Engineering Resilience into Multi-UAV Systems

ScienceDirect ScienceDirect Procedia Computer Science 00 (2019) 000–000

Available online at www.sciencedirect.com

Available online at www.sciencedirect.com Procedia Computer Science 00 (2019) 000–000

ScienceDirect

www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia

Procedia Computer Science 153 (2019) 9–16

17th Annual Conference on Systems Engineering Research (CSER) 17th Annual Conference on Systems Engineering Research (CSER)

Engineering Resilience into Multi-UAV Systems Engineering into Multi-UAV a EdwinResilience Ordoukhanian *, Azad M. Madnia Systems a Department of Astronautical Engineering,a Systems Architecting and Engineering, Viterbi School of Engineering, University of Southern California, Los Angeles, CA, 90089, USA a Department of Astronautical Engineering, Systems Architecting and Engineering, Viterbi School of Engineering, University of Southern California, Los Angeles, CA, 90089, USA a

Abstract

Edwin Ordoukhanian *, Azad M. Madni

Abstract Multi-UAV Operations are an area of great interest in government, industry, and research community. In multi-UAV operations, a group of unmanned aerial vehicles (UAVs) are deployed to carry out missions such as search and rescue, or disaster relief. This Multi-UAV Operations are ansystem area of in great government, industry, and research community. In multi-UAV paper discusses multi-UAV theinterest face ofindisruptions from a system-of-systems perspective. Each UAV,operations, with somea group of unmanned aerial vehicles (UAVs) aretodeployed carry aout missions such as search or disaster This decision-making capability, is assigned a task performtowithin multi-UAV system. When and theserescue, vehicles operate relief. in an open paper discusses multi-UAV the face disruptions frombecomes a system-of-systems perspective. Eachwords, UAV, multi-UAV with some operational environment, the system ability in to cope withofdisruptive events especially important. In other decision-making is methodological assigned a task framework to performtowithin a multi-UAV system. When these vehicles an open system needs to becapability, resilient. A dynamically evaluate resilience alternatives duringoperate missioninoperation operational to copeallows with system disruptive eventsappropriate becomes especially other context. words, multi-UAV is discussed environment, in this paper. the Thisability framework to select alternativeimportant. given the In current Simulation system needsthat to besuch resilient. A methodological framework dynamically alternativesfor during missionsystem. operation results show framework can be created and used to dynamically to evaluate resilience resilient alternatives multi-UAV is discussed in this paper. This framework allows system to select appropriate alternative given the current context. Simulation results that such framework can be created and B.V. used dynamically to evaluate resilient alternatives for multi-UAV system. © 2019show Edwin Ordoukhanian, Published by Elsevier This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) © 2019 The Authors. Published by Elsevier B.V. © 2019 Edwin Ordoukhanian, Published by Elsevier B.V. Peer-review under responsibility of the scientific committee of the 17th Annual Conference on Systems Engineering Research This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) This is an open access article under the CC BY-NC-ND license (CSER) Peer-review under responsibility of the scientific committee of the (https://creativecommons.org/licenses/by-nc-nd/4.0/) 17th Annual Conference on Systems Engineering Research (CSER). Peer-review under responsibility of the scientific committee of the 17th Annual Conference on Systems Engineering Research Keywords: (CSER) Multi-UAV Systems; Resilience; Continous Decision Making; Keywords: Multi-UAV Systems; Resilience; Continous Decision Making;

1. Introduction

1. Introduction Unmanned Aerial Vehicle (UAVs) are typically deployed as group for a variety of application domains such as military reconnaissance and surveillance, search and rescue, science data collection, and payload delivery [1,2]. A Unmanned Aerialis Vehicle (UAVs) are typically group for a variety and of application domains such as Multi-UAV system essentially a network of agentsdeployed in which as managing interactions dependencies are important military reconnaissance and surveillance, search and rescue, science data collection, and payload delivery [1,2]. A to achieve successful mission. Operating multiple UAVs simultaneously has several advantages. It enables flexible Multi-UAV system is essentially a network of agents in which managing interactions and dependencies are important allocation of requirements to multiple vehicles, thereby reducing operational complexity while increasing overall to achieve successful mission. Operating multiple UAVs simultaneously has several advantages. It enables flexible allocation of requirements to multiple vehicles, thereby reducing operational complexity while increasing overall * Corresponding author. Tel.: +1-818-720-2682; E-mail address: [email protected] * Corresponding author. Tel.: +1-818-720-2682; E-mail address: [email protected] 1877-0509 © 2019 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) Peer-review ofPublished the scientific committee of the 17th Annual Conference on Systems Engineering Research (CSER) 1877-0509 ©under 2019 responsibility The Author(s). by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) Peer-review under responsibility of the scientific committee of the 17th Annual Conference on Systems Engineering Research (CSER) 1877-0509 © 2019 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) Peer-review under responsibility of the scientific committee of the 17th Annual Conference on Systems Engineering Research (CSER). 10.1016/j.procs.2019.05.050

10 2

Edwin Ordoukhanian et al. / Procedia Computer Science 153 (2019) 9–16 Edwin Ordoukhanian, Azad M. Madni/ Procedia Computer Science 00 (2019) 000–000

mission coverage. Component systems collect information from multiple sources (using onboard sensors), share with other members in the system and execute action in coordinated fashion at different locations [3]. This capability brings more time efficiency into mission execution as vehicles perform assigned (or negotiated) tasks in parallel to fulfill mission objective [1-4]. Multi-UAV systems, typically operating in open environments, are susceptible to disruptions [1,2]. Uncertainties and unexpected factors in the environment can disrupt system’s operation and adversely impact overall system performance [1-4]. The system needs to handle disruptions in dynamic environments. The ability to maintain acceptable level of performance through flexibility and adaptability in the face of disruptions is the hallmark of resilient systems [4-6]. Resilience is an important capability needed in 21st century systems and system-of-systems (SoS) [7]. Current resilience methods are rooted in safety and risk-based methods [6-8]. They require anticipating and proper planning for circumventing disruptions or recovering from potential disruptions [8]. However, since real world systems operate in open, dynamic environments, anticipating and planning for all potential disruptions is a dauting task [2,5,7]. To this end, it is desirable to achieve resilience during mission execution when system or SoS already has some degree of robustness [2]. This in fact requires high level decision-making to select proper course of action to handle disrupting event. There are many ways a disruption can be handled by the system; however, which alternative is appropriate is something that current is not addressed [1,2,4]. To overcome this shortcoming, a methodological framework that enables evaluating and selecting resilience alternatives is needed [2]. This is the focus of the preformed and presented research in this paper. The framework combines concepts from systems modeling, decision science, and military science to address the gap. This paper is organized as follow. Section 2 presents the background for multi-UAV systems and resilience in this context. Section 3 discusses current gaps in the literature and research questions. Section 4 discusses the solution approach. Section 5 presents simulation results and section 6 provides the conclusion and implications of the finding. 2. Background 2.1. Multi-UAV Systems and Operation A multi-UAV system is essentially a system-of-systems that satisfies the requirements defined by Maier [9]. Each UAV has operational independence as it performs an assigned function while also participating in the SoS. Vehicles can have different governance while participating in the SoS, which can have an impact on the interaction and communication protocols among vehicles. Furthermore, multi-UAV system can evolve with functions and purposes added, removed, and modified with experience and with changing needs or mission objectives. Multi-UAV systems exhibit emergent behavior as overall functionality of the system does not reside within any single UAV. In a multiUAV system, UAVs are geographically distributed; they primarily exchange information – not mass or energy [9,10]. Madni et al [5] identified four operational phases for multi-UAV systems. During Deployment (or Takeoff) phase multi-UAV system is put into operation. Various take-off methods can be observed in this phase. In En-Route (or Cruise) phase already deployed system flies towards target area to start the main phase of the mission. In this phase, path-planning and navigation plays a major role. Actions on Objective phase is the key part of overall multi-UAV operation. In this phase system performs the main mission. During Redeployment phase, UAVs returns to the deployment site. Additionally, within each operational phase, UAVs maneuvers fall into five patterns [2]. For instance, a quadrotor has following maneuvers. Vertical Take-off and Land, which is flying vertically to a specified altitude/ or land from a certain altitude. Hover is to stay stationary at a specified location for a certain period. Going Straight Path with/without angle is flying from point A to point B in specified time. Flying in an arc is going between two locations with an arc of radius R. Combined Maneuvers which are maneuvers such as Zig-Zag or Cuban 8 (Figure 8). Other types of UAVs can demonstrate variations of these maneuvers depending on their physical capabilities.



Edwin Ordoukhanian et al. / Procedia Computer Science 153 (2019) 9–16 Edwin Ordoukhanian, Azad M. Madni/ Procedia Computer Science 00 (2019) 000–000

11 3

2.2. Resilience in Multi-UAV Systems Resilience is a key requirement in multi-UAV system operations [1,2,4]. As these systems operate in open and dynamic environment, they are exposed to various forms of disruptions. Uncertainties and unexpected factors in the environment disrupt system’s operation and adversely impact overall performance. A multi-UAV system is called resilient if it is able to accomplish the original mission within acceptable level of performance in the face of disruptions [2]. In general, there are three categories of disruptions: external, systemic, and human-triggered [6]. External disruptions are largely associated with environmental obstacles and incidents. These disruptions are often random, and their severity and duration cannot be predicted. Systemic disruptions happen when internal component’s functionality, capability, or capacity causes performance degradation. It is perhaps the most easily detectable in technological systems when a component failure affects the functionality of the node or overall system. Humantriggered disruptions are caused by human operators inside or outside of the system boundary [2,6]. There are multiple resilience alternatives to deal with a disruption, however, not all are affordable and compatible with system/SoS constraints [2]. Therefore, before selecting an alternative the capabilities and constraints of the system and the context of the environment should be considered [1,2]. Table 1 represents list of applicable resilience alternative in multi-UAV systems which are adopted from Madni and Jackson [6]. Table 1: Applicable Resilient Alternative to Multi-UAV Systems • • • • • • •

Human Backup if system is unable to handle a disruption Pre-planned protocols to deal with known (and some unknown) disruptions using previously developed plans Physical Redundancy to have another physical component (e.g., UAV or subsystem) take over Functional Redundancy to achieve same functionality by other means Function re-allocation to re-distribute overall functionalities among remaining UAVs upon disruption Circumvention to avoid a disrupting event Reconfiguration to change SoS structure

3. Current Gaps and Research Questions Current resilient methods are rooted in safety-based analysis and are primarily limited to system design or verification stages [2,7,8]. Most of these methods require anticipating disrupting events and/or are very subjective to system designer's opinion [8]. This forces system designer to identify when, where, and how disruptions can occur and plan resilience in specific areas [7,11]. If systems were to operate in a closed and static environment, this would be quite appropriate. However, most real-life systems operate in open, dynamic, and uncertain environments [7]. Identifying all possible disruptions and mitigating them during design stage seems to be a daunting task [2]. It is important to ensure system can make reasonable high-level decisions to deal with unexpected disruptions during the operation. Limited number of techniques address resilience in real-time (i.e. during operation stage) where appropriate resilience response is chosen by the system to handle the disrupting event [2]. Safety and risk-based analysis techniques are suitable for single systems, however, their application to system-ofsystems are limited [12]. It is rarely the case that one designs constituent systems within SoS from scratch [9,10]. Existing systems are usually integrated together under SoS umbrella to satisfy mission requirement. Often time, constituent systems are fully developed systems with some degree of fault tolerance and robustness [9,10]. When integrating constituent systems and forming the SoS, the systems engineer can ensure that SoS is robust and can withstand disrupting events, however, anticipating every disruption that can occur is again a daunting task [2]. Therefore, it is important to show resilience during SoS operation to unexpected events when each system has already some degree of fault-tolerance and robustness. This also requires high level decision making to select an appropriate resilience response to disruptions. As number of alternatives to employ during system operation to cope with a disruption can be quite large, specifically which alternative is appropriate given the system condition is not addressed in current literature. Thus, main questions that presented research aims to answer are: Is there an organizing framework

12 4

Edwin Ordoukhanian et al. / Procedia Computer Science 153 (2019) 9–16 Edwin Ordoukhanian, Azad M. Madni/ Procedia Computer Science 00 (2019) 000–000

that would enable selection of appropriate resilience alternatives? Can system resilience be achieved during operation by evaluating resilience alternative's impact? If so, how? Previously, there have been attempts to categorize resilience techniques for acknowledged system-of-systems in space domain [13]. MITRE's cyber resiliency framework [14] was extended to include non-cyber disruption and main resilience attributes were identified [13]. Then resilience techniques were mapped to resilience goals and objectives. Furthermore, conflicts and harmonies among resilience techniques were identified. However, the analysis was primarily qualitative and did not have quantitative bases. Some key questions such as why one resilience technique (mechanism) is preferred over the other or how trade-off can be performed among competing resilience techniques were not addressed. The research presented in this paper aims to address these questions as well. 4. Solution Approach 4.1. Research Framework, Methodology, and Hypothesis The research framework combines elements from military science, systems modeling, and decision science. Tactics and Strategies, from military science, provide underlying construct to deal with dynamic context and changing conditions. Tactic are short time-horizon configurations or actions that contribute toward overall mission [15]. Utility Theory and Utility Function, from decision science, provide necessary theoretical underpinning for evaluating and selecting appropriate resilience alternative. Utility functions are mathematical functions that evaluate usefulness of an alternative or an option [16]. Systems Modeling provides underlying construct for alternative construction, as well as, providing necessary information to support decision making. The overall research framework is shown in figure 1. It is hypothesized that resilience of multi-UAV system during operation can be achieved if resilience alternatives (mechanisms) are evaluated by a utility function with attributes concerning safety, resources available, and mission objective where priorities of these attributes change during mission execution. For example, physical redundancy and function (task) re-allocation are two possible alternatives to deal with a disruption (e.g. loss of a member). However, specifically, which alternative is appropriate given the current context depends on multiple factors. If one alternative is chosen without proper considerations, it can potentially lead to unintended consequences such as waste of resources, improper task re-allocation, or even jeopardizing safety of vehicles.

Figure 1. Research Framework

Since resilience during operation is the focus of the performed research, simulation-based research methodology has been utilized to investigate research hypothesis. The goal is to evaluate resilience alternatives during simulation using a decision-making framework and determine best course of action during disrupting events. Furthermore, as more simulations are performed a mapping between resilience alternatives and class of disruptions can be created. Consequently, a set of heuristics can be generated to know when specific alternatives should be executed given the current situation.



Edwin Ordoukhanian et al. / Procedia Computer Science 153 (2019) 9–16 Edwin Ordoukhanian, Azad M. Madni/ Procedia Computer Science 00 (2019) 000–000

13 5

4.2. Decision Making Framework Every decision-making process includes activities such as assessing current situation, generating set of alternatives, evaluating each alternative, then selecting the best alternative [17]. The decision-making process shown in figure 2 include these components. System tactics block determines appropriate tactic based on the current situation. Once system tactic is defined, set of possible alternatives and the weighting parameters for the utility function are then selected from their respective databases. The "Construct Alternative" module is responsible for constructing the alternatives. It pulls in set of the possible alternatives from the list of resilience alternative and constructs the details of the alternative using combination of models and algorithms. For example, for task re-allocation, this block calculates time to complete the mission, covered area, amount of utilized resources, remaining resources, and safety concerns such as possible collision between vehicles. This information is then sent to “Normalizing Values” block. This block is responsible for calculating appropriate “score” for each variable mentioned earlier. These scores along with the weightings are then passed to “Evaluate Alternative” module. In this block, the utility function calculates the utility of each alternative. Once utility of each alternative is calculated, the list of alternatives with their respective utility is sent to "Select Alternative" block. In this block the alternative with highest utility is selected. Based on the selected alternative, the "Command" module issues high level command for the multi-UAV system. If this framework were to be used as a decision support system, the last two modules would be replaced by a human decision maker.

Figure 2 Decision Making Framework

4.3. Normalizing Values and Calculating Utilities Utility functions are mathematical equations that are used to rank alternative based on their usefulness [16]. Utility functions can be used to make informed decisions about a situation [16]. Hence, they can be applied to make an informed decision about which resilience mechanism (alternative) to execute to deal with a disruption [2]. To construct a useful and measurable utility function, three criteria needs to be considered: safety, goal, and resource constraints [18]. Assuming utility function is the linear sum of these attributes, it takes the following form (eq. 1) 𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈𝑈 = 𝑤𝑤𝑤𝑤1 𝑀𝑀𝑀𝑀𝑈𝑈𝑈𝑈𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑈𝑈𝑈𝑈𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 + 𝑤𝑤𝑤𝑤2 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 + 𝑤𝑤𝑤𝑤3 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅

(1)

Mission is sum of scores associated with Mission-ETC and Accomplished-Mission. Mission-ETC (Estimated Time of Completion) is the score associated with completing mission on time. Depending on the mission type, two types of functions can be used to calculate the score ETC. Assuming optimum value (desired ETC) and deviations from this value is defined by the mission planner, a bell shape curve can be constructed to score mission completion time. In this bell shape curve, the score associated with the optimum value is equal to 1 and at the end points (the deviations)

Edwin Ordoukhanian et al. / Procedia Computer Science 153 (2019) 9–16 Edwin Ordoukhanian, Azad M. Madni/ Procedia Computer Science 00 (2019) 000–000

14 6

the score is equal to zero. To this end, Wymore’s Standard Scoring Function #6 can be used [19]. The reason for choosing SSF6 is because of its shape, as well as, being a continuous function and easily modifiable. For time critical missions such as finding a lost hiker, the sooner the mission is accomplished the better. Therefore, even if the mission is completed before desired ETC, the score will still be equal to 1. Assuming the maximum allowable time to complete the mission is also given, a smooth curve can be constructed between the desired value and the maximum allowable value. To this end, Wymore's Standard Scoring Function #10 [19] can be chosen for this purpose due to its shape. This function is furthermore modified and combined with a step function to construct the overall desired shape. Accomplished-Mission is the score associated with the amount of accomplished mission. For example, if total number of waypoints to cover an area is known then essentially the score is the ratio of covered waypoints over total number of waypoints. Resource is the score associated with utilizing resources (i.e. deploying UAVs on reserve to replace an incapacitated vehicle). Assuming UAVs on reserve have equal functionalities and worth, then each vehicle's worth is essentially 1/(total number of UAVs on reserve). Therefore, a linear curve can be constructed where score associated with zero utilization is equal to 1 and the score associated with maximum utilization is equal to zero. Safety is the score associated with vehicles safety (i.e. collision). If an alternative puts vehicle on collision course, it receives score of zero. Alternatively, if it does not, it receives score of 1. Therefore, the scoring function for safety is essentially a step function. 4.4. Multi-UAV System Tactics A multi-UAV system dealing with a disruption is very similar to a commander or a chess master. To the chess master or the commander in battlefield, the moves made by the opponent or enemy can be viewed as disruptions. Tactics are then used to deal with these disruptions. A multi-UAV system can also benefit from similar tactics when dealing with a disruption. In general, tactics can be categorized as offensive, defensive, or trade-off tactics [2]. The tactics for multi-UAV system are defined in table 2. Table 2: Multi-UAV System Tactics • • • •

Execute Mission: The priority is to accomplish the mission within the required performance or better. Safety and resource consumption have equal low priorities. Resource Conservation - Execute Mission: The priority is to conserve resources. Executing the mission and safety have second and third priorities, respectively. Safe/Secure - Execute Mission: The priority is safety and security. Execution of the mission and resource consumption have second and third priorities, respectively. Safe/Secure - Resource Conservation: Safety and security, and resource conservation have both equal high priority while executing the intended mission has low priority

The first tactic is an offensive tactic where accomplishing the mission is the priority. The second and third tactics are trade-off tactics where system is trying to balance the mission execution with resource utilization and safety concerns. The last tactic is a defensive tactic where accomplishing the mission is not a priority and surviving is more important. To switch between tactics, set of transition logics needs to be defined. For brevity, a subset of transition logic is presented in table 3, where R is available resource, D is amount of damage to the system, and A is amount of accomplished mission. All the values in table 3 are scaled between 0 and 1. Table 3: Transition Logic between Tactics Tactic Execute Mission

Transition Logic IF (R > 0.75) & (D < 0.25) & (A < 1); IF (0.25 < R < 0.75) & (A <=0.25) & (D<= 0.25)

Resource Conservation-Execute Mission

IF (R < 0.25) & (A <=0.25) & (D <=0.5)

Safe/Secure-Execute Mission

IF (D < 0.75 & D > 0.25) & (R > 0.25) & (A > 0.25)

Safe/Secure-Resource Conservation

IF (D > 0.75) & (R < 0.25) & (A > 0.25)



Edwin Ordoukhanian et al. / Procedia Computer Science 153 (2019) 9–16 Edwin Ordoukhanian, Azad M. Madni/ Procedia Computer Science 00 (2019) 000–000

15 7

5. Simulation Setup and Results The simulation platform is created using commonly used programming languages: Python and Java. Java being an object-oriented programing language lends itself nicely into the problem domain as main class of vehicles can be defined and multiple agents (vehicles) can be instantiated dynamically during simulation. Python being a popular programming language provides many libraries for data visualization. In this research, an agent-based modeling technique is used to model multi-UAV system. Each UAV (i.e. a quadcopter) is an agent. To run simulation with requisite fidelity, each agent is equipped with dynamics model of the vehicles. To this end, system of differential equations that mimic actual vehicle’s behavior has been implemented in Java. To ensure Ordinary Differential Equations (ODE) solver within Java produces results with minimum errors, simulation runs at 100Hz. Additionally, system tactics and resilience alternatives are also determined and evaluated at every simulation time step. A monitoring scenario has been considered to demonstrate the framework. The goal of the simulation is to show the proposed framework can evaluate resilience alternatives in real-time. In this scenario there are N number of UAVs on reserve. The mission planner designs the mission and utilizes some of the vehicles while keeping the reset at reserve. During the mission execution, the multi-UAV system loses one of the UAVs. This is viewed as a disruption to the commander (leader UAV). A decision must be made to either continue the mission without taking any actions, deploy a new vehicle to replace the incapacitated vehicle, or re-allocate the remaining task(s) to satisfy the mission requirement. Figure 3 shows the result of a simulation run for two separate cases. In Case I, 2 UAVs are deployed, and 1 UAV is on reserve (N=3). UAV1 is considered the leader and UAV2 the subordinate. They exchange messages every step of the simulation. Every message includes information such as their current states and the decision-making criteria. 30 seconds into the mission UAV1 loses communication with the other vehicle. This is a disruption that puts the mission success at risk. The leader now must decide. In this case, the leader decides to re-allocate the remaining task and essentially take over the remaining task of the UAV2. It successfully finishes the mission in about 2.5 minutes which is well within the acceptable time boundaries. Even though there is a spare UAV that can be deployed to replace the missing UAV, within the current context the best decision is to re-allocate the task. The mission starts with “offensive” tactic to ensure some portion of the mission is accomplished. After multi-UAV system loses a member and re-allocates the task, it changes the tactic to “Safe/Secure-Execute Mission”. This means the system has now become somewhat risk averse (as it lost 50% of its original size) and it is going to put safety and security first when faced with future disruptions since it might be operating in a hostile environment.

Figure 3 Simulation Results

16 8

Edwin Ordoukhanian et al. / Procedia Computer Science 153 (2019) 9–16 Edwin Ordoukhanian, Azad M. Madni/ Procedia Computer Science 00 (2019) 000–000

Case II in figure 3 shows result of simulation run where 2 UAVs are deployed and there is no additional resource (N=2). Consequently, the multi-UAV system is starting the mission with “Resource Conservation- Execute Mission” tactic. In this case, “Deploy” alternative is not present in the alternative evaluation pool and it does not get evaluated. Therefore, the decision is between continuing the mission or re-allocating remaining task. Results of these simulations can be summarized by the following simple heuristic, even if there are enough resources, utilizing them is not always the right choice. 6. Summary Multi-UAV operations are important for a variety of applications in both military and civilian domains. In this paper, multi-UAV systems are discussed from a system-of-systems perspective. As these systems operate in an open environment, the need for resilience is emphasized. This paper addressed the gap associated with developing a decision-making framework for real-time evaluation of resilience mechanisms to handle disrupting events. It focused on the resilience of multi-UAV systems and demonstrated that the resilience of these systems can be achieved by evaluating resilience alternatives (mechanisms) during system operation. It introduced the notion of multi-UAV system tactics to dynamically change the priorities of the system based on the current state of the mission, remaining resources, and the damage(s) caused by the disrupting events. The simulation results validated the research hypothesis that such real-time evaluation is possible and demonstrated that presented framework is viable. References [1] Ordoukhanian, E.; Madni, A.M. Resilient Multi-UAV Operation: Key Concepts and Challenges. 54th AIAA Aerospace Sciences Meeting 2016, pp. 1–7. doi:10.2514/6.2016-0475. [2] Ordoukhanian E, Madni AM. Model-Based Approach to Engineering Resilience in Multi-UAV Systems. Systems. 2019; 7(1):11. [3] Almeida, P.; Bencatel, R.; Gonçalves, G.; Sousa, J. Multi-UAV integration for coordinated missions. Encontro Científico de Robótica, Guimarães 2006. [4] Ordoukhanian, E.; Madni, A.M. Toward Development of Resilient Multi-UAV System-of-Systems. AIAA Space 2016; American Institute of Aeronautics and Astronautics: Reston, Virginia, 2016; pp. 1–8. doi:10.2514/6.2016-5414. [5] Madni, A.M.; Sievers, M.; Humann, J.; Ordoukhanian, E.; Boehm, B.W.; Lucero, S. Formal Methods in Resilient Systems Design: Application to Multi-UAV System-of-Systems Control. Conference on Systems Engineering Research (CSER); Springer: Redondo Beach, 2018 [6] Madni, A.M.; Jackson, S. Towards a Conceptual Framework for Resilience Engineering. IEEE Systems Journal 2009, 3, 181–191. doi:10.1109/jsyst.2009.2017397. [7] Madni, A.M. Transdisciplinary systems engineering: exploiting convergence in a hyper-connected world; Springer, 392 2018. [8] Uday, P.; Marais, K.B. Resilience-based System Importance Measures for System-of-Systems. Procedia Computer. Sci. 2014, 28, 257–264. doi:10.1016/j.procs.2014.03.033. [9] Maier, M.W. Architecting principles for systems-of-systems. INCOSE International Symposium. Wiley Online Library, 1996, Vol. 6, pp. 565– 573. [10] Madni, A.M.; Sievers, M. System of Systems Integration: Key Considerations and Challenges. Systems Engineering 2013, 17, 330–347. doi:10.1002/sys.21272. [11] Haghnevis, M.; Askin, R.G. A Modeling Framework for Engineered Complex Adaptive Systems. IEEE Syst. J. 2012, 6, 520–530. doi:10.1109/JSYST.2012.2190696. [12] Gorod, A.; Gandhi, S.J.; Sauser, B.; Boardman, J. Flexibility of System of Systems. Glob. J. Flex. Syst. Manag. 2008, 9, 21–31. [13] Bodeau, D.; Brtis, J.; Graubart, R.; Salwen, J. Resiliency Techniques for Systems-of-Systems; The MITRE Corporation: Bedford, MA, USA, 2013. [14] Bodeau, D.; Graubart, R. Cyber Resiliency Engineering Framework; MTR110237; The MITRE Corporation: Bedford, MA, USA, 2011. [15] Training UA, Command D. comp. Army Doctrine Publication 1-01: Doctrine Primer. [16] Abbas, A.E. Constructing Multi-Attribute Utility Functions for Decision Analysis. In Risk and Optimization in an Uncertain World; INFORMS: Maryland, USA , 2010; pp. 62–98. [17] Rouse, W.B.; Rouse, S.H. A Framework for Research on Adaptive Decision Aids; Technical Report; Air Force Aerospace Medical Research Laboratory, USA , 1983. [18] Madni, A.; Freedy, A. Decision aids for airborne intercept operations in advanced aircrafts 1981. [19] Wymore, A.W. Model-Based Systems Engineering; CRC Press, 1993.