Method, device and medium for modeling and predicting availability of drone swarm storage

Abstract

The application belongs to the technical field of unmanned aerial vehicle clusters, and specifically discloses a method and device for modeling and predicting the storage availability of unmanned aerial vehicle clusters and a medium, wherein the modeling method comprises the following steps: acquiring storage period availability influencing factors, wherein the storage period availability influencing factors comprise maintenance support attributes, unmanned aerial vehicle self attributes and cluster task information; and establishing an unmanned aerial vehicle cluster storage availability model according to the storage period availability influencing factors. The application comprehensively considers the influencing factors during the storage of unmanned aerial vehicle clusters, covers multiple general quality characteristics such as reliability, maintainability and supportability, and the availability evaluation effect is more in line with the actual work requirements of the cluster.

Description

Technical Field

[0001] This invention belongs to the field of unmanned aerial vehicle (UAV) swarm technology, and more specifically, relates to a method, apparatus and medium for modeling and predicting the storage availability of UAV swarms. Background Technology

[0002] With the development of artificial intelligence technology, drone swarms are being widely used in logistics, rescue, and air quality monitoring. Depending on actual needs, only a portion of the drones in the swarm participate in each mission. The remaining large number of drones are stored as inventory, in a tiered and inactive state, until needed for a mission. The characteristics of modern high technology require drone swarms to be deployed quickly. For long-term stored drone swarms, the success rate in handling unexpected missions depends not only on the performance of the equipment itself, but also on the availability of the stored drones. Availability is a concentrated reflection of various general quality characteristics such as reliability, maintainability, and supportability. Therefore, establishing accurate and effective methods for modeling and predicting the availability of drone swarm storage is of great significance to its technological development.

[0003] In fact, UAV swarms differ significantly from traditional equipment in terms of structure, storage conditions, and mission patterns. First, UAV swarms are dynamically combinable and highly adaptive, resulting in a more complex and variable system structure. Second, the types and interactions of factors affecting the storage life of UAV swarms are complex. Even for a single payload, the impact of configuration, performance, and attributes on its health must be considered, and various interactions exist between elements with different functions and dimensions. Third, UAVs are in a non-operational state during storage, requiring a comprehensive consideration of multiple factors to fully describe the storage availability of the UAV swarm. Therefore, developing a scientific and effective modeling and prediction method for UAV swarm storage availability, tailored to the characteristics and storage features of UAV swarms, is an urgent problem to be solved.

[0004] Currently, research on long-term storage availability modeling and prediction for UAV swarms is relatively limited. However, significant research has been achieved in storage studies of pumped-storage hydropower station equipment, aviation electromagnetic relays, communication equipment, and in the analysis and prediction of degradation states of other systems. These findings can provide valuable insights for research on long-term storage availability modeling and prediction of UAV swarms. Commonly used methods in related research include function modeling, Markov methods, and system state modeling. Function modeling generally uses functional equations to quantify component processes, often simplifying interactions into functional expressions, leading to deficiencies in accuracy and precision. Markov methods use discretized multi-state models to describe the evolution of multi-component systems and are widely used in pathology, industrial systems, computer networks, and maintenance strategy optimization, but are not suitable for long-term prediction. In fact, the collaborative relationships between individual UAVs in a UAV swarm are not simple summations; the evaluation process involves numerous interactions. System state modeling methods, including reliability block diagrams and Monte Carlo numerical simulations, struggle to meet the interaction and accuracy requirements of complex system evaluations. Multi-agent simulation methods can flexibly simulate individual behaviors and interactions, and are well-suited for complex systems with large-scale, concurrent, and polymorphic organizational structures. Therefore, in this invention, we consider introducing a multi-agent simulation method to model and predict the availability of UAV swarms during their storage period. In this method, the intelligent interaction process can be implemented using search algorithms or reinforcement learning, which can well meet the requirements of intelligence, dynamic adaptability, interactivity, and real-time performance for UAV swarm models during storage.

[0005] This invention addresses the practical need for assessing the storage status of UAV swarms. Taking into account both swarm characteristics and storage features, it proposes a UAV swarm storage status metric that covers the entire storage cycle. A multi-agent simulation method is used to build a swarm storage availability model, ultimately enabling the prediction of swarm storage availability. Therefore, the proposed method can scientifically and rapidly measure the usability of UAV swarms, providing a reference for rapid response and decision-making from storage to use. Summary of the Invention

[0006] This invention addresses the aforementioned problems in the prior art. Therefore, there is a need for a method, apparatus, and medium for modeling and predicting the storage availability of unmanned aerial vehicle (UAV) swarms. This invention comprehensively considers factors such as health degradation during UAV storage, maintenance, and support, establishing a UAV swarm storage availability model. Utilizing multi-agent simulation technology, it achieves the modeling and prediction of UAV swarm storage availability. Thus, this invention scientifically and rapidly measures the usability of UAV swarms, providing a reference for rapid response and decision-making from storage to use.

[0007] According to a first aspect of the present invention, a method for modeling the storage availability of unmanned aerial vehicle (UAV) swarms is provided, the modeling method comprising:

[0008] The factors affecting availability during the storage period are identified, including maintenance and support attributes, UAV attributes, and cluster mission information.

[0009] A drone cluster storage availability model was established based on the factors affecting storage availability.

[0010] Furthermore, a UAV cluster storage availability model is established based on the aforementioned factors affecting storage availability using the following method:

[0011] The expression for determining availability is shown in the following formula:

[0012]

[0013] In the formula: T U To enable working hours; T T Total owned time; T DW For periods when work is not possible; T O For working hours; T S Standby time; T M Maintenance time includes preventative maintenance time and restorative maintenance time; T OS For the time of use; T D To delay time;

[0014] Based on the specific mission and usage of the drone swarm, the storage period availability index, modified to meet storage task requirements, is expressed as follows:

[0015]

[0016] When assessing the availability of a drone swarm, the swarm's size characteristics are additionally considered, resulting in a drone swarm storage availability index A. S :

[0017]

[0018] In the formula: and , where represents the working time, maintenance time, and usage support time of the i-th drone, respectively; n represents the total number of drones in the cluster.

[0019] Furthermore, after obtaining the availability index A during the storage period of the drone swarm... S Subsequently, the method further includes:

[0020] During storage, the stresses caused by environmental factors, tasks, and management that reduce cluster availability are collectively referred to as degradation stresses, and denoted as... It is a parameter vector, and i = 1, 2, ..., n;

[0021] The maintenance time T is calculated using the following formula. M :

[0022]

[0023] It is a type feature, and:

[0024]

[0025] Use Z i (t)=(Z i1 (t),Z i2 (t),L,Z imi (t) as a single-machine system U i The fault vector;

[0026] If the working load C ij ∈C p If the device is in a working state at time t, then Z = ... ij (t) = 1;

[0027] When the system is in an inoperable state at time t, we have Z ij (t) = 0, j = 1, 2, ..., m i Then, the number of drones in the cluster that are in a working state at time t is:

[0028]

[0029] During the period from time point a to time point b, the single U-type drone i The degradation stress it bears is

[0030] The number of drones in the cluster that are operational between time point a and time point b is ∑ s (a,b)={∑ s (t): a<t≤b};

[0031] Up to time t, within the storage period (0,t), the instantaneous availability index of the UAV swarm storage period is expressed as:

[0032]

[0033]

[0034] According to a second aspect of the present invention, a device for modeling the storage availability of unmanned aerial vehicle (UAV) swarms is provided, the modeling device comprising:

[0035] The data acquisition unit is configured to acquire factors affecting storage availability, including maintenance and support attributes, UAV self-attributes, and cluster mission information.

[0036] The model building unit is configured to build a UAV cluster storage availability model based on the storage availability influencing factors.

[0037] Furthermore, the model building unit is further configured as follows:

[0038] The expression for determining availability is shown in the following formula:

[0039]

[0040] In the formula: T U To enable working hours; T T Total owned time; T DW For periods when work is not possible; T O For working hours; T S Standby time; T M Maintenance time includes preventative maintenance time and restorative maintenance time; T OS For the time of use; T D To delay time;

[0041] Based on the specific mission and usage of the drone swarm, the storage period availability index, modified to meet storage task requirements, is expressed as follows:

[0042]

[0043] When assessing the availability of a drone swarm, the swarm's size characteristics are additionally considered, resulting in a drone swarm storage availability index A. S :

[0044]

[0045] In the formula: and , where represents the working time, maintenance time, and usage support time of the i-th drone, respectively; n represents the total number of drones in the cluster.

[0046] Furthermore, the model building unit is further configured as follows:

[0047] During storage, the stresses caused by environmental factors, tasks, and management that reduce cluster availability are collectively referred to as degradation stresses, and denoted as... It is a parameter vector, and i = 1, 2, ..., n;

[0048] The maintenance time T is calculated using the following formula. M :

[0049]

[0050] It is a type feature, and:

[0051]

[0052] use As a standalone system U i The fault vector;

[0053] If the working load C ij ∈C p If the device is in a working state at time t, then Z = ... ij (t) = 1;

[0054] When the system is in an inoperable state at time t, we have Z ij (t) = 0, j = 1, 2, ..., m i Then, the number of drones in the cluster that are in a working state at time t is:

[0055]

[0056] During the period from time point a to time point b, the single U-type drone i The degradation stress it bears is

[0057] The number of drones in the cluster that are operational between time point a and time point b is ∑ s (a,b)={∑ s (t): a<t≤b};

[0058] Up to time t, within the storage period (0,t), the instantaneous availability index of the UAV swarm storage period is expressed as:

[0059]

[0060]

[0061] According to a third aspect of the present invention, a method for predicting the storage availability of a drone swarm is provided, based on a drone swarm storage availability model established by the modeling method described above. The prediction method includes:

[0062] Set the basic conditions for the availability simulation model;

[0063] Configure the availability simulation model logic;

[0064] Configure the parameters of the availability simulation model.

[0065] Furthermore, the basic conditions for setting the availability simulation model are as follows:

[0066] A swarm of n drones is stored at a certain location, and each drone is in the same environmental conditions.

[0067] Each drone can carry 0 to q types of payloads. Each drone cannot carry the same payload repeatedly, but there is no limit to the number of drones carrying the same payload combination.

[0068] Each type of load has different characteristics, including but not limited to the time and cost of load maintenance and replacement, and the load's own storage degradation patterns. Specific characteristics can be set according to actual conditions;

[0069] Once the load degrades to a certain extent, repair or replacement methods are used to restore it to its original usable state, meaning that all repair and replacement methods are perfect repairs;

[0070] Regardless of the number of payloads carried by a single drone, the failure of any payload will render the drone unable to complete the corresponding task, thus rendering the drone unusable on its own.

[0071] Furthermore, the setting of the availability simulation model logic specifically includes:

[0072] The objects in the abstract simulation scenario are called intelligent agents, and the interaction rules among these intelligent agents are defined, as well as the action time and state transitions of each intelligent agent are synchronized.

[0073] Furthermore, the availability simulation model parameters are determined based on the factors affecting availability during the storage period. These availability simulation model parameters include one or a combination of the following: number of UAVs, payload type, payload combination, failure time, failure rate, maintenance level, spare parts replacement time, spare parts arrival time, mean time between failures, preventive maintenance and replacement interval, mission configuration, and loss rate.

[0074] According to a fourth aspect of the present invention, a device for predicting the storage availability of unmanned aerial vehicle (UAV) swarms is provided, the device comprising:

[0075] The above-described device for modeling the availability of storage in drone swarms;

[0076] The first setting unit is configured to set the basic conditions for the availability simulation model;

[0077] The second setting unit is configured to set the availability simulation model logic;

[0078] The third setting unit is configured to set the availability simulation model parameters.

[0079] Furthermore, the first setting unit is further configured as follows:

[0080] The basic conditions for setting the availability simulation model are as follows:

[0081] A swarm of n drones is stored at a certain location, and each drone is in the same environmental conditions.

[0082] Each drone can carry 0 to q types of payloads. Each drone cannot carry the same payload repeatedly, but there is no limit to the number of drones carrying the same payload combination.

[0083] Each type of load has different characteristics, including but not limited to the time and cost of load maintenance and replacement, and the load's own storage degradation patterns. Specific characteristics can be set according to actual conditions;

[0084] Once the load degrades to a certain extent, repair or replacement methods are used to restore it to its original usable state, meaning that all repair and replacement methods are perfect repairs;

[0085] Regardless of the number of payloads carried by a single drone, the failure of any payload will render the drone unable to complete the corresponding task, thus rendering the drone unusable on its own.

[0086] Furthermore, the second setting unit is further configured as follows:

[0087] The objects in the abstract simulation scenario are called intelligent agents, and the interaction rules among these intelligent agents are defined, as well as the action time and state transitions of each intelligent agent are synchronized.

[0088] Furthermore, the third setting unit is further configured as follows:

[0089] The availability simulation model parameters are determined based on the factors affecting availability during the storage period. These parameters include one or a combination of the following: number of UAVs, payload type, payload combination, failure time, failure rate, maintenance level, spare parts replacement time, spare parts arrival time, mean time between failures (MTBF), preventive maintenance and replacement interval, mission configuration, and loss rate.

[0090] According to a fifth aspect of the present invention, a readable storage medium is provided, characterized in that the readable storage medium stores one or more programs, which can be executed by one or more processors to implement the modeling method and / or the prediction method as described above.

[0091] The present invention has at least the following beneficial effects:

[0092] (1) Taking into account the influencing factors during the storage of UAV clusters, the availability assessment results are more in line with the actual working needs of the clusters, covering multiple general quality characteristics such as reliability, maintainability and supportability.

[0093] (2) The static availability model of the cluster is extended into a dynamic model by using multi-agent simulation technology, which reflects the availability change law in the process of UAV cluster storage and maintenance. It also provides a solution for availability modeling and prediction for cluster systems with complex emergent characteristics. Attached Figure Description

[0094] Figure 1 This diagram illustrates the design concept of a method for modeling and predicting the storage availability of unmanned aerial vehicle (UAV) clusters according to an embodiment of the present invention.

[0095] Figure 2 The following are the factors affecting the storage availability of a drone swarm according to an embodiment of the present invention;

[0096] Figure 3 A schematic diagram of the drone cluster architecture according to an embodiment of the present invention is shown;

[0097] Figure 4 A state machine diagram of a single UAV during storage according to an embodiment of the present invention is shown;

[0098] Figure 5 A state machine diagram of the payload agent during storage according to an embodiment of the present invention is shown;

[0099] Figure 6 A state machine diagram of the manager's intelligent agent during storage is shown according to an embodiment of the present invention;

[0100] Figure 7 A structural diagram of a drone cluster storage availability modeling device according to an embodiment of the present invention is shown;

[0101] Figure 8 A structural diagram of a drone swarm storage availability prediction device according to an embodiment of the present invention is shown. Detailed Implementation

[0102] To enable those skilled in the art to better understand the technical solutions of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. The embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and specific examples, but this is not intended to limit the present invention. If there is no necessary sequential relationship between the various steps described herein, the order in which they are described as examples should not be considered a limitation. Those skilled in the art should understand that the order can be adjusted, as long as it does not disrupt the logical consistency between them and render the entire process impossible.

[0103] This invention provides a method for modeling the storage availability of unmanned aerial vehicle (UAV) clusters, which includes steps (1) and (2).

[0104] Step (1) Analyze the factors affecting shelf life availability:

[0105] A review of factors influencing storage availability across the entire lifecycle of drone swarms, such as... Figure 2 As shown, factors that directly affect cluster storage availability include: drone attributes, maintenance and support factors, and cluster mission information.

[0106] Step (2) Establish a storage availability model for the UAV cluster:

[0107] The basic mathematical expression for availability is:

[0108]

[0109] In the formula: T U To enable working hours; T T Total owned time; T DW For periods when work is not possible; T O For working hours; T S Standby time; T M Maintenance time includes preventative maintenance time and restorative maintenance time; T OS For the time of use; T D To delay time.

[0110] For drone swarms stored during the storage period, they are in a non-operational state for the vast majority of the time, and the overhaul or scrapping period is generally long. Based on the specific mission and usage of the drone swarm, the storage period availability index, adjusted to meet storage task requirements, is as follows:

[0111]

[0112] During the mission, the swarm relies on a large number of drones to achieve economies of scale and complete the task, such as... Figure 3 Let S be a swarm of n drones, where S = {U1, U2, L, U...} n-1 U n Each single U i Equipped with one or more different working loads C p p = 1, 2, L, q, single-machine U i The number of working payloads carried is m i In other words, when using availability to assess the status of a drone swarm, it is necessary to additionally consider the swarm's size characteristics. Based on equation (2), a drone swarm storage availability index is proposed, denoted as A. S :

[0113]

[0114] In the formula and , where represents the working time, maintenance time, and usage support time of the i-th drone, respectively; n represents the total number of drones in the cluster.

[0115] During storage, the environmental stress borne by each drone per unit time gradually reduces its availability. All stresses, including those related to the environment, mission, and management, that may cause a decrease in cluster availability are collectively referred to as degradation stress, and denoted as... It is a parameter vector (taking the gamma distribution as an example). (These can be shape and rate parameters) and i = 1, 2, L, n.

[0116] During maintenance, the maintenance time required for each individual machine is related to the number of maintenance operations for each of its payloads. Other inherent properties such as mean time between failures (MTBF) If related to, then:

[0117]

[0118] It is a type feature, and:

[0119]

[0120] In a swarm, drones with the same payload each undertake the same sub-task. If a payload fails, the drone carrying that payload is considered to have failed. While the failure of a single drone does not directly cause the entire swarm to fail, over time, some individual drones may become unable to perform the corresponding tasks in the future. The task allocation within the swarm will change. Although the swarm may continue to function, other drones undertaking the same sub-tasks as the failed drone will need to work harder to achieve the same results.

[0121] Therefore, a single machine failure increases the load on other machines in the same configuration, causing the cumulative downtime to gradually increase to an unacceptable level. As a standalone system U i The fault vector, if the working load C ij ∈C p If the device is in a working state at time t, then Z = ... ij If (t) = 1, and the system is in an inoperable state at time t, then Z = 1. ij (t) = 0, j = 1, 2, ..., m i Then, the number of drones in the cluster that are in a working state at time t is:

[0122]

[0123] During the period from time point a to time point b, the single U-type drone i The degradation stress it bears is Similarly, the number of drones in the cluster that are operational during the period from time a to time b is . Likewise, up to time t, the instantaneous availability index of the drone cluster during the storage period (0,t) can be expressed as:

[0124]

[0125]

[0126] This invention also provides a method for predicting the storage availability of unmanned aerial vehicle (UAV) clusters. Based on the UAV cluster storage availability model established by the modeling method described above, the prediction method further includes steps (3) to (5) after completing steps (1) and (2).

[0127] Step (3) Set the basic conditions for the availability simulation model

[0128] The basic scenario of the drone swarm availability simulation model during storage is as follows: drones are stored together in a non-working state, and all drones in the swarm can carry different quantities and types of payloads, i.e., the swarm is a heterogeneous drone swarm. Due to differences in payload structure, materials, and other characteristics, their health degradation patterns and maintenance time costs are also different.

[0129] The basic conditions for the simulation are set as follows:

[0130] ① A cluster of n drones is stored at a certain location, and each drone is in the same environmental conditions;

[0131] ② Each drone can carry 0 to q types of payloads. Each drone cannot carry the same payload repeatedly, but there is no limit to the number of drones carrying the same payload combination.

[0132] ③ The characteristics of each load are not entirely the same, including but not limited to the time cost of load maintenance and replacement, and the storage degradation law of the load itself. Specific characteristics can be set according to the actual situation;

[0133] ④ After the load degrades to a certain extent, repair or replacement measures are taken to restore it to its original usable state, that is, all repair and replacement measures are perfect repairs;

[0134] ⑤ Regardless of the number of payloads carried by a single drone, the failure of any payload will cause the drone to lose its ability to complete the corresponding task, and the drone can be considered unusable on its own.

[0135] (4) Configure the availability simulation model logic

[0136] In the scenario of drone swarm storage, the system-level individual drones, the equipment-level payloads, and the maintenance and support personnel will spontaneously generate certain behaviors and undergo state transitions during the storage period. The representation rules of the state machine diagram are shown in Table 1.

[0137] Table 1. Representation rules for state machine diagrams

[0138]

[0139] The state machine diagram during the storage of a single UAV is as follows: Figure 4 As shown. During storage, drones may experience failures, require repair, or need replacement, resulting in both available and unavailable states. The drone's own structure can also be considered a payload. When a drone receives a failure message from any payload, it transitions from an available state to an unavailable state. Additionally, the drone will send a message indicating its unavailability to the administrator; once the drone receives a message that all payloads have been restored to their original state, it immediately transitions back to an available state.

[0140] The state machine diagram during the storage of the payload intelligent agent is as follows: Figure 5 As shown, the initial state of the loads is normal storage. After a period of time, the health of the components falls below a threshold. Two scenarios may occur: either the component's condition deteriorates over time and can be restored with simple repairs, or the degradation is irreversible and replacement is necessary to restore its original state. When either of these scenarios occurs, the load must both transmit the information requiring repair or replacement to the management personnel and receive a notification of the personnel's arrival. Only when the management personnel arrive and begin work can the load enter the repair or replacement phase. Finally, after a certain period of repair or replacement work, the load is restored to its original condition and returns to normal storage.

[0141] The state machine diagram of the manager's intelligent agent during storage is as follows: Figure 6 As shown, the management agent is in a routine maintenance state. Following a pre-defined maintenance strategy, it conducts a unified inspection of the drone swarm at regular intervals and performs reactive maintenance. The time interval between the manager's arrival at the swarm location and the start of maintenance is negligible; however, the time interval before replacing the drone's payload is taken into account. Once the manager receives a message that the drones are back to usable status, they can return for maintenance and await the next maintenance period.

[0142] Considering the storage requirements of drone swarms, other intelligent agent factors such as spare parts and spare parts transport vehicles can be regarded as having sufficient quantity in the model and are not the main objects of the drone swarm storage availability model behavior modeling.

[0143] Step (5) Set the parameters of the availability simulation model

[0144] For drone swarms, availability-influencing factors should be represented as parameters in the simulation model. To ensure the simulation logic, process, and parameters are reasonable and reflect the characteristics of the swarm, the specific index settings for the simulation model are shown in Table 2. These parameters are determined and input into the model before the simulation begins, and are continuously accessed as needed during the simulation.

[0145] Table 2 Parameters of the UAV Cluster Storage Availability Model

[0146]

[0147]

[0148] This invention also provides a device for modeling the storage availability of unmanned aerial vehicle (UAV) clusters, such as... Figure 7 As shown, the modeling device 700 includes:

[0149] The data acquisition unit 701 is configured to acquire factors affecting storage availability, including maintenance and support attributes, UAV self-attributes, and cluster mission information.

[0150] The model building unit 702 is configured to build a UAV cluster storage availability model based on the storage availability influencing factors.

[0151] In some embodiments, the model building unit is further configured to:

[0152] The expression for determining availability is shown in the following formula:

[0153]

[0154] In the formula: T U To enable working hours; T T Total owned time; T DW For periods when work is not possible; T O For working hours; T S Standby time; T M Maintenance time includes preventative maintenance time and restorative maintenance time; T OS For the time of use; T D To delay time;

[0155] Based on the specific mission and usage of the drone swarm, the storage period availability index, modified to meet storage task requirements, is expressed as follows:

[0156]

[0157] When assessing the availability of a drone swarm, the swarm's size characteristics are additionally considered, resulting in a drone swarm storage availability index A. S :

[0158]

[0159] In the formula: and , where represents the working time, maintenance time, and usage support time of the i-th drone, respectively; n represents the total number of drones in the cluster.

[0160] In some embodiments, the model building unit is further configured to:

[0161] During storage, the stresses caused by environmental factors, tasks, and management that reduce cluster availability are collectively referred to as degradation stresses. It is a parameter vector, and i = 1, 2, ..., n;

[0162] The maintenance time T is calculated using the following formula. M :

[0163]

[0164] It is a type feature, and:

[0165]

[0166] use As a standalone system U i The fault vector;

[0167] If the working load C ij ∈C p If the device is in a working state at time t, then Z = ... ij (t) = 1;

[0168] When the system is in an inoperable state at time t, we have Z ij (t) = 0, j = 1, 2, ..., m i Then, the number of drones in the cluster that are in a working state at time t is:

[0169]

[0170] During the period from time point a to time point b, the single U-type drone i The degradation stress it bears is

[0171] The number of drones in the cluster that are operational between time point a and time point b is ∑ s (a,b)={∑ s(t): a<t≤b};

[0172] Up to time t, within the storage period (0,t), the instantaneous availability index of the UAV swarm storage period is expressed as:

[0173]

[0174]

[0175] It should be noted that the UAV swarm storage availability modeling device proposed in this embodiment belongs to the same technical concept as the previously described UAV swarm storage availability modeling method. It has the same working principle and can achieve the same beneficial effect, which will not be elaborated here.

[0176] This invention also provides a device for predicting the storage availability of unmanned aerial vehicle (UAV) clusters, such as... Figure 8 As shown, the prediction device 800 includes:

[0177] The drone cluster storage availability modeling device 801 as described in any of the above embodiments;

[0178] The first setting unit 802 is configured to set the basic conditions of the availability simulation model;

[0179] The second setting unit 803 is configured to set the availability simulation model logic;

[0180] The third setting unit 804 is configured to set the availability simulation model parameters.

[0181] In some embodiments, the first setting unit is further configured to set the basic conditions of the availability simulation model as follows:

[0182] A swarm of n drones is stored at a certain location, and each drone is in the same environmental conditions.

[0183] Each drone can carry 0 to q types of payloads. Each drone cannot carry the same payload repeatedly, but there is no limit to the number of drones carrying the same payload combination.

[0184] Each type of load has different characteristics, including but not limited to the time and cost of load maintenance and replacement, and the load's own storage degradation patterns. Specific characteristics can be set according to actual conditions;

[0185] Once the load degrades to a certain extent, repair or replacement methods are used to restore it to its original usable state, meaning that all repair and replacement methods are perfect repairs;

[0186] Regardless of the number of payloads carried by a single drone, the failure of any payload will render the drone unable to complete the corresponding task, thus rendering the drone unusable on its own.

[0187] In some embodiments, the second setting unit is further configured to: abstract objects in the simulation scenario as intelligent agents, formulate mutual interaction rules among the intelligent agents, and synchronize the action time and state transitions of each intelligent agent.

[0188] In some embodiments, the third setting unit is further configured to: determine availability simulation model parameters based on factors affecting availability during storage period, wherein the availability simulation model parameters include one or a combination of the following: number of UAVs, payload type, payload combination, failure time, failure rate, maintenance level, spare parts replacement time, spare parts arrival time, mean time between failures, preventive maintenance and replacement interval, mission configuration, and loss rate.

[0189] It should be noted that the UAV swarm storage availability prediction device proposed in this embodiment belongs to the same technical concept as the previously described UAV swarm storage availability prediction method. It has the same working principle and can achieve the same beneficial effect, which will not be elaborated here.

[0190] This invention provides a readable storage medium that stores one or more programs, which can be executed by one or more processors to implement the UAV swarm storage availability modeling and prediction method as described in any of the above embodiments.

[0191] Furthermore, although exemplary embodiments have been described herein, their scope includes any and all embodiments based on the invention that have equivalent elements, modifications, omissions, combinations (e.g., schemes involving intersections of various embodiments), adaptations, or alterations. Elements in the claims will be interpreted broadly based on the language used in the claims and are not limited to the examples described in this specification or during the implementation of this application, and such examples will be interpreted as non-exclusive. Therefore, this specification and examples are intended to be considered illustrative only, and the true scope and spirit are indicated by the following claims and the full scope of their equivalents.

[0192] The above description is intended to be illustrative and not restrictive. For example, the above examples (or one or more of them) can be used in combination with each other. Other embodiments can be used by those skilled in the art when reading the above description. Furthermore, in the above detailed description, various features may be grouped together to simplify the invention. This should not be construed as an intention that a feature of an unclaimed invention is necessary for any claim. Rather, the subject matter of the invention may be less than all the features of a particular embodiment of the invention. Thus, the following claims are incorporated herein by reference as examples or embodiments, wherein each claim is an independent, separate embodiment, and these embodiments are contemplated to be combined with each other in various combinations or arrangements. The scope of the invention should be determined by reference to the appended claims and the full scope of their equivalents.

Claims

1. A method for modeling the storage availability of unmanned aerial vehicle (UAV) swarms, characterized in that, The modeling method includes: The factors affecting availability during the storage period are identified, including maintenance and support attributes, UAV attributes, and cluster mission information. A drone cluster storage availability model is established based on the factors affecting storage availability during the storage period: The expression for determining availability is shown in the following formula: （1） In the formula: T U To enable working hours; T T Total owned time; T DW This refers to the time during which one cannot work. T O Working hours; T S Standby time; T M Maintenance time includes preventative maintenance time and restorative maintenance time; T OS For the time of use; T D To delay time; Based on the specific mission and usage of the drone swarm, the storage period availability index, modified to meet storage task requirements, is expressed as follows: （2） When assessing the availability of drone swarms, the swarm size characteristics are additionally considered to obtain a drone swarm storage availability index. A S : （3） In the formula: , and The first i The operating time, maintenance time, and usage support time of the drone; n The total number of drones in the cluster; Availability index of drone cluster storage period obtained A S Subsequently, the method further includes: During storage, the stresses caused by environmental factors, tasks, and management that reduce cluster availability are collectively referred to as degradation stresses, and denoted as... , It is a parameter vector, and ; Repair time is calculated using the following formula. T M : （4） in, For the number of repairs, Mean time to repair (MTBL); For standalone Number of payloads carried; It is a type feature, and: （5） use As a stand-alone system The fault vector; If the working load At any moment t If it is in a working state, then it has If the working load exist t When constantly in an unusable state, there is , ; t The number of drones in the cluster that are currently operational is: (6) At the point of time a At the appointed time b During this period, single drone The degradation stress it bears is ; At the point of time a At the appointed time b During this period, the number of drones in the cluster that were operational was: ; Up to time t, during this storage period (0,t), the instantaneous availability index of the UAV swarm storage period is expressed as: (7) (8)。 2. A device for modeling the availability of storage in a drone swarm, used to implement the modeling method as described in claim 1, characterized in that, The modeling device includes: The data acquisition unit is configured to acquire factors affecting storage availability, including maintenance and support attributes, UAV self-attributes, and cluster mission information. The model building unit is configured to build a UAV cluster storage availability model based on the storage availability influencing factors.

3. A method for predicting the availability of storage in a drone swarm, characterized in that, Based on the UAV swarm storage availability model established by the modeling method described in claim 1, the prediction method includes: Set the basic conditions for the availability simulation model; Configure the availability simulation model logic; Configure the parameters of the availability simulation model.

4. The method for predicting the availability of UAV swarm storage according to claim 3, characterized in that, The logic for setting up the availability simulation model specifically includes: The objects in the abstract simulation scenario are called intelligent agents, and the interaction rules among these intelligent agents are defined, as well as the action time and state transitions of each intelligent agent are synchronized.

5. The method for predicting the availability of UAV swarm storage according to claim 3, characterized in that, The availability simulation model parameters are determined based on the factors affecting availability during the storage period. These availability simulation model parameters include one or a combination of the following: number of UAVs, payload type, payload combination, failure time, failure rate, maintenance level, spare parts replacement time, spare parts arrival time, mean time between failures (MTBF), preventive maintenance and replacement interval, mission configuration, and loss rate.

6. A device for predicting the availability of storage in a drone swarm, characterized in that, The device includes: The drone cluster storage availability modeling device as described in claim 2; The first setting unit is configured to set the basic conditions for the availability simulation model; The second setting unit is configured to set the availability simulation model logic; The third setting unit is configured to set the availability simulation model parameters.

7. A readable storage medium, characterized in that, The readable storage medium stores one or more programs, which can be executed by one or more processors to implement the modeling method as described in claim 1 and / or the prediction method as described in any one of claims 3 to 5.

Images