Unmanned aerial vehicle group task reliability simulation method based on effective task loop

Abstract

The invention discloses an unmanned aerial vehicle group task reliability simulation method based on an effective task loop, and relates to the field of unmanned aerial vehicle task reliability calculation. According to the method, firstly, the unmanned aerial vehicle group heterogeneous network is constructed according to the characteristics by analyzing the composition and association relationship of the unmanned aerial vehicle group system; secondly, establishing an unmanned aerial vehicle group effective task loop mathematical model on the basis of the constructed heterogeneous network; and finally, by analyzing the failure type and the dynamic reconstruction process of the unmanned aerial vehicle group, constructing a task reliability evaluation model, and providing a Monte Carlo-based unmanned aerial vehicle group task reliability simulation method. Scientific theoretical reference is provided for current unmanned aerial vehicle task reliability simulation and even other aircraft group task reliability simulation, and meanwhile certain engineering applicability is achieved.

Description

Technical Field

[0001] The present invention relates to the field of unmanned aerial vehicle (UAV) mission reliability modeling and simulation, and in particular to a UAV swarm mission reliability simulation method based on an effective mission loop. Background Art

[0002] With the improvement of the ability and standards of drones to perform tasks, drones contain more complex systems. The increase in the number of drones has led to more complex and prominent problems such as degradation and failure of drone swarm systems. For a long time, the industry has used reliability to measure the degradation and failure of the system, and used it to evaluate the possibility of the system performing the corresponding functions without failure within a specified time period. However, with the complexity of the system and the diversification of the tasks to be performed, the traditional reliability evaluation indicators are not enough to fully characterize the operating mechanism and dynamic characteristics of complex systems. Therefore, the indicator of mission reliability has been widely used in complex systems such as manufacturing systems and drone systems. The definition of mission reliability is the ability of the system to complete the specified functions within the specified mission profile. It reflects the system's requirements for mission success and is the ability of the system to complete the task in the natural environment in peacetime and the hostile environment in wartime, without considering human hostile factors. In view of the characteristics of the system complexity, individual heterogeneity and task coordination of drone swarms, mission reliability can more appropriately describe its system status, reflect more intuitive and valuable indicators, and also be more in line with engineering practice.

[0003] Considering that drone swarms have network structure properties, some researchers have used complex network theory to study drone swarms, but generally speaking, there are several problems. First, the characteristics of the research object are inconsistent with drones, and the characteristics of drones containing multiple mission payloads are not considered; second, the established indicator system cannot directly reflect the mission reliability of the swarm, but is measured through several indirect indicators; third, some simulation methods are divorced from engineering reality and lack certain practicality; fourth, some simulation methods are cumbersome to calculate and are not conducive to practical application.

[0004] In view of the above problems, this paper will combine previous research and study the mission reliability simulation method of drone swarms. First, this paper will describe the drone swarm system, analyze its unique system composition and association relationship, and construct a drone swarm heterogeneous network based on its characteristics. Secondly, a drone swarm mission loop mathematical model is established based on the constructed heterogeneous network. Finally, by analyzing the failure type and dynamic reconstruction process of the drone swarm, its mission reliability evaluation model is constructed, and a drone swarm mission reliability simulation method based on Monte Carlo is proposed. Summary of the invention

[0005] The purpose of the present invention is to propose a UAV swarm mission reliability simulation method based on an effective mission loop to address the problems of difficult calculation of UAV swarm mission reliability, inaccurate and unreasonable calculation, and few calculation methods.

[0006] To achieve the above objectives, the present invention constructs a drone swarm heterogeneous network according to its characteristics by analyzing the unique system composition and association relationship of the drone swarm; establishes a drone swarm mission loop mathematical model based on the constructed heterogeneous network; constructs its mission reliability evaluation model by analyzing the failure type and dynamic reconstruction process of the drone swarm, and implements a drone swarm mission reliability simulation method based on Monte Carlo. Therefore, the technical solution of the present invention is: a drone swarm mission reliability simulation method based on an effective mission loop, the method comprising:

[0007] Step 1: UAV swarm system description and network model establishment;

[0008] Step 1.1: Analyze and clarify the composition of the drone swarm system and the interaction between its functional systems;

[0009] Step 1.2: Define the UAV swarm heterogeneous network model G;

[0010] Step 2: drone swarm mission loop definition;

[0011] Step 2.1: Define the task loop: When a drone swarm performs a task, the loop composed of nodes that can perform subtasks is called a task loop.

[0012] Step 2.2: Combine the network model G to describe four different types of mission loops of the drone swarm;

[0013] Step 3: Failure type analysis and dynamic reconstruction process;

[0014] Step 3.1: Analyze the failure types of the drone swarm network, including node failure and edge failure;

[0015] Step 3.2: Reconstruction process of node failure, including communication node failure and task load node failure;

[0016] Step 3.3: Reconstruction process of edge failure;

[0017] Step 4: Constructing a task reliability evaluation model based on effective task loops;

[0018] Step 4.1: Establish an effective task loop model;

[0019] Step 4.2: Construct task reliability indicators based on effective task loops.

[0020] Step 5: Monte Carlo-based mission reliability simulation algorithm is used to solve the mission reliability of the drone swarm.

[0021] The present invention firstly analyzes the composition and correlation of the drone swarm system, and constructs a drone swarm heterogeneous network according to its characteristics; secondly, a mathematical model of the drone swarm's effective mission loop is established based on the constructed heterogeneous network; finally, by analyzing the failure type and dynamic reconstruction process of the drone swarm, its mission reliability assessment model is constructed, and a drone swarm mission reliability assessment method based on Monte Carlo simulation is proposed. It provides a scientific theoretical reference for the current drone mission reliability assessment and even other swarm mission reliability assessment, and has certain engineering applicability. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 A diagram showing the composition of the drone system provided by the present invention;

[0023] Figure 2 A schematic diagram of the interaction relationship of the drone group provided by the present invention;

[0024] Figure 3 Schematic diagram of four types of task loops provided by the present invention;

[0025] Figure 4 A schematic diagram of the task reconstruction process provided by the present invention;

[0026] Figure 5 A flow chart of the Monte Carlo-based UAV swarm mission reliability simulation algorithm provided by the present invention. DETAILED DESCRIPTION

[0027] Step 1: UAV swarm system description and network model establishment;

[0028] Step 1.1: Analyze and clarify the composition of the drone swarm system and the interactions between its functional systems.

[0029] First, we need to clarify the composition of a single UAV system and divide it into three layers according to its system function, namely, the UAV body structure and flight system layer (hereinafter referred to as the UAV platform layer), the UAV communication module layer, and the UAV mission load layer. The three layers can equivalently represent the characteristics of the UAV's mission execution process. The mission load layer includes multiple mission loads such as the perception system, decision-making system, and operation system. This article will mainly consider the above three most important mission loads. UAV system composition is as follows: Figure 1 shown.

[0030] Secondly, combined with the functional system attributes of drones, the interactive relationship of drone groups is clarified. The interactive relationship diagram of drone groups is as follows: Figure 2 shown.

[0031] Step 1.2: Define the UAV swarm heterogeneous network model as G = (V, E), where V is the point set and E is the edge set. The node V includes the communication node (abbreviated as C), the perception node (abbreviated as S), the decision node (abbreviated as D), the operation node (abbreviated as W), and the target node in the task (abbreviated as T). The nodes of the heterogeneous network G can be obtained as follows:

[0032]

[0033] Where nc, nd, ns, nw, nt are the number of the above five types of nodes respectively. At the same time, the edge set of the heterogeneous network G can also be obtained:

[0034] E={e TS ,e SD ,e DW ,e WT ,e SC ,e DC ,e CC ,e CD ,e CW} (2)

[0035] The meaning of each edge in the edge set is shown in Table 1:

[0036] Step 2: drone swarm mission loop definition;

[0037] Table 1 Meaning of edges in heterogeneous networks

[0038]

[0039]

[0040] Step 2.1: Give the definition of the task loop: When a drone swarm performs a task, the loop composed of nodes that can perform subtasks is called a task loop.

[0041] Step 2.2: Combine the network model G to describe four different types of mission loops for drone swarms:

[0042] (1) A typical single-machine task loop is “T→S1→D1→W1→T”, which includes all aspects of a single-target strike mission, including perception, decision-making, and strike.

[0043] (2) The task loop of the interaction of perception information in collaborative work has an effective operation loop of “T→S1→C1→C2→D2→W2→T”, which means that when the decision-making system of UAV 1 fails, the perception information can be given to UAV 2 through the communication system to form a task loop.

[0044] (3) The task loop for decision-making information exchange in collaborative operations is “T→S1→D1→C1→C2→W2→T”, which means that when the operating system of UAV 1 fails, the decision-making information can be given to UAV 2 through the communication system to form a task loop.

[0045] (4) The task loop of perception and decision-making information interaction in collaborative operation is “T→S1→C1→C2→D2→C2→C3→W3→T”, which means that even if UAV 1 has only the perception system working, UAV 2 has only the decision-making system working, and UAV 3 has only the operation system working, when the three machines work collaboratively, a task loop can still be formed.

[0046] The above four task loops are as follows Figure 3 shown.

[0047] Step 3: Failure type analysis and dynamic reconstruction process

[0048] Step 3.1: Analyze the failure types of the drone swarm network, including node failure and edge failure. Node failure includes communication node, perception node, decision node and operation node failure, as well as platform node failure. When the platform node fails, it means that the entire drone loses the ability to perform the mission. Edge failure means that the communication path between two drones is subject to electromagnetic interference, obstruction, etc., causing the communication system to be unable to communicate with each other.

[0049] Step 3.2: Reconstruction process for node failure

[0050] (1) When a communication node fails, it means that the UAV cannot communicate with other UAVs and cannot participate in collaborative operations. Therefore, the UAV will perform tasks independently, forming a separate task loop and no longer participate in the dynamic reconstruction of the fleet. Figure 4 As shown in (b), when C2 fails, the connection lines ① and ② between UAV 2 and UAV 1 and UAV 3 are disconnected. At this time, UAV 2 can only perform the mission alone without adopting the reconstruction strategy.

[0051] (2) When a mission payload node fails, it means that some functions of the machine are damaged and it needs to rely on the corresponding functions of other machines to form a mission loop. Therefore, the machine needs to use its communication system to contact and interact with other machines with corresponding functions. From the perspective of the network model, the communication node of the machine needs to connect to a communication node of another machine with corresponding functions to form a mission loop. At the same time, the drones that were previously connected to the machine will also be damaged due to the damage to the machine's functions, resulting in a reduction in mission loops. In particular, when the other machine does not have this function, it will have a greater impact on the collaborative mission loops in which the other machine participates. Therefore, when a mission payload node fails, in order to increase the effective mission loop in the network model, the communication node of the machine should randomly connect to a communication node of another machine with the failed function. At the same time, for the drones that were previously connected to the machine, it is also assumed that their communication nodes will also randomly connect to a communication node of another machine with the failed function. Figure 4 As shown in (c), when the sensing node S1 fails, the entire fleet loses the ability to perceive the target node T1. At this time, UAV 2 will use its sensing node S2 to connect to T1 to ensure that all targets can form a mission loop.

[0052] Step 3.3: Reconstruction process for edge failure

[0053] The mapping of the edge failure characteristics in the network model is easy to understand, that is, the connection edge of the communication nodes of two drones is disconnected due to external interference. When the connection edge is disconnected, it will directly affect the task loop of the fleet. Therefore, in the present invention, when a certain edge fails, the two nodes will randomly connect to the communication node of another drone with the same or more functions to ensure that all targets can be covered by the fleet. Figure 4 As shown in (a), when the communication path ① or ② is blocked by interference, the coordination capability of the cluster will be seriously affected.

[0054] Step 4: Construction of task reliability evaluation model based on effective task loop

[0055] Step 4.1: Build an effective task loop model

[0056] First, a mathematical model of the task loop is established. Assume that the number of drones is n, the number of target nodes T is t, and the swarm network contains n communication nodes C, perception nodes S, decision nodes D, and operation nodes W. At the same time, the information transfer matrix P is defined TS , P SD , P DW , P WT , P CC , P SC , P CD , P DC , P CW .

[0057] P TS For example, the meaning of the information transfer matrix is: the target node in the network model Knowledge Node If there is a link in the model that connects the target node T i and sensing node S j If P TS (i,j)=1, otherwise P TS (i,j) = 0. The same applies to other information transfer matrices.

[0058] Combined with the drone swarm network model, we can infer that there are four task loop connectivity formulas:

[0059]

[0060] Based on the non-repetitive path characteristics, the connectivity formula for all task loops can be obtained as follows:

[0061]

[0062] P TT The element P of the matrix TT (i,i) represents the cluster’s response to the target node T i The number of task loops of the cluster to the target node can be recorded as N i for:

[0063] N i =P TT (i,i),i∈{1,2,…,t} (8)

[0064] After obtaining the cluster's target node T i Number of task loops N i Then, define the cluster for the target node T i The effective task loop E i for:

[0065]

[0066] in The minimum number of task loops required by the operator when performing the task. This value can be adjusted according to the importance or difficulty of the task. For tasks with high importance or difficulty, it can be reasonably increased. Thus, all valid operation loops N in the entire drone swarm network can be obtained. E for:

[0067]

[0068] Step 4.2: Mission reliability index construction

[0069] In the actual mission process, the determination of whether the mission is successful does not necessarily require the destruction or interference of all targets, but allows certain errors to exist. Therefore, the present invention defines the mission completion rate M a for:

[0070] M a =N E / t (11)

[0071] And define the conditions for task success as:

[0072]

[0073] Among them, M b is the mission baseline. The above formula means that when the proportion of all targets destroyed or interfered with by the fleet during the mission is greater than or equal to the mission baseline, the mission is considered successful, otherwise the mission fails. b The value can also be determined by the operator according to the actual situation.

[0074] At the same time, considering that the mission reliability of the UAV swarm is its ability to complete the specified task on the mission baseline, this paper defines the mission reliability of the UAV swarm as R m for:

[0075]

[0076] Where N is the total number of simulations performed by the fleet under the mission profile, s is the number of successful tasks of the cluster in the simulation.

[0077] Step 5: Monte Carlo-based mission reliability simulation algorithm

[0078] Combined with the corresponding models constructed above, a task reliability simulation algorithm considering dynamic reconstruction is proposed using the Monte Carlo method, which can reasonably evaluate the reliability of UAV swarm tasks and has better engineering applicability. The algorithm flow is as follows Figure 5 As shown. The whole algorithm includes the following steps:

[0079] Step 5.1: Construct the initial heterogeneous network G = (V, E) according to the actual mission plan, including the number of UAVs participating in the mission, the number of mission targets, and the relationship between UAVs;

[0080] Step 5.2: Enter the number of simulations N, the single simulation time T, and the minimum number of task loops Reconstruction threshold N b , task baseline value M b , and the failure rate λ J , sensing node failure rate λ S , decision node failure rate λD , weapon node failure rate λ W , communication node failure rate λ C , the edge failure rate λ between communication nodes CC ;

[0081] Step 5.3: Set the simulation number variable n, simulation time variable t, and task success number variable N s , and assign initial values ​​to the three variables, let n = 1, t = 0, N s =0;

[0082] Step 5.4: Start the nth simulation;

[0083] Step 5.5: Start the tth time of the nth simulation;

[0084] Step 5.6: Using Monte Carlo simulation to simulate the failure link of the drone, first input the failure rate of the airframe node for sampling to obtain the failure situation of the drone, and then input the failure rate of the communication node, the failure rate of the perception node, the failure rate of the decision node, and the failure rate of the weapon node for sampling according to the result to obtain the failure situation of all nodes in the heterogeneous network and update the heterogeneous network;

[0085] Step 5.7: Calculate the effective task loop N based on the updated heterogeneous network structure E And the task completion rate M a ;

[0086] Step 5.8: Determine whether the drone swarm network reaches the reconstruction threshold for the first time, that is, N E Is it less than N for the first time? b , if yes, go to step 5.9, if no, go to step 5.11;

[0087] Step 5.9: Analyze the type of failed nodes, and dynamically reconstruct the above network using a reconstruction method to obtain a repaired heterogeneous network;

[0088] Step 5.10: Compute the effective task loop N of the heterogeneous network again E And the task completion rate M a ;

[0089] Step 5.11: Determine whether the simulation time T has been reached. If not, return to step 5.5; otherwise, execute step 5.12.

[0090] Step 5.12: Determine the task completion rate M a Is it greater than or equal to M? b , if yes, go to step 5.10, otherwise go to step 5.13;

[0091] Step 5.13: Calculate the number of successful tasks N s =Ns +1;

[0092] Step 5.14: Determine whether the simulation times N is reached. If not, return to step 5.4. Otherwise, execute step 5.14.

[0093] Step 5.15: Output the task reliability value R m =N s / N.

Claims

1. A UAV swarm mission reliability simulation method based on an effective mission loop, the method comprising: Step 1: Establish a drone swarm system and a drone swarm heterogeneous network; Step 1.1: Divide the drone into three layers according to system functions: drone platform layer, drone communication module layer, and drone mission payload layer; Step 1.2: Define the UAV swarm heterogeneous network G = (V, E), where V is the node set and E is the edge set; V includes the communication node set C, the perception node set S, the decision node set D, the weapon node set W, and the target node set T in the mission. The nodes of the heterogeneous network G are: Among them, nc, nd, ns, nw, nt are the number of the above five types of nodes respectively, and the edge set of the heterogeneous network G is obtained: And={and TS ,And SD ,And DW ,And WT ,And SC ,And DC ,And CC ,And CD ,And CW } (2) e TS The target information is obtained by the sensing node, e SD The perception node gives information to the decision node, e DW The decision node gives the command to the weapon node, e WT The weapon node successfully attacks the target, e SC For collaborative tasks, the sensing node gives information to the communication node, e DC For collaborative tasks, the decision node sends command information to the communication node. CC For collaborative tasks, the information transmission between the two communication nodes is CD For collaborative tasks, the communication node sends the received information to the local decision node. CW For collaborative tasks, the communication node will pass the received information to the local weapon node; Step 2: Define the drone swarm mission loop; Step 2.1: Define the mission loop: When the drone swarm performs a mission, the loop composed of nodes that can perform subtasks is called a mission loop; Step 2.2: Combine the heterogeneous network G and define different types of mission loops for the drone swarm; Step 3: Define the failure type and the dynamic reconstruction process after failure; Step 3.1: Define the failure types of the drone swarm network, including node failure and edge failure; Node failures include: communication node failure, perception node failure, decision node failure, weapon node failure, and platform node failure. When a platform node fails, it means that the entire drone loses the ability to perform its mission. Edge failure means that the communication path between two drones fails. Step 3.2: Reconstruction process of node failure, including communication node failure and mission payload node failure; mission payload node failure includes: perception node failure, decision node failure, weapon node failure; Step 3.3: Reconstruction process of edge failure; When an edge fails, the two nodes will randomly connect to another communication node of a drone with the same or more functions to ensure that all targets can be covered by the fleet; Step 4: Construct a task reliability evaluation model based on effective task loops; Step 4.1: Establish an effective task loop model; Step 4.1.1: Establish a mathematical model of the mission loop. Assume that the number of drones is n, the number of target nodes T is t, and the swarm network contains n communication nodes C, perception nodes S, decision nodes D and weapon nodes W; meanwhile, define the information transfer matrix P TS , P SD , P DW , P WT , P CC , P SC , P CD , P DC , P CW ,The subscript of the information transfer matrix indicates the information transmission between the corresponding nodes; Step 4.1.2: Information transfer matrix P TS Represents the target node T in the heterogeneous network G i ,i∈{1,2,…,t} and the sensor node S j ,j∈{1,2,…,n}, if there is a connection target node T in the model i and sensing node S j If P TS (i,j)=1, otherwise P TS (i,j)=0; According to the heterogeneous network G, the following four task loop connectivity formulas are obtained: The formula for connecting all task loops based on the non-repetitive path characteristics is: P TT The element P of the matrix TT (i,i) represents the cluster’s response to the target node T i The number of task loops of the cluster to the target node is N i for: N i =P TT (i,i),i∈{1,2,…,t} (8) t represents the total number of target nodes. i Number of task loops N i Then, define the cluster for the target node T i The effective task loop E i for: in The minimum number of task loops required by personnel when performing tasks. This value is adjusted according to the importance or difficulty of the task, and all valid operating loops N in the entire drone swarm network are obtained. E for: Step 4.1.3, use the method in step 4.1.2 to obtain the information transfer matrix P SD , P DW , P WT , P CC , P SC , P CD , P DC , P CW All valid operating loops; Step 4.2: Construction of task reliability index based on effective task loop; Define the task completion rate M a for: M a =N E / t (11) And define the conditions for task success as: Among them, M b is the mission baseline. Formula (12) indicates that when the proportion of all targets destroyed or interfered by the fleet during the mission is greater than or equal to the mission baseline, the mission is considered successful, otherwise the mission fails. The mission baseline M b The value of depends on the actual situation; Defining the mission reliability R of the drone swarm m for: Where N is the total number of simulations performed by the fleet under the mission profile, s is the number of successful missions of the fleet in the simulation; Step 5: The Monte Carlo-based mission reliability evaluation algorithm is used to solve the mission reliability of the drone swarm.

2. A method for simulating the reliability of a drone swarm mission based on an effective mission loop as claimed in claim 1, characterized in that: The task loop in step 2.2 includes: (1) Single-machine mission loop: Its mission loop is "T→S1→D1→W1→T", which includes all the links of a single-target strike mission, including perception, decision-making and strike; S1, D1, W1 represent a node in the corresponding node set; (2) The task loop of the interaction of perception information in the collaborative task, whose effective operation loop is "T→S1→C1→C2→D2→W2→T", means that when the decision system of UAV 1 fails, the perception information is given to UAV 2 through the communication system to form a task loop; (3) The task loop for decision information interaction in collaborative tasks is "T→S1→D1→C1→C2→W2→T", which means that when the weapon system of UAV 1 fails, the decision information is given to UAV 2 through the communication system to form a task loop; (4) The task loop of the interaction of perception and decision-making information in the collaborative task is "T→S1→C1→C2→D2→C2→C3→W3→T". This means that even if UAV 1 has only the perception system working, UAV 2 has only the decision-making system working, and UAV 3 has only the weapon system working, when the three drones perform collaborative tasks, a task loop can still be formed.

3. The method for simulating the reliability of a drone swarm mission based on an effective mission loop as claimed in claim 1, characterized in that: The specific method of step 3.2 is: (1) When a communication node fails, it means that the UAV cannot communicate with other UAVs and cannot participate in the collaborative mission. Therefore, the UAV will perform the mission independently, forming a separate mission loop and no longer participating in the dynamic reconstruction of the swarm. (2) When a mission payload node fails, it means that some functions of the aircraft are damaged and it needs to rely on the corresponding functions of other aircraft to form a mission loop. Therefore, when a mission payload node fails, the communication node of the aircraft randomly connects to a communication node of another aircraft that has the failed function. At the same time, for the drone that was previously connected to the aircraft, its communication node also randomly connects to a communication node of another aircraft that has the failed function.

4. The method for simulating the reliability of a drone swarm mission based on an effective mission loop as claimed in claim 1, characterized in that: The specific method of step 5 is: Step 5.1: Construct the initial heterogeneous network G = (V, E) according to the actual mission plan, including the number of UAVs participating in the mission, the number of mission targets, and the relationship between UAVs; Step 5.2: Enter the number of simulations N, the single simulation time T, and the minimum number of task loops Reconstruction threshold N b , task baseline value M b , and the failure rate λ J , sensing node failure rate λ S , decision node failure rate λ D , weapon node failure rate λ W , communication node failure rate λ C , the edge failure rate λ between communication nodes CC ; Step 5.3: Set the simulation number variable n, simulation time variable t, and task success number variable N s , and assign initial values ​​to the three variables, let n = 1, t = 0, N s =0; Step 5.4: Start the nth simulation; Step 5.5: Start the tth time of the nth simulation; Step 5.6: Using Monte Carlo simulation to simulate the failure link of the drone, first input the failure rate of the airframe node for sampling to obtain the failure situation of the drone, and then input the failure rate of the communication node, the failure rate of the perception node, the failure rate of the decision node, and the failure rate of the weapon node for sampling according to the result to obtain the failure situation of all nodes in the heterogeneous network and update the heterogeneous network; Step 5.7: Calculate the effective task loop N based on the updated heterogeneous network structure E And the task completion rate M a ; Step 5.8: Determine whether the drone swarm network reaches the reconstruction threshold for the first time, that is, N E Is it less than N for the first time? b , if yes, go to step 5.9, if no, go to step 5.11; Step 5.9: Analyze the type of failed nodes, and dynamically reconstruct the above network using a reconstruction method to obtain a repaired heterogeneous network; Step 5.10: Compute the effective task loop N of the heterogeneous network again E And the task completion rate M a ; Step 5.11: Determine whether the simulation time T has been reached. If not, return to step 5.5; otherwise, execute step 5.

12. Step 5.12: Determine the task completion rate M a Is it greater than or equal to M? b , if yes, go to step 5.10, otherwise go to step 5.13; Step 5.13: Calculate the number of successful tasks N s =N s +1; Step 5.14: Determine whether the simulation times N is reached. If not, return to step 5.

4. Otherwise, execute step 5.

14. Step 5.15: Output the task reliability value R m =N s / N.

Images