Elsevier

Ad Hoc Networks

Volume 68, January 2018, Pages 33-47
Ad Hoc Networks

The CUSCUS simulator for distributed networked control systems: Architecture and use-cases

https://doi.org/10.1016/j.adhoc.2017.09.004Get rights and content

Abstract

The current merging of networking and control research fields within the scope of robotic applications is creating fascinating research and development opportunities. However, the tools for a proper and easy management of experiments still lag behind. Although different solutions have been proposed to simulate and emulate control systems and, more specifically, fleets of Unmanned Aerial Vehicles (UAVs), still they do not include an efficient and detailed network-side simulation, which is usually available only on dedicated software. On the other hand, current advancements in network simulations suites often do not include the possibility to include an accurate description of controlled systems. In the middle 2010s, integrated solutions of networking and control for fleets of UAVs are still lacking. In this paper, we fill such gap by presenting a simulation architecture for networked control systems which is based on two well-known solutions in both the fields of networking simulation (the NS-3 tool) and UAV control simulation (the FL-AIR tool). Three main research contributions are provided: (i) first, we show how the existing tools can be integrated on a closed-loop architecture, so that the network propagation model (NS-3 side) is influenced by the drone mobility and by the 3D scenario map (FL-AIR side); (ii) second, we implement a novel module, which allows modeling realistic 3D environments by importing city-wide characteristics by the popular OpenStreetMap service; (iii) third, we demonstrate the modeling capabilities of the CUSCUS framework on two realistic use-cases, corresponding to well-known application scenarios of UAVs, i.e. dynamic formation control and static coverage of a target area.

Introduction

Aerial networks composed by Unmanned Aerial Vehicles (UAVs) constitute emerging cooperative systems characterized by unique features such as distributed coordination, autonomous 3D mobility, and context-awareness through the sensing capabilities [1]. In the next few years, the pervasive diffusion of UAVs is expected to pave the way to novel scenarios integrating IoT devices, aerial communications and mobile/multimedia applications. At the same time, the state of art of UAVs already includes a wide range of real-case deployments, from disaster recovery to surveillance and precision agriculture [2], [3], [4].

A key issue in most of the mentioned scenarios is the management of flying nodes’ autonomous mobility in order to meet the Quality of Service (QoS) requirements of the applications [1]. In absence of a centralized controller, the fleet mobility is determined by decisions performed at each UAV, hence consensus-based or distributed coordination protocols are needed to avoid collisions, keep the network connected and achieve the mission-specific goals [5]. At the same time, the communication among nodes is strongly affected by the propagation conditions of the environment, so far that the effect of packet loss must be taken into account in networked robotic architecture design [6], [7]. Finally, since the micro-mobility of each UAV involves complex electromechanical dynamics, robust controllers are required for tuning the parameters governing the position and orientation of the flying node (e.g. the Proportional, Integrative and Derivative terms of the controller) [8]. The merging of networking and control fields is a natural consequence of the above mentioned issues: several communication-aware mobility schemes have been proposed for fleet creation and management [9], [10]. Similarly, there exists plenty of communication protocols at the MAC, network and transport layers, which are specifically tailored to the UAVs scenarios, in order to cope with the dynamic topology and, at the same time, take maximum benefit from the self-placement capabilities of the nodes [11], [12]. The growth of research in this area poses a fundamental question: which methodology to adopt in order to evaluate the performance of distributed networked control systems, producing reliable and accurate results? Several studies rely on small-case test-beds, e.g. [13]. However, experimental studies, in order to be meaningful, should consider many UAVs at the same time, and this might easily introduce excessive costs or present safety problems. Similarly, analytical models might likely become infeasible due to the large number of parameters to take into account, and the unknown correlations among them. Vice versa, simulation tools can provide a cost-effective solution in order to model the UAV applications before their effective deployment on a real scenario. However, although there are several tools enabling to model flight control [14], [15] or network protocols [16], [17], no software addresses the issues of both the fields at the same time.

In this paper, we fill such gap by proposing a novel simulation framework for networked control system, called CommUnicationS-Control distribUted Simulator (CUSCUS). Differently from the state of the art, CUSCUS allows simulating both the UAV networking and formation phases, via the integration of two existing tools: the Framework Libre AIR (FL-AIR) simulator [18] and the mainstream network simulator NS-3 [19]. Using FL-AIR, a real-time and fine-grained simulation of the micro-mobility of each UAV can be achieved, including the modeling of virtual sensors/actuators, the PID regulations and the drone stability. Moreover, it is possible to create UAV applications and test them on a simulated control environment before the actual deployment, since the same code can also be plugged in real drones. More specifically, we provide three main research contributions in this paper:

  • First, we describe how to integrate the FL-AIR and NS-3 simulation with a closed-loop control, so that the fleet mobility is influenced by the propagation conditions and networking protocols. As a result, we are able to perform real-time accurate simulations of the wireless communication among the UAVs, and analyze the impact of the propagation phenomena on the algorithms used for fleet control.

  • Second, we add a Scenario Module in both FL-AIR and NS-3, in order to make a step towards the usage of fleets of UAVs in Smart city scenarios. The Scenario Module allows modeling realistic 3D environments, by importing the scenario description directly from OpenStreetMaps and by taking into account the location of buildings and the street topology.

  • Third, we demonstrate the capabilities of the CUSCUS framework on two use-cases, corresponding to well-known application scenarios of UAVs, i.e. dynamic formation control [20], and Static Coverage of the target area [21], [22]. More specifically, we implement a reference algorithm for each use-case in CUSCUS, and we show the impact of micro-mobility control parameters, beaconing frequency, propagation conditions and scenario characteristics, on the application performance. Furthermore, as in CUSCUS it is possible to define the underlying control model, we take into consideration different physical parameters of the drone, such as the length of its arms and its total weight, which are information of utmost importance when it comes to define accurate movement dynamics.

Finally, we show by experimental results the scalability of the CUSCUS framework in terms of resource utilization (e.g. CPU and memory), and its fine-grained ability to model complex UAVs dynamics, characterized by the interplay between network-side configuration, control-side configuration and 3-D scenario characteristics.

The rest of the paper is structured as follows. Section 2 reviews the state-of-art of simulation tools for UAVs. Section 3 illustrates the CUSCUS framework, describing the logical architectures, the three main components (i.e. FL-AIR, NS-3 and the Scenario Module), and their interworking. Section 4 introduces the use-cases and the mobility algorithms implemented in CUSCUS. Section 5 shows the performance of the CUSCUS framework, and demonstrates the modeling capabilities on the use-case previously mentioned. Finally, conclusions follow in Section 6.

Section snippets

Related works

While pertaining to robotic research, our main purpose in this work is to give the possibility to simulate networked control algorithms on UAVs. From a broader point of view, the literature includes examples of simulation suites that attempt to integrate objectives that belong to the robotics research field along with objectives of other research domains. However, at the best of our knowledge, all the existing solutions lack the ability to simulate UAV flight models.

Historically, the field of

The CUSCUS platform

This Section is devoted to the presentation of the CUSCUS platform: we will first introduce the logical architecture and present separately the two main simulators, FL-AIR and NS-3, on which we have built our tool, and then we will show how we put these two blocks to interwork towards the first integrated control-network simulator specific for fleets of UAVs.

Use cases

To demonstrate the validity and the effectiveness of CUSCUS, we have chosen to showcase its features by analyzing its behavior in two relevant use cases. The two use cases implemented in CUSCUS are: (i) a UAV dynamic formation control that leverages Corrective Consensus, and (ii) the implementation of a Static Coverage algorithm. Both of them represent important scientific challenges for the worlds of networking and control. In the first use case a networked fleet of UAVs is required to follow

Performance evaluation

We performed a four-folded simulation campaign in order to display the features of CUSCUS and evaluate the feasibility of its deployment. The first campaign aims at showing the impact of CUSCUS on its host system. The second campaign aims at evaluating the simulator’s ability in integrating accurate control models. The capability to incorporate real-world UAV parameters into network-oriented simulations is the object of the third simulation campaign. The last campaign summarizes the simulator

Conclusion and future works

In this paper, we have presented CUSCUS, a novel framework for modeling and simulation distributed Networked Control Systems, and more specifically fleets of Unmanned Aerial Vehicles (UAVs). Differently from the existing tools, our software is able to take into account both realistic UAV micro-mobility, drone dynamics and wireless communications, via the integration of the FL-AIR suite with the mainstream network simulator NS-3. Furthermore, CUSCUS enables realistic 3D simulations by importing

Acknowledgments

This work has been carried out in the framework of the DIVINA Challenge Team, which is funded by the Labex MS2T program. Labex MS2T is supported by the French Government, through the program “Investments for the future”, managed by the French National Agency for Research (Reference ANR-11-IDEX-0004-02).

Nicola Roberto Zema was born in Italy on 11 January, 1986. He received his B.S. and M.S. degree from University “Mediterranea” of Reggio Calabria, Italy in 2009 and 2011, respectively. Since then he has been a Ph.D. Student in University “Mediterranea” of Reggio Calabria while spending several abroad periods in Université de Technologie de Compiègne and Inria Lille. Currently he is a Research Fellow at Université de Technologie de Compiègne, France. His current research activities include

References (52)

  • A. Trotta et al.

    Connectivity recovery in post-disaster scenarios through cognitive radio swarms

    Comput. Netw.

    (2015)
  • M. Kudelski et al.

    Robonetsim: an integrated framework for multi-robot and network simulation

    Robot. Auton. Syst.

    (2013)
  • Federal Communication Commission (fCC) Report, Deployable Aerial Communications Architecture in Emergency...
  • Cisco Systems, Cisco Aironet 802.11a/b/g Wireless CardBus Adapter, Data Sheet, available on line at....
  • L. Gupta et al.

    Survey of important issues in UAV communication networks

    IEEE Commun. Surv. Tutor.

    (2016)
  • M. Erdelj et al.

    Uav-assisted disaster management: applications and open issues

    Proceedings of the 2016 International Conference on Computing, Networking and Communications (ICNC)

    (2016)
  • S. Hayat et al.

    Survey on unmanned aerial vehicle networks for civil applications: a communications viewpoint

    IEEE Commun. Surv. Tutor.

    (2016)
  • M. Erdelj et al.

    Help from the sky: leveraging UAVs for disaster management

    IEEE Pervasive Comput.

    (2017)
  • S. Rao et al.

    Sliding mode control-based autopilots for leaderless consensus of unmanned aerial vehicles

    IEEE Trans. Control Syst. Technol.

    (2014)
  • M. Yajnik et al.

    Measurement and modelling of the temporal dependence in packet loss

    Proceedings of the Eighteenth Annual Joint Conference of the IEEE Computer and Communications Societies (INFOCOM ’99)

    (1999)
  • ZengY. et al.

    Wireless communications with unmanned aerial vehicles: opportunities and challenges

    IEEE Commun. Mag.

    (2016)
  • T.T. Mac et al.

    AR.Drone uav control parameters tuning based on particle swarm optimization algorithm

    Proceedings of the 2016 IEEE International Conference on Automation, Quality and Testing, Robotics (AQTR)

    (2016)
  • K. Daniel et al.

    Cognitive agent mobility for aerial sensor networks

    IEEE Sens. J.

    (2011)
  • S. Morgenthaler et al.

    UAVNet: a mobile wireless mesh network using unmanned aerial vehicles

    Proceedings of the 2012 IEEE Globecom Workshops

    (2012)
  • S. Rosati et al.

    Dynamic routing for flying ad hoc networks

    IEEE Trans. Veh. Technol.

    (2016)
  • CaiY. et al.

    Medium access control for unmanned aerial vehicle (UAV) ad-hoc networks with full-duplex radios and multipacket reception capability

    IEEE Trans. Veh. Technol.

    (2013)
  • E. Yanmaz et al.

    Achieving air-ground communications in 802.11 networks with three-dimensional aerial mobility

    Proceedings of IEEE INFOCOM

    (2013)
  • Runtime Games Ltd, Phoenix RC,...
  • H-SIM, Simdrone,...
  • A. Varga

    Omnet++

    Modeling and Tools for Network Simulation

    (2010)
  • Various Academic Subjects, Network Simulator (Version 2), available from...
  • Heudiasyc Laboratory, UMR CNRS 7253, FL-AIR: Framework Libre air, available from...
  • P. Fuxjaeger et al.

    Validation of the ns-3 interference model for ieee802.11 networks

    Proceedings of the 2015 8th IFIP Wireless and Mobile Networking Conference (WMNC)

    (2015)
  • S. Manfredi et al.

    A cooperative packet-loss-tolerant algorithm for wireless networked robots rendezvous

    Proceedings of the 2017 International Conference on Computing, Networking and Communications (ICNC)

    (2017)
  • M.D. Felice et al.

    Self-organizing aerial mesh networks for emergency communication

    Proceedings of the 2014 IEEE 25th Annual International Symposium on Personal, Indoor, and Mobile Radio Communication (PIMRC)

    (2014)
  • C. Sommer et al.

    Progressing toward realistic mobility models in vanet simulations

    IEEE Commun. Mag.

    (2008)
  • Cited by (17)

    • Cooperative UAVs framework for Mobile Target Search and tracking

      2022, Computers and Electrical Engineering
      Citation Excerpt :

      The main drawback of FlyNetSim is its lack of accuracy when many events occur in a short period of time. In [11], the authors have proposed the CUSCUS co-simulator that integrates FL-AIR and NS-3. The communication between the two simulators is done through the tap bridges and the Linux containers.

    • Data collection using unmanned aerial vehicles for Internet of Things platforms

      2019, Computers and Electrical Engineering
      Citation Excerpt :

      As we can see, when the flying height of the UAV is between 5 m and 25 m, the UAV can achieve an optimal velocity of 25 m/s. In addition, we performed several tests using a steady speed propagation delay model [25]. The MAC layer is set to ad-hoc and non-QoS configurations.

    • Enclosing and monitoring of disaster area boundary using multi-UAV network

      2023, Journal of Ambient Intelligence and Humanized Computing
    • A Co-Simulation Framework for Communication and Control in Autonomous Multi-Robot Systems

      2023, IEEE International Conference on Intelligent Robots and Systems
    View all citing articles on Scopus

    Nicola Roberto Zema was born in Italy on 11 January, 1986. He received his B.S. and M.S. degree from University “Mediterranea” of Reggio Calabria, Italy in 2009 and 2011, respectively. Since then he has been a Ph.D. Student in University “Mediterranea” of Reggio Calabria while spending several abroad periods in Université de Technologie de Compiègne and Inria Lille. Currently he is a Research Fellow at Université de Technologie de Compiègne, France. His current research activities include Controlled Mobility, Wireless Sensor Networks, Epidemic and Autonomic Networks, Distributed Networked Control.

    Angelo Trotta received his Bachelor Degree (summa cum Laude) in Computer Science in 2008, from the University of Bologna, Italy. He received his Master Degree (summa cum Laude) in Computer Science in 2011, from the University of Bologna, Italy. From October 2012 to October 2013 he was a Research Student for the italian PRIN 2009 project “STEM-Net” working on the study, modeling and simulation of cognitive radio technologies and self-organizing wireless networks. From January 2013 he is a PhD student at the Department of Computer Science and Engineering of the University of Bologna, Italy. From October 2015 to April 2016 he was a visiting researcher in the Heudiasyc laboratory at the Sorbonne Universits, Universit de technologie de Compigne, France. From January 2017 he is a research fellow on “Design and performance evaluation of swarm mobility algorithms for self-organizing wireless networks” at the University of Bologna, Italy. His research activity is focused on the design, the development and the analysis of protocols and architectures for wireless networks in multi-robot systems.

    Enrico Natalizio (Member IEEE) is currently an associate professor with Université de Technologie de Compiègne, France. He obtained his masters degree magna cum laude and his Ph.D. in omputer engineering at the University of Calabria in 2000 and 2005, respectively. In 2005–2006 he was a visiting researcher at the BWN (Broadband Wireless Networking) Lab at Georgia Tech in Atlanta, GA, USA. From 2006 till 2010, he was a research fellow at the Titan Lab of the Universit della Calabria, Italy. In October 2010, he joined POPS team at Inria Lille Nord Europe, France as a postdoc researcher. His research interest include robot & sensor networks and swarm communications with applications on networking technologies for disaster prevention and management. He is currently an associated editor of Elsevier Ad hoc Networks, and Elsevier Digital Communications and Networks.

    Marco di Felice received the Laurea (summa cum laude) and Ph.D. degrees in computer science from the University of Bologna, Italy, in 2004 and 2008, respectively. In 2007, he was a visiting researcher with the Broadband Wireless Networking Laboratory, Georgia Institute of Technology, Atlanta, GA, USA. In 2009, he was a visiting researcher with Northeastern University, Boston, MA, USA. Currently, he is an Associate Professor in computer science with the University of Bologna. His research interests include self-organizing wireless networks, cognitive radio and vehicular systems, mobile applications and services. Prof. Di Felice currently serves on the editorial board of Elseviers Ad Hoc Networks journal. He authored more than 80 papers on wireless and mobile systems. He joint several national and international research projects. He received the Best Paper Award at the Association for Computing Machinery International Symposium on Mobility Management and Wireless Access (MOBIWAC) in 2012 and at the IEEE Annual Mediterranean Ad Hoc Networking Workshop (MED-HOC-NET) in 2013.

    Luciano Bononi received the Laurea in Computer Science (Summa cum laude) and Ph.D. in Computer Science from the University of Bologna. He is currently Associate Professor at the Department of Computer Science and Enginenering of the University of Bologna. His research activity includes the design and analysis of protocol architectures for wireless networks (wireless ad hoc, vehicular, mesh, sensor, cognitive radio), Networks on Chip (NoC), modeling and simulation of complex systems, Internet of Things and applications/services based on iOS/Android mobile devices for smart and sustainable mobility and Internet of Energy. He authored 7 book chapters and more than 90 publications on journals and conferences, including several best paper awards, and he is Associate Editor of 7 international journals. He had Chair roles in more than 12 international Conferences and Workshops.

    View full text