Air-side operation management method and system, storage medium and electronic equipment

Abstract

The invention provides an air-side operation management method and system, a storage medium and electronic equipment, and relates to the technical field of air-side operation management. According to the method, firstly, multiple pieces of motion positioning track data corresponding to multiple motion objects can be obtained; secondly, a plurality of regional image data corresponding to a plurality of image acquisition devices can be obtained; then, based on the multiple pieces of motion positioning track data and the multiple pieces of regional image data, motion conflict prediction processing can be carried out on the multiple motion objects, and corresponding motion conflict prediction results are obtained; and finally, when the motion conflict prediction result reflects that the plurality of motion objects have the risk of motion conflict in the target operation management area, generating corresponding motion conflict early warning information, and pushing the motion conflict early warning information to corresponding motion conflict early warning terminal equipment. On the basis of the content, the problem that in the prior art, the reliability of motion conflict early warning is relatively low can be solved.

Description

Technical Field

[0001] The present application relates to the technical field of airside operation management, and in particular, to an airside operation management method, system, storage medium and electronic device. Background Art

[0002] The Airfield area Operation Management System (AOMS) focuses on the relevant functions of airfield business management and adopts emerging technologies to strengthen the intelligent interconnection and reasonable scheduling between aircraft, vehicles, runways and facilities. It promotes the level of intelligent coordination and operation process reengineering of the airfield from the aspects of airfield surface operation perception, resource allocation, path planning, safety supervision, etc., improves the thorough perception, agile response, data decision-making and operation evaluation capabilities of airfield operations, and realizes efficient coordination of support entities and coordinated operation of the entire airport.

[0003] However, the inventors have discovered through research that in the above-mentioned management system, the monitoring of the operational safety of moving objects such as aircraft and vehicles has a relatively low reliability problem. Summary of the invention

[0004] In view of this, the purpose of the present application is to provide an airside operation management method, system, storage medium and electronic device to improve the problem of relatively low reliability of motion conflict warning in the prior art.

[0005] To achieve the above objectives, this application adopts the following technical solutions:

[0006] An airside operation management method, comprising:

[0007] For each moving object in the target operation management area, acquiring the motion positioning trajectory data of the moving object, so as to obtain a plurality of motion positioning trajectory data corresponding to a plurality of moving objects, wherein the equipment types of the plurality of moving objects at least include aircraft and vehicles;

[0008] For each image acquisition device in the target operation management area, regional image data formed by the image acquisition device performing an image acquisition operation on the corresponding image acquisition area is acquired to obtain multiple regional image data corresponding to the multiple image acquisition devices, wherein the image acquisition areas corresponding to every two image acquisition devices have at least partially different areas;

[0009] Based on the multiple motion positioning trajectory data and the multiple regional image data, motion conflict prediction processing is performed on the multiple moving objects to obtain corresponding motion conflict prediction results, wherein the motion conflict prediction results are used to reflect whether there is a risk of motion conflict between the multiple moving objects in the target operation management area;

[0010] When the motion conflict prediction result reflects that there is a risk of motion conflict among the multiple moving objects in the target operation management area, corresponding motion conflict warning information is generated, and the motion conflict warning information is pushed to the corresponding motion conflict warning terminal device, so that the conflict management user corresponding to the motion conflict warning terminal device can obtain the motion conflict warning information.

[0011] In a preferred embodiment of the present application, in the above-mentioned airside operation management method, the step of performing motion conflict prediction processing on the multiple moving objects based on the multiple motion positioning trajectory data and the multiple regional image data to obtain corresponding motion conflict prediction results includes:

[0012] Using a first feature mining sub-network in a target motion conflict prediction network, feature mining processing is performed on the plurality of motion positioning trajectory data to output corresponding target motion positioning trajectory features, wherein the target motion conflict prediction network is a trained neural network;

[0013] Using the second feature mining sub-network in the target motion conflict prediction network, the plurality of regional image data are subjected to feature mining processing to output corresponding target regional image features;

[0014] The feature fusion output subnetwork in the target motion conflict prediction network is used to perform feature fusion output processing on the target motion positioning trajectory features and the target area image features to achieve corresponding motion conflict prediction processing and obtain corresponding motion conflict prediction results.

[0015] In a preferred embodiment of the present application, in the above-mentioned airside operation management method, the step of using the first feature mining subnetwork in the target motion conflict prediction network to perform feature mining processing on the plurality of motion positioning trajectory data and outputting the corresponding target motion positioning trajectory features includes:

[0016] Using the embedding unit included in the first feature mining sub-network in the target motion conflict prediction network, each of the plurality of motion positioning trajectory data is respectively embedded to obtain a motion positioning trajectory embedding feature corresponding to each motion positioning trajectory data;

[0017] Using the first significant feature mining unit included in the first feature mining sub-network, for each piece of motion positioning trajectory data, a significant feature mining process is performed on the motion positioning trajectory embedding feature corresponding to the motion positioning trajectory data, and the corresponding motion positioning trajectory significant feature is output;

[0018] Utilizing the second significant feature mining unit included in the first feature mining subnetwork, for each of the motion positioning trajectory data, among the other motion positioning trajectory data other than the motion positioning trajectory data, determining other motion positioning trajectory data having the maximum correlation with the motion positioning trajectory data, and, based on the significant features of the motion positioning trajectory corresponding to the other motion positioning trajectory data, performing significant feature association mining processing on the significant features of the motion positioning trajectory corresponding to the motion positioning trajectory data, outputting the motion positioning trajectory association features corresponding to the motion positioning trajectory data, and based on the maximum correlation, performing weighted superposition processing on the motion positioning trajectory association features and the motion positioning trajectory significant features corresponding to the motion positioning trajectory data, and outputting the local motion positioning trajectory features corresponding to the motion positioning trajectory data;

[0019] The first feature fusion unit included in the first feature mining sub-network is used to perform feature fusion processing on the local motion positioning trajectory features corresponding to each of the motion positioning trajectory data, and output the corresponding target motion positioning trajectory features.

[0020] In a preferred embodiment of the present application, in the above-mentioned airside operation management method, the step of using the second feature mining subnetwork in the target motion conflict prediction network to perform feature mining processing on the plurality of regional image data and outputting corresponding target regional image features includes:

[0021] Using the convolution unit included in the second feature mining subnetwork in the target motion conflict prediction network, respectively perform convolution processing on each of the regional image data, and output regional image convolution features corresponding to each of the regional image data;

[0022] Using the third significant feature mining unit included in the second feature mining subnetwork, for each of the regional image data, a significant feature mining process is performed on the regional image convolution feature corresponding to the regional image data, and the corresponding regional image significant feature is output;

[0023] Using the fourth significant feature mining unit included in the second feature mining subnetwork, for each of the regional image data, other regional image data having the greatest correlation with the regional image data are determined from other regional image data other than the regional image data, and, based on the regional image significant features corresponding to the other regional image data, significant feature association mining processing is performed on the regional image significant features corresponding to the regional image data, and the regional image association features corresponding to the regional image data are output, and, based on the maximum correlation, the regional image association features corresponding to the regional image data and the regional image significant features are weighted superimposed, and the local regional image features corresponding to the regional image data are output;

[0024] The second feature fusion unit included in the second feature mining subnetwork is used to perform feature fusion processing on the local area image features corresponding to each of the area image data, and the corresponding target area image features are output.

[0025] In a preferred embodiment of the present application, in the above-mentioned airside operation management method, the step of utilizing the feature fusion output subnetwork in the target motion conflict prediction network to perform feature fusion output processing on the target motion positioning trajectory features and the target area image features to implement corresponding motion conflict prediction processing and obtain the corresponding motion conflict prediction results includes:

[0026] Using the fifth significant feature mining unit included in the feature fusion output subnetwork in the target motion conflict prediction network, based on the target motion positioning trajectory feature, the target area image feature is subjected to significant feature association mining processing, and the corresponding first association fusion feature is output;

[0027] Using the sixth significant feature mining unit included in the feature fusion output subnetwork, based on the target area image feature, the target motion positioning trajectory feature is subjected to significant feature association mining processing, and the corresponding second association fusion feature is output;

[0028] Using the feature fusion unit included in the feature fusion output subnetwork, perform feature fusion processing on the first associated fusion feature and the second associated fusion feature, and output a corresponding target associated fusion feature;

[0029] The conflict prediction unit included in the feature fusion output subnetwork is used to perform conflict prediction processing on the target associated fusion feature and output a corresponding motion conflict prediction result.

[0030] In a preferred embodiment of the present application, in the above-mentioned airside operation management method, the fifth significant feature mining unit included in the feature fusion output subnetwork in the target motion conflict prediction network performs significant feature association mining processing on the target area image features based on the target motion positioning trajectory features, and outputs the corresponding first associated fusion feature, including:

[0031] Using the first spatial parameter, the second spatial parameter and the third spatial parameter included in the fifth significant feature mining unit included in the feature fusion output subnetwork in the target motion conflict prediction network, respectively perform spatial mapping processing on the target motion positioning trajectory feature and the target area image feature, and output the first trajectory spatial feature, the first image spatial feature and the second image spatial feature;

[0032] Determine a feature association parameter between the first trajectory space feature and the first image space feature, and perform weighted processing on the second image space feature based on the feature association parameter to output a corresponding first association fusion feature.

[0033] In a preferred embodiment of the present application, in the above-mentioned airside operation management method, before executing the step of performing motion conflict prediction processing on the multiple moving objects based on the multiple motion positioning trajectory data and the multiple regional image data to obtain corresponding motion conflict prediction results, the airside operation management method further includes:

[0034] Acquire sample data, wherein the sample data includes sample motion positioning trajectory data, sample area image data, and motion conflict risk labels;

[0035] Using the first feature mining sub-network in the candidate motion conflict prediction network, the sample motion positioning trajectory data is subjected to feature mining processing, and the corresponding sample motion positioning trajectory features are output;

[0036] Using the second feature mining sub-network in the candidate motion conflict prediction network, perform feature mining processing on the sample region image data, and output corresponding sample region image features;

[0037] Using the feature fusion output subnetwork in the candidate motion conflict prediction network, the sample motion positioning trajectory features and the sample area image features are subjected to feature fusion output processing to achieve corresponding motion conflict prediction processing and obtain corresponding sample motion conflict prediction results;

[0038] An error between the sample motion conflict prediction result and the motion conflict risk label is determined, and network parameters of the candidate motion conflict prediction network are updated and adjusted in a direction to reduce the error, so as to form the target motion conflict prediction network.

[0039] This application also provides an airside operation management system, including:

[0040] A trajectory data acquisition module is used to acquire the motion positioning trajectory data of each moving object in the target operation management area, so as to obtain a plurality of motion positioning trajectory data corresponding to a plurality of moving objects, wherein the device types of the plurality of moving objects at least include aircraft and vehicles;

[0041] An image data acquisition module is used to acquire, for each image acquisition device in the target operation management area, regional image data formed by the image acquisition device performing an image acquisition operation on a corresponding image acquisition area, so as to obtain a plurality of regional image data corresponding to the plurality of image acquisition devices, wherein the image acquisition areas corresponding to every two image acquisition devices have at least partially different areas;

[0042] A motion conflict prediction module, used to perform motion conflict prediction processing on the multiple moving objects based on the multiple motion positioning trajectory data and the multiple regional image data, and obtain corresponding motion conflict prediction results, wherein the motion conflict prediction results are used to reflect whether there is a risk of motion conflict between the multiple moving objects in the target operation management area;

[0043] A motion conflict warning module is used to generate corresponding motion conflict warning information when the motion conflict prediction result reflects that there is a risk of motion conflict between the multiple moving objects in the target operation management area, and push the motion conflict warning information to the corresponding motion conflict warning terminal device, so that the conflict management user corresponding to the motion conflict warning terminal device can obtain the motion conflict warning information.

[0044] Based on the above, the present application also provides an electronic device, including:

[0045] Memory for storing computer programs;

[0046] The processor connected to the memory is used to execute the computer program stored in the memory to implement the above-mentioned airside operation management method.

[0047] On the basis of the above, the present application also provides a computer-readable storage medium, in which a computer program is stored, and when the computer program is run, each step of the above-mentioned airside operation management method is executed.

[0048] The airside operation management method, system, storage medium and electronic device provided by the present application can obtain multiple motion positioning trajectory data corresponding to multiple moving objects; secondly, multiple regional image data corresponding to multiple image acquisition devices can be obtained; then, based on the multiple motion positioning trajectory data and the multiple regional image data, motion conflict prediction processing can be performed on multiple moving objects to obtain corresponding motion conflict prediction results; finally, when the motion conflict prediction results reflect that there is a risk of motion conflict between multiple moving objects in the target operation management area, corresponding motion conflict warning information can be generated, and the motion conflict warning information can be pushed to the corresponding motion conflict warning terminal device. Based on the above content, since multiple types of moving objects will be predicted as a whole, and the basis for the prediction includes motion positioning trajectory data and regional image data, multiple types of moving objects can be jointly predicted based on multi-dimensional data. Therefore, compared with the conventional scheme of predicting a single type of moving object based on single-dimensional data, it can have higher reliability, thereby improving the problem of relatively low reliability of motion conflict warning in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings.

[0050] Figure 1 A structural block diagram of an electronic device provided in an embodiment of the present application.

[0051] Figure 2 A flow chart of the airside operation management method provided in an embodiment of the present application.

[0052] Figure 3 A schematic diagram of the network architecture of a target motion conflict prediction network provided in an embodiment of the present application.

[0053] Figure 4 A block diagram of the airside operations management system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations.

[0055] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application for which protection is sought, but merely represents selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0056] like Figure 1 As shown, an embodiment of the present application provides an electronic device, wherein the electronic device may include a memory, a processor, and an airside operation management system.

[0057] In detail, the memory and the processor are electrically connected directly or indirectly to achieve data transmission or interaction. For example, the memory and the processor can be electrically connected through one or more communication buses or signal lines. The airside operation management system includes at least one software function module stored in the memory in the form of software or firmware. The processor is used to execute an executable computer program stored in the memory, for example, the software function module and computer program included in the airside operation management system, so as to implement the airside operation management method provided in the embodiment of the present application.

[0058] Optionally, the memory can be, but is not limited to, random access memory (RAM), read only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable read-only memory (Electric Erasable Programmable Read-Only Memory, EEPROM), etc.

[0059] Furthermore, the processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), a system on chip (SoC), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0060] Understandably, Figure 1 The structure shown is for illustration only, and the electronic device may also include Figure 1 More or fewer components as shown, or with Figure 1 The different configurations shown, for example, may also include a communication unit for exchanging information with other devices (such as an image acquisition device).

[0061] Combination Figure 2 The embodiment of the present application also provides an airside operation management method applicable to the above electronic device. The method steps defined in the process related to the airside operation management method can be implemented by the electronic device. Figure 2 The specific process shown is explained in detail.

[0062] Step S110 : acquiring the motion positioning trajectory data of each moving object in the target operation management area, so as to obtain a plurality of motion positioning trajectory data corresponding to the plurality of moving objects.

[0063] In an embodiment of the present application, the electronic device can obtain the motion positioning trajectory data of each moving object in the target operation management area to obtain multiple motion positioning trajectory data corresponding to multiple moving objects. Among them, the device types of the multiple moving objects include at least aircraft and vehicles, and can also include personnel or other movable devices. Exemplarily, each of the moving objects can be configured with a positioning tag (such as a GPS positioning tag), so that the position of the corresponding moving object can be determined by the positioning tag, thereby collecting the corresponding motion positioning trajectory data, that is, a position sequence formed in chronological order.

[0064] Step S120, for each image acquisition device in the target operation management area, obtain regional image data formed by the image acquisition device performing an image acquisition operation on the corresponding image acquisition area, so as to obtain multiple regional image data corresponding to multiple image acquisition devices.

[0065] In an embodiment of the present application, the electronic device can obtain, for each image acquisition device in the target operation management area, regional image data formed by the image acquisition device performing an image acquisition operation on the corresponding image acquisition area, so as to obtain multiple regional image data corresponding to multiple image acquisition devices. Among them, the image acquisition areas corresponding to every two image acquisition devices have at least partially different areas, that is, the corresponding image acquisition areas between every two image acquisition devices can be completely different or partially different, as long as they are not completely the same. In addition, the image acquisition device can be a camera or other device, so that the image acquisition of the corresponding area is performed by deploying multiple cameras at multiple locations in the target operation management area. In addition, it should be noted that one regional image data can guarantee multiple frames of images, and the multiple frames of images can be sorted in the order of acquisition time to form a corresponding image sequence.

[0066] Step S130 , based on the multiple motion positioning trajectory data and the multiple regional image data, motion conflict prediction processing is performed on the multiple moving objects to obtain corresponding motion conflict prediction results.

[0067] In an embodiment of the present application, the electronic device can perform motion conflict prediction processing on the multiple moving objects based on the multiple motion positioning trajectory data and the multiple regional image data to obtain corresponding motion conflict prediction results. The motion conflict prediction results are used to reflect whether there is a risk of motion conflict between the multiple moving objects in the target operation management area. In other words, based on the movement of multiple moving objects in the current period of time, it is possible to predict whether there will be a possibility of conflict (such as collision problems) in the next period of time.

[0068] Step S140, when the motion conflict prediction result reflects that there is a risk of motion conflict between the multiple moving objects in the target operation management area, corresponding motion conflict warning information is generated, and the motion conflict warning information is pushed to the corresponding motion conflict warning terminal device, so that the conflict management user corresponding to the motion conflict warning terminal device can obtain the motion conflict warning information.

[0069] In an embodiment of the present application, the electronic device can generate corresponding motion conflict warning information when the motion conflict prediction result reflects that there is a risk of motion conflict among the multiple moving objects in the target operation management area, and push the motion conflict warning information to the corresponding motion conflict warning terminal device (such as the corresponding conflict management user's mobile phone, computer and other devices), so that the conflict management user corresponding to the motion conflict warning terminal device can obtain the motion conflict warning information, thereby performing corresponding conflict intervention to avoid conflict among the moving objects.

[0070] Based on the above content, since multiple types of moving objects will be predicted as a whole, and the basis for the prediction includes motion positioning trajectory data and regional image data, multiple types of moving objects can be jointly predicted based on multi-dimensional data. Therefore, compared with the conventional scheme of predicting a single type of moving object based on single-dimensional data, it can have higher reliability, thereby improving the problem of relatively low reliability of motion conflict warning in the prior art.

[0071] It should be noted that for step S130 , the specific manner of performing motion conflict prediction processing on the multiple moving objects is not limited and can be selected according to actual needs.

[0072] For example, in an alternative implementation, in order to ensure the reliability of motion conflict prediction processing so that a more reliable motion conflict prediction result can be obtained, the above-mentioned step S130 may further include step S131, step S132 and step S133, the specific contents of which are described as follows.

[0073] Step S131, using the first feature mining sub-network in the target motion conflict prediction network, performing feature mining processing on the plurality of motion positioning trajectory data, and outputting corresponding target motion positioning trajectory features.

[0074] In an embodiment of the present application, the first feature mining subnetwork in the target motion conflict prediction network can be used to perform feature mining processing on the multiple motion positioning trajectory data, and output corresponding target motion positioning trajectory features. Among them, the target motion conflict prediction network belongs to a trained neural network, that is, the powerful learning ability and expression ability of the neural network (that is, artificial intelligence technology) are used to improve the reliability of feature mining and conflict prediction. Specifically, the first feature mining subnetwork can be used to mine the rich semantic information in the multiple motion positioning trajectory data, thereby obtaining the corresponding target motion positioning trajectory features.

[0075] Step S132, utilizing the second feature mining sub-network in the target motion conflict prediction network to perform feature mining processing on the plurality of region image data, and outputting corresponding target region image features.

[0076] In the embodiment of the present application, the second feature mining sub-network in the target motion conflict prediction network can be used to perform feature mining processing on the multiple regional image data and output corresponding target regional image features. In other words, the second feature mining sub-network can be used to mine rich semantic information in the multiple regional image data, thereby obtaining target regional image features.

[0077] Step S133, utilizing the feature fusion output subnetwork in the target motion conflict prediction network to perform feature fusion output processing on the target motion positioning trajectory features and the target area image features to implement corresponding motion conflict prediction processing and obtain corresponding motion conflict prediction results.

[0078] In an embodiment of the present application, after obtaining the target motion positioning trajectory features and the target area image features, the feature fusion output sub-network in the target motion conflict prediction network can be used to perform feature fusion output processing on the target motion positioning trajectory features and the target area image features to achieve corresponding motion conflict prediction processing and obtain corresponding motion conflict prediction results. In other words, the feature fusion output sub-network can first fuse the target motion positioning trajectory features and the target area image features, and then perform prediction output to obtain the corresponding motion conflict prediction results.

[0079] It can be understood that, in the above step S131, the specific method of performing feature mining processing on the multiple motion positioning trajectory data is not limited and can be selected according to actual needs.

[0080] For example, in an alternative embodiment, in order to improve the semantic representation capability of the mined target motion positioning trajectory features, the above-mentioned step S131 may further include step S131a, step S131b, step S131c and step S131d, and the specific contents of each step are described as follows.

[0081] Step S131a, using the embedding unit included in the first feature mining subnetwork in the target motion conflict prediction network, respectively embed each of the multiple motion positioning trajectory data to obtain a motion positioning trajectory embedding feature corresponding to each motion positioning trajectory data.

[0082] In the embodiment of the present application, the embedding unit (such as Figure 3), respectively embedding each of the multiple motion positioning trajectory data to obtain the motion positioning trajectory embedding features corresponding to each motion positioning trajectory data. It should be noted that the embedding unit can be a word embedding model, which is used to embed each word corresponding to the motion positioning trajectory data to obtain the corresponding word embedding features, and then splicing the various word embedding features to obtain the corresponding motion positioning trajectory embedding features. In addition, considering that the sequence relationship between each position in the motion positioning trajectory data plays an important role in the semantic expression of the trajectory, therefore, the embedding features of each position can also be fused into the motion positioning trajectory embedding features to obtain more semantically rich and reliable motion positioning trajectory embedding features. Specifically, the position embedding features and the word embedding features can be first fused by splicing or superimposing, and then splicing them again to obtain the motion positioning trajectory embedding features.

[0083] Step S131b, using the first significant feature mining unit included in the first feature mining sub-network, for each motion positioning trajectory data, performs significant feature mining processing on the motion positioning trajectory embedded features corresponding to the motion positioning trajectory data, and outputs the corresponding motion positioning trajectory significant features.

[0084] In an embodiment of the present application, after obtaining the embedded features of the motion positioning trajectory, the first significant feature mining unit included in the first feature mining subnetwork can be used to perform significant feature mining processing on the embedded features of the motion positioning trajectory corresponding to each motion positioning trajectory data, and output the corresponding significant features of the motion positioning trajectory. In other words, significant feature mining processing can be performed on each motion positioning trajectory embedded feature respectively, so as to obtain each corresponding significant feature of the motion positioning trajectory, so that the important features in the embedded features of the motion positioning trajectory can be focused on, thereby improving the semantic expression ability. It should be noted that performing significant feature mining processing on the embedded features of the motion positioning trajectory corresponding to the motion positioning trajectory data actually means that, based on the embedded features of the motion positioning trajectory corresponding to the motion positioning trajectory data, significant feature association mining processing is performed on the embedded features of the motion positioning trajectory, wherein the specific processing method of the significant feature association mining processing can refer to the relevant description below.

[0085] Step S131c, utilizing the second significant feature mining unit included in the first feature mining subnetwork, for each of the motion positioning trajectory data, among the other motion positioning trajectory data other than the motion positioning trajectory data, determine other motion positioning trajectory data having the maximum correlation with the motion positioning trajectory data, and, based on the motion positioning trajectory significant features corresponding to the other motion positioning trajectory data, perform significant feature association mining processing on the motion positioning trajectory significant features corresponding to the motion positioning trajectory data, output the motion positioning trajectory associated features corresponding to the motion positioning trajectory data, and based on the maximum correlation, perform weighted superposition processing on the motion positioning trajectory associated features and the motion positioning trajectory significant features corresponding to the motion positioning trajectory data, and output the local motion positioning trajectory features corresponding to the motion positioning trajectory data.

[0086] In an embodiment of the present application, the second significant feature mining unit included in the first feature mining subnetwork can be used to determine, for each of the motion positioning trajectory data, other motion positioning trajectory data with the maximum correlation with the motion positioning trajectory data among other motion positioning trajectory data other than the motion positioning trajectory data, and, based on the significant features of the motion positioning trajectory corresponding to the other motion positioning trajectory data, perform significant feature association mining processing on the significant features of the motion positioning trajectory corresponding to the motion positioning trajectory data, output the motion positioning trajectory association features corresponding to the motion positioning trajectory data, and based on the maximum correlation, perform weighted superposition processing on the motion positioning trajectory association features and the motion positioning trajectory significant features corresponding to the motion positioning trajectory data, and output the local motion positioning trajectory features corresponding to the motion positioning trajectory data. Exemplarily, the correlation between the motion positioning trajectory data can be determined based on the overlap between the trajectories, for example, the overlap can be used as the correlation (when there is no overlap between the trajectories, the minimum distance between the trajectories can be determined as the corresponding correlation). In addition, when performing weighted superposition processing, the maximum correlation degree can be multiplied by the motion positioning trajectory correlation feature to obtain the weighted feature, and then the weighted feature can be superimposed with the motion positioning trajectory significance feature to achieve weighted superposition processing and obtain the corresponding local motion positioning trajectory feature.

[0087] Step S131d, using the first feature fusion unit included in the first feature mining sub-network, perform feature fusion processing on the local motion positioning trajectory features corresponding to each of the motion positioning trajectory data, and output the corresponding target motion positioning trajectory features.

[0088] In an embodiment of the present application, the first feature fusion unit included in the first feature mining subnetwork can be used to perform feature fusion processing on the local motion positioning trajectory features corresponding to each of the motion positioning trajectory data, and output the corresponding target motion positioning trajectory features. Exemplarily, the local motion positioning trajectory features corresponding to each of the motion positioning trajectory data can be spliced, and then convolution, pooling and other processing can be performed to obtain the target motion positioning trajectory features.

[0089] It can be understood that, in the above step S132, the specific manner of performing feature mining processing on the plurality of regional image data is not limited and can be selected according to actual needs.

[0090] For example, in an alternative embodiment, in order to improve the semantic representation capability of the mined target area image features, the above-mentioned step S132 may further include step S132a, step S132b, step S132c and step S132d, and the specific content of each step is described as follows.

[0091] Step S132a, using the convolution unit included in the second feature mining subnetwork in the target motion conflict prediction network, respectively perform convolution processing on each of the regional image data, and output the regional image convolution features corresponding to each of the regional image data.

[0092] In the embodiment of the present application, the convolution unit (such as Figure 3 ), convolution processing is performed on each of the regional image data respectively, and the regional image convolution features corresponding to each of the regional image data are output. It should be noted that, in some embodiments, each of the regional image data may include multiple images, so each image can be convolved separately, and then they can be spliced ​​to form the corresponding regional image convolution features. Similarly, the order relationship between multiple images also has important semantic information. Therefore, the embedded features of the order position of each image can be superimposed or spliced ​​with the convolution result of the image, and then the fused features are spliced ​​to obtain the regional image convolution features corresponding to the multiple images.

[0093] Step S132b, using the third significant feature mining unit included in the second feature mining subnetwork, for each of the regional image data, performs significant feature mining processing on the regional image convolution features corresponding to the regional image data, and outputs the corresponding regional image significant features.

[0094] In an embodiment of the present application, after obtaining the regional image convolution features, the third significant feature mining unit included in the second feature mining subnetwork can be used to perform significant feature mining processing on the regional image convolution features corresponding to each of the regional image data, and output the corresponding regional image significant features. In other words, significant feature mining processing can be performed on each of the regional image convolution features respectively, so as to obtain each corresponding regional image significant feature, so that the important features in the regional image convolution features can be focused on, thereby improving the semantic expression ability. It should be noted that performing significant feature mining processing on the regional image convolution features corresponding to the regional image data actually means performing significant feature association mining processing on the regional image convolution features based on the regional image convolution features corresponding to the regional image data, wherein the specific processing method can refer to the relevant description below.

[0095] Step S132c, using the fourth significant feature mining unit included in the second feature mining subnetwork, for each of the regional image data, among the other regional image data other than the regional image data, determine other regional image data having the maximum correlation with the regional image data, and, based on the regional image significant features corresponding to the other regional image data, perform significant feature association mining processing on the regional image significant features corresponding to the regional image data, and output the regional image association features corresponding to the regional image data, and, based on the maximum correlation, perform weighted superposition processing on the regional image association features and regional image significant features corresponding to the regional image data, and output the local regional image features corresponding to the regional image data.

[0096] In an embodiment of the present application, the fourth significant feature mining unit included in the second feature mining subnetwork can be used to determine, for each of the regional image data, other regional image data with the greatest correlation with the regional image data among other regional image data other than the regional image data, and, based on the regional image significant features corresponding to the other regional image data, perform significant feature association mining processing on the regional image significant features corresponding to the regional image data, and output the regional image association features corresponding to the regional image data, and, based on the maximum correlation, perform weighted superposition processing on the regional image association features and regional image significant features corresponding to the regional image data, and output the local regional image features corresponding to the regional image data. Exemplarily, the correlation between regional image data can refer to the overlap between corresponding image acquisition devices regarding the image acquisition areas (when the image acquisition areas do not overlap, the correlation can be determined based on the distance between the image acquisition areas, and the smaller the distance, the greater the correlation). In addition, when performing weighted superposition processing, the maximum correlation degree can be multiplied by the regional image correlation feature to obtain the weighted feature, and then the weighted feature can be superimposed with the regional image saliency feature to achieve weighted superposition processing and obtain the corresponding local area image feature.

[0097] Step S132d, using the second feature fusion unit included in the second feature mining sub-network, perform feature fusion processing on the local area image features corresponding to each of the area image data, and output the corresponding target area image features.

[0098] In the embodiment of the present application, the second feature fusion unit included in the second feature mining subnetwork can be used to perform feature fusion processing on the local area image features corresponding to each of the regional image data, and output the corresponding target area image features. Exemplarily, the local area image features corresponding to each of the regional image data can be spliced, and then convolution, pooling and other processing can be performed to obtain the target area image features.

[0099] It can be understood that, in the above step S133, the specific method of performing feature fusion output processing on the target motion positioning trajectory features and the target area image features is not limited and can be selected according to actual needs.

[0100] For example, in an alternative embodiment, in order to improve the reliability of feature fusion output processing so that the obtained motion conflict prediction result has a higher reliability, the above-mentioned step S133 may further include step S133a, step S133b, step S133c and step S133d, and the specific content of each step is described as follows.

[0101] Step S133a, utilizing the fifth significant feature mining unit included in the feature fusion output subnetwork in the target motion conflict prediction network, based on the target motion positioning trajectory features, performs significant feature association mining processing on the target area image features, and outputs the corresponding first associated fusion features.

[0102] In the embodiment of the present application, the fifth significant feature mining unit (such as Figure 3 ), based on the target motion positioning trajectory feature, the target area image feature is subjected to significant feature association mining processing, and the corresponding first associated fusion feature is output. In other words, features that are associated with the target motion positioning trajectory feature can be mined from the target area image feature, so that this part of the features can be focused on, thereby obtaining the first associated fusion feature. In this way, the first associated fusion feature can have the semantic information of the target motion positioning trajectory feature and the semantic information of the target area image feature.

[0103] Step S133b, using the sixth significant feature mining unit included in the feature fusion output subnetwork, based on the target area image features, performs significant feature association mining processing on the target motion positioning trajectory features, and outputs the corresponding second association fusion features.

[0104] In the embodiment of the present application, the sixth significant feature mining unit included in the feature fusion output subnetwork can be used to perform significant feature association mining processing on the target motion positioning trajectory feature based on the target area image feature, and output the corresponding second associated fusion feature. In other words, features that are associated with the target area image feature can be mined from the target motion positioning trajectory feature, so that this part of the feature can be focused on, thereby obtaining the second associated fusion feature, so that the second associated fusion feature can have the semantic information of the target area image feature and the semantic information of the target motion positioning trajectory feature.

[0105] Step S133c, utilizing the feature fusion unit included in the feature fusion output subnetwork to perform feature fusion processing on the first associated fusion feature and the second associated fusion feature, and outputting the corresponding target associated fusion feature.

[0106] In an embodiment of the present application, the feature fusion unit included in the feature fusion output subnetwork can be used to perform feature fusion processing on the first associated fusion feature and the second associated fusion feature, and output a corresponding target associated fusion feature. Exemplarily, the first associated fusion feature and the second associated fusion feature can be superimposed or spliced ​​to obtain a target associated fusion feature.

[0107] Step S133d, utilizing the conflict prediction unit included in the feature fusion output subnetwork to perform conflict prediction processing on the target associated fusion feature, and outputting the corresponding motion conflict prediction result.

[0108] In an embodiment of the present application, the conflict prediction unit included in the feature fusion output subnetwork can be used to perform conflict prediction processing on the target-associated fusion feature and output the corresponding motion conflict prediction result. Exemplarily, the target-associated fusion feature can be fully connected to obtain the corresponding fully connected feature, and then the fully connected feature can be output-processed based on the corresponding output function to obtain the corresponding motion conflict prediction result. The output function can be a binary classification function, such as a Sigmoid function.

[0109] It can be understood that, in the above-mentioned step S133a, the specific method of performing significant feature association mining processing is not limited and can be selected according to actual needs. For example, in an alternative implementation, in order to ensure the reliability of significant feature association mining processing, the above-mentioned step S133a can further include step a1 and step a2. The specific content of each step is as follows.

[0110] Step a1, using the first spatial parameter, the second spatial parameter and the third spatial parameter included in the fifth significant feature mining unit included in the feature fusion output subnetwork in the target motion conflict prediction network, respectively perform spatial mapping processing on the target motion positioning trajectory feature and the target area image feature, and output the first trajectory spatial feature, the first image spatial feature and the second image spatial feature.

[0111] In an embodiment of the present application, the first spatial parameter, the second spatial parameter and the third spatial parameter included in the fifth significant feature mining unit included in the feature fusion output subnetwork in the target motion conflict prediction network can be used to perform spatial mapping processing on the target motion positioning trajectory feature and the target area image feature, and output the first trajectory spatial feature, the first image spatial feature and the second image spatial feature. Exemplarily, the first spatial parameter, the second spatial parameter and the third spatial parameter can all be parameter matrices, and can all be used as network parameters of the fifth significant feature mining unit, and can be updated and formed in the corresponding training process. In addition, the target motion positioning trajectory feature and the first spatial parameter can be multiplied to obtain the first trajectory spatial feature, and the target area image feature can be multiplied with the second spatial parameter and the third spatial parameter respectively to obtain the second trajectory spatial feature and the third trajectory spatial feature respectively.

[0112] Step a2: determining a feature association parameter between the first trajectory space feature and the first image space feature, and performing weighted processing on the second image space feature based on the feature association parameter to output a corresponding first association fusion feature.

[0113] In an embodiment of the present application, a feature association parameter between the first trajectory space feature and the first image space feature can be determined, and the second image space feature can be weighted based on the feature association parameter to output a corresponding first associated fusion feature. Exemplarily, the first image space feature can be transposed and then multiplied with the first trajectory space feature to obtain a corresponding feature association parameter, and then the feature association parameter can be multiplied with the second image space feature to achieve weighted processing, thereby obtaining a first associated fusion feature.

[0114] It should be noted that for the above step S130, in order to obtain a reliable motion conflict prediction result through the target motion conflict prediction network when executing step S130, that is, to ensure that the target motion conflict prediction network can have a high prediction accuracy, before executing step S130, corresponding training can be performed first. That is, before executing step S130, the airside operation management method can further include the following steps:

[0115] First, sample data may be obtained, wherein the sample data includes sample motion positioning trajectory data, sample area image data, and motion conflict risk labels; illustratively, the sample motion positioning trajectory data and the sample area image data may be data formed historically in the target operation management area, or may be historical data in other operation management areas, and the motion conflict risk label may be generated by performing a labeling operation based on whether a conflict actually occurs;

[0116] Secondly, the first feature mining sub-network in the candidate motion conflict prediction network can be used to perform feature mining processing on the sample motion positioning trajectory data, and the corresponding sample motion positioning trajectory features can be output. The processing method of the motion positioning trajectory data in the previous article can be referred to;

[0117] Then, the second feature mining sub-network in the candidate motion conflict prediction network can be used to perform feature mining processing on the sample region image data, and the corresponding sample region image features can be output, and the processing method of the sample region image data in the previous text can be referred to;

[0118] Afterwards, the feature fusion output subnetwork in the candidate motion conflict prediction network can be used to perform feature fusion output processing on the sample motion positioning trajectory features and the sample area image features to achieve corresponding motion conflict prediction processing and obtain the corresponding sample motion conflict prediction results, which can refer to the processing methods of the target motion positioning trajectory features and the target area image features;

[0119] Finally, the error between the sample motion conflict prediction result and the motion conflict risk label can be determined (the error calculation method can refer to the relevant existing technology and is not specifically limited here), and the network parameters of the candidate motion conflict prediction network are updated and adjusted in the direction of reducing the error until the error converges (such as the error is reduced to a preset value or the error reduction is less than a preset value, etc.), so as to form the target motion conflict prediction network.

[0120] Combination Figure 4 The embodiment of the present application also provides an airside operation management system applicable to the above electronic device. The airside operation management system may include a trajectory data acquisition module, an image data acquisition module, a motion conflict prediction module, and a motion conflict warning module.

[0121] The trajectory data acquisition module is used to acquire the motion positioning trajectory data of each moving object in the target operation management area, so as to obtain a plurality of motion positioning trajectory data corresponding to a plurality of moving objects, wherein the device types of the plurality of moving objects at least include aircraft and vehicles. In the embodiment of the present application, the trajectory data acquisition module can be used to execute Figure 2 As shown in step S110 , the relevant contents of the trajectory data acquisition module may refer to the above description of step S110 .

[0122] The image data acquisition module is used to acquire the regional image data formed by the image acquisition operation of the image acquisition device on the corresponding image acquisition area for each image acquisition device in the target operation management area, so as to obtain multiple regional image data corresponding to multiple image acquisition devices, wherein the image acquisition areas corresponding to every two image acquisition devices have at least partially different areas. In the embodiment of the present application, the image data acquisition module can be used to perform Figure 2 As shown in step S120, the relevant contents of the image data acquisition module can refer to the above description of step S120.

[0123] The motion conflict prediction module is used to perform motion conflict prediction processing on the multiple moving objects based on the multiple motion positioning trajectory data and the multiple regional image data to obtain corresponding motion conflict prediction results, wherein the motion conflict prediction results are used to reflect whether there is a risk of motion conflict between the multiple moving objects in the target operation management area. In the embodiment of the present application, the motion conflict prediction module can be used to perform Figure 2 As shown in step S130, the relevant contents of the motion conflict prediction module can refer to the above description of step S130.

[0124] The motion conflict warning module is used to generate corresponding motion conflict warning information when the motion conflict prediction result reflects that there is a risk of motion conflict between the multiple moving objects in the target operation management area, and push the motion conflict warning information to the corresponding motion conflict warning terminal device, so that the conflict management user corresponding to the motion conflict warning terminal device can obtain the motion conflict warning information. In the embodiment of the present application, the motion conflict warning module can be used to execute Figure 2 As shown in step S140, the relevant contents of the motion conflict warning module can refer to the above description of step S140.

[0125] In an embodiment of the present application, corresponding to the above-mentioned airside operation management method applied to the electronic device, a computer-readable storage medium is also provided, in which a computer program is stored, and when the computer program is run, each step of the airside operation management method is executed.

[0126] Among them, the steps executed when the aforementioned computer program is running will not be described one by one here, and reference may be made to the explanation of the aforementioned airside operation management method in the previous text.

[0127] In summary, the airside operation management method, system, storage medium and electronic device provided by the present application can obtain multiple motion positioning trajectory data corresponding to multiple moving objects; secondly, multiple regional image data corresponding to multiple image acquisition devices can be obtained; then, based on the multiple motion positioning trajectory data and the multiple regional image data, motion conflict prediction processing can be performed on multiple moving objects to obtain corresponding motion conflict prediction results; finally, when the motion conflict prediction results reflect that there is a risk of motion conflict between multiple moving objects in the target operation management area, corresponding motion conflict warning information can be generated, and the motion conflict warning information can be pushed to the corresponding motion conflict warning terminal device. Based on the above content, since multiple types of moving objects will be predicted as a whole, and the basis for the prediction includes motion positioning trajectory data and regional image data, multiple types of moving objects can be jointly predicted based on multi-dimensional data. Therefore, compared with the conventional scheme of predicting a single type of moving object based on single-dimensional data, it can have higher reliability, thereby improving the problem of relatively low reliability of motion conflict warning in the prior art.

[0128] In several embodiments provided in the embodiments of the present application, it should be understood that the disclosed device and method can also be implemented in other ways. The device and method embodiments described above are merely schematic, for example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the device, method and computer program product according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0129] In addition, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0130] If the function is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, an electronic device, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a disk or an optical disk. It should be noted that, in this article, the term "include", "include" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements that are not explicitly listed, or also includes elements inherent to such a process, method, article or device. Without more constraints, an element defined by the phrase "comprising a..." does not exclude the existence of other identical elements in the process, method, article or apparatus comprising the element.

[0131] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. An airside operation management method, characterized in that: include: For each moving object in the target operation management area, acquiring the motion positioning trajectory data of the moving object, so as to obtain a plurality of motion positioning trajectory data corresponding to a plurality of moving objects, wherein the equipment types of the plurality of moving objects at least include aircraft and vehicles; For each image acquisition device in the target operation management area, regional image data formed by the image acquisition device performing an image acquisition operation on the corresponding image acquisition area is acquired to obtain multiple regional image data corresponding to the multiple image acquisition devices, wherein the image acquisition areas corresponding to every two image acquisition devices have at least partially different areas; Based on the multiple motion positioning trajectory data and the multiple regional image data, motion conflict prediction processing is performed on the multiple moving objects to obtain corresponding motion conflict prediction results, wherein the motion conflict prediction results are used to reflect whether there is a risk of motion conflict between the multiple moving objects in the target operation management area; When the motion conflict prediction result reflects that there is a risk of motion conflict among the multiple moving objects in the target operation management area, corresponding motion conflict warning information is generated, and the motion conflict warning information is pushed to the corresponding motion conflict warning terminal device, so that the conflict management user corresponding to the motion conflict warning terminal device can obtain the motion conflict warning information.

2. The airside operation management method according to claim 1, characterized in that: The step of performing motion conflict prediction processing on the multiple moving objects based on the multiple motion positioning trajectory data and the multiple regional image data to obtain corresponding motion conflict prediction results includes: Using a first feature mining sub-network in a target motion conflict prediction network, feature mining processing is performed on the plurality of motion positioning trajectory data to output corresponding target motion positioning trajectory features, wherein the target motion conflict prediction network is a trained neural network; Using the second feature mining sub-network in the target motion conflict prediction network, the plurality of regional image data are subjected to feature mining processing to output corresponding target regional image features; The feature fusion output subnetwork in the target motion conflict prediction network is used to perform feature fusion output processing on the target motion positioning trajectory features and the target area image features to achieve corresponding motion conflict prediction processing and obtain corresponding motion conflict prediction results.

3. The airside operation management method according to claim 2, characterized in that: The step of using the first feature mining sub-network in the target motion conflict prediction network to perform feature mining processing on the plurality of motion positioning trajectory data and outputting corresponding target motion positioning trajectory features comprises: Using the embedding unit included in the first feature mining sub-network in the target motion conflict prediction network, each of the plurality of motion positioning trajectory data is respectively embedded to obtain a motion positioning trajectory embedding feature corresponding to each motion positioning trajectory data; Using the first significant feature mining unit included in the first feature mining subnetwork, for each piece of motion positioning trajectory data, a significant feature mining process is performed on the motion positioning trajectory embedding feature corresponding to the motion positioning trajectory data, and the corresponding motion positioning trajectory significant feature is output; Utilizing the second significant feature mining unit included in the first feature mining subnetwork, for each of the motion positioning trajectory data, among the other motion positioning trajectory data other than the motion positioning trajectory data, determining other motion positioning trajectory data having the maximum correlation with the motion positioning trajectory data, and, based on the significant features of the motion positioning trajectory corresponding to the other motion positioning trajectory data, performing significant feature association mining processing on the significant features of the motion positioning trajectory corresponding to the motion positioning trajectory data, outputting the motion positioning trajectory association features corresponding to the motion positioning trajectory data, and based on the maximum correlation, performing weighted superposition processing on the motion positioning trajectory association features and the motion positioning trajectory significant features corresponding to the motion positioning trajectory data, and outputting the local motion positioning trajectory features corresponding to the motion positioning trajectory data; The first feature fusion unit included in the first feature mining sub-network is used to perform feature fusion processing on the local motion positioning trajectory features corresponding to each of the motion positioning trajectory data, and output the corresponding target motion positioning trajectory features.

4. The airside operation management method according to claim 2, characterized in that: The step of using the second feature mining sub-network in the target motion conflict prediction network to perform feature mining processing on the plurality of regional image data and outputting corresponding target regional image features comprises: Using the convolution unit included in the second feature mining subnetwork in the target motion conflict prediction network, respectively perform convolution processing on each of the regional image data, and output regional image convolution features corresponding to each of the regional image data; Using the third significant feature mining unit included in the second feature mining subnetwork, for each of the regional image data, a significant feature mining process is performed on the regional image convolution feature corresponding to the regional image data, and the corresponding regional image significant feature is output; Using the fourth significant feature mining unit included in the second feature mining subnetwork, for each of the regional image data, other regional image data having the greatest correlation with the regional image data are determined from other regional image data other than the regional image data, and, based on the regional image significant features corresponding to the other regional image data, significant feature association mining processing is performed on the regional image significant features corresponding to the regional image data, and the regional image association features corresponding to the regional image data are output, and, based on the maximum correlation, the regional image association features corresponding to the regional image data and the regional image significant features are weighted superimposed, and the local regional image features corresponding to the regional image data are output; The second feature fusion unit included in the second feature mining subnetwork is used to perform feature fusion processing on the local area image features corresponding to each of the area image data, and the corresponding target area image features are output.

5. The airside operation management method according to claim 2, characterized in that: The step of utilizing the feature fusion output subnetwork in the target motion conflict prediction network to perform feature fusion output processing on the target motion positioning trajectory features and the target area image features to implement corresponding motion conflict prediction processing and obtain corresponding motion conflict prediction results includes: Using the fifth significant feature mining unit included in the feature fusion output subnetwork in the target motion conflict prediction network, based on the target motion positioning trajectory feature, the target area image feature is subjected to significant feature association mining processing, and the corresponding first association fusion feature is output; Using the sixth significant feature mining unit included in the feature fusion output subnetwork, based on the target area image feature, the target motion positioning trajectory feature is subjected to significant feature association mining processing, and the corresponding second association fusion feature is output; Using the feature fusion unit included in the feature fusion output subnetwork, perform feature fusion processing on the first associated fusion feature and the second associated fusion feature, and output a corresponding target associated fusion feature; The conflict prediction unit included in the feature fusion output subnetwork is used to perform conflict prediction processing on the target associated fusion feature and output a corresponding motion conflict prediction result.

6. The airside operation management method according to claim 5, characterized in that: The step of utilizing the fifth significant feature mining unit included in the feature fusion output subnetwork in the target motion conflict prediction network to perform significant feature association mining processing on the target area image features based on the target motion positioning trajectory features and outputting the corresponding first associated fusion features comprises: Using the first spatial parameter, the second spatial parameter and the third spatial parameter included in the fifth significant feature mining unit included in the feature fusion output subnetwork in the target motion conflict prediction network, respectively perform spatial mapping processing on the target motion positioning trajectory feature and the target area image feature, and output the first trajectory spatial feature, the first image spatial feature and the second image spatial feature; Determine a feature association parameter between the first trajectory space feature and the first image space feature, and perform weighted processing on the second image space feature based on the feature association parameter to output a corresponding first association fusion feature.

7. The airside operation management method according to any one of claims 2 to 6, characterized in that: Before executing the step of performing motion conflict prediction processing on the multiple moving objects based on the multiple motion positioning trajectory data and the multiple regional image data to obtain corresponding motion conflict prediction results, the airside operation management method further includes: Acquire sample data, wherein the sample data includes sample motion positioning trajectory data, sample area image data, and motion conflict risk labels; Using the first feature mining sub-network in the candidate motion conflict prediction network, the sample motion positioning trajectory data is subjected to feature mining processing, and the corresponding sample motion positioning trajectory features are output; Using the second feature mining sub-network in the candidate motion conflict prediction network, perform feature mining processing on the sample region image data, and output corresponding sample region image features; Using the feature fusion output subnetwork in the candidate motion conflict prediction network, the sample motion positioning trajectory features and the sample area image features are subjected to feature fusion output processing to achieve corresponding motion conflict prediction processing and obtain corresponding sample motion conflict prediction results; An error between the sample motion conflict prediction result and the motion conflict risk label is determined, and network parameters of the candidate motion conflict prediction network are updated and adjusted in a direction to reduce the error, so as to form the target motion conflict prediction network.

8. An airside operation management system, characterized in that: include: A trajectory data acquisition module is used to acquire the motion positioning trajectory data of each moving object in the target operation management area, so as to obtain a plurality of motion positioning trajectory data corresponding to a plurality of moving objects, wherein the device types of the plurality of moving objects at least include aircraft and vehicles; An image data acquisition module is used to acquire, for each image acquisition device in the target operation management area, regional image data formed by the image acquisition device performing an image acquisition operation on a corresponding image acquisition area, so as to obtain a plurality of regional image data corresponding to the plurality of image acquisition devices, wherein the image acquisition areas corresponding to every two image acquisition devices have at least partially different areas; A motion conflict prediction module, used to perform motion conflict prediction processing on the multiple moving objects based on the multiple motion positioning trajectory data and the multiple regional image data, and obtain corresponding motion conflict prediction results, wherein the motion conflict prediction results are used to reflect whether there is a risk of motion conflict between the multiple moving objects in the target operation management area; A motion conflict warning module is used to generate corresponding motion conflict warning information when the motion conflict prediction result reflects that there is a risk of motion conflict between the multiple moving objects in the target operation management area, and push the motion conflict warning information to the corresponding motion conflict warning terminal device, so that the conflict management user corresponding to the motion conflict warning terminal device can obtain the motion conflict warning information.

9. An electronic device, characterized in that: include: Memory for storing computer programs; A processor connected to the memory is used to execute the computer program stored in the memory to implement the airside operation management method described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed, executes the airside operation management method according to any one of claims 1 to 7.

Images