Multi-device cooperative scheduling processing method and system of mobile transparent display device

Abstract

The invention relates to the technical field of equipment collaborative scheduling, and discloses a multi-equipment collaborative scheduling processing method and system for mobile transparent display equipment, and the method comprises the steps: collecting the real-time state data of a plurality of mobile transparent display equipment in an exhibition hall, and calculating the task compatibility data between the real-time state data and an exhibition task; deadlock risk prediction is carried out based on the real-time state data and the task compatibility data, and an inter-device space conflict probability and a task execution deadlock risk value are obtained; and performing cooperative control on the plurality of mobile transparent display devices based on the space conflict probability between the devices and the task execution deadlock risk value, and outputting a device movement instruction and a transparency coordination instruction. Continuous and stable operation of the multiple mobile transparent display devices in the exhibition hall is ensured, the optimal visual display effect is ensured, and meanwhile energy waste is avoided.

Description

Technical Field

[0001] This invention relates to the field of equipment collaborative scheduling technology, and in particular to a multi-device collaborative scheduling processing method and system for a mobile transparent display device. Background Technology

[0002] Mobile transparent display devices are widely used in museums, exhibition centers, and commercial displays due to their unique visual permeability and spatial flexibility. These devices can maintain spatial transparency while displaying digital content, providing viewers with an immersive, multi-layered information display experience. However, when multiple mobile transparent display devices operate simultaneously in a shared workspace, it is difficult to effectively handle the unique visual layering effect and content compatibility issues of transparent displays. In exhibition environments with high device density, this can easily lead to problems such as spatial location conflicts, content interference, and task execution blockages. Summary of the Invention

[0003] This invention provides a multi-device collaborative scheduling method and system for mobile transparent display devices. Compared with traditional passive detection methods, this invention reduces the probability of deadlock, ensures the continuous and stable operation of multiple mobile transparent display devices in the exhibition hall, ensures the optimal visual display effect, and avoids energy waste.

[0004] The first aspect of this invention provides a multi-device collaborative scheduling method for a mobile transparent display device, the method comprising: Collect real-time status data of multiple mobile transparent display devices in the exhibition hall, and calculate task compatibility data between the display task and the task. Based on the real-time status data and the task compatibility data, deadlock risk prediction is performed to obtain the probability of spatial conflict between devices and the risk value of task execution deadlock. Based on the spatial conflict probability between the devices and the deadlock risk value of the task execution, the multiple mobile transparent display devices are coordinated and controlled to output device movement commands and transparency coordination commands.

[0005] In conjunction with the first aspect, in a first implementation of the first aspect of the present invention, the step of collecting real-time status data of multiple mobile transparent display devices within the exhibition hall and calculating task compatibility data with the display task includes: The exhibition hall is divided into three-dimensional grid units to construct a spatial coordinate system for the exhibition hall; Based on the exhibition hall spatial coordinate system, motion parameters of multiple mobile transparent display devices are collected to obtain motion state parameters; The transparent display function parameters of the multiple mobile transparent display devices are collected to obtain the function status parameters; The motion state parameters and the functional state parameters are used as real-time state data; Based on the real-time status data, a transparent display compatibility analysis is performed on the displayed tasks to obtain task compatibility data.

[0006] In conjunction with the first aspect, in a second implementation of the first aspect of the present invention, the step of performing a transparent display compatibility analysis on the display task based on the real-time status data to obtain task compatibility data includes: The task attribute parameters, including task priority, execution duration, content data volume, display area, and preceding dependent tasks, are parsed from the real-time status data. Based on the preceding dependent tasks in the task attribute parameters, construct a dependency graph of the task execution order; The content data volume and display area in the task attribute parameters are used to calculate the transparency display conflict degree, and a compatibility matrix is ​​obtained. By combining the dependency graph and the compatibility matrix, task scheduling constraint analysis is performed to obtain task compatibility data.

[0007] In conjunction with the first aspect, in a third implementation of the first aspect of the present invention, the step of predicting deadlock risk based on the real-time status data and the task compatibility data to obtain the inter-device spatial conflict probability and the task execution deadlock risk value includes: Based on the real-time status data, a device status graph containing device nodes and spatial relationship edges is constructed. The device nodes contain position coordinates, movement speed, and transparency information, and the spatial relationship edges contain distance and relative motion information between devices. The device state diagram is input into a graph attention network for spatial conflict analysis to obtain the spatial conflict probability between devices. The historical motion trajectory sequence of the device is extracted from the real-time status data, and the historical motion trajectory sequence of the device is input into the time series transformation network for trajectory prediction to obtain the task execution deadlock risk value.

[0008] In conjunction with the first aspect, in a fourth implementation of the first aspect of the present invention, the step of inputting the device state diagram into a graph attention network for spatial conflict analysis to obtain the inter-device spatial conflict probability includes: The device state graph is input into the first graph attention layer of the graph attention network for primary feature extraction to obtain the first latent feature vector; The first hidden feature vector is input into the second graph attention layer of the graph attention network for intermediate feature fusion to obtain the second hidden feature vector; The second hidden feature vector is input into the third graph attention layer of the graph attention network for high-level semantic abstraction processing to obtain the third hidden feature vector; The probability mapping transformation of the third hidden feature vector is performed to obtain the spatial conflict probability between devices.

[0009] In conjunction with the first aspect, in a fifth implementation of the first aspect of the present invention, the step of coordinating the control of the plurality of mobile transparent display devices based on the spatial conflict probability between the devices and the task execution deadlock risk value, and outputting device movement commands and transparency coordination commands, includes: Calculate the deadlock resolution cost based on the inter-device spatial conflict probability and the task execution deadlock risk value; The deadlock resolution cost is set based on a multi-objective optimization function that includes constraints on safe distances between devices, overlap of transparent display areas, and integrity of task time-dependent constraints. The multi-objective optimization function is solved iteratively using the particle swarm optimization algorithm to obtain the optimal parameter combination; Based on the optimal parameter combination, a deadlock prevention scheduling scheme is generated, which includes the target movement position of the device, movement path planning, transparency adjustment value, and task execution timing arrangement. The multiple mobile transparent display devices are coordinated and controlled according to the anti-deadlock scheduling scheme, and device movement commands and transparency coordination commands are output.

[0010] In conjunction with the first aspect, in the sixth implementation of the first aspect of the present invention, the step of calculating the deadlock resolution cost based on the inter-device spatial conflict probability and the task execution deadlock risk value includes: Based on the spatial conflict probability between devices, the energy consumption cost of device movement is calculated for the current position coordinates of each mobile transparent display device and the position coordinates of the avoidance target. Based on the deadlock risk value of the task execution, the waiting time of the displayed task is calculated to obtain the task waiting time cost; The transparency of each mobile transparent display device is adjusted to obtain the cost of transparency adjustment. The deadlock resolution cost is obtained by weighted summing of the device mobility energy consumption cost, the task waiting time cost, the transparency adjustment cost, and the task switching penalty cost.

[0011] In conjunction with the first aspect, in the seventh implementation of the first aspect of the present invention, the step of coordinating the control of the plurality of mobile transparent display devices according to the anti-deadlock scheduling scheme and outputting device movement commands and transparency coordination commands includes: The device control parameters, including the target location coordinates, expected arrival time, and transparency setting value, are parsed from the aforementioned deadlock prevention scheduling scheme. Based on the device control parameters, a distributed collaborative control law is calculated for each mobile transparent display device to obtain the device control output vector. When the overlapping area of ​​the display area of ​​the mobile transparent display device exceeds a preset threshold, the transparency negotiation protocol is initiated to perform a balanced solution and obtain the negotiated transparency parameters. Based on the device control output vector, a device movement command containing the movement direction, movement speed, and arrival time is generated, and a transparency coordination command is generated based on the negotiated transparency parameter.

[0012] In conjunction with the first aspect, in the eighth implementation of the first aspect of the present invention, the step of calculating the distributed cooperative control law for each mobile transparent display device based on the device control parameters to obtain the device control output vector includes: Based on the target position coordinates in the device control parameters, the position error of each mobile transparent display device is calculated to obtain the position control component; The repulsive force field is calculated based on the real-time distance relationship between each mobile transparent display device to obtain the repulsive force field control component. The transparency gradient force is calculated based on the current transparency value of each mobile transparent display device and the importance weight of the displayed content to obtain the transparency control component; The position control component, the repulsive field control component, and the transparency control component are weighted and synthesized using a proportional-differential control law to obtain the device control output vector.

[0013] A second aspect of the present invention provides a multi-device collaborative scheduling and processing system for a mobile transparent display device, the multi-device collaborative scheduling and processing system for the mobile transparent display device comprising: The data acquisition module is used to collect real-time status data from multiple mobile transparent display devices in the exhibition hall and to calculate task compatibility data between the data acquisition module and the display tasks. The deadlock risk prediction module is used to predict deadlock risk based on the real-time status data and the task compatibility data, and to obtain the spatial conflict probability between devices and the task execution deadlock risk value. The collaborative control module is used to collaboratively control the multiple mobile transparent display devices based on the spatial conflict probability between the devices and the task execution deadlock risk value, and output device movement commands and transparency coordination commands.

[0014] Compared to traditional passive detection methods, this invention reduces the probability of deadlock and ensures the continuous and stable operation of multiple mobile transparent display devices in an exhibition hall. Based on the transparent display task compatibility matrix, it can accurately assess the visual compatibility of different display content in overlapping areas, effectively solving the content conflict problem caused by traditional scheduling methods ignoring the characteristics of transparent displays. By simultaneously optimizing four objective functions—movement cost, waiting time, transparency adjustment, and task switching—it achieves the optimal balance between deadlock avoidance and performance maintenance, avoiding local optima problems caused by single-objective optimization. Employing a hierarchical distributed control structure combined with a transparency negotiation protocol, it possesses stronger robustness and real-time response capabilities. Compared to centralized control methods, it can better handle device failures and network latency. When display areas overlap, the transparency negotiation protocol automatically coordinates the transparency settings between devices based on the Nash equilibrium principle of game theory, ensuring optimal visual display effects while avoiding energy waste. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] The structures, proportions, sizes, etc., shown in the accompanying drawings of this specification are only for the purpose of assisting those skilled in the art in understanding and reading the content disclosed in the specification, and are not intended to limit the conditions under which the present invention can be implemented. Therefore, they have no substantial technical significance. Any modifications to the structure, changes in the proportions, or adjustments to the size, without affecting the effects and objectives that the present invention can produce, should still fall within the scope of the technical content disclosed in the present invention.

[0017] Figure 1 This is a flowchart illustrating the multi-device collaborative scheduling method for a mobile transparent display device provided in an embodiment of the present invention. Figure 2 This is a schematic block diagram of the structure of the multi-device collaborative scheduling and processing system for a mobile transparent display device provided in an embodiment of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be performed in the described order. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0020] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0021] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items, and all possible combinations, and includes such combinations. See also Figure 1 One embodiment of the multi-device collaborative scheduling processing method for mobile transparent display devices in this invention includes: Step 100: Collect real-time status data of multiple mobile transparent display devices in the exhibition hall, and calculate the task compatibility data between the display task and the task. Specifically, the entire exhibition hall space is divided into three-dimensional mesh units to form a spatial coordinate system with a unified scale benchmark. This coordinate system is discretized using equidistant cubic meshes, with each mesh unit uniquely numbered and labeled with its three-dimensional spatial position. Based on this coordinate system, a positioning module, inertial navigation unit, and visual positioning system are deployed to frequently collect key physical pose information such as the three-dimensional coordinates, velocity vectors, and orientation of multiple mobile transparent display devices within the exhibition hall. The collected results are then uniformly converted into the coordinate system, forming motion state parameters that include position, velocity, and orientation information. Simultaneously, the current functional operating status of the transparent display devices is acquired through the device controller interface or sensing module, including parameters such as the current transparency setting, remaining battery power, display task identifier, and content display intensity coefficient. Content characterizing the device's visual output features is categorized as functional state parameters. The motion state parameters and functional state parameters are then time-series aligned and structurally integrated to generate a real-time status dataset. Based on the set of display tasks preset in the task scheduling module, and relying on the content characteristics, display position, display area and expected transparency requirements of the transparent display tasks, combined with the current position, transparency value and current task load information of each device in real-time status data, the compatibility measurement calculation between tasks and between devices and tasks is performed. By calculating comprehensive indicators such as content semantic similarity difference, transparency difference and spatial overlap, task compatibility data is generated.

[0022] Step 200: Based on real-time status data and task compatibility data, predict deadlock risk to obtain the probability of spatial conflict between devices and the deadlock risk value of task execution. Specifically, a device state diagram oriented towards the dynamic interaction relationships of devices is constructed based on real-time state data. The device state diagram consists of a set of nodes representing each mobile transparent display device and a set of edges describing the spatial relationships between devices. Each device node embeds its three-dimensional position coordinates in the exhibition hall's spatial coordinate system, its current motion velocity vector, and its transparency value, reflecting the device's spatial occupancy and visual output status at the current moment. The spatial relationship edges connecting these nodes use Euclidean distance, velocity direction angle, and relative velocity magnitude between devices as key feature parameters, characterizing the geometric relationship and potential interaction risks between devices in terms of motion trends. The device state diagram is input into a trained graph attention network model. The graph attention network model automatically extracts spatial interaction features and weights and fuses the adjacency relationships between devices through a multi-layer attention mechanism. It identifies high-risk device combinations that may cause physical collisions or spatial occlusion and outputs a spatial conflict probability matrix between devices, where the probability value between each pair of devices reflects the likelihood of them experiencing spatial conflict within a short-term prediction window. Simultaneously, the historical motion trajectory sequences of each device within the most recent time period are parsed from real-time status data. These sequences include continuous features such as device position, velocity, acceleration, task progress, and transparency changes corresponding to multiple frame timestamps, forming a temporal input dataset. This historical motion trajectory sequence is then input into a temporal transformation network constructed based on a multi-layer encoder-decoder structure. The temporal transformation network maintains temporal order information through a positional encoding mechanism and utilizes a multi-head self-attention mechanism to model the dynamic evolution of device trajectories. This predicts the spatial position change trends of each device and its task state evolution path over a future period. Based on this, task compatibility data and temporal conflict analysis indicators are combined to calculate the task execution deadlock risk value. The deadlock risk value quantifies the probability of system-level logical deadlock that may result from the combined effects of factors such as task dependency conflicts, delayed transparency adjustment, and overlapping resource usage.

[0023] Step 300: Based on the probability of spatial conflict between devices and the risk value of task execution deadlock, coordinate the control of multiple mobile transparent display devices and output device movement commands and transparency coordination commands.

[0024] Specifically, a deadlock resolution cost function is established to assess the overall system operational risk level by integrating the spatial conflict probability matrix and the task execution deadlock risk value. This function quantifies the deadlock cost level of the entire system in its current state by weightedly integrating the spatial interference probability between devices, the probability of potential temporal defaults in task scheduling, and the severity of transparency resource allocation conflicts. Using deadlock resolution cost as the core evaluation metric, a multi-objective optimization function with multiple constraints is constructed. This function sets a minimum safe distance threshold between devices to ensure spatial non-interference, sets an upper limit threshold for the visual overlap between transparent display areas to avoid content occlusion, and preserves the temporal integrity of display tasks through task topology dependencies. This ensures that the generated scheduling scheme satisfies the deadlock resolution objective without disrupting the continuity and integrity of the display logic. The multi-objective optimization function is iteratively solved using a particle swarm optimization algorithm. In each iteration, the particle positions, velocities, and dynamic trade-offs between historical and global best solutions are updated to gradually converge to the optimal parameter combination under the current constraints. This parameter combination specifically includes the target movement coordinates of each device, path node selection, transparency adjustment targets, and the precise time sequence of task execution. Based on the optimal parameter combination, a deadlock-prevention scheduling scheme is constructed, specifying the target position and path planning node sequence for each moving transparent display device in the next stage. This also includes the time points, target values, and adjustment speed control values ​​during transparency adjustment, and simultaneously outputs a task execution schedule table conforming to the task dependency sequence. The deadlock-prevention scheduling scheme is parsed into specific control instructions by a central scheduling module and distributed to each device in a structured format, including device movement instructions and transparency coordination instructions.

[0025] In one specific embodiment, the process of performing step 100 may specifically include the following steps: The exhibition hall is divided into three-dimensional grid units to construct a spatial coordinate system for the exhibition hall; Motion parameters of multiple mobile transparent display devices are collected based on the exhibition hall spatial coordinate system to obtain motion state parameters; The transparent display function parameters of multiple mobile transparent display devices are collected to obtain the function status parameters; Use motion state parameters and functional state parameters as real-time state data; Based on real-time status data, a transparent display compatibility analysis is performed on the displayed tasks to obtain task compatibility data.

[0026] Specifically, the entire exhibition hall space is divided into three-dimensional grid units. A unified spatial discretization strategy is adopted to construct an exhibition hall spatial coordinate system with defined three-dimensional coordinates. The exhibition hall spatial coordinate system uses fixed-size cubic units as basic building blocks, for example, with a grid resolution of 0.5m × 0.5m × 0.2m. The entire exhibition hall area is rasterized, and each three-dimensional unit is assigned a unique number and spatial location identifier, so that any spatial point can be mapped to the corresponding grid unit under the coordinate system, establishing a unified spatial reference framework for equipment positioning, task display area mapping, and path planning. Based on the exhibition hall spatial coordinate system, a multi-source sensor cluster consisting of an inertial navigation module, UWB positioning nodes, high frame rate visual tracking equipment, and wireless communication units is deployed to monitor the spatial motion characteristics of all mobile transparent display devices in the exhibition hall in real time. Through multi-source data fusion algorithms, the original position data is synchronized in time, error corrected, and trajectory fitted to extract motion state parameters, including three-dimensional position coordinates, velocity vectors, acceleration information, motion orientation angle, and current path node number, reflecting the operating status and evolution trend of each device in space. The device's internal control module collects transparent display-related functional parameters in parallel, including reading and encoding data such as the current set transparency value, content display brightness level, current task ID, remaining battery level, display content size, border occlusion status, and screen content semantic weight, to obtain functional status parameters. Motion status parameters and functional status parameters are aligned on the timeline and combined into real-time status data based on a unified structural specification. This real-time status data is then standardized and encapsulated using a unified data structure. Based on the real-time status data, a transparent display compatibility analysis is performed on the currently scheduled set of display tasks. Task description parameters for each display task are extracted, including semantic tags of the displayed content, recommended transparency range, display area range, content type (image, text, motion graphics, etc.), semantic focus position, and display priority level. These task description parameters are compared one by one with the device's functional status parameters. Simultaneously, the device's current motion status is considered to determine its spatial coverage and adjustability. A compatibility evaluation model is constructed based on content semantic similarity calculation, transparency difference function, and spatial overlap function. The compatibility assessment model calculates the compatibility score between each device and each task by setting content semantic difference weight, transparency matching weight and spatial overlap penalty coefficient. At the same time, it generates a task parallel execution matrix and a conflict warning index matrix. When the compatibility score exceeds the set threshold, the device is considered to have the scheduling potential to undertake the task. Otherwise, it is marked as a combination relationship that needs to avoid conflict and outputs task compatibility data.

[0027] In one specific embodiment, the process of performing a transparent display compatibility analysis on the display task based on real-time status data to obtain task compatibility data can specifically include the following steps: Parse task attribute parameters from real-time status data, including task priority, execution duration, content data volume, display area, and preceding dependent tasks; A dependency graph is constructed based on the preceding dependent tasks in the task attribute parameters to determine the order of task execution. The content data volume and display area area in the task attribute parameters are used to calculate the transparency display conflict degree, and a compatibility matrix is ​​obtained. Task scheduling constraint analysis was performed by combining dependency graphs and compatibility matrices to obtain task compatibility data.

[0028] Specifically, the task instruction set is parsed in a structured manner. Based on the task control fields contained in the real-time status data, key attribute parameters associated with each display task are extracted, including task priority, estimated execution time, content data volume, display area, and the corresponding set of pre-dependent task identifiers. The task priority is represented by discrete integer values ​​to reflect the urgency of the task scheduling or the importance of the display. The execution time is determined by the time budget value set when the task is created. The content data volume is estimated based on the graphic complexity, resolution, and information carrying density of the display content. The display area is determined by calculating the size of the three-dimensional bounding box mapped to the exhibition hall spatial coordinate system. Pre-dependent tasks form a dependency chain through identifier references, which is used to constrain the task to be scheduled and executed only after certain tasks are completed. These structured task attribute parameters are uniformly extracted and stored in the task description vector set in the preprocessing module. Based on task attribute parameters, the execution order relationship between tasks is handled by introducing a directed acyclic graph (DAG) structure to construct a task dependency graph. Each node in the graph corresponds to a specific display task, and the direction of the edges represents the dependency relationship of the execution order, i.e., from a task node to its successor task node, indicating that the completion of the current task is a necessary prerequisite for the execution of subsequent tasks. By scanning the set of pre-dependent dependencies defined in the task attributes, an incoming edge set is constructed for each task, and an explicit temporal dependency path is established between tasks. After the entire graph structure is constructed, a topological sorting algorithm is used to verify whether there are cycles in the task graph to avoid logical circular dependency deadlocks. After confirming the legality of the graph structure, a set of task sequences is obtained. The transparent display compatibility between tasks is evaluated and analyzed based on two dimensions: the amount of content data and the area of ​​the display area in the task attribute parameters. The amount of content data and the content type of each task are jointly embedded in a vector space to extract the semantic feature vector of each display task. At the same time, the area of ​​the display area of ​​each display task is mapped to a three-dimensional spatial bounding box model in the exhibition hall coordinate system. A conflict evaluation function between task pairs is constructed by combining content semantic difference calculation and spatial overlap analysis. The conflict assessment function quantifies the semantic differences between two display tasks using cosine similarity calculations. Simultaneously, it calculates the spatial overlap ratio between display areas using the intersection of 3D bounding box volumes. These two quantities are then weighted and fused to output a comprehensive transparency conflict score, which fills the transparency display compatibility matrix between display tasks. Based on the task dependency graph and the transparency display compatibility matrix, a scheduling feasibility constraint dataset is formed. This dataset is input into the scheduling optimization module. The output task compatibility data includes a Boolean decision result indicating whether display tasks can be scheduled in parallel, as well as the recommended execution time interval for each display task, a list of allocable transparent display device numbers, and corresponding transparency configuration weight parameters.

[0029] In one specific embodiment, the process of performing step 200 may specifically include the following steps: A device state graph containing device nodes and spatial relationship edges is constructed based on real-time status data. Device nodes include location coordinates, movement speed and transparency information, while spatial relationship edges include distance and relative motion information between devices. The device state diagram is input into a graph attention network for spatial conflict analysis to obtain the spatial conflict probability between devices; The historical motion trajectory sequence of the device is extracted from the real-time status data, and the historical motion trajectory sequence of the device is input into the time series transformation network for trajectory prediction to obtain the deadlock risk value of task execution.

[0030] Specifically, a device state diagram is constructed to depict the motion state and spatial relationships of all mobile transparent display devices within the current exhibition space. The device state diagram is represented by a graph structure, consisting of a set of device nodes and a set of spatial relationship edges. In the device state diagram, each device node represents a specific mobile transparent display device. Each device node contains three key attributes obtained from real-time state data: the device's three-dimensional position coordinates in the exhibition space coordinate system, the device's movement speed vector, and the device's transparency value. The three-dimensional position coordinates describe the device's absolute spatial position, the movement speed describes the device's motion trend and dynamic behavior, and the transparency value reflects the current visual output state and affects the visual occlusion relationship between devices. Simultaneously, each spatial relationship edge connects two device nodes with potential interaction relationships. The spatial relationship edge contains parameters such as the Euclidean distance between the two devices, the relative motion direction angle, and the relative velocity magnitude, expressing the geometric relationships and dynamic coupling characteristics between devices in terms of spatial distribution, motion trajectory, and visual interference. The device state graph is input into a graph attention network, which consists of multi-layer graph neural network modules. Each layer employs a multi-head attention mechanism to adaptively extract key spatial interaction features between different devices. The network dynamically assigns weights to the contribution of edges in different spatial relationships by introducing a learnable adjacency weighting mechanism. This enables the model to accurately identify high-risk device combinations that may experience spatial interference, display overlap, or physical collisions in the short term, and outputs the spatial conflict probability value between each pair of devices in the form of a probability matrix, resulting in a spatial conflict probability matrix. To complete the temporal dimension analysis in deadlock prediction, the historical motion trajectory sequence of each mobile transparent display device over a past period is extracted from the real-time state data. The historical motion trajectory sequence consists of state vectors corresponding to multiple timestamps. The state vectors contain time-series data of multiple dimensions, including the device's three-dimensional position coordinates, velocity vector, acceleration, transparency, task execution progress, and display content status at each moment. The sequence data reflects the dynamic behavior evolution path of the device and the changing trends during task execution. The historical motion trajectory sequence of each device is input into a temporal transformation network built on a multi-layer encoder-decoder architecture. The temporal transformation network uses position encoding to enhance the representation of temporal information and captures the long-distance dependencies and evolutionary patterns between states in the time dimension through a multi-head self-attention mechanism. After multi-layer feature extraction and sequence modeling, the predicted trajectory of the device state is obtained. Based on the predicted trajectory, combined with the task dependency graph, the current task execution state, and the spatial evolution trend between devices, the comprehensive risk that may lead to task blocking, resource competition, transparency conflict, and spatial occlusion is calculated. The quantified task execution deadlock risk value is output to represent the probability intensity of a display task-level deadlock event occurring for a certain device or combination of devices within a given time window.

[0031] In one specific embodiment, the process of performing spatial conflict analysis on the input device state diagram into a graph attention network to obtain the spatial conflict probability between devices can specifically include the following steps: The device state graph is input into the first graph attention layer of the graph attention network for primary feature extraction to obtain the first latent feature vector; The first hidden feature vector is input into the second graph attention layer of the graph attention network for intermediate feature fusion to obtain the second hidden feature vector; The second hidden feature vector is input into the third graph attention layer of the graph attention network and subjected to high-level semantic abstraction processing to obtain the third hidden feature vector. By performing a probability mapping transformation on the third hidden feature vector, the spatial conflict probability between devices can be obtained.

[0032] Specifically, the device state graph is input into the graph attention network. The device state graph contains a set of devices with moving transparent display devices as nodes and a set of edges with spatial relationships between devices as edges. Each device node contains the spatial coordinates of its corresponding device, its motion velocity vector, its transparency setting, and a brief state code for the current display task. The edge attributes include geometric and semantic fusion features such as the physical distance between adjacent devices, the relative velocity direction angle, and the area of ​​overlapping display areas. The device state graph is then input into the first graph attention layer, which acts as the primary feature extraction module of the graph neural network. This layer performs weighted aggregation processing on the original node features within their local neighborhood. The graph attention mechanism concatenates and combines the feature vectors of each device node with those of its neighboring device nodes. A trainable attention weight function is used to calculate the weight coefficient of each neighboring node in the aggregation operation. The weights are obtained by linear mapping, LeakyReLU activation, and softmax normalization of the input feature vectors, thus modeling the importance between device nodes with spatial proximity and dynamic correlation. After processing by the first graph attention layer, the first latent feature vector of each device node is obtained, containing preliminary aggregated local spatial geometric structure information and low-level transparency motion feature codes. The first latent feature vectors of all device nodes are input into the second graph attention layer. The second graph attention layer serves as an intermediate feature fusion layer, which plays a role in completing cross-regional and multi-scale feature interactions within a wider neighborhood receptive field. This enhances the ability of device nodes to identify potential conflict patterns in more complex display scenarios. At this stage, the graph attention mechanism retains the feature aggregation relationship between each device node and its first-order neighbors and introduces a multi-head attention mechanism to compute feature weight paths in multiple different subspaces in parallel. This captures the intermediate structural relationships between devices in terms of transparency adjustment trends, speed direction differences, and spatial center of gravity shifts. The outputs of all attention heads are concatenated and integrated to generate the second latent feature vector, which represents a higher-resolution joint representation of device nodes in both structural and semantic spaces. The second latent feature vector is input into the third graph attention layer for processing. As a high-level semantic abstraction processing module of the graph neural network, the third graph attention layer performs global pattern modeling and high-order semantic integration on the potential conflict risk relationship between device nodes. By encoding second-order neighbors and cross-graph structural relationships into the node representation, it analyzes the nonlinear conflict risk caused by high overlap of display areas, convergence of movement trends, or conflict of transparency settings in complex display scenarios. The third graph attention layer uses residual connections and nonlinear activation functions to improve information propagation efficiency and improves sensitivity to large-scale structural changes through global context modeling, outputting the third latent feature vector of each device node.Numerical mapping processing of spatial conflict probability is performed on all third hidden feature vectors. The third hidden feature vector pair of each pair of device nodes is used as input, and cross-fusion calculation is performed through a shared probability mapping module. The probability mapping module is composed of a fully connected neural network. It calculates the output value of the feature vector combination of each pair of devices after nonlinear transformation, and normalizes it to the range of [0,1] through the sigmoid or softmax function. The spatial conflict probability between each pair of devices is output. The spatial conflict probability represents the likelihood that any two mobile transparent display devices will have a display conflict in the short term due to spatial proximity, intersection of movement paths, or visual overlap.

[0033] In one specific embodiment, the process of performing step 300 may specifically include the following steps: Calculate the deadlock resolution cost based on the probability of spatial conflicts between devices and the risk value of deadlock during task execution; A multi-objective optimization function is set based on deadlock resolution cost, including constraints on safe distance between devices, overlap of transparent display areas, and integrity of task time-dependent constraints. The particle swarm optimization algorithm is used to iteratively solve the multi-objective optimization function to obtain the optimal parameter combination; Based on the optimal parameter combination, a deadlock prevention scheduling scheme is generated, which includes the target movement position of the device, movement path planning, transparency adjustment value, and task execution timing arrangement. Based on the deadlock prevention scheduling scheme, multiple mobile transparent display devices are coordinated and controlled, and device movement commands and transparency coordination commands are output.

[0034] Specifically, the system uses the spatial conflict probability matrix between devices and the task execution deadlock risk value as inputs. These represent the probability level of physical conflicts that may occur between multiple mobile transparent display devices in a short period of time, and the risk intensity of the display task failing to complete due to dependency conflicts, resource competition, or transparency configuration conflicts, respectively. By setting a set of weighting coefficients, the two risk indicators are weighted and fused to construct a comprehensive deadlock risk scenario evaluation function. A deadlock resolution cost is defined, which quantifies the minimum resource adjustment and task reordering cost required for the system to eliminate potential deadlocks in the current scheduling state. The calculation method uses spatial conflict cost, transparency adjustment cost, task waiting cost, and task switching cost as the main components, and combines the current scheduling state and predicted trajectory to generate a total deadlock resolution cost function. The value of this function is used as the evaluation benchmark for the optimization objective function. A scheduling optimization model with multi-dimensional physical and logical constraints is constructed, setting three types of core constraints: inter-device safety distance constraints, transparent display area overlap constraints, and task temporal dependency integrity constraints. The safety distance constraint between devices requires that the three-dimensional spatial distance between any two devices at any given time is not less than the minimum safety distance threshold set by the system. The minimum safety distance threshold is determined based on the device size and inertial safety belt, and is used to avoid physical collisions and interference from laser projection. The overlap constraint of transparent display area is used to control the total visual transparency of devices with display tasks in the same area to not exceed a specific threshold, thereby ensuring that the audience can clearly identify the display content without visual saturation. The task temporal dependency integrity constraint requires that the logical dependency relationship of all display tasks must be preserved in the scheduling, and any task must not be executed before its predecessor task is completed. The constraint relationship is derived from the topology of the task dependency graph and remains unchanged in the scheduling. Based on the cost function and constraints, a multi-objective optimization function is constructed, which is to minimize the deadlock resolution cost function under the premise of satisfying all constraints. The optimization objectives cover multiple optimization indicators such as shortest path adjustment, minimum energy consumption, minimum waiting time, minimum transparency adjustment range, and minimum number of task switching. To efficiently solve the multi-objective optimization function, a particle swarm optimization algorithm is used as the global search optimizer. The particle swarm population is initialized, with each particle representing a complete scheduling scheme. Its position vector contains adjustable parameters such as the target movement position of all devices, transparency adjustment value, task execution time point, and path number. Control parameters such as particle swarm size, inertia weight, learning factor, and maximum number of iterations are set. By continuously updating the particle swarm's position and velocity vectors, global and local information are fused based on the current scheduling state and historical best solutions. During the iteration process, the particle swarm continuously searches for the optimal solution to the total cost function for deadlock resolution, and ultimately converges to a set of optimal parameter combinations under the premise of satisfying all constraints. The optimal parameter combination represents the optimal configuration scheme for multi-device coordinated scheduling in the current demonstration environment.Based on the optimal parameter combination output by the particle swarm optimization algorithm, a structured deadlock-prevention scheduling scheme is constructed. This scheme specifies for each mobile transparent display device the 3D spatial target position to be reached in the next stage, the recommended movement path node sequence, the target transparency adjustment value and adjustment rate, as well as the corresponding start and end times and priority markers for the display task execution. The deadlock-prevention scheduling scheme is converted into a standardized control instruction set, encoded into structured device instructions by a central scheduling platform, and sent to the controller modules of each mobile transparent display device. The instruction set includes device movement instructions and transparency coordination instructions. The device movement instructions specify the 3D movement trajectory and time window, while the transparency coordination instructions set the target transparency value and adjustment strategy for each display period. Each device controller executes these instructions in real time based on its local cooperative control law, completing target position tracking, speed adjustment, and transparency control within each control cycle.

[0035] In one specific embodiment, the process of calculating the deadlock resolution cost based on the inter-device spatial conflict probability and the task execution deadlock risk value can specifically include the following steps: Calculate the energy consumption cost of device movement based on the current position coordinates of each mobile transparent display device and the position coordinates of the target to be avoided, according to the probability of spatial conflict between devices. The waiting time of the displayed task is calculated based on the deadlock risk value of the task execution, and the waiting time cost of the task is obtained. The transparency of each mobile transparent display device is adjusted to obtain the cost of transparency adjustment. The deadlock resolution cost is obtained by weighting and summing the energy consumption cost of device movement, the task waiting time cost, the transparency adjustment cost, and the task switching penalty cost.

[0036] Specifically, based on the spatial conflict probability matrix between devices output by the graph attention network, the system identifies potential pairs of devices with spatial conflict relationships within the current scheduling cycle. It then calculates the three-dimensional distance between the current position coordinates of each mobile transparent display device and the recommended avoidance target position coordinates. Combining this with the unit movement energy consumption coefficient of each device, such as the electricity consumed per unit movement distance, driving power, or energy index, and introducing a path impedance function to account for path complexity factors such as obstacle avoidance and turning, the system calculates the total energy consumption value required for the device to migrate from its current position to the avoidance target position, i.e., the device movement energy consumption cost, reflecting the movement cost incurred to avoid potential conflicts under the current spatial scheduling state. Meanwhile, considering the impact of deadlock risk at the task level, the deadlock risk value of task execution generated by the time-series transformation network is used as the input basis for task scheduling delay. Based on the number of pre-dependencies of each task in the task dependency graph, the current task's scheduling queuing status, and the execution availability of the allocated devices, the cumulative waiting time experienced by the current task from being requested to being allowed to execute is dynamically calculated. A task priority weight coefficient is introduced to modulate the waiting time, so that the waiting time cost of high-priority tasks has a higher weight in the overall cost function, forming the task waiting time cost, which is used to reflect the scheduling delay cost caused by task logic conflicts, ensuring that the scheduling strategy prioritizes the continuity and time sensitivity requirements of the task dependency chain. At the transparency adjustment level, based on the spatial overlap and visual conflict relationships between display tasks, and combining the current transparency values ​​of each mobile transparent display device with the recommended transparency target value, the required transparency change range for each device is calculated. The semantic importance score of the displayed content is considered as an adjustment weight, and the transparency change is multiplied by the visual priority factor of the corresponding task to form the transparency adjustment cost. This cost is used to quantify the dynamic adjustment cost of display parameters made by the device to achieve visual coordination, and reflects the cost incurred by the system in maintaining visual recognizability and display consistency in the transparent display space. Considering task migration, device reallocation, or task reordering after interruption caused by deadlock avoidance during scheduling, the number of task switching times for each device in the current scheduling cycle is counted, and a fixed task switching penalty coefficient is introduced as a cost factor to model the cost penalty of the device's task switching behavior. The penalty term aims to constrain the scheduling optimization algorithm to avoid frequent changes in task allocation strategies when pursuing spatial avoidance and visual coordination, thereby improving system stability while reducing task loading overhead and operation latency caused by content switching.The energy consumption cost of device movement, the waiting time cost of tasks, the cost of transparency adjustment, and the cost of task switching penalties are each assigned a weight coefficient, and a unified deadlock resolution cost function is constructed by weighted summation. The deadlock resolution cost function serves as an evaluation index for minimizing the objective function during the optimization phase. By dynamically adjusting the target position of devices, transparency settings, and task execution order, the overall deadlock resolution cost is minimized as the criterion. This achieves a fusion scheduling decision that integrates multiple objectives, such as conflict avoidance between devices, guarantee of task logic integrity, visual output coordination, and task switching efficiency control, effectively preventing deadlock from occurring during system operation.

[0037] In one specific embodiment, the process of coordinating the control of multiple mobile transparent display devices according to the deadlock prevention scheduling scheme and outputting device movement commands and transparency coordination commands may specifically include the following steps: The device control parameters, including target location coordinates, expected arrival time, and transparency settings, are extracted from the deadlock prevention scheduling scheme. Based on the device control parameters, a distributed collaborative control law is calculated for each mobile transparent display device to obtain the device control output vector. When the overlapping area of ​​the display area of ​​the mobile transparent display device exceeds a preset threshold, the transparency negotiation protocol is initiated to perform a balanced solution and obtain the negotiated transparency parameters. Based on the device control output vector, generate device movement instructions containing movement direction, movement speed, and arrival time, and generate transparency coordination instructions based on negotiated transparency parameters.

[0038] Specifically, the control parameter set for each device is extracted from the anti-deadlock scheduling scheme. This scheme originates from the global optimal solution of the particle swarm optimization algorithm, internally encoding the execution strategy that each mobile transparent display device should follow in the current scheduling cycle. Therefore, the anti-deadlock scheduling scheme is structurally analyzed, extracting three core control parameters for each mobile transparent display device: target location coordinates, expected arrival time, and transparency setting value. The target location coordinates indicate the three-dimensional spatial coordinate point that the device must ultimately reach; the expected arrival time is the time constraint boundary of the target location; and the transparency setting value reflects the display output intensity that the device should maintain after reaching the designated display area. The transparency setting value is determined comprehensively based on the visibility requirements of the displayed content and the display task status of adjacent devices, ensuring the visual clarity, layering, and information recognition of the overall display layout. Based on the device control parameters, a distributed collaborative control law calculation is performed for each mobile transparent display device. The distributed collaborative control law module executes the solution process for the control output vector based on the current real-time state of each device and its corresponding control parameters. The displacement vector between the current position and the target position of the device is calculated, and the required average moving speed and rate of change of speed are determined by combining the target arrival time. Simultaneously, the transparency setpoint is compared with the current transparency value to calculate the dynamic adjustment coefficient for transparency changes. The entire control law consists of a proportional-derivative structure, where the position error term serves as the dominant component of the control input, working in conjunction with the speed and transparency error terms to affect the device's motor control, attitude adjustment, and screen display modules. The calculated output of the control law is the device control output vector, including elements such as the target movement direction, the desired speed curve, and the transparency adjustment rate. As the device status is continuously updated and the movement path advances, multiple devices may simultaneously gather in adjacent areas and perform overlapping display tasks during the execution of the display task. In this case, the display areas form a geometric intersection in the exhibition hall's spatial coordinate system. When the overlapping area exceeds the set transparency conflict threshold, a transparency negotiation protocol is automatically triggered to achieve visual output balance between devices in order to avoid visual occlusion, overlapping display content, or brightness overload. The transparency negotiation protocol is based on the game equilibrium mechanism. Each pair of overlapping display devices exchanges the current transparency value, target value, and importance coefficient of the displayed content as input information. In the process of multiple rounds of iteration for optimal response, each device solves the maximum visual utility function that it can obtain under the transparency setting conditions of the other until the two converge to a Nash equilibrium transparency combination. This combination is the negotiated transparency parameter, which represents the most suitable transparency setting strategy for each device under the current display overlap conditions.The device control output vector and negotiated transparency parameters are structured, encapsulated, and converted into a standard instruction set for the device actuator. The displacement direction information, velocity curve, and target time point in the device control output vector together constitute the device movement instruction. The device movement instruction consists of a three-dimensional movement direction vector, velocity magnitude, and time label, which is sent to the device's path control module to drive it to complete spatial movement. The negotiated transparency parameters are used as input to generate transparency coordination instructions. The transparency coordination instructions include a transparency target value, adjustment rate, and negotiation reference weight, which are used to drive the device screen display module to dynamically adjust the visual output state, ensuring that the transparency output intensity and readability in overlapping areas meet the system's set visual priority logic.

[0039] In one specific embodiment, the process of performing distributed cooperative control law calculations on each mobile transparent display device based on device control parameters to obtain the device control output vector can specifically include the following steps: The position error of each mobile transparent display device is calculated based on the target position coordinates in the device control parameters to obtain the position control component. The repulsive force field is calculated based on the real-time distance relationship between each mobile transparent display device to obtain the repulsive force field control component. The transparency gradient force is calculated based on the current transparency value of each mobile transparent display device and the importance weight of the displayed content to obtain the transparency control component; The position control component, repulsion field control component, and transparency control component are weighted and synthesized using a proportional-differential control law to obtain the device control output vector.

[0040] Specifically, based on the target position coordinates, transparency setpoint, and importance indicators of the current task content in the device control parameters, and combined with the real-time spatial distribution relationship and transparency status between devices, multi-source control components are constructed and dynamically fused through proportional-derivative control laws to ultimately form a highly responsive device control output vector. First, the system calculates the real-time position error based on the difference between the current three-dimensional spatial position of each mobile transparent display device and the target position coordinates explicitly specified in the device control parameters. The position error vector consists of three-dimensional coordinate differences, specifically calculating the offset of the current device from the target point in the x, y, and z axes. This position error vector not only reflects the degree of deviation of the current device but also provides necessary basic data for subsequent path vector planning and speed control. Based on this, the system combines the current speed status of the equipment, arrival time requirements, and task execution time window. In the proportional controller, the current position error is used as the input variable, and the error is amplified through a set position proportional gain coefficient to form a primary position control component characterizing the desired movement direction and corrected speed of the equipment. Simultaneously, a differential term is introduced, using the time derivative of the current position error to reflect its rate of change. The differential gain adjusts the response speed of the equipment's movement, thereby improving the system's dynamic stability and path tracking accuracy. While calculating the position control component, the system also implements mutual constraints and conflict avoidance for the operational behavior of multiple devices within a confined space, introducing a repulsive field calculation mechanism based on real-time spatial distance relationships. In each control cycle, spatial coordinate data between all devices is acquired, and a distance matrix between devices is constructed in real time. Each element in the matrix represents the Euclidean distance between two devices. When the spatial distance between any two devices is less than a set safety threshold, a dynamic intervention mechanism based on repulsive field theory is automatically triggered. A repulsive vector is calculated for the two devices, with the direction of the repulsive vector being the unit direction vector between the two devices. Its magnitude is modeled using an inverse square function, meaning the closer the distance, the greater the repulsive force; when the distance is greater than the safety distance, the repulsive force is zero. The repulsive vectors generated by all devices spatially close to the current device are summed and synthesized to form the total repulsive field control component experienced by the current device. This component is used to dynamically adjust the device's movement direction and speed, preventing physical collisions or occlusions caused by path overlap or insufficient navigation accuracy in space. To achieve adaptive dynamic control of the transparency adjustment strategy, the transparency gradient force is calculated based on the device's current transparency value and the importance weighting factor of the content of the current display task.At any given moment, the current transparency status of the device is extracted, and the visual presentation importance score of the current task is obtained from the task scheduling parameters. This score represents the degree to which the task content is perceived and recognized by the audience in the exhibition hall scene. The gradient direction is calculated based on the difference between the transparency control target value and the current transparency value. The gradient direction is used to guide the upward or downward trend of transparency, while the rate of change of transparency is adjusted by the weighted factor of the content importance factor. That is, the more critical the task content is, the stronger the influence of its corresponding transparency change, forming a transparency control component. The position control component, repulsion field control component, and transparency control component are input into a proportional-differential control law for weighted and synthesized calculations. In the control law, the position control component is given the main weight for basic path tracking to ensure that the device can reach the display position on time, while the repulsion field control component is used to dynamically counteract the tendency of adjacent devices to approach each other and maintain a safe distance. The transparency control component is used to compensate and adjust in the visual interference zone. All components are adjusted in the control law with gain coefficients to adjust the intensity of their influence on the final control result, and the final device control output vector is output through vector weighting. This vector contains the three-dimensional velocity direction, path offset correction amount, and transparency adjustment rate that the device should execute at the current moment, which is used to drive the actuator to complete the path adjustment and visual parameter adjustment operations.

[0041] The multi-device collaborative scheduling processing method for the mobile transparent display device in the embodiments of the present invention has been described above. The multi-device collaborative scheduling processing system for the mobile transparent display device in the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 2 One embodiment of the multi-device collaborative scheduling and processing system for mobile transparent display devices in this invention includes: The data acquisition module 11 is used to collect real-time status data of multiple mobile transparent display devices in the exhibition hall and calculate task compatibility data with the display task. The deadlock risk prediction module 12 is used to predict deadlock risk based on real-time status data and task compatibility data, and obtain the probability of spatial conflict between devices and the deadlock risk value of task execution. The collaborative control module 13 is used to collaboratively control multiple mobile transparent display devices based on the probability of spatial conflict between devices and the risk value of task execution deadlock, and output device movement commands and transparency coordination commands.

[0042] Through the collaborative efforts of the aforementioned components, this invention reduces the probability of deadlock compared to traditional passive detection methods, ensuring the continuous and stable operation of multiple mobile transparent display devices in the exhibition hall. Based on the transparent display task compatibility matrix, it accurately assesses the visual compatibility of different display content in overlapping areas, effectively solving the content conflict problem caused by traditional scheduling methods ignoring transparent display characteristics. By simultaneously optimizing four objective functions—movement cost, waiting time, transparency adjustment, and task switching—it achieves the optimal balance between deadlock avoidance and performance maintenance, avoiding local optima problems caused by single-objective optimization. Employing a hierarchical distributed control structure combined with a transparency negotiation protocol, it possesses stronger robustness and real-time response capabilities. Compared to centralized control methods, it better handles device failures and network latency. When display areas overlap, the transparency negotiation protocol automatically coordinates the transparency settings between devices based on the Nash equilibrium principle of game theory, ensuring optimal visual display effects while avoiding energy waste.

[0043] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0044] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause an electronic device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0045] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for multi-device cooperative scheduling of moving transparent display devices, characterized in that, The method comprises the following steps: Collecting real-time state data of a plurality of mobile transparent display devices in an exhibition hall and calculating task compatibility data between the display tasks; Based on the real-time state data and the task compatibility data, a deadlock risk prediction is performed to obtain a spatial conflict probability between devices and a task execution deadlock risk value; Based on the spatial conflict probability between devices and the task execution deadlock risk value, the plurality of mobile transparent display devices are cooperatively controlled to output device movement instructions and transparency coordination instructions. 2.The method of claim 1, wherein, The method comprises the following steps: Dividing the exhibition hall into three-dimensional grid units to construct an exhibition hall space coordinate system; Based on the exhibition hall space coordinate system, motion parameter collection is performed on the plurality of mobile transparent display devices to obtain motion state parameters; Transparent display function parameter collection is performed on the plurality of mobile transparent display devices to obtain function state parameters; The motion state parameters and the function state parameters are taken as real-time state data; According to the real-time state data, transparent display compatibility analysis is performed on the display tasks to obtain task compatibility data. 3.The method of claim 2, wherein, The method comprises the following steps: Task attribute parameters including task priority, execution time, content data volume, display area, and pre-dependent tasks are parsed from the real-time state data; A dependency graph of task execution sequence is constructed based on the pre-dependent tasks in the task attribute parameters; Transparent display conflict degree calculation is performed on the content data volume and the display area in the task attribute parameters to obtain a compatibility matrix; Task scheduling constraint analysis is performed in combination with the dependency graph and the compatibility matrix to obtain task compatibility data. 4.The method of claim 1, wherein, The method comprises the following steps: Based on the real-time state data, a device state graph containing device nodes and spatial relationship edges is constructed, wherein the device nodes contain position coordinates, movement speed, and transparency information, and the spatial relationship edges contain device-to-device distance and relative motion information; The device state graph is input into a graph attention network for spatial conflict analysis to obtain a spatial conflict probability between devices; Device historical motion trajectory sequences are extracted from the real-time state data, and the device historical motion trajectory sequences are input into a time series transformation network for trajectory prediction to obtain a task execution deadlock risk value. 5.The method of claim 4, wherein, The method comprises the following steps: The device state graph is input into a first graph attention layer of the graph attention network for primary feature extraction to obtain a first hidden feature vector; The first hidden feature vector is input into a second graph attention layer of the graph attention network for intermediate feature fusion to obtain a second hidden feature vector; The second hidden feature vector is input into a third graph attention layer of the graph attention network for high-level semantic abstraction processing to obtain a third hidden feature vector; performing probability mapping transformation on the third implicit feature vector to obtain an inter-device spatial conflict probability. 6.The method of claim 1, wherein, The inter-device spatial conflict probability and the task execution deadlock risk value are used to cooperatively control the plurality of mobile transparent display devices, and device movement instructions and transparency coordination instructions are output, including: calculating a deadlock resolution cost according to the inter-device spatial conflict probability and the task execution deadlock risk value; setting a multi-objective optimization function including an inter-device safety distance constraint, a transparency display area overlap constraint, and a task timing dependency integrity constraint based on the deadlock resolution cost; iteratively solving the multi-objective optimization function using a particle swarm optimization algorithm to obtain an optimal parameter combination; generating a deadlock prevention scheduling scheme including device target movement positions, movement path plans, transparency adjustment values, and task execution timing arrangements based on the optimal parameter combination; cooperatively controlling the plurality of mobile transparent display devices according to the deadlock prevention scheduling scheme, and outputting device movement instructions and transparency coordination instructions.

7. The method of claim 6, wherein, The deadlock resolution cost is calculated according to the inter-device spatial conflict probability and the task execution deadlock risk value, including: calculating a device movement energy consumption cost based on the inter-device spatial conflict probability for the current position coordinates and the avoidance target position coordinates of each mobile transparent display device; calculating a task waiting time based on the task execution deadlock risk value to obtain a task waiting time cost; adjusting the transparency of each mobile transparent display device to obtain a transparency adjustment cost; weighting and summing the device movement energy consumption cost, the task waiting time cost, the transparency adjustment cost, and a task switching penalty cost to obtain the deadlock resolution cost. 8.The method of claim 7, wherein, The plurality of mobile transparent display devices are cooperatively controlled according to the deadlock prevention scheduling scheme, and device movement instructions and transparency coordination instructions are output, including: parsing device control parameters including target position coordinates, expected arrival times, and transparency setting values from the deadlock prevention scheduling scheme; performing distributed cooperative control law calculation on each mobile transparent display device based on the device control parameters to obtain a device control output vector; when the overlap area of the display regions of the mobile transparent display devices exceeds a preset threshold, starting a transparency negotiation protocol to obtain negotiated transparency parameters; generating device movement instructions including movement direction, movement speed, and arrival time based on the device control output vector, and generating transparency coordination instructions based on the negotiated transparency parameters. 9.The method of claim 8, wherein, The device control output vector is obtained by performing distributed cooperative control law calculation on each mobile transparent display device based on the device control parameters, including: performing position error calculation on each mobile transparent display device according to the target position coordinates in the device control parameters to obtain a position control component; performing repulsive field calculation based on the real-time distance relationship between each mobile transparent display device to obtain a repulsive field control component; performing transparency gradient force calculation based on the current transparency value of each mobile transparent display device and the importance weight of the display content to obtain a transparency control component; The position control component, the repulsive field control component and the transparency control component are proportionally and differentially controlled and weighted to obtain a device control output vector.

10. A multi-device cooperative scheduling processing system of moving transparent display devices, characterized in that, The method for performing the multi-device cooperative scheduling process of the mobile transparent display device as claimed in any one of claims 1-9 comprises: A collection module is configured to collect real-time state data of a plurality of mobile transparent display devices in an exhibition hall and calculate task compatibility data between the real-time state data and a display task; A deadlock risk prediction module is configured to predict a deadlock risk based on the real-time state data and the task compatibility data to obtain a spatial conflict probability between devices and a task execution deadlock risk value; A cooperative control module is configured to cooperatively control the plurality of mobile transparent display devices based on the spatial conflict probability between devices and the task execution deadlock risk value, and output a device movement instruction and a transparency coordination instruction.

Images