that consists of a non-linear cascade configuration of the variable structure control between first order sliding mode based on exponential reaching law and modified super-twisting second order sliding mode algorithm. The developed method is tested on simulation on a quadrotor system, the results obtained demonstrate good performance for trajectory tracking and as well as other non-linear controller options, it is robust against unmodeled dynamics and disturbances.","PeriodicalId":351115,"journal":{"name":"2019 Workshop on Research, Education and Development of Unmanned Aerial Systems (RED UAS)","volume":"204 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129279355","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-11-01DOI: 10.1109/REDUAS47371.2019.8999686
Aarón López Luna, J. Martínez-Carranza, I. Vega
We propose an aerial Interaction system based on a micro aerial vehicle (MAV) with a two degree of freedom (DOF) arm incorporated in the lower part of the frame. There are numerous potential benefits of physical interaction with the environment. However, the process of approaching a MAV to an object or a surface also brings challenging control problems. Additionally, potential scenarios are usually inadequate for the use of Global Positioning System (GPS) technology, usually used for the MAV’s pose measurement in the control algorithms. In this work, a Gain-Scheduling (GS) approach is incorporated into a conventional Proportional-Integral-Derivative (PID) control algorithm to solve the problem of successfully touch a rigid surface in flight mode with no collisions. The Simultaneous Localization and Mapping (SLAM) approach is incorporated in the control loop as an alternative to the GPS. Visual SlAM utilizes images captured from a monocular camera onboard the MAV. Once captured, images are passed to an RGB-Depth SLAM system. Thus, MAV’s pose can be estimated with a metric, which is then considered into the interaction control. Experimental testing results demonstrate satisfactory performance of the proposed control strategy.
{"title":"Towards Aerial Interaction of MAVs in GPS-Denied Environments","authors":"Aarón López Luna, J. Martínez-Carranza, I. Vega","doi":"10.1109/REDUAS47371.2019.8999686","DOIUrl":"https://doi.org/10.1109/REDUAS47371.2019.8999686","url":null,"abstract":"We propose an aerial Interaction system based on a micro aerial vehicle (MAV) with a two degree of freedom (DOF) arm incorporated in the lower part of the frame. There are numerous potential benefits of physical interaction with the environment. However, the process of approaching a MAV to an object or a surface also brings challenging control problems. Additionally, potential scenarios are usually inadequate for the use of Global Positioning System (GPS) technology, usually used for the MAV’s pose measurement in the control algorithms. In this work, a Gain-Scheduling (GS) approach is incorporated into a conventional Proportional-Integral-Derivative (PID) control algorithm to solve the problem of successfully touch a rigid surface in flight mode with no collisions. The Simultaneous Localization and Mapping (SLAM) approach is incorporated in the control loop as an alternative to the GPS. Visual SlAM utilizes images captured from a monocular camera onboard the MAV. Once captured, images are passed to an RGB-Depth SLAM system. Thus, MAV’s pose can be estimated with a metric, which is then considered into the interaction control. Experimental testing results demonstrate satisfactory performance of the proposed control strategy.","PeriodicalId":351115,"journal":{"name":"2019 Workshop on Research, Education and Development of Unmanned Aerial Systems (RED UAS)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125700239","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-11-01DOI: 10.1109/REDUAS47371.2019.8999724
J. J. Acevedo, Á. Castaño, J. L. Andrade-Pineda, A. Ollero
This paper proposes an algorithm to detect conflicts among UAVs from a strategic point of view. The method is based on the representation of the airspace as a 4D grid of cells. Given a set of UAVs with their scheduled 4D trajectories (their flight plans), the whole scenario is discretized as a 4D grid and the problem is solved by filling the appropriate cell for each way-point from the trajectories and checking the neighboring cells. The proposed method is tested and compared against a traditional algorithm getting a significantly better performance. The proposed method also scales very well with increasing number of UAVs and way-points per trajectory.
{"title":"A 4D grid based approach for efficient conflict detection in large-scale multi-UAV scenarios","authors":"J. J. Acevedo, Á. Castaño, J. L. Andrade-Pineda, A. Ollero","doi":"10.1109/REDUAS47371.2019.8999724","DOIUrl":"https://doi.org/10.1109/REDUAS47371.2019.8999724","url":null,"abstract":"This paper proposes an algorithm to detect conflicts among UAVs from a strategic point of view. The method is based on the representation of the airspace as a 4D grid of cells. Given a set of UAVs with their scheduled 4D trajectories (their flight plans), the whole scenario is discretized as a 4D grid and the problem is solved by filling the appropriate cell for each way-point from the trajectories and checking the neighboring cells. The proposed method is tested and compared against a traditional algorithm getting a significantly better performance. The proposed method also scales very well with increasing number of UAVs and way-points per trajectory.","PeriodicalId":351115,"journal":{"name":"2019 Workshop on Research, Education and Development of Unmanned Aerial Systems (RED UAS)","volume":"64 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123804312","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-11-01DOI: 10.1109/REDUAS47371.2019.8999681
S. R. Nekoo, J. Acosta, A. E. Gomez-Tamm, A. Ollero
Variable-pitch propellers quadrotors possess nonlinear algebraic relations between force/moment of the system and thrust factors of the rotor dynamics. The nonlinear relations make the thrust allocation a challenging topic in overall control design. The state-dependent Riccati equation (SDRE) is selected as a controller for regulation task in fully coupled six degree-of-freedom (DoF) mode. Common designs of the SDRE fail to deliver a fully coupled six-DoF control due to under-actuation. Virtual constraints are used to deliver a position and orientation control in a cascade design. Within the structure of the SDRE, four thrust allocation methods are proposed to compute the thrust factors based on the output results of control system. Practical implementation has been the main reason to generate such allocations. The use of Mean Value Theorem makes it possible to find an implementable formalism for thrust factors since they can be categorized as non-affine systems. Agile and aggressive maneuver is one of the application of the variable-pitch propellers quadrotors; so, flip maneuver is studied to highlight the advantages of the thrust allocation methods. Analysis of the four methods and comparisons are carried out to present the advantages and disadvantages of the proposed structures.
{"title":"Optimized Thrust Allocation of Variable-pitch Propellers Quadrotor Control: A Comparative Study on Flip Maneuver","authors":"S. R. Nekoo, J. Acosta, A. E. Gomez-Tamm, A. Ollero","doi":"10.1109/REDUAS47371.2019.8999681","DOIUrl":"https://doi.org/10.1109/REDUAS47371.2019.8999681","url":null,"abstract":"Variable-pitch propellers quadrotors possess nonlinear algebraic relations between force/moment of the system and thrust factors of the rotor dynamics. The nonlinear relations make the thrust allocation a challenging topic in overall control design. The state-dependent Riccati equation (SDRE) is selected as a controller for regulation task in fully coupled six degree-of-freedom (DoF) mode. Common designs of the SDRE fail to deliver a fully coupled six-DoF control due to under-actuation. Virtual constraints are used to deliver a position and orientation control in a cascade design. Within the structure of the SDRE, four thrust allocation methods are proposed to compute the thrust factors based on the output results of control system. Practical implementation has been the main reason to generate such allocations. The use of Mean Value Theorem makes it possible to find an implementable formalism for thrust factors since they can be categorized as non-affine systems. Agile and aggressive maneuver is one of the application of the variable-pitch propellers quadrotors; so, flip maneuver is studied to highlight the advantages of the thrust allocation methods. Analysis of the four methods and comparisons are carried out to present the advantages and disadvantages of the proposed structures.","PeriodicalId":351115,"journal":{"name":"2019 Workshop on Research, Education and Development of Unmanned Aerial Systems (RED UAS)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122814532","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-11-01DOI: 10.1109/REDUAS47371.2019.8999705
Jose A. Millan-Romera, J. J. Acevedo, Á. Castaño, Héctor Pérez-Leon, C. Capitán, A. Ollero
This paper presents a simulation framework for the development and testing of UTM functionalities. The simulator is based on the U-space functional architecture and services. U-space includes pre-flight services (before the UAV is allowed to take-off) and in-flight services (when the UAV is flying). There are many in-flight services that are not currently implemented (or only partialy) and, as far as we know, there is not a simulation framework that tackle this. Our UTM simulator is based on the Robot Operating System (ROS) and Gazebo and it allows to implement and test U-space in-flight services, speeding up their development and allowing to implement new functionalities. We show the implementation of a use-case of the EU-funded project GAUSS that involves rotary and fix-wing UAVs flying on a shared environment.
{"title":"A UTM simulator based on ROS and Gazebo","authors":"Jose A. Millan-Romera, J. J. Acevedo, Á. Castaño, Héctor Pérez-Leon, C. Capitán, A. Ollero","doi":"10.1109/REDUAS47371.2019.8999705","DOIUrl":"https://doi.org/10.1109/REDUAS47371.2019.8999705","url":null,"abstract":"This paper presents a simulation framework for the development and testing of UTM functionalities. The simulator is based on the U-space functional architecture and services. U-space includes pre-flight services (before the UAV is allowed to take-off) and in-flight services (when the UAV is flying). There are many in-flight services that are not currently implemented (or only partialy) and, as far as we know, there is not a simulation framework that tackle this. Our UTM simulator is based on the Robot Operating System (ROS) and Gazebo and it allows to implement and test U-space in-flight services, speeding up their development and allowing to implement new functionalities. We show the implementation of a use-case of the EU-funded project GAUSS that involves rotary and fix-wing UAVs flying on a shared environment.","PeriodicalId":351115,"journal":{"name":"2019 Workshop on Research, Education and Development of Unmanned Aerial Systems (RED UAS)","volume":"210 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127602831","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-11-01DOI: 10.1109/REDUAS47371.2019.8999703
L. Rojas-Perez, J. Martínez-Carranza
Convolutional Neural Networks (CNN) and Deep Learning (DL) have become a popular tool to address all sorts of artificial intelligent challenges. The Autonomous Drone Racing is a challenge consisting of developing an autonomous drone capable of beating a human in a drone race, and DL is a tool that has been included in state of the art solutions to address this problem. Current works have proposed to use CNN and DL to detect the gates, whereas other works have proposed to use a CNN to obtain drone’s control commands and a goal point, with all of these approaches using a single image as input. In this work we propose a CNN based on the well known pose-net network. Originally used for camera relocalisation, we propose to use pose-net to provide control commands to drive the drone towards and to cross the gate autonomously. In contrast to previous works, we also propose to use a temporal set of images as input for the network. In specific, we use 6 images captured every 166 milliseconds in one second to create a mosaic. The latter is used as input of the CNN to predict the control commands. We compare this proposed temporal approach against using a single image as input for the CNN. Our results, although in simulation, demonstrate that the using only our temporal approach is feasible, less noisy and more effective than the single image approach, enabling the drone to autonomously cross a set of gates placed randomly, and even under the scenario where the gate moves dynamically.
{"title":"A Temporal CNN-based Approach for Autonomous Drone Racing","authors":"L. Rojas-Perez, J. Martínez-Carranza","doi":"10.1109/REDUAS47371.2019.8999703","DOIUrl":"https://doi.org/10.1109/REDUAS47371.2019.8999703","url":null,"abstract":"Convolutional Neural Networks (CNN) and Deep Learning (DL) have become a popular tool to address all sorts of artificial intelligent challenges. The Autonomous Drone Racing is a challenge consisting of developing an autonomous drone capable of beating a human in a drone race, and DL is a tool that has been included in state of the art solutions to address this problem. Current works have proposed to use CNN and DL to detect the gates, whereas other works have proposed to use a CNN to obtain drone’s control commands and a goal point, with all of these approaches using a single image as input. In this work we propose a CNN based on the well known pose-net network. Originally used for camera relocalisation, we propose to use pose-net to provide control commands to drive the drone towards and to cross the gate autonomously. In contrast to previous works, we also propose to use a temporal set of images as input for the network. In specific, we use 6 images captured every 166 milliseconds in one second to create a mosaic. The latter is used as input of the CNN to predict the control commands. We compare this proposed temporal approach against using a single image as input for the CNN. Our results, although in simulation, demonstrate that the using only our temporal approach is feasible, less noisy and more effective than the single image approach, enabling the drone to autonomously cross a set of gates placed randomly, and even under the scenario where the gate moves dynamically.","PeriodicalId":351115,"journal":{"name":"2019 Workshop on Research, Education and Development of Unmanned Aerial Systems (RED UAS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133570153","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：