CN113344473B - Optimization method and device for ocean target cooperative detection device - Google Patents

Abstract

The invention provides an optimization method and equipment for a cooperative detection device of an ocean target, wherein the method comprises the following steps: determining a management decision period and acquiring at least one dynamic event modeling model; the dynamic event modeling model comprises dynamic events of new target appearance or distance, equipment relay and supplement, over-large co-detection degree, data abnormity or instruction issuing of a user; in the time window, if a dynamic event occurs, a first adjusting instruction is generated immediately, and if any dynamic event does not occur, a second adjusting instruction is generated aiming at the current time window; responding to the adjustment instruction, and calculating the deviation of the actual detection efficiency and a reference value according to a target and detection equipment cooperative scheduling detection efficiency evaluation model based on a target function; and generating a device scheduling optimization scheme according to the deviation so as to adjust the detection device to the device scheduling optimization scheme to perform target cooperative detection immediately or in the next time window. By the scheme, the cooperative detection efficiency can be improved, and the scheduling response capability can be improved.

Description

Optimization method and device for ocean target cooperative detection device

Technical Field

The invention relates to the technical field of computers, in particular to an optimization method and equipment of ocean target cooperative detection equipment.

Background

With the increasing development of multi-sensor network communication technology, sensor cooperation technology and data fusion analysis technology and the remarkable improvement of the detection performance of the sensors, the target detection system can monitor the motion situation and the characteristic attributes of the target by using various devices. In the target detection System, large-sized devices may include an Automatic Identification System (AIS) for a ship, a radar, etc., and small-sized devices may include infrared tracking, a photo camera, etc.

The increase of well-jet detection information provides a basis for comprehensively knowing target information and analyzing target behaviors, so that a detection system actively distributes and integrates multi-sensing equipment to carry out cooperative detection of multiple detection means, performs detection tasks in a combined mode, and achieves advantage superposition and capability complementation. In addition, a large number of moving targets have characteristic diversity and behavior burstiness, uncertain events such as detection environment change occur occasionally, a detection system is difficult to predict in advance, and the safety of equipment and the system is possibly threatened. From the viewpoints of ensuring system safety, expanding data dimensionality and the like, only a single device is used, continuous detection cannot be carried out, and detection information with more dimensionalities cannot be obtained. At the same time, the user wants each target to be stably detected by multiple devices. However, the limitation of insufficient resource of the detection equipment exists in the practical environment, and the problem of solving the equipment allocation problem is solved due to the uneven supply and demand allocation.

Therefore, how to allocate and schedule limited device resources for a plurality of dynamic targets, develop and use advanced detection means in a targeted manner and further improve the performance of the targets, meet the detection requirements of users on the targets of being good and fast, fully utilize the characteristics and advantages of independent or integrated operation of a plurality of devices, and are the key problems of perfecting the detection method and strengthening the detection effect.

Disclosure of Invention

In view of this, the present invention provides an optimization method and device for a cooperative detection device for an ocean target, so as to solve one or more problems in the prior art.

In order to achieve the purpose, the invention is realized by adopting the following scheme:

according to an aspect of the embodiments of the present invention, there is provided an optimization method for a cooperative detection device for a marine target, including:

determining a management decision period and acquiring at least one dynamic event modeling model; the at least one dynamic event modeling model comprises at least one of a modeling model of a new target appearing or far away from a regional detection range, a modeling model of relay and supplement of detection equipment, a modeling model of excessive total detection for representing the quantity of the detection equipment allocated to the target, a modeling model of abnormal information entropy increment for representing detection data of the detection equipment, and a modeling model of instruction issuing by a user;

in a time window of the management decision period, if a dynamic event corresponding to any dynamic event modeling model occurs, a first detection device scheduling scheme adjusting instruction is generated immediately, and if a dynamic event corresponding to any dynamic event modeling model does not occur, a second detection device scheduling scheme adjusting instruction is generated aiming at the current time window;

responding to the first detection equipment scheduling scheme adjusting instruction or the second detection equipment scheduling scheme adjusting instruction, and according to an objective function-based objective and detection equipment cooperative scheduling detection efficiency evaluation model, calculating the deviation between the actual detection efficiency and a set efficiency reference value of the detection equipment by using the detection data acquired by the detection equipment in the current time window; the objective function in the objective function-based objective and detection device cooperative scheduling detection efficiency evaluation model comprises a weighted sum of decision factors influencing a detection device scheduling scheme adjustment decision;

and generating a first detection device scheduling optimization scheme according to the deviation calculated according to the adjustment instruction of the first detection device scheduling scheme, so that the detection device is immediately adjusted to perform target cooperative detection according to the first detection device scheduling optimization scheme, or generating a second detection device scheduling optimization scheme according to the deviation calculated according to the adjustment instruction of the second detection device scheduling scheme, so that the detection device is adjusted to perform target cooperative detection according to the second detection device scheduling optimization scheme in the next time window of the management decision period.

According to another aspect of the embodiments of the present invention, there is also provided an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method according to any of the above embodiments when executing the computer program.

According to another aspect of the embodiments of the present invention, there is also provided a computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the steps of the method of any of the above embodiments.

The optimization method of the ocean target cooperative detection equipment, the electronic equipment and the computer readable storage medium provided by the embodiment of the invention design a detection time decomposition strategy and a rolling equipment scheduling mechanism, define the time and scene for triggering the dynamic scheduling of equipment resources, enhance the real-time responsiveness, realize the closed-loop feedback of the detection efficiency and improve the detection efficiency.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts. In the drawings:

FIG. 1 is a schematic flow chart of an optimization method of a cooperative detection device for marine targets according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of a framework structure of a cooperative-management-scheduling closed loop according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of a mathematical model and detectability evaluation system in accordance with an embodiment of the present invention;

FIG. 4 is a diagram illustrating periodic window and dynamic scheduling in accordance with an embodiment of the present invention;

fig. 5 is a schematic flow chart of a visualization implementation according to an embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the embodiments of the present invention are further described in detail below with reference to the accompanying drawings. The exemplary embodiments and descriptions of the present invention are provided to explain the present invention, but not to limit the present invention.

It should be noted in advance that the features described in the following embodiments or examples or mentioned therein can be combined with or replace the features in other embodiments or examples in the same or similar manner to form a possible implementation. In addition, the term "comprises/comprising" as used herein refers to the presence of a feature, element, step or component, but does not preclude the presence or addition of one or more other features, elements, steps or components.

Although the existing research has realized the access of equipment and the storage of detection data, the existing research has many defects in the aspects of reasonably scheduling detection resources and optimizing the detection performance of a system, for example, the evaluation and feedback of the final efficiency of the system are lacked, a detection closed-loop framework is not formed, and the improvement of long-term accumulative benefit is not facilitated; the analysis of the dynamic scheduling opportunity and the influence factors is insufficient, and the dynamic response of the scheduling decision is not ensured.

In view of the above problems, embodiments of the present invention provide an optimization method for a cooperative detection device for a marine target, so as to form a detection closed-loop control and ensure a dynamic response of a scheduling decision.

Fig. 1 is a schematic flow chart of an optimization method of a cooperative detection device for marine targets according to an embodiment of the present invention. As shown in fig. 1, the method for optimizing the cooperative detection device for the marine target according to the embodiments may include the following steps S110 to S140.

Specific embodiments of steps S110 to S140 will be described in detail below.

Step S110: determining a management decision period and acquiring at least one dynamic event modeling model; the at least one dynamic event modeling model comprises at least one of a modeling model of new target occurrence or far-area detection range, a modeling model of detecting device relay and supplement, a modeling model of excessive total detection for representing the quantity of the detecting devices allocated to the target, a modeling model of abnormal information entropy increment for representing the detection data of the detecting devices, and a modeling model of command issuing by a user.

In step S110, the target to be detected according to the embodiment of the present invention may be a small ocean target. Coordinated probing may be performed with multiple probing devices. The management decision period can be determined by designing a time decomposition strategy, and after the management decision period is determined, the length of the period can be flexibly adjusted by a user. In addition, the dynamic event modeling model is defined, so that the subsequent identification of accidents, exceptions and other events can be facilitated, and the rapid response can be carried out when the detection scheme is no longer reasonable.

For example, a modeling model of the appearance of a new object or the detection range of a distant region may include: when a new target enters the area detection range, allocating the detection resources of the detection device to the new target in the generated detection device scheduling scheme adjusting instruction, and when the target leaves the detection area, releasing the detection resources occupied by the target in the generated detection device scheduling scheme adjusting instruction.

In this embodiment, the modeling model defining the detection range of the new object in the detection area or the detection area away from the new object can be adjusted accordingly when the new object is in the detection area or when the object is out of the detection area.

For example, the modeling model of detecting device relay and complement may include: when the target enters the detection range of a second detection device from the detection range of a first detection device, the detection device of the target is switched from the first detection device to the second detection device in the generated detection device scheduling scheme adjustment instruction.

In this embodiment, when the detection capability of one detection device cannot meet the requirement, switching to another detection device is possible. Thereby, missed detection can be avoided.

For example, a modeling model for indicating that the total detection degree of the target allocated detection device quantity is too large may include: when the number of the target allocated detection devices is larger than the set number or the occupation ratio of all the detection devices exceeds the set proportion, determining that the common detection degree of the target is too large, and dynamically adjusting by issuing a scheduling instruction, namely reducing the number of the detection devices to which the target is allocated in the generated detection device scheduling scheme.

In this embodiment, by defining the co-detection measure, if it is too large, it may indicate that there may be a waste of resources. Therefore, the resource waste can be reduced by adjusting the device scheduling scheme at this time.

For another example, the modeling model for the user to issue the instruction may include: and when receiving a manual instruction decision given by a user, performing corresponding deviation control and correction in the generated detection equipment scheduling scheme adjusting instruction.

In this embodiment, the user may perform deviation control and correction by manual command decision based on the performance feedback value. Therefore, if the instruction issued by the user is received, the corresponding adjustment can be carried out.

In some embodiments, the management decision period, i.e. the decision period for the sounding scheme scheduling optimization, is determined by designing a time decomposition policy.

For example, in the step S110, determining the management decision period may specifically include the steps of: s1111, determining a set target detection process according to a mode that a target appears in a detection range from the first appearance target to the last appearance target leaves the detection range; and segmenting the set target detection process according to a set time interval to obtain a series of management decision periods.

In this embodiment, the entire detection process (from the first object appearing to the last object far from the detection range) is decomposed at time intervals to generate a cluster of continuous and continuously rolling device scheduling decision periods.

In some embodiments, the information entropy concept is applied to a device detection environment, and as a detection time window slides, according to the definition that entropy in shannon information theory represents complexity and chaos, if the information entropy is increased back and forth, the information entropy may represent that a certain degree of data variation occurs in a device during detection or measurement, the complexity and chaos of data or information are increased, and the accuracy and stability of information are reduced, the system and a user may be required to further confirm and identify the device, so as to eliminate the risk of device failure or data abnormality.

For example, in step S110, the method for calculating information entropy in the modeling model of information entropy increment anomaly indicating data anomaly detected by the detection device may specifically include the steps of:

s1121, in the time window of each management decision period, acquiring an original detection data queue generated by the detection equipment, and counting the occurrence frequency of each original detection data value in the queue;

s1122, according to the number of times of the value of each original detection data appearing in the queue, the ratio of the value of each original detection data to the length of the queue is the probability that the value of each original detection data is randomly selected by the system; according to the formula

An information entropy of the values of the corresponding raw detection data is determined, wherein,hrepresenting the entropy of the information, the original data queue being

，iIndicating the sequence number of the data in the queue,p _i representing a data value probability;

wherein the probability may be calculated by

The original data queue is

，x _i Representing the data in the queue, count () represents the count.

In step S1122, the ratio of the number of times the value of each original probe data appears in the queue to the queue length is the probability that the value of each original probe data is randomly selected by the system according to the classical probability.

S1123, forming an information entropy sequence based on a time sequence according to the information entropy corresponding to the original detection data queue in each time window;

s1124, the relation between the value of each information entropy in the entropy sequence and the set threshold is judged successively, if the value of the information entropy is larger than the set threshold, the information entropy is marked;

in step S1124, the system operator may determine the size of the information entropy threshold according to past experience and expert judgment, and may dynamically change to gradually judge the relationship between the value of each information entropy in the entropy sequence and the set threshold;

s1125, if the number of the marked information entropies in the entropy sequence accounts for more than half of the total number of the information entropies, it is determined that the detection data has sudden change and disturbance, and the detection device is considered to have lost the capability of continuously executing the detection task, and the allocation plan of the detection device is cancelled in the generated detection device scheduling scheme adjustment instruction (i.e. corresponding detection device scheduling allocation scheme adjustment is performed).

In this embodiment, if it is determined that data is suddenly changed or disturbed, the possibility of a device failure is high, and the device may have lost the capability of continuously executing the probe task, so that the system may perform re-solution and update of the scheduling assignment scheme. By using the information entropy increment operation, the abnormal phenomenon of the detection sensing data can be found, and the probability of uncertain risks such as unexpected faults of detection equipment and the like is reduced.

In a specific embodiment, in the step S110, the method for calculating the co-detection degree in the modeling model with the excessive co-detection degree indicating the number of the target allocated detection devices may specifically include the steps of:

s1131, acquiring a time period set of the target detected by each detection device in the motion cycle of the target, wherein time period elements in the time period include a start time and an end time;

s1132, extracting the starting time and the ending time of all the time periods in the time period set, removing the repetition of the time points, and sequencing the time points in an ascending order to form a new set;

s1133, traversing the time periods in the detection time period set of each detection device in the motion cycle of the target and the sub-time periods in the new set, if the traversed sub-time period is in the detection time period set, adding one to the flag value of the traversed sub-time period, and obtaining the final flag value of each sub-time period in the new set after the traversal is completed, which is used as the co-detection measure (i.e., the co-detection measure) of the target in the corresponding sub-time period.

In this embodiment, the number of detection devices to which the target is allocated may be calculated, and the number of detection devices to which the target is allocated and the priority of the target detection task may be subjected to weighting operation, so as to perform the common-detection-degree evaluation. If the priority of a certain target task is higher, a plurality of detection devices are required to carry out omnibearing and multidimensional detection on the target task, and the acceptable common detection degree is higher; if the priority of a certain target is small, in order to reduce the device loss and load, more detection resources should not be allocated to the target in view of global performance. If the weighted quantization values of the task priority and the common probing measure at the moment are not reasonable, the system should propose a scheduling scheme again to properly allocate the target holding device to other probing tasks.

Step S120: and in a time window of the management decision period, if a dynamic event corresponding to any dynamic event modeling model occurs, immediately generating a first detection device scheduling scheme adjusting instruction, and if no dynamic event corresponding to any dynamic event modeling model occurs, generating a second detection device scheduling scheme adjusting instruction aiming at the current time window.

In step S120, the occurrence of the dynamic event may be used as the timing of the trigger scheme adjustment, or the timing of the trigger scheme adjustment may be determined according to the management decision period. If a dynamic event occurs, the scheme adjustment is triggered immediately, and the scheme adjustment can be made by quickly responding when the scheduling scheme is no longer reasonable. If no dynamic event occurs in a period or a certain period of the previous period in the period, an adjustment scheme can be calculated according to the detection condition of the period time window, and can be used for adjustment in the next period. By designing the timing for triggering the scheme adjustment, on one hand, the method can quickly respond to the sudden abnormality, and on the other hand, the scheme adjustment can be continuously carried out.

In a specific implementation, in step S120, if a dynamic event corresponding to any one of the dynamic event modeling models occurs in a time window of the management decision period, the method may include the steps of: and S121, in a time window of the management decision cycle, if the dynamic event modeling model is a modeling model which is used for representing the excessive total detection degree of the quantity of the target allocated detection equipment, performing weighted operation on the quantity of the target allocated detection equipment and the detection priority of the target, which are obtained according to the modeling model which is used for representing the excessive total detection degree of the quantity of the target allocated detection equipment, and instantly generating a first detection equipment scheduling scheme adjusting instruction according to the weighted operation result.

The determination method of the detection priority of the target comprises the following steps: if the target identity is definite and is a nationality-set or friend target, the corresponding detection priority is low; if the identity information of the target is more fuzzy, the corresponding detection priority is higher; if the speed of the target is higher, the corresponding detection priority is higher; if the physical distance between the target and the detection device is shorter, the detection priority corresponding to the target is higher.

In the embodiment, when the trigger time is generated according to the common detection degree, the priority of the target detection task is considered at the same time, and the target detection can be performed at a more reasonable time. The number of target-allocated probe devices can be determined by obtaining the final flag value in step S1133.

Step S130: responding to the first detection equipment scheduling scheme adjusting instruction or the second detection equipment scheduling scheme adjusting instruction, and according to an objective function-based objective and detection equipment cooperative scheduling detection efficiency evaluation model, calculating the deviation between the actual detection efficiency and a set efficiency reference value of the detection equipment by using the detection data acquired by the detection equipment in the current time window; the objective function in the objective function-based target and probing device cooperative scheduling probing performance evaluation model comprises a weighted sum of decision factors influencing a probing device scheduling scheme adjustment decision.

In step S130, if the scheme adjustment triggered by the dynamic event is performed, the deviation may be calculated according to the detection data obtained from the time window of the period to the triggering time. If there is no dynamic time, for example, the deviation may be calculated according to the detection data in a set proportion of time before the time window of the cycle, so that it may be used to perform scheme adjustment in the time window of the next cycle.

The objective function-based objective and probe device co-scheduling probing performance evaluation model may be implemented using existing or improved models, and may include, for example, device probing capability evaluation and objective functions and constraints. The device detection capability evaluation function can be used for measuring the static or dynamic performance of the detection device, and can be used as a tool for demand analysis and top-level design to provide guidance for overall efficiency evaluation, follow-up feedback and improvement; the objective function and the constraint condition can be used for ensuring the detection effect and the user satisfaction degree of the scheduling scheme, serve as the necessary constraint condition of model feasibility and high efficiency, and simultaneously perform linear or nonlinear optimization under the constraint condition to obtain the maximum equipment detection efficiency. The target function is used as a fitness function of a heuristic search algorithm-improved genetic algorithm, the detection efficiency of the equipment is used as a fitness value, the equipment and the target to be detected are correspondingly coded according to the general phenomenon of the nature, an initial detection scheme population is generated, the fitness of each scheme is calculated by using the target function, the iteration and the generation optimization are continuously carried out through the steps of meeting the actual cross variation and excellent gene communication of the equipment detection, excellent distribution schemes are continuously reserved, and the proportion of individuals with higher target function values in the population is increased. And when the threshold value of the evolution iteration times is reached or the optimal allocation scheme of successive generations is kept unchanged, finishing the algorithm, and obtaining the optimal solution which is the optimal target detection device cooperative scheduling scheme.

The prior art only considers unilaterally or over emphasizes a certain factor for determining the detection efficiency of the equipment, and a complete detection efficiency evaluation system cannot be built. In this regard, the performance evaluation system can be further optimized.

For example, the deciding factors influencing the adjustment decision of the scheduling scheme of the probing device in the objective function may include: the detection task priority and threat degree, the single body task execution capability and tracking precision of the detection equipment, the detection equipment coordination degree depending on tasks or independent tasks, the detection cost and loss of the detection equipment, and one or more of the target and detection equipment pairing coefficients. The quantization formula of the objective function can be as follows:

the maximum value of the total detection capability of the equipment detection system to the target is the target priority in the scheduling periodTTarget-device adaptation factormDevice co-ordination levelhWeighted integral value of the cost of detection, in the formula

、

、

、

All are weighting coefficients of each part, and are assigned by a decision maker or a user to meet the requirement

。fIndicating the total detection capability. Establishing a 0-1 coding solution matrix with distribution decision variable of 0 or 1 and rows and columns in common, indicating whether a device detection target is distributed or not, wherein the matrix is as follows:

if it isx _ij =1, representing a detection deviceiIn thatkTime of day to targetjCarry out the detection task, otherwise ifx _ij =0, i.e. detection devicesiIn thatkTime of day to targetjNo probing task is performed.

Further, for example, the constraints in the objective function-based objective and probe device co-scheduling probing performance evaluation model may include environmental constraints. Still further, the environmental limiting factors may include: the method comprises the necessary limiting condition for establishing a target and detection device cooperative scheduling detection efficiency evaluation model based on an objective function and the unnecessary limiting condition for efficiently scheduling the detection efficiency evaluation model based on the target and detection device cooperative scheduling of the objective function. The necessary constraints may include: one or more of an operating temperature of the detection device, a communication link constraint, an energy supply constraint, a solar irradiance constraint, and the like. Non-essential limitations may include: target coverage constraints, probe capacity constraints, probe capability limits, and scheduling real-time constraints.

The target coverage constraint may mean that each target should have at least one probing device observing it. A probe device capacity constraint may refer to the probe task's demand for a device that cannot be greater than the total amount of devices available. Probing a device capability limit may refer to the number of objects that each device is able to probe at the same time cannot exceed its own capabilities and capacities. Scheduling real-time constraints may mean that the start time and end time of any probe task must be completed within a specified time window.

The determination method of the detection task priority and the threat degree may include: if the target identity is definite and is a nationality-set or friend target, the corresponding detection priority is low; if the identity information of the target is more fuzzy, the corresponding detection priority is higher; if the speed of the target is higher, the corresponding detection priority is higher; if the physical distance between the target and the detection device is shorter, the detection priority corresponding to the target is higher.

The determination method of the single body task execution capability and the tracking accuracy of the detection device may include: and a detection index represented by one or more of a distance measurement, an angle measurement, a detection resolution, and a detection positioning accuracy.

The determination mode of the detection device cooperation degree depending on the task or the independent task may include: if the target detection task to be distributed is regarded as a dependent task, preferentially selecting heterogeneous model detection equipment with high coordination level for distribution so as to complete the target detection task; and if the target detection task to be distributed is regarded as an independent task, preferentially selecting the detection equipment with the same type and high cooperation level for distribution so as to complete the target detection task.

The determination method of the detection cost and the loss of the detection equipment can comprise the following steps: calculating the product of the inherent cost and the cost loss of the detection equipment caused by the prediction error and the event occurrence probability of the corresponding prediction error to obtain a risk value caused by the prediction evaluation error of the threat level of the target; the cost of the inherent cost and loss of the detection equipment comprises the cost of the inherent cost and loss of the equipment, interference of non-friendly targets, energy consumption and time overhead.

Further, the costs of intrinsic cost and loss of the probing device may include costs of intrinsic cost and loss of the device (which may be quantified by equipment manufacturer definitions, operator and expert experience summaries, and may include unstable and variable communication link performance), interference of non-friendly targets, energy consumption, and time overhead that may exist when switching probing targets multiple times. Because of the uncertainty and the difficult predictability of the target, the uncertainty may be conducted to the evaluation of the threat level in the detection process, so that the accuracy of the threat level prediction is deviated, and the risk of false alarm or missing alarm is generated, therefore, the risk caused by the misestimation of the threat level prediction of the target is calculated, and the value of the risk is equal to the product of the loss caused by the misestimation and the occurrence probability of the false alarm or missing alarm event. And the method is easy to obtain, and corresponding risks cannot be brought if the threat level prediction judgment is correct.

Due to the characteristics of the multi-source heterogeneous sensor equipment, the capabilities of different equipment individuals for executing detection tasks are greatly different, and common measurement indexes comprise searching capability, tracking capability, mapping capability, target positioning capability, tracking identification capability and the like. The device with the detection capability greatly different from the detection requirement cannot meet the requirement of task execution. The pairing coefficient of the target and the specific detection equipment is defined, the respective capability pairing score of each to-be-allocated equipment is calculated and integrated, and whether the equipment plays a positive and positive role in executing the detection task or not can be reflected.

The method for determining the pairing coefficient of the target and the detection device can include: the actual measurement precision of detection equipment and the expected precision of a target are represented by one or more detection precisions including ranging, angle measurement, resolution, positioning and positioning precisions, a covariance matrix of the actual measurement precision and a covariance matrix of the expected precision of the target under an inertial rectangular coordinate system are obtained, and a filter distance function after normalization processing of difference measurement of the two covariance matrices is calculated to obtain a matching coefficient of the target and the detection equipment.

Specifically, for example, the actual measurement accuracy of the equipment is defined as P_nDesired accuracy to target is P_tAnd both are covariance matrices established according to the inertial rectangular coordinate system. The difference measure of two covariance matrices can be calculated by using a formulaf：

f (P_n,P_t) = tr(P_t-P_n)

Wherein, tr (P)_t-P_n) Representing a calculation matrix P_t-P_nThe trace of (c).

The metric C of the ability value can be calculated using the following formula_n：

C_n = α×[1-Δf(P_n,P_t)] + β×‖ΔP_n‖

In the formula 1-Deltaf(P_n,P_t) For normalizing the processed filter distance function, | Δ P_nAnd II, the normalized filtering error covariance norm is shown, and the weighting coefficients alpha and beta respectively reflect the weights of the two parts. According to the formula, if the error between the current measurement precision and the expected precision is smaller and the measurement is closer, the detection capability of the sensor device is stronger, and the high-precision task can be better executed.

For probe tasksJThe required set of capabilities TaskCap (A), (B), (C) a, aJ) Can be represented by the formula:

TaskCap(J)=<(C ₁ ,JL ₁ ),(C ₂ ,JL ₂ ),…, (C _n ,JL _n )>

wherein the content of the first and second substances,C _n indicating the presence of the device itselfkThe capability of the seed to be planted,JL _n representing device correspondence capabilitiesC _n The capacity value of (a), wherein,itaking an integer from 1 to n.

When the capability value of the deviceL _n When the requirement of the detection task on the capability value is higher, the capability of the equipment is considered to be saturated, the equipment is fully capable of executing the current task, but negative effects such as capability or resource waste exist. Recording deviceuExecuteJCapability overflow value for scheduling tasksSpill(u,J)Can be obtained from the following formula:

wherein the content of the first and second substances,L _t to representtThe value of the capacity of the device at the moment,nindicating the number of capabilities the device itself possesses.

Definition ofuPerforming tasksJCapability adaptation coefficient C (C)u, J) As shown in the following formula:

step S140: and generating a first detection device scheduling optimization scheme according to the deviation calculated according to the adjustment instruction of the first detection device scheduling scheme, so that the detection device is immediately adjusted to perform target cooperative detection according to the first detection device scheduling optimization scheme, or generating a second detection device scheduling optimization scheme according to the deviation calculated according to the adjustment instruction of the second detection device scheduling scheme, so that the detection device is adjusted to perform target cooperative detection according to the second detection device scheduling optimization scheme in the next time window of the management decision period.

In step S140, according to the deviation, the corresponding adjustment can be performed in combination with the adjustment of the trigger scheme. The adjustment can be performed in a direction to satisfy the solution result of the objective function, and the specific adjustment content can be determined according to a trigger factor and the like. For example, when a dynamic event occurring in a new target triggers adjustment, the detection device may be allocated, and when the target leaves the detection area, the detection resource of the detection device may be released; for another example, when relay and supplement are needed, switching of corresponding detection equipment is performed; for another example, when the co-detection degree is too large, the number of detection devices that detect the target can be reduced; for example, when the information entropy increment is abnormal, the resource of the detection device with abnormal detection data may be released.

In a further embodiment, information in any embodiment of the present invention may also be visually displayed.

In some embodiments, the method for optimizing the cooperative detection device for marine targets shown in fig. 1 may further include the steps of: s150, generating a GIS map, displaying the geographic background of the target detection area of the detection equipment on the GIS map, and rendering the detection track of each detection equipment to the GIS map.

In specific implementation, in step S150, the method for generating a GIS map includes the following steps:

s151, defining an equipment detection range, cutting a target motion area map, determining a map service type (such as data analysis and three-dimensional display functions), setting the scale precision and resolution of the map, uploading to a SuperMapiServer and issuing to a GIS tile map service so as to be shared and called; the method comprises the steps that an interface provided by SuperMap iClient can be introduced through a script tag, and attribute parameters such as url, layer, style, format, resolutions and the like are set so as to be in butt joint with map service;

s152, creating a SuperMap.Map class as a container for storing a GIS Map, initializing the Map and configuring a Controller, calling an addControl () method to add a ScaleLine control as a scale, using a LayerSwitcher control as a switching layer control and switching a detection effect layer in real time, and using Navigation as a control for monitoring mouse events;

s153, setting visibility, converting coordinates and setting a central point through true or false;

in the step, the map can switch two map coordinate representation methods at any time due to the difference of the geographic coordinate system and the projection coordinate system.

S154, adding addLayers to the Map object, and creating a Marker mark or a Vector layer.

In other embodiments, in step S150, rendering the detection track of each detection device to the GIS map may include the following steps:

s1, setting a timer for track drawing by a setTimeout (function, i) method, where the parameter function is a function executed in the timer, and i can be used as a unique identifier of the timer; the method can send the mature AJAX () packaged in the jQuery every 1 second, and requests the latest latitude and longitude of the target detected by the automatic identification system AIS equipment and the radar equipment of the ship from a background server through an appointed API (application program interface);

s2, if the latest position accords with logic and is not repeated, associating the two detection devices to the same target, and respectively representing the two detection devices by different colors and track point sizes;

s3, extracting the longitude and latitude of the track point through a Supermap, LonLat class, using the longitude and latitude as the geometric information of the Geometry attribute, and packaging the geometric information into a Vector element class (or called as a graph on a layer); different colors and shapes can be set for the Vector according to different data source devices (AIS and radar), and a Marker marking object is created for each detection target through a Supermap.

S4, respectively mounting the vector image and the marker mark on the corresponding image layer to complete visual rendering; when the target leaves the detection range of the detection equipment or the user selects to stop the display of the detection track, the detection process can be ended; by means of the clearTimeout (i), the data request process of the timer is stopped according to the unique identifier of the timer, detection is finished, and meanwhile, the Marker layer, the point diagram layer and the track point set can be cleared.

In other embodiments, during the visualization process, device allocation and multi-dimensional data information presentation may be performed.

In other embodiments, visual data entry may be performed. Illustratively, a method of performing targeted multidimensional input data may include the steps of:

s1, adding a new detection device into the detection system, simultaneously storing a static information file as an XML file in a sensors path under a server public folder, and distributing by a scheduling center to be controlled;

s2, the scheduling center loading file reading module reads all file names in the sensors path, and judges whether the file name 'equipment number, xml' of the equipment to be distributed is in the file name;

s3, if the file of the device exists, constructing an XML and JSON format parsing object parser, so that static information of the XML is conveniently parsed into a uniform and readable JSON structure;

s4, mapping the XML structure into a JSON structure through the analytic object, and using the JSON structure as original data for calculating the capability matching coefficient and the device cooperation level of the device;

s5, sending Axios short polling to the background server and the database in a set request format (for example, http:// [ server address ]: monitoring port ]/routing module/specific route), and continuously requesting each device in the database for real-time dynamic information of target detection.

In a further embodiment, the method for optimizing the cooperative detection device for marine targets shown in fig. 1 may further include the steps of: and S160, generating an information map of the target and the detection device so as to trigger the analysis of the coordination capacity of the target and the detection device when the capacity accumulation of the information map exceeds a set capacity threshold value.

In step S160, as different devices continuously detect the target at different time periods, the capacity of the information map is continuously accumulated, the association between nodes is continuously expanded, and the complexity is increased. And when the accumulation degree is about to exceed the threshold value, triggering a capability extraction and collaborative analysis algorithm to quantify the collaborative capability of the target-equipment in detail. The calculation method can be as follows:

quantization definition apparatusiAnd apparatusjThe degrees of the message interaction, the capability complementation and the capability of the control cooperation are respectively alpha, beta and gamma, and then the equipmentiAnd apparatusjLevel of synergy betweenC _ij The quantization formula can be as follows:

C _ij =ω₁α_ij+ω₂β_ij+ω₃γ_ij

wherein, ω is₁、ω₂、ω₃Denotes the coefficient, α_ij、β_ij、γ_ijRespectively representing devicesiAnd apparatusjMessage interaction, capability complementation and capability degree of control cooperation.

In a specific implementation, in step S160, generating an information map of the target and the detection device may specifically include the steps of:

s161, initializing a target node to store the unique identifier of the target corresponding to the target node; sequencing according to time from morning to night, and connecting the detection equipment of the corresponding target by taking the target node as a center so as to distribute equipment nodes;

s162, when the latest equipment node is generated, the nodes are connected in series according to a tail insertion method, and the tail of the linked list is inserted; wherein the target node and the device node are in an observed and observed relationship;

s163, if two detection devices detect the same target at the same time, the detection relation of the two detection devices points to the corresponding same target node, so as to form transverse association;

s164, different detection devices continuously detect the target at different time periods, the capacity of the information map formed by the target node and the device node is continuously accumulated, a mesh structure between the target and each device is formed by point-to-tree networking and tree networking, and when the capacity accumulation degree is about to exceed a preset information map size threshold, an analysis algorithm of detection capability extraction and the cooperation capability of the target and the detection devices is triggered to analyze the cooperation capability of the target and the detection devices.

In this embodiment, a graph database Neo4j is used to store an information graph associating targets with assigned device bindings. The information graph may be represented using G = (V, E), where V represents a device node and E represents a connection relationship between nodes. And taking the target node as a center, and continuously accumulating the target equipment distribution conditions to form equipment distribution nodes. With the advance of the detection process, multi-source heterogeneous equipment is continuously allocated with detection tasks and continuously establishes association with a central target node, so that multi-round iterative association, incremental superposition and combing integration are triggered.

In a further embodiment, the method for optimizing the cooperative detection device for marine targets shown in fig. 1 may further include the steps of: s170, dividing a target detection area in the GIS map into a plurality of cellular grid units, calculating detection efficiency values of detection equipment in areas corresponding to the cellular grid units, abstracting the detection efficiency values of the detection equipment to map vector points falling on the grid unit areas, summarizing and aggregating label information of the number of point objects falling in the grid unit areas in the center of the grid unit areas to obtain efficiency values, summarizing and counting area efficiency characteristics based on the map vector points of the cellular grid units, and sending out detection equipment detection scheme adjustment prompt messages under the condition that the efficiency does not meet set efficiency requirements.

The step S170 may specifically include the following steps:

s171, abstracting data after the efficiency evaluation of the equipment scheduling result into the number of points falling into an area (which is convenient for statistics and summarization), creating Supermap.

S172, creating a sub-object array of the grid, setting parameters influencing point data aggregation calculation in the area, such as grid width and height pixels, upper and lower limits of data quantity and the like, and setting appearance attributes such as a frame, transparency and the like in the area range;

s173, setting a corresponding mapping relation between the numerical values and the colors of the areas, mapping different efficacies into the colors of different areas, and then filling the colors from light to dark;

s174, through the distribution characteristics (color composition) and the statistical characteristics (grid unit center label information) of the regional efficiency, the user can be reminded to make further decision and feedback, and the real-time improvement of the system efficiency is promoted.

In addition, an embodiment of the present invention further provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor executes the computer program to implement the steps of the method according to any of the above embodiments.

Embodiments of the present invention further provide a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps of the method according to any of the above embodiments.

The embodiment of the invention designs a detection time decomposition strategy and a rolling equipment scheduling mechanism, defines the time and scene for triggering the dynamic scheduling of equipment resources and enhances the real-time responsiveness. A coordination-management-scheduling closed-loop framework is formed, the real-time scheduling concept in the traditional open-loop mode is broken through, and a feedback control process and a real-time scheduling decision are organically combined; the dynamic scheduling of the equipment is fixed, the dynamic and burstiness of the equipment can be high according to the regional environment and the object target of the detection demand, the dynamic scheduling method is used for combining the scheduling decision period of the rolling continuous equipment with the dynamic event scheduling, and the real-time responsiveness is enhanced while the scheduling scheme is adjusted through continuous closed-loop feedback.

In a cooperation-management-scheduling closed-loop framework, in the equipment scheduling process, the control center dynamically allocates available detection equipment resources, can control the optimization of performance, and embodies the process of resource scheduling for controlling application service; in the feedback control process, the performance evaluation is introduced into the dynamic process in a negative feedback or positive feedback mode, and the feedback control provides services for resource scheduling.

In the dynamic scheduling analysis process, a time decomposition strategy is utilized to decompose the whole detection process (the first target appears to the last target is far away from the detection range) according to time intervals to generate a cluster of continuous and continuous rolling equipment scheduling decision period; the dynamic scheduling event analysis process can ensure the dynamic property and the quick response property of the scheduling distribution of the detection equipment, and the scheduling scheme is dynamically adjusted or updated in real time when the original equipment distribution scheme is no longer reasonable and is no longer applicable.

Furthermore, a general mathematical model for analyzing and evaluating the actual detection effect is established, and a detectability index system and an objective function are evaluated according to the optimization criterion. And the target-equipment cooperative scheduling model based on the target function is used for forming a general model for analyzing and evaluating the detection effect according to the measuring standard of the static or dynamic performance of the detection equipment.

Further, the system visualization module takes the optimized dispatching control center as a system brain which is used for presenting the condition of target detection of the equipment in the whole regional environment in real time. An interactive friendly visual detection System is built, a Geographic Information System (GIS) map and a read frame are used as a front-end visual basis, a browser/server architecture and a componentization design idea can be adopted, a visual interface of a target detection situation and a dynamic equipment scheduling result is realized, and feasibility and high efficiency are achieved.

The above method is described below with reference to a specific example, however, it should be noted that the specific example is only for better describing the present application and is not to be construed as limiting the present application.

In a specific embodiment, in the first aspect, a closed-loop management framework for device cooperative scheduling is established, and the framework has the characteristic of combining global and unit control with feedback. The control, that is, the resource manager of the device, adjusts and updates the device allocation plan using a corresponding control function or dynamic mechanism, and issues a control command to the device again, so as to compensate the performance deficiency on line, and continuously make the current detection effect approach the reference value all the time. The feedback, that is, the performance evaluation is introduced into the dynamic process in a negative feedback or positive feedback mode, can play a role in strengthening correction or deviation control, and promotes the optimization adjustment and iteration of the real-time running performance of the whole system.

In a second aspect, a scenario and an opportunity for triggering target detection rescheduling are defined, which include a periodic scheduling window and a dynamic scheduling analysis:

in the periodic scheduling window, the whole detection process (the first target appears to the last target is far away from the detection range) is decomposed according to a certain time interval, a cluster of continuous and continuous rolling equipment scheduling decision periods is generated, and each period is used as a time window for the scheduling decision to take effect;

the dynamic scheduling analysis refers to modeling aiming at various dynamic events to ensure the dynamic property and the quick response property of the scheduling distribution of the detection equipment, defining the scene that the original equipment distribution scheme is no longer reasonable and is no longer applicable after the events occur, recording the occurrence time of the events as an equipment dynamic scheduling point, and dynamically adjusting or updating the scheduling scheme in real time.

Further, the dynamic scheduling analysis may be operable to:

time window size value delta of scheduling periodTThe speed of generation of a new target or the load condition of equipment is positively correlated, and can be flexibly set by a user according to the actual situation. If it is presentkThe time period window is about to end, and the next continuous window is enteredk+ΔTAnd previously, completing the calculation of the next window length and the scheduling scheme.

Further, the periodic scheduling window may be used to:

if no defined dynamic event occurs, only one time of algorithm solving calculation and equipment scheduling is needed in each scheduling window, and compared with the dynamic scheduling method from beginning to end, the switching times of the equipment are reduced. The defined rules and scenarios for triggering dynamic scheduling of devices have the following five aspects:

(1) appearance and remoteness of new targets: when a new target appears or drives away from the detection range of the area, detecting resources occupied by the target are distributed or released in time;

(2) detecting equipment relay and supplement: when a target enters the detection range of another device, the front device and the rear device are relayed, the target is rapidly switched and captured, and missed detection and wrong detection are avoided;

(3) the co-exploration degree is too large: defining the quantity of the detection equipment distributed to the target as a common detection degree, and if the quantity of the detection equipment distributed to the target is too large, the resource waste condition may exist;

(4) information entropy increment anomaly: when equipment is suddenly damaged and fails or an environmental interference event occurs, the uncertain factors are conducted, and equipment data are abnormal;

(5) the user issues an instruction: and the user can make a decision to control and correct the deviation through a manual instruction according to the efficiency feedback value.

In the third aspect, the universal mathematical model for analyzing and evaluating the actual detection effect perfects an evaluation detectability index system, and comprises an objective function and constraint conditions:

the evaluation index system refers to that factors such as detection task priority and threat degree, single task execution capability and tracking precision of sensor equipment, equipment coordination degree depending on tasks or independent tasks, equipment detection cost and loss, target-equipment pairing coefficient and the like are taken as decision factors influencing scheduling scheme decision and are jointly brought into the index system. According to a mathematical method, each determinant is quantitatively calculated in the form of a membership function;

the objective function is to flexibly set reasonable weights and perform weighted summation on the values or functions of the parts to form an objective function which is easy to realize in a quantitative manner by taking the best optimization as a target. The constraint condition refers to a precondition for simulating an actual environment condition by limiting parameter values, and the precondition is used for embodying the rationality of the cooperative scheduling scheme.

In order to meet the all-weather and all-coverage detection monitoring requirements of users, when the target detection equipment optimization collaborative simulation system is displayed at the browser end, the space of a human-computer interaction interface is limited, so that the functions of all parts of the user system are presented in a highlighted, concise and clear mode. The optimized dispatching control center is used as the brain of the system to present the target detection condition of the equipment in the whole area environment in real time, and simultaneously, a result visualization interface for the user to execute the dispatching algorithm in real time or trigger dynamic dispatching is provided.

The system visualization module includes:

the control scheduling center input submodule is used for inputting dynamic detection information such as multi-dimensional data and motion tracks, static information and performance parameters of equipment, manual instructions of users and various detection evaluation feedbacks into the optimization scheduling control center;

and the control scheduling center output submodule is used for issuing an optimized cooperative detection instruction to the equipment to form an equipment scheduling scheme and simultaneously providing a result visualization interface for a user to execute the system in real time or trigger dynamic scheduling.

The control scheduling center input submodule may include:

the GIS map implementation unit is used for building a geographical background of a target detection area and is also used as a carrier of a visual interface of the simulation system;

the target multi-dimensional input data unit is used as an input interface of a detection scene and an optimized scheduling model frame, and target real-time multi-dimensional data acquired by equipment is used as a target function to calculate original data;

and the target-equipment information map unit is used for triggering multi-round iterative association, increment superposition and combing integration, and quantifying the capabilities of message interaction, capability complementation and control cooperation between equipment.

Data entry into the control center may include:

(1) and (3) realizing a GIS map: generating a geographical background of a target detection area, and simultaneously, generating a carrier of a visual interface of the simulation system;

(2) target multidimensional input data implementation: reading a corresponding static detection equipment configuration file and target real-time multi-dimensional data acquired by equipment as input;

(3) target-device information map implementation: extracting and analyzing the association between the multi-source equipment and target binding detection, and excavating equipment cooperative measurement through horizontal and longitudinal series connection and repeated iteration;

(4) the equipment detection track is visually realized: connecting a plurality of track points in series to form a track, distinguishing detection tracks of different devices, and displaying a target position and a motion situation;

(5) equipment allocation and multi-dimensional data visualization are realized: presenting target data information acquired by optimizing scheduling and collaborative detection of six types of detection equipment in a table visualization mode;

(6) and (3) visual implementation of efficiency evaluation: and comprehensively analyzing the reasonability and the high efficiency of the final detection efficiency of the system by combining geographical visualization, and giving a quantitative value or conclusion evaluation.

The control scheduling center output submodule may include:

the device detection track visualization unit is used for rendering track points on a real-time track layer of the GIS map in a certain size and color mode, and finally, a plurality of track points are connected in series to form a track;

the equipment distribution and multi-dimensional data information display unit is used for integrating the multi-dimensional detection data of six types of equipment commonly used in the offshore detection environment and the time and reason criterion for triggering dynamic scheduling;

and the efficiency evaluation visualization unit is used for comprehensively analyzing the solved equipment distribution scheme and the final detection efficiency reasonableness and high efficiency of the system and giving a quantitative value or conclusion evaluation.

In the embodiment, the application is used as the guide, the ocean detection environment is used as the application scene, the deep cooperation and linkage capacity among all the devices is improved, the automation degree of target detection tracking cooperative early warning detection, cooperative resource allocation and cooperative command control is improved, and theoretical support and technical support are provided for the construction of a detection system in a complex environment.

Fig. 2 is a schematic structural diagram of a cooperation-management-scheduling closed-loop framework in an embodiment of the present invention, and referring to fig. 2, the cooperation-management-scheduling closed-loop framework in the method for optimizing cooperative detection of marine small targets in the embodiment includes a device network subsystem, a control center, and a detection task subsystem. Wherein, the device network subsystem comprises a plurality of detection devices (device 1, device 2, device 3, etc.) at the bottom layer; the control center comprises a controller, a device scheduler unit (device scheduler) and the like; the detection task subsystem is used for executing a plurality of detection tasks, such as a detection task 1, a detection task 2, a detection task 3 and the like.

The detection device at the bottom layer is used as a 'producer' nearest to the target to be detected, is used for generating original sensing data, and can be used for driving the device to execute a detection task after receiving a detection scheme scheduled and allocated by the dynamic device.

And the controller is used as a resource manager of the equipment, compares the performance evaluation after the detection task is executed with the performance reference value to obtain deficiency or error, adjusts and updates the equipment distribution plan by using a corresponding control function or a dynamic mechanism, resends a control instruction to the equipment, compensates the deficiency of the performance on line, continuously enables the current detection effect to be close to the reference value all the time, and drives the equipment scheduler to form a new distribution scheme.

And the equipment scheduler determines a management decision period, generates an equipment allocation scheme in a time window in the period, and realizes resource allocation of the equipment, information interaction among the equipment and relay cooperative execution of tasks.

Fig. 3 is a schematic diagram of a mathematical model and a detection capability evaluation system in an embodiment of the present invention, referring to fig. 3, in this embodiment, a detection efficiency evaluation index system further includes a detection task priority and a threat degree, a single task execution capability and tracking accuracy of a sensor device, a device cooperation degree depending on a task or an independent task, a device detection cost and loss, a target-device pairing coefficient, and the like, and a constraint condition mainly considers a limitation of an actual environment:

and the priority evaluation refers to the quantitative sorting of the threat degree of the detection target and the task priority generated by the threat degree so as to distinguish the detection task priority and the emphasis degree. Further, if the target identity is known and is a Chinese or friend target, the priority ranking is lower. The more fuzzy the identity information is for a target, the higher the priority of its task of probing. If the target speed is faster, the threat degree is higher, and the priority is higher. The closer the physical distance between the target and the equipment is, the greater the pressure on the geographical position brought by the target is, and the higher the threat degree and priority of the target are;

the target-device adaptive coefficient refers to the capability of different individual devices to execute detection tasks with great difference, and defines a covariance matrix established according to an inertial rectangular coordinate system, which quantifies the actual measurement accuracy of the device and the expected accuracy of the target, by using detection accuracy (such as ranging, angle measurement, resolution and positioning accuracy), and calculates the normalized filtering distance function of the difference measurement of the two matrixes as the capability coefficient. Has the significance of whether the device improves the detection precision. In a certain detection period, the difference between the measurement precision and the estimation precision must be within a certain range, and if the difference between the measurement precision and the estimation precision is too large, the data accuracy of the equipment in the detection process is poor, and detection loss may be caused; if the actual error and the expected error are smaller, the device can capture the target in time and continuously detect the target. In addition, the load rate and the resource consumption rate of each device tend to be balanced and reasonable, the overall utilization rate of all devices in the system is improved, and individual devices are not fatigued due to overuse;

the device cooperation level refers to whether different models are in the same model or different models, and the devices are not isolated, but have time sequence, function, logic or obvious or hidden association, and are embodied in the processes of motor coordination, message transmission and guide and evidence. If the devices with strong coordination level are selected to be combined together to perform the detection task, the group inevitably improves the detection final benefit due to strong cohesion. Defining the task tight cooperation degree to distinguish the dependent task from the independent task: when the task to be allocated is regarded as a dependent task, selecting heterogeneous type equipment with high coordination level as much as possible and allocating the heterogeneous type equipment to finish detection work; when the tasks to be distributed are regarded as independent tasks, equipment with the same model and high coordination level is selected as far as possible when the equipment is distributed;

the detection cost and cost means that unnecessary resource loss of equipment and even the whole detection array is reduced, and even serious resource loss is caused. And (4) comprehensively considering the reward and punishment relationship between the benefit and the cost. And calculating the risk caused by prediction evaluation error of the threat level of the target, wherein the value of the risk is equal to the product of the loss caused by prediction error and the occurrence probability of the false alarm and the false alarm event. And counting Bayes prior probability values of the target threats, namely the total probability that all historical nodes are special nodes. And calculating the posterior prediction probability of the occurrence of the target threat event according to the Bayesian rule, and multiplying the posterior prediction probability by the prediction error cost matrix to generate a risk expected value. The cost of the inherent cost and loss of the device mainly includes communication link loss, interference of non-friendly targets, energy consumption, time overhead and the like.

The content focus of the constraints comes from both objective and subjective: objective constraints such as operating temperature of certain equipment, communication link constraints are necessary constraints that the model can hold; subjective constraints are efficient, non-essential constraints that the model can be, such as capacity, capability, limits and operations on allocation, spatial constraints in location and logic, temporal constraints, resource constraints, task constraints, and so on.

Target coverage constraints: each target should have at least one device observing it; and (3) equipment capacity constraint: the demand of a certain detection task on the equipment cannot be larger than the available total amount of the equipment; device capability limitation: the number of targets simultaneously detected by each device cannot exceed the capacity and capacity of the device; scheduling real-time constraints: the starting time and the ending time of any detection task must be completed within a specified time window;

further, a 0-1 coding solution matrix with an allocation decision variable of 0 or 1 is formed to indicate whether the device detection target is allocated.

And reasonably quantizing the weight of each part to form an optimized objective function, wherein the objective is the maximum total detection capability of the equipment detection system to the objective, and the equipment management scheduling of the detection work is executed by using the equipment set, and the practical significance is as follows: the larger the threat degree and priority of the detected target, the stronger the detection capability of the device, and the smaller the use cost and detection risk, the larger the objective function value.

Fig. 4 is a schematic diagram of periodic window and dynamic scheduling in an embodiment of the present invention, and referring to fig. 4, fig. 4 provides a strategy for planning periodic scheduling and dynamic scheduling by using a short time window as a scale in another embodiment of the present invention. And periodically driving the scheduling window to move forward along with the time lapse and the feedback of the detection result, staticizing the dynamic problem and reducing the solving scale of the problem. By means of threshold judgment and other methods, if a corresponding dynamic disturbance event occurs and the number and influence of dynamic changes exceed a certain limit, a dynamic scheduling decision is triggered in time, an original scheduling scheme in a scheduling window is adjusted in real time, and the capability of the system for flexibly coping with sudden critical events is gradually optimized. The method specifically comprises the following steps:

the detection action of the detection device on the target takes the moving target entering the detection range as a starting point and takes the target position far away from the detection range of the device as an end point. When a new target appears in or enters the area detection range, allocating the detection equipment resources to the target; when the target leaves the detection area, the equipment detection loses effectiveness, so that the detection resource occupied by the target is released;

when the detection capability of a certain device can not meet the task requirement, the device is timely handed over to realize capability supplement. The detection range of the single-type equipment cannot completely cover the area, when a target crosses over a sea, land and air medium and enters the detection range of the other equipment, the front equipment and the rear equipment are relayed to quickly switch and capture the target;

the definition of the common detection degree represents the number of the detection devices which are commonly distributed to a certain target at a certain moment. And comprehensively judging by combining the task priority of the target: when the detection priority is lower, the device is detected by excessive devices, unreasonable factors such as uneven device distribution, resource waste, unclear primary and secondary factors and the like may exist in the scheme of scheduling the device, and the task with higher priority cannot obtain enough device resources and needs to be distributed again in the scheduling scheme; on the other hand, when the detection priority is higher, more detection devices are acceptable and necessary to be allocated in terms of improving the multi-dimension and accuracy of detection information and providing richer criteria for analysis decision;

further, the method for calculating the co-detection measure comprises the following steps:

s1, settingTsIs a set of time periods, elements of the set, detected by the apparatus during a movement cycle of an objecttRepresents a time period having start time and end time attributes;

s2, mixingTsExtracting the starting time point and the ending time point of all time periods in the time frame, removing the duplication, and sequencing the time points in ascending order to form a new setls；

S3, traverseTsThe detection time period of each device andlsin a sub-period of time, iflsAt which the sub-period of timeTsIn the detection time period set of the device, 1 is added to the flag value of the former time interval, and the final result is the co-detection degree.

The information entropy concept is applied to the equipment detection environment, along with the sliding transition of a detection time window, the information entropy is increased, the data change occurs when the equipment is detected or measured, the information complexity and the disorder degree are increased, the accuracy and the stability of the information are reduced, a system and a user are required to further confirm and identify the equipment, and the risk of equipment failure or data abnormality is eliminated. The method for calculating the information entropy is as follows:

s1, acquiring an original data queue uploaded by the detection equipment in each time window, and counting the occurrence frequency of each original value in the queue;

s2, forming the probability of each detection data value being selected, and calculating the information entropy;

s3, sequentially calculating the information entropy of the data queue in each time window along with the backward time, and forming a dynamically-changing information entropy sequence based on the time lapse;

s4, calculating an information entropy increment set, and gradually judging the relationship between the value of each entropy increment and a preset threshold value: if the entropy increase value is larger than the threshold value, marking the element, and when the total number of the entropy increase values larger than the threshold value accounts for more than half of the information entropy increment set, determining that the data has sudden change and disturbance, the equipment has higher possibility of failure, the equipment loses the capability of continuously executing the detection task, and the system performs re-solving and updating of the scheduling distribution scheme.

Fig. 5 is a schematic flow chart of visualization implementation in an embodiment of the present invention, and referring to fig. 5, fig. 5 provides a visualization module, which may specifically include: the system comprises a GIS map realization sub-module, a target multi-dimensional input data realization sub-module, a target-equipment information map sub-module, an equipment detection track visualization sub-module, an equipment distribution and multi-dimensional data information display sub-module, a scheduling analysis sub-module and one or more of efficiency evaluation visualization sub-modules.

The GIS map implementation submodule uses the following steps in the system:

and S1, defining the detection range of the equipment, cutting a target moving area map, determining the service type of the map, setting the scale precision and resolution of the map, uploading the map to a SuperMap iServer and issuing the map to a GIS tile map service for convenient sharing and calling. Introducing an interface provided by SuperMap iClient through a script tag, and setting attribute parameters such as url, layer, style, format, resolutions and the like to be in butt joint with services;

s2, creating a SuperMap.Map class as a container for storing a GIS Map, initializing the Map and configuring a Controller, calling an addControl () method to add a scalLine control as a scale, using a LayerSwitcher control as a switching layer control and switching a detection effect layer in real time, and using Navigation as a control for monitoring mouse events;

s3, switching a longitude and latitude 4326 coordinate system and a 3857 coordinate system, setting visibility through true or false, converting coordinates, and setting a central point;

s4, adding the addLayers to the map object, and creating a Marker mark or Vector layer.

The target multi-dimensional input data implementation submodule uses the following steps in the system:

s1, adding a new detection device into the detection system, simultaneously storing a static information file as an XML file in a sensors path under a server public folder, and distributing by a scheduling center to be controlled;

s2, the scheduling center loading file reading module reads all file names in the sensors path, and judges whether the file name 'equipment number, xml' of the equipment to be distributed is in the file name;

s3, if the file of the device exists, constructing an XML and JSON format parsing object parser, so that static information of the XML is conveniently parsed into a uniform and readable JSON structure;

s4, mapping the XML structure into a JSON structure through the analytic object, and using the JSON structure as original data for calculating the capability matching coefficient and the device cooperation level of the device;

and S5, sending Axios short polling to a background server and a database in a http:// [ address of a service end ]: monitoring port ]/routing module/specific routing request format, and continuously requesting real-time dynamic information of each device in the database for target detection.

The target-device information map sub-module uses the map database Neo4j to store an information map associated with the target and the assigned device binding. The information graph may be represented using G = (V, E), where V represents a device node and E represents a connection relationship between nodes. And taking the target node as a center, and continuously accumulating the target equipment distribution conditions to form equipment distribution nodes. With the advance of the detection process, multi-source heterogeneous equipment is continuously allocated with detection tasks and continuously establishes association with a central target node, so that multi-round iterative association, incremental superposition and combing integration are triggered. The method comprises the following steps:

and S1, initializing the target node and storing the unique identification of the target. Distributing nodes by equipment which is connected with a target by taking the target node as a center according to the sequence of time from morning to night;

and S2, when the latest equipment node is generated, the latest equipment node is connected in series according to a tail insertion method, and the tail end of the linked list is inserted. The target node and the equipment node are in observation and observed relation;

s3, at the same time, if the two devices detect the same target, the detection relation of the two devices points to the same target node, so that the two devices are transversely associated;

and S4, as different devices continuously detect the target at different time periods, the capacity of the information map is continuously accumulated, the association between the nodes is continuously expanded, and the complexity is increased. And when the accumulation degree is about to exceed the threshold value, triggering a capability extraction and collaborative analysis algorithm to quantify the collaborative capability of the target-equipment in detail.

The equipment detection track visualization submodule renders track points on a real-time track layer of a GIS map in a certain size and color mode according to the target position, and finally a plurality of track points are connected in series to form a track. The method comprises the steps of displaying the detection tracks of different devices in different colors, and displaying a picture of a ship on the tracks as a mark. And when the latitude and longitude are changed due to the movement of the target, the track of the map is updated in real time to reflect the latest movement position of the target. The method comprises the following steps:

s1, setting a timer for track plotting by a setTimeout (function, i) method, where the parameter function is a function executed in the timer, and i is a unique identifier of the timer. Sending a mature AJAX () method packaged in the jQuery every 1 second, and requesting the latest latitude and longitude of a target detected by AIS equipment and radar equipment from a background server through an appointed API (application programming interface);

s2, if the latest position accords with logic and is not repeated, associating the two detection devices to the same target, and respectively representing the two detection devices by different colors and track point sizes;

and S3, extracting the longitude and latitude of the track point through a Supermap LonLat class, using the longitude and latitude as the geometric information of the Geometry attribute, and packaging the geometric information into a Vector element class (graph on the layer). Setting different colors and shapes for a Vector according to different data source equipment (AIS and radar), and creating a Marker marking object for each detection target through Supermap.

And S4, respectively mounting the vector image and the marker mark on the corresponding image layer to complete the visual rendering. And when the target leaves the detection range of the detection equipment or the user selects to stop displaying the detection track, ending the detection process. The system stops the data request process of the timer according to the unique identifier of the timer by a clearTimeout (i) method, and simultaneously carries out emptying operation on the Marker layer, the point diagram layer and the track point set.

The equipment distribution and multi-dimensional data information display submodule integrates AIS, radar, sonar, overwater photoelectric, underwater photoelectric and acoustic detection equipment commonly used in the offshore detection environment, and presents target data information acquired by optimized scheduling and cooperative detection in a Table form visualization mode. The AIS dimension information comprises a target ship name, an mmsi number (unique identification number), nationality, a type, a length width, a heading to ground, a ship heading direction and the like, and the radar dimension information comprises acquisition time, a frame number, an absolute batch number, a distance, a radial speed, an azimuth, a pitch angle, trace characteristics and the like. The sonar dimension information comprises target acquisition time, distance, angle, depth, type and the like, and the photoelectricity comprises an overwater or underwater photoelectric camera for capturing shot target pictures;

note that Table defines column attributes, and defines dataSouce attributes to fill each column with a numerical value in a self-adaptive manner, taking the column attributes as column objects. And the multi-dimensional data information is integrated into a DIV label to form a Card component, and a content container in the component provides an area for displaying data, behaviors and equipment details for a Tab component and a TabPane component. Because the invention displays the target data information detected by six devices, and the TabPane area is limited, for the sake of concise and clear top-down browsing, an accordion-type Collapse folding component is used in the TabPane, which is convenient for the expansion and the storage of the information.

And the scheduling analysis submodule displays a time line of equipment scheduling. In the initial state, all moving objects do not start to actually move, the holding detection device is zero, and the display content of the six-dimensional data table is empty. And then, in the movement process, the detection equipment monitors the equipment for a long time and can trigger a dynamic scheduling decision of the equipment, and after the equipment scheduling scheme is issued to the equipment, the detection data of the equipment can appear in the six-dimensional data table. In a TimeLine component form, the module shows a TimeLine of detection behavior change and detection target switching of the moving target detection equipment along with the time lapse, time for triggering dynamic scheduling each time for a user, reason criteria and the number of equipment for detecting the target at present;

the efficiency evaluation visualization submodule displays the size relation of the internal efficiency data of different areas in a honeycomb aggregation mode, and divides a map area into a plurality of hexagonal honeycomb grid units which are fixed in size and are adjacent to each other according to the characteristics that the hexagonal shape is close to a circle and the adjacent centers are equidistant. And evaluating the rationality of the equipment scheduling scheme according to a performance evaluation algorithm, abstracting the performance value quantization data into map vector points with equal number, and summarizing and counting the map vector points to a unit area. The region center tag information indicates the number of point objects that fall into a cell, which is aggregated to equal the performance value. The method comprises the following steps:

s1, abstracting data after the performance evaluation of the equipment scheduling result into the number of points falling into the area, creating Supermap.

S2, creating a sub-object array of the grid, setting parameters influencing point data aggregation calculation in the area, such as grid width and height pixels, upper and lower limits of data quantity and the like, and setting appearance attributes such as a frame, transparency and the like in the area range;

s3, setting a corresponding mapping relation between the values and the region colors, mapping different efficacies into colors of different regions, and then filling the colors from light to dark;

and S4, the user can be reminded to make further decision and feedback through the distribution characteristics and the statistical characteristics of the regional efficiency, and the real-time improvement of the system efficiency is promoted. When the overall system performance is insufficient, according to three types of evaluation indexes of system performance evaluation: multidimensional information integrity, algorithm evaluation and equipment access indexes are analyzed, the reason of poor efficiency is analyzed, the efficiency limiting factor is obtained, and then, the efficiency limiting factor can be fed back from top to bottom in a multi-stage mode, so that corresponding means are called to purposefully check missing and fill up, and the system efficiency is improved.

The invention provides a device optimization cooperative detection method for small ocean targets, which is based on the multi-target multi-device detection environment requirements and takes friendly user experience, perfect system functions and excellent system performance as the guide. The detection capability is evaluated through a general mathematical model and an objective function, and when a disturbance event occurs or the environment changes, the detection scheduling scheme can be automatically updated and adaptively adjusted dynamically, so that the dynamic operation performance of the system is improved. The GIS map carrier is combined with the geographical visualization technology, and the ocean island detection environment is taken as a scene, so that interactive friendly visualization of target detection scheduling and optimization cooperative system is realized.

In the description herein, reference to the description of the terms "one embodiment," "a particular embodiment," "some embodiments," "for example," "an example," "a particular example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. The sequence of steps involved in the various embodiments is provided to schematically illustrate the practice of the invention, and the sequence of steps is not limited and can be suitably adjusted as desired.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are only exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.

Claims (9)

1. An optimization method for a marine target cooperative detection device is characterized by comprising the following steps:

determining a management decision period and acquiring at least one dynamic event modeling model; the at least one dynamic event modeling model comprises at least one of a modeling model of a new target appearing or far away from a regional detection range, a modeling model of relay and supplement of detection equipment, a modeling model of excessive total detection for representing the quantity of the detection equipment allocated to the target, a modeling model of abnormal information entropy increment for representing detection data of the detection equipment, and a modeling model of instruction issuing by a user;

in a time window of the management decision period, if a dynamic event corresponding to any dynamic event modeling model occurs, a first detection device scheduling scheme adjusting instruction is generated immediately, and if a dynamic event corresponding to any dynamic event modeling model does not occur, a second detection device scheduling scheme adjusting instruction is generated aiming at the current time window;

responding to the first detection equipment scheduling scheme adjusting instruction or the second detection equipment scheduling scheme adjusting instruction, and according to an objective function-based objective and detection equipment cooperative scheduling detection efficiency evaluation model, calculating the deviation between the actual detection efficiency and a set efficiency reference value of the detection equipment by using the detection data acquired by the detection equipment in the current time window; the objective function in the objective function-based objective and detection device cooperative scheduling detection efficiency evaluation model comprises a weighted sum of decision factors influencing a detection device scheduling scheme adjustment decision;

generating a first detection device scheduling optimization scheme according to the deviation calculated according to the adjustment instruction of the first detection device scheduling scheme, so that the detection device is immediately adjusted to perform target cooperative detection according to the first detection device scheduling optimization scheme, or generating a second detection device scheduling optimization scheme according to the deviation calculated according to the adjustment instruction of the second detection device scheduling scheme, so that the detection device is adjusted to perform target cooperative detection according to the second detection device scheduling optimization scheme in the next time window of the management decision period;

a modeling model of new object appearance or far-field detection range, comprising: when a new target enters a regional detection range, allocating detection resources of a detection device to the new target in a generated detection device scheduling scheme adjusting instruction, and when the target leaves the detection region, releasing the detection resources occupied by the target in the generated detection device scheduling scheme adjusting instruction;

the modeling model for detecting the relay and supplement of the equipment comprises the following steps: when an object enters the detection range of a second detection device from the detection range of a first detection device, switching the detection device of the object from the first detection device to the second detection device in the generated detection device scheduling scheme adjustment instruction;

a modeling model for representing a collective excess of probabilistically of the number of target-allocated probe devices, comprising: when the number of the target allocated detection devices is larger than the set number or the ratio of all the detection devices exceeds the set ratio, determining that the common detection degree of the target is too large, and reducing the number of the detection devices to which the target is allocated in the generated detection device scheduling scheme adjusting instruction;

the modeling model for issuing the instruction by the user comprises the following steps: when receiving a manual instruction decision given by a user, performing corresponding deviation control and correction in the generated detection equipment scheduling scheme adjusting instruction;

a method for computing entropy in a modeling model of incremental information entropy anomalies representing anomalies in instrumentation probe data, comprising:

in a time window of each management decision period, acquiring an original detection data queue generated by detection equipment, and counting the occurrence frequency of each original detection data value in the queue;

according to the number of times of the value of each original detection data appearing in the queue, the ratio of the number of times to the length of the queue is the probability that the value of each original detection data is randomly selected by the system, so as to obtain the information entropy of the value of the corresponding original detection data; forming an information entropy sequence based on a time sequence according to the information entropy corresponding to the original detection data queue in each time window;

successively judging the size relation between the value of each information entropy in the entropy sequence and a set threshold, and marking the information entropy if the value of the information entropy is larger than the set threshold;

and if the number of the marked information entropies in the entropy sequence accounts for more than half of the total number of the information entropies, determining that the detection data has sudden change and disturbance, and if the detection equipment loses the capability of continuously executing the detection task, canceling the distribution plan of the detection equipment in the generated detection equipment scheduling scheme adjusting instruction.

2. The method for optimizing a cooperative detection facility for marine targets as claimed in claim 1, wherein determining a management decision cycle comprises:

determining a set target detection process according to a mode from the first appearing target appearing in the detection range to the last appearing target leaving the detection range; and segmenting the set target detection process according to a set time interval to obtain a series of management decision periods.

3. The method for optimizing a cooperative detection apparatus for marine objects as claimed in claim 1,

according to the formula

An information entropy of the values of the corresponding raw detection data is determined, wherein,hthe entropy of the information is represented and,iindicating the sequence number of the data in the queue,p _i representing a data value probability;

and/or the presence of a gas in the gas,

a method for computing a co-detection metric in a modeling model of co-detection overabundance representing an amount of target allocated detection devices, comprising:

acquiring a time period set of a target detected by each detection device in a motion cycle of the target, wherein time period elements in the time period comprise a start time and an end time;

extracting the starting time and the ending time of all the time periods in the time period set, removing the repetition of the time points, and sequencing the time points in an ascending time mode to form a new set;

and traversing the time period in the detection time period set of each detection device in the motion cycle of the target and the sub-time period in the new set, if the traversed sub-time period is in the detection time period set, adding one to the mark value of the traversed sub-time period, and obtaining the final mark value of each sub-time period in the new set after traversal is finished, namely the co-detection degree.

4. The method according to claim 1, wherein in a time window of the management decision cycle, if a dynamic event corresponding to any one of the dynamic event modeling models occurs, generating a first probe scheduling scheme adjustment instruction in real time, includes:

in a time window of the management decision period, if the dynamic event modeling model is a modeling model which is used for representing the quantity of the target allocated detection equipment and has the excessive total detection degree, performing weighted operation on the quantity of the target allocated detection equipment and the detection priority of the target, which are obtained according to the modeling model which is used for representing the quantity of the target allocated detection equipment and has the excessive total detection degree, and instantly generating a first detection equipment scheduling scheme adjusting instruction according to the weighted operation result;

the determination method of the detection priority of the target comprises the following steps: if the target identity is definite and is a nationality-set or friend target, the corresponding detection priority is low; if the identity information of the target is more fuzzy, the corresponding detection priority is higher; if the speed of the target is higher, the corresponding detection priority is higher; if the physical distance between the target and the detection device is shorter, the detection priority corresponding to the target is higher.

5. The method for optimizing a cooperative detection apparatus for marine objects as claimed in claim 1,

the deciding factors influencing the adjustment decision of the dispatching scheme of the detection equipment in the objective function comprise: one or more of detection task priority and threat degree, single body task execution capability and tracking precision of detection equipment, detection equipment coordination degree depending on tasks or independent tasks, detection cost and loss of the detection equipment, and pairing coefficient of targets and the detection equipment;

constraint conditions in the objective function-based objective and detection device cooperative scheduling detection efficiency evaluation model comprise environment limiting factors; the environmental limiting factors include: a necessary limiting condition for establishing a target and detection device cooperative scheduling detection efficiency evaluation model based on an objective function and a non-necessary limiting condition for efficiently scheduling the detection efficiency evaluation model based on the target and detection device cooperative scheduling of the objective function; the necessary limitations include: detecting one or more of an operating temperature of the device, a communication link constraint, an energy supply constraint, and a solar irradiance constraint; non-essential limitations include: one or more of target coverage constraints, probe capacity constraints, probe capability limits, and scheduling real-time constraints;

the method for determining the priority and the threat degree of the detection task comprises the following steps: if the target identity is definite and is a nationality-set or friend target, the corresponding detection priority is low; if the identity information of the target is more fuzzy, the corresponding detection priority is higher; if the speed of the target is higher, the corresponding detection priority is higher; if the physical distance between the target and the detection equipment is shorter, the detection priority corresponding to the target is higher;

the method for determining the single body task execution capability and the tracking accuracy of the detection equipment comprises the following steps: a detection index represented by one or more of a distance measurement, an angle measurement, a detection resolution, and a detection positioning accuracy;

the method for determining the detection device cooperation degree depending on the tasks or the independent tasks comprises the following steps: if the target detection task to be distributed is regarded as a dependent task, preferentially selecting heterogeneous model detection equipment with high coordination level for distribution so as to complete the target detection task; if the target detection task to be distributed is regarded as an independent task, preferentially selecting detection equipment with the same type and high cooperation level for distribution so as to complete the target detection task;

the method for determining the detection cost and the loss of the detection equipment comprises the following steps: calculating the product of the inherent cost and the cost loss of the detection equipment caused by the prediction error and the event occurrence probability of the corresponding prediction error to obtain a risk value caused by the prediction evaluation error of the threat level of the target; the detection equipment inherent cost and loss cost comprises the equipment inherent cost and loss cost, interference of a non-friendly target, energy consumption and time overhead existing when the detection target is switched for multiple times;

the method for determining the pairing coefficient of the target and the detection equipment comprises the following steps: the actual measurement precision of detection equipment and the expected precision of a target are represented by one or more detection precisions including ranging, angle measurement, resolution, positioning and positioning precisions, a covariance matrix of the actual measurement precision and a covariance matrix of the expected precision of the target under an inertial rectangular coordinate system are obtained, and a filter distance function after normalization processing of difference measurement of the two covariance matrices is calculated to obtain a matching coefficient of the target and the detection equipment.

6. The method for optimizing a cooperative detection apparatus for marine objects as claimed in claim 1, further comprising:

generating a GIS map, displaying the geographic background of a target detection area of the detection equipment on the GIS map, and rendering the detection track of each detection equipment to the GIS map;

generating an information map of the target and the detection equipment so as to trigger and analyze the cooperative capacity of the target and the detection equipment when the capacity accumulation of the information map exceeds a set capacity threshold;

dividing a target detection area in a GIS map into a plurality of cellular grid units, calculating detection efficiency values of detection equipment in areas corresponding to the cellular grid units, abstracting the detection efficiency values of the detection equipment to map vector points falling onto the grid unit areas, summarizing and aggregating label information of the number of point objects falling into the grid unit areas in the center of the grid unit areas to obtain efficiency values, summarizing area efficiency characteristics based on the map vector points of the cellular grid units, and sending out detection equipment detection scheme adjustment prompt messages under the condition that the efficiency does not meet set efficiency requirements.

7. The method for optimizing a cooperative detection apparatus for marine objects as claimed in claim 6,

generating an information map of the target and the detection device, comprising:

initializing a target node to store a unique identifier of a target corresponding to the target node; sequencing according to time from morning to night, and connecting the detection equipment of the corresponding target by taking the target node as a center so as to distribute equipment nodes;

when the latest equipment node is generated, the nodes are connected in series according to a tail insertion method, and the tail of the linked list is inserted; wherein the target node and the device node are in an observed and observed relationship;

if two detection devices detect the same target at the same time, the detection relation of the two detection devices points to the corresponding same target node, so that transverse association is formed;

different detection devices continuously detect the target at different time periods, the capacities of the information maps formed by the target nodes and the device nodes are continuously accumulated, a mesh structure between the target and each device is formed from points to trees and through tree networking, and when the capacity accumulation degree is about to exceed a preset information map size threshold, an analysis algorithm of detection capacity extraction and the cooperative capacity of the target and the detection devices is triggered to analyze the cooperative capacity of the target and the detection devices.

8. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method according to any of claims 1 to 7 are implemented when the processor executes the program.

9. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 7.

Images