Multi-scene photoelectric display cooperative control method and system

Abstract

The invention provides a multi-scene photoelectric display cooperative control method and system, and the method comprises the steps: obtaining the state data of photoelectric equipment, and constructing an equipment topology state diagram according to the state data of the photoelectric equipment; based on the equipment topology state diagram, an iterative greedy search strategy is adopted to generate a fault recovery plan, then fault recovery is carried out on the photoelectric equipment according to the fault recovery plan, and a new equipment topology state diagram is generated; and obtaining a user instruction, deconstructing the user instruction into a target constraint set, generating an optimal conflict resolution strategy by using a graph search algorithm based on the target constraint set and the new equipment topology state graph, and controlling the optoelectronic equipment according to the optimal conflict resolution strategy. The method solves the problems that in the prior art, a control instruction depends on static configuration, the instruction cannot be executed easily due to various configuration conflicts, and the configuration conflicts cannot be solved automatically.

Description

Technical Field

[0001] This invention relates to the field of intelligent display control technology, and in particular to a multi-scene optoelectronic display collaborative control method and system. Background Technology

[0002] Most existing multi-scenario optoelectronic display systems (such as smart conference rooms and exhibition centers) adopt an integrated mode of "central control system + device drive". Their typical workflow is: a fixed set of device control instructions ("macro commands") is pre-configured for each scenario and executed sequentially when switching.

[0003] The system architecture lacks a unified digital description of device capabilities, connection topology, and real-time status. The formulation of control commands relies on the experience of integrators and static configurations, and cannot be verified and adjusted according to the actual system status at runtime. It is prone to switching failures and display abnormalities due to configuration conflicts such as device capability mismatch (e.g., resolution exceeding limits), resource conflicts (e.g., signal port preemption), and fault conflicts. Furthermore, it cannot resolve configuration conflicts on its own and cannot adapt to changing business needs. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a multi-scenario optoelectronic display collaborative control method, which solves the problems in existing technologies where control commands rely on static configuration, are prone to failure to execute commands due to various configuration conflicts, and cannot resolve configuration conflicts on their own.

[0005] According to an embodiment of the present invention, a multi-scene optoelectronic display collaborative control method includes: Acquire the status data of optoelectronic devices and construct a device topology status diagram based on the status data; Based on the device topology state diagram, an iterative greedy search strategy is used to generate a fault recovery plan. Then, the optoelectronic device is repaired according to the fault recovery plan, and a new device topology state diagram is generated. The system acquires user commands, deconstructs them into a set of target constraints, and uses a graph search algorithm to generate an optimal conflict resolution strategy based on the target constraint set and a new device topology state diagram. The system then controls the optoelectronic device according to the optimal conflict resolution strategy.

[0006] Preferably, the method for generating a fault recovery plan based on an iterative greedy search strategy using a device topology state diagram includes: S1: Determine the fault node and the root cause node based on the equipment topology state diagram, calculate the initial impact score of the fault node, and generate a candidate action set based on the root cause node and the equipment topology state diagram. The candidate action set includes multiple recovery actions. S2: Initialize the device topology state diagram and create a blank fault recovery action sequence; S3: Randomly select a recovery action from the candidate action set, adjust the device topology state diagram according to the recovery action, and then recalculate the global impact score of the faulty node; S4: If the global impact score of this iteration is lower than the global impact score or the initial impact score of the previous iteration, then add the recovery action to the fault recovery action sequence; otherwise, discard the action and repeat steps S3-S4 until all recovery actions are taken. S5: Repeat steps S2-S4 to obtain multiple different fault recovery action sequences. Then calculate the global recovery degree of each fault recovery action sequence and select the fault recovery action sequence with the highest global recovery degree as the fault recovery plan.

[0007] Preferably, the method for determining the root cause node of a fault based on the equipment topology state diagram includes: A fault propagation graph is generated based on the device topology state diagram. The fault propagation graph includes multiple normal nodes, faulty nodes, and multiple fault-affected edges. Based on the fault impact edges, potential root cause nodes that will affect the faulty nodes are selected from the normal nodes, and then the impact probability of each potential root cause node is calculated based on the fault impact table. The node with the highest probability of impact is selected as the root cause node of the failure.

[0008] Preferably, if the faulty node cannot be determined based on the equipment topology status diagram, no fault recovery plan will be generated.

[0009] Preferably, the method for generating the optimal conflict resolution strategy using a graph search algorithm based on the target constraint set and the new device topology state diagram includes: A1: Construct a node state space and an action space based on the device topology state diagram. The node state space includes multiple nodes, and the action space includes multiple transition actions. Construct a corresponding local cost for each transition action. A2: Construct the evaluation function, action set, and global cost function; A3: Use the A* algorithm to extract a transition action from the action space that minimizes the evaluation function, add the extracted transition action to the action sequence, and generate the current node's transition state graph; The evaluation function is: in, UC(n) is the sum of the local costs of the transition actions in the action set from the initial state to the current state n; UC(n) is the number of constraints in the current node's transition state graph that do not satisfy the target constraint set; and NAC(n) is the number of nodes in the current node's transition state graph that need to be adjusted. A4: Repeat step A3 until the current node's transition state diagram satisfies the objective constraint set and the global cost function is minimized; A5: The action set obtained from the last iteration is used as the optimal conflict resolution strategy.

[0010] Preferably, in step A2, if multiple transition actions are equivalent, the transition action that minimizes the evaluation function is retained, and other equivalent transition actions are removed from the action space.

[0011] On the other hand, according to embodiments of the present invention, a multi-scene optoelectronic display collaborative control system is also provided. This system uses the above-described multi-scene optoelectronic display collaborative control method, including: The state perception module is used to acquire the state data of the optoelectronic device and user instructions, and to construct the device topology state diagram based on the optoelectronic device state data and deconstruct the user instructions into a target constraint set. The fault recovery module is used to generate a fault recovery plan based on the device topology state diagram and an iterative greedy search strategy, and then repair the optoelectronic device according to the fault recovery plan. The decision module is used to generate the optimal conflict resolution strategy based on the target constraint set and the new device topology state diagram using a graph search algorithm. A control module is provided for controlling the optoelectronic device according to an optimal conflict resolution strategy.

[0012] Compared with the prior art, the present invention has the following beneficial effects: This invention transforms the status data of optoelectronic devices into a device topology state diagram, then performs self-checks and identifies root cause nodes of faults. Based on the root cause nodes, it generates corresponding fault recovery plans to automatically repair the optoelectronic devices and generate a new device topology state diagram. Subsequently, it deconstructs user instructions to generate a target constraint set, finds conflict points by comparing the target constraint set with the device topology state diagram, and uses a graph search algorithm to generate an optimal conflict resolution strategy. Thus, based on the current state of the optoelectronic devices, it automatically and dynamically generates a strategy that can resolve the current conflict and correctly executes user instructions. Attached Figure Description

[0013] Figure 1 This is a diagram of a multi-scene optoelectronic display collaborative control method according to an embodiment of the present invention. Detailed Implementation

[0014] The technical solutions of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0015] like Figure 1As shown, this embodiment of the invention proposes a multi-scene optoelectronic display collaborative control method, including: Acquire the status data of optoelectronic devices and construct a device topology status diagram based on the status data; To address the "state awareness" problem in the dynamic collaboration of heterogeneous devices in optoelectronic display systems, this invention constructs and maintains a device topology state diagram (which is a weighted directed graph): G(t)=(V,E,A(t)), where t represents time, and the model is dynamic.

[0016] (1) Node V includes two types of entities: Physical entities: such as display device D_i, signal source S_j, matrix switch port P_k, image processor, etc.

[0017] Logical entities include "display wall" (Wall_m, which is aggregated from multiple D_i through specific splicing logic) and "signal path" (Path_n, which describes the complete signal link from S_j through P_k to D_i). Logical entities are abstractions at the business level, simplifying high-level decision-making.

[0018] (2) Edge E describes the structural and logical relationships between entities, mainly including: Connection edges: S_j→P_k (signal source connected to matrix port), P_k→D_i (matrix port output to display). These indicate the physical or logical direction of signal flow.

[0019] Dependency edge: D_i→Wall_m (the physical display belongs to a certain logical display wall), used for aggregation calculation and state inference at the business level.

[0020] Failure-affected edges: P_k.Failure -- [probability 0.9] --> Di_i.Signal_Loss, Di_i.Failure -- [impact weight 0.7] --> Wall_m.Function_Degrade. The probability or weight of the edge can be obtained through statistical analysis and learning from historical failure data. (3) Attribute A(t), each node v has a set of attributes A_v(t) that change over time: Static capability attributes include Res_max (maximum resolution), Ifc_set (set of supported interfaces), and Power_Model (power consumption model). These attributes are typically defined when the device is added to the library.

[0021] Dynamic state attributes: such as Res_cur(t) (current resolution), PowerState(t) (power state), Health(t)∈[0,1] (health status, calculated based on temperature, error logs, runtime, etc.).

[0022] Constraint attributes: Define the conditions that must be met for effective interaction between entities using mathematical expressions or logical rules. For example, for a valid "signal channel" logical entity, its core constraint is: S_j.OutputFormat∈P_k.SupportedFormats∩D_i.SupportedFormats (the signal format must be supported at all stages of the path).

[0023] Based on the device topology state diagram, an iterative greedy search strategy is used to generate a fault recovery plan. Then, the optoelectronic device is repaired according to the fault recovery plan, and a new device topology state diagram is generated. Since each node corresponds to a photoelectric device, the status of each node in the device topology state diagram can be used to determine whether the corresponding photoelectric device has experienced an abnormal event. If all photoelectric devices are normal, this step is skipped and no fault recovery plan is generated. If there is a node with an abnormal event, such as D_i black screen, a fault propagation graph is generated after removing the connecting edges and dependent edges according to the device topology state diagram. At this time, the node with the abnormal event in the fault propagation graph is defined as the fault node, and other nodes are defined as normal nodes.

[0024] Starting from the fault node, backtrack along the fault impact edge to filter out potential root cause nodes that will affect the fault node (i.e., all upstream nodes of the fault node). Then, combine the real-time health status Health(t) and historical fault probability in the fault impact table of each potential root cause node to calculate the impact probability P(R_c|E) of each potential root cause node R_c leading to the abnormal event E. Then, the potential root cause node with the highest impact probability is taken as the fault root cause node R.

[0025] Based on the equipment topology diagram, the impact of any node I affected by the root cause node is calculated using the following formula, where each factor is automatically obtained from the system model or real-time data: Wherein, F_dec is the functional availability degradation value. If the faulty node object O is completely unavailable (such as a black screen), the value is 1; if it is partially available (such as a screen in a display wall failing), the value is (number of failed sub-components / total number of sub-components).

[0026] P_dec represents the performance degradation percentage, calculated as (performance baseline value - current or predicted performance value) / performance baseline value. For example, for a display wall, the performance baseline is the preset target resolution; the current / predicted value may be degraded to the actual achievable resolution due to a fault (calculated from the capability attributes of devices on the path in the device topology status diagram).

[0027] T_int is the predicted service interruption time, which can be predicted based on the repair time records of similar historical faults (MTTR database) or by summing the estimated execution time (EDL) of the recovery plan.

[0028] λ is the scenario sensitivity coefficient, which is dynamically adjusted according to the current business scenario (such as emergency command, conference presentation, exhibition playback). A larger value is taken for emergency scenarios to amplify the impact of the combined degradation of function and performance. This formula uses a nonlinear combination of product and exponential decay, avoiding the subjective parameter tuning problem of linear weighting, and can more accurately reflect the coupling effect between failure factors: when function and performance are severely degraded at the same time, the impact increases exponentially, which is consistent with the sensitivity of display business to multiple failures.

[0029] The values ​​of each parameter in the above formula are obtained through real-time calculation or prediction of the device attributes stored in the device topology state diagram: Functional availability degradation value F_dec: If the faulty node object is a single display and is completely unavailable, then F_dec=1; if some sub-displays in the logical display wall fail, then F_dec=(number of failed sub-displays) / (total number of sub-displays in the display wall).

[0030] Performance degradation percentage P_dec: P_dec = (Performance baseline value - Current or predicted performance value) / Performance baseline value. The performance baseline value is determined based on the target resolution in the user command or a system-preset scenario template; the current or predicted performance value is obtained by querying the maximum resolution, supported interface set, and other static capability attributes of all devices along the signal path in the device topology status diagram, and then taking the intersection to obtain the actual achievable resolution.

[0031] Business interruption time prediction value T_int: When the fault is a hardware failure, T_int is obtained from the mean time to repair (MTTR) database of similar historical faults; when the fault can be resolved by signal reconstruction, T_int is obtained by accumulating the estimated execution time (EDL) of the recovery action.

[0032] Scene sensitivity coefficient λ: Dynamically set according to the current display business scene type. λ=2.0 for emergency command scenes, λ=1.0 for conference presentation scenes, and λ=0.5 for exhibition playback scenes. The λ value is pre-stored in the scene configuration table and read by the controller at runtime.

[0033] If the failure at node O is caused by the propagation of the fault influence edge e(R→O) from the upstream root cause node, then its final influence degree needs to be multiplied by the propagation weight w(e) of that edge: Ultimately, the initial impact score / global impact score of the faulty node O is the sum of the weighted impact scores of all affected nodes I that are relevant to the current business: .

[0034] The system takes the diagnosed root cause node R and the current device topology state diagram as input, and automatically enumerates the feasible candidate action set {Action_i}. The enumeration rules are based on the device operation knowledge base and SDRG constraints. For device failures: Actions include rebooting, isolating, and switching to a backup device. Backup devices are automatically discovered by querying the SDRG to find entities of the same device type, with good health, and currently idle.

[0035] For connectivity / path failures: Actions include signal path reconstruction (ReroutePath). Based on the SDRG topology, the system automatically identifies other reachable paths from the source to the destination (which must meet constraints such as format and bandwidth), and each path is a candidate action.

[0036] For performance degradation: Actions include degrading display parameters (DegradeParams), such as reducing resolution or refresh rate. The target degradation value is automatically determined by querying the SDRG's list of parameters supported by the affected device that are below the current value.

[0037] It should be noted that the above fault recovery strategy is not a simple application of a general greedy algorithm. The "global impact score" in this invention is an evaluation function specifically designed for optoelectronic display systems, which fully considers the business characteristics of the display device. First, the determination of the functional availability degradation value F_dec is deeply tied to the aggregation logic of the display device. For a logical display wall composed of multiple physical displays, this value is dynamically calculated based on the ratio of the number of failed sub-components to the total number of sub-components, rather than simply judging it as "faulty" or "normal". This allows the recovery strategy to accurately distinguish between different scenarios of partial functional degradation and complete failure.

[0038] Second, the assessment of the performance degradation rate P_dec is related to the actual capability attributes of the display device. This value is calculated based on static capability attributes such as the maximum resolution and supported interface set stored in each node of the device topology state diagram, as well as dynamic status attributes such as the current resolution and health status. It can accurately reflect the true degree of damage to display services (such as resolution requirements and signal format matching).

[0039] Third, the introduction of the business interruption time prediction value T_int ensures that the selection of recovery actions not only considers the impact elimination effect but also takes into account the recovery efficiency, thus avoiding long-term business interruptions in actual operations due to the pursuit of the theoretical optimal solution.

[0040] In summary, the global impact scoring of this invention quantifies the domain knowledge of optoelectronic display systems (display wall aggregation logic, equipment capability constraints, and service interruption tolerance) into a computable objective function, enabling the iterative greedy search strategy to generate recovery plans that meet the actual needs of display services, rather than a rigid transplantation of general algorithms into the field of equipment failure.

[0041] Then, an iterative greedy search strategy is used to automatically generate candidate solutions P_i: (1) Initialization: Create a blank fault recovery action sequence P_i.

[0042] (2) Iterative generation: A recovery action is randomly selected from the candidate action set {Action_i} without replacement, and the device topology state diagram is adjusted according to the recovery action. Then, the global impact score of the faulty node O is recalculated. If the global impact score of this iteration is lower than the global impact score or the initial impact score of the previous iteration, then the recovery action is added to the fault recovery action sequence; otherwise, the action is discarded, and the above steps are repeated until all recovery actions are taken.

[0043] At the same time, each recovery action is simulated and executed on the device topology state diagram to check whether it violates physical or safety constraints. If it does, the action is discarded and the next one is tried.

[0044] (3) Output: After all recovery actions are retrieved and a fault recovery action sequence is generated, the above iterative generation steps are executed again. In this execution, the first recovery action retrieved is different from the previous execution. Thus, after executing the iterative generation steps multiple times, multiple feasible fault recovery action sequences that can effectively reduce the impact can be generated. Each sequence is a candidate recovery plan P_i.

[0045] For each automatically generated contingency plan P_i, its global recovery degree is calculated using a comprehensive benefit function as follows: Here, R_ratio represents the recoverable impact ratio (normalized to [0,1]), C_exec represents the execution cost (the smaller the normalized value, the better), T_recover represents the recovery time (normalized), and R_risk represents the risk score for introducing new faults (normalized). This function places the recovery effect, cost, time, and risk within the same non-linear framework: the better the recovery effect and the lower the execution cost, the higher the recovery degree; the longer the recovery time or the higher the risk of new faults, the faster the recovery degree decreases. Through the construction of multiplication and denominator, a balance between multiple objectives is automatically achieved, avoiding the problem that the weight coefficients in linear weighting are difficult to adapt to different fault types. At the same time, the function's range is [0,1], which facilitates direct comparison between different contingency plans.

[0046] Then, the fault recovery action sequence with the highest global recoverability is selected as the fault recovery plan, and the optoelectronic equipment is repaired. After the repair is completed, a new equipment topology state diagram is generated.

[0047] The system acquires user commands, deconstructs them into a set of target constraints, and uses a graph search algorithm to generate an optimal conflict resolution strategy based on the target constraint set and a new device topology state diagram. The system then controls the optoelectronic device according to the optimal conflict resolution strategy.

[0048] After obtaining the user's instructions, they need to be deconstructed into instructions that the computer can recognize. Taking "start 4K exhibition mode" as an example, it can be deconstructed into the following set of target constraints: C1:Wall_exhibition.Res_cur=3840x2160 C2:S_content.Path exists&S_content.Path.endpoint=Wall_exhibition C3:All D in Wall_exhibition.PowerState=ON.

[0049] The device topology state diagram is compared with the target constraint set. If there is a hard conflict that directly violates the physical or capability constraints (such as the resolution required by C1 being greater than the Res_max of a certain display in Wall_exhibition), then it is directly reported as infeasible or a degradation negotiation is initiated.

[0050] If there is no hard conflict, then construct a global cost function: Where Resource_Reconfig_Cost is the cost of resource reconfiguration, State_Perturbation_Cost is the cost of state perturbation, Estimated_Duration is the estimated duration, and λ is the time penalty coefficient.

[0051] For complex conflicts that rules cannot cover or resolve at excessive cost, this invention transforms the state transition problem into a graph search problem. Starting with the device topology state graph as the initial nodes, and all states satisfying the target constraint set C as the target node set, each atomic device operation (such as switching, signal switching, or parameter tuning) is used as the transition action and state transition edge. The local cost of this action is used as the edge weight, and heuristic search algorithms such as the A* algorithm are employed to find the transition path with the minimum global cost function. This path corresponds to the physically executable optimal conflict resolution strategy.

[0052] The complete configuration of the optoelectronic device at a certain moment is defined as the node state space S. The node state space S includes multiple nodes, which correspond to all nodes V in the device topology state diagram at that moment. It also includes the set of all node attributes {A_v(t)} and the edge set E of the nodes. In addition, the target state set {S_goal} is defined as all state nodes that satisfy the target constraint set C.

[0053] Action space A includes multiple transition actions. A transition action 'a' is defined as an atomic, minimally executable change to the device topology state diagram, which causes the system to transition from one state S to another adjacent state. The action space A is automatically generated based on the current node state space S and the physical and logical constraints of the optoelectronic devices, including but not limited to: changing the working state of a single device (such as PowerState:OFF->ON), establishing or releasing a connection edge between a signal source and an idle port, and modifying the current resolution, refresh rate, and other parameters of a display device.

[0054] Each transition action a is associated with a local cost c(S,a), and the local cost function of the i-th transition action is as follows: Where R_cost is the resource reconfiguration cost (e.g., the number of signal link switching times, the number of device reconfiguration times), P_cost is the state disturbance cost (e.g., the number of affected running tasks, the number of flickering times perceived by the user), and D_time is the expected execution delay (normalized to the [0,1] interval). κ is the scenario penalty coefficient, dynamically calculated based on the current system load and business priority: when the system is under high load or in a critical business scenario, κ increases, causing the cost function to impose additional penalties when both resource reconfiguration and state disturbances are large, thus guiding the search algorithm to prioritize paths with more balanced costs. This formula achieves nonlinear coupling between multiple factors through a combination of product and normalized denominator, avoiding the problem of objectively calibrating weight coefficients in linear weighting.

[0055] In addition, a heuristic function h(n) is constructed to efficiently estimate the minimum remaining cost from the current state S to any target state n, guiding the search direction. This invention designs a heuristic function that fits the field of display control: Where UC(n) is the number of constraints in the current node transition state graph that do not meet the target constraint set, and NAC(n) is the number of nodes in the current node transition state graph that need to be adjusted.

[0056] This heuristic function employs a non-linear form with product and logarithmic decay, avoiding the subjective parameter tuning problem of linear weighting: when the number of unsatisfied constraints UC(n) is large, h(n) is mainly dominated by UC(n), prioritizing the resolution of business-level conflicts; when UC(n) is small, the logarithmic term ln(1+NAC(n)) plays a major role, guiding the search to prioritize paths with fewer adjustment nodes. This design enables the heuristic function to adaptively balance the two objectives of "resolving constraint conflicts" and "reducing operational costs" without requiring manual setting of weight coefficients.

[0057] The heuristic function h(n) of this invention is not a geometric metric such as Euclidean distance or Manhattan distance commonly used in the general A* algorithm, but a domain heuristic function specifically designed for the business characteristics of optoelectronic display collaborative control.

[0058] Specifically, UC(n) (the number of constraints that do not satisfy the target constraint set) reflects the gap between the current state and the target state in key business indicators. For optoelectronic display systems, these constraints typically include core business requirements such as resolution matching (e.g., C1), signal path integrity (e.g., C2), and device power status (e.g., C3). UC(n) directly quantifies "how many business-level conflicts still need to be resolved," providing a clear business orientation for the search.

[0059] NAC(n) (the number of nodes whose states need to be adjusted) estimates the remaining workload from the perspective of operational costs. In optoelectronic display systems, adjusting the state of a device (such as switching signal sources or modifying resolution) often involves resource reconfiguration costs and state disturbance costs. As part of a heuristic function, NAC(n) can guide the search to prioritize paths with fewer nodes to be adjusted, thereby minimizing interference with the existing state of the system while ensuring that business objectives are achieved.

[0060] The design of this heuristic function fully reflects a deep understanding of the state space characteristics of optoelectronic display systems: there is usually a positive correlation between the number of constraints and the number of nodes that need to be adjusted, but they are not entirely equivalent. For example, a constraint of resolution mismatch (UC=1) may require adjusting multiple devices (NAC>1) to resolve, while a constraint of missing signal path (UC=1) may only require establishing a new connection edge (NAC=1). Using a weighted combination of the two as a heuristic function can more accurately estimate the residual cost, significantly improve the search efficiency of the A* algorithm, and effectively address the state space explosion problem caused by the large number of devices and complex constraint relationships in optoelectronic display systems.

[0061] Next, the evaluation function is constructed, which consists of a heuristic function and accumulated local cost, as follows: Where n is the number of transition actions in the action sequence.

[0062] The optimal conflict resolution strategy obtained by using the A* algorithm includes multiple transition actions, and each transition action accumulates its corresponding local cost.

[0063] When using the A* algorithm for iterative search, in each iteration, a transition action that minimizes the evaluation function is extracted from the action space. The extracted transition action is added to the action sequence, and a transition state graph of the current node is generated. To address the issue of potentially large state space, this invention introduces a state pruning strategy. If multiple transition actions are equivalent in core business indicators (such as the status of key signal paths or the layout of core display walls) when extracting transition actions, the transition action that minimizes the evaluation function is retained, and other transition actions are deleted from the action space.

[0064] The A* algorithm ends when the current node transition state diagram satisfies the target constraint set and the global cost function is minimized. At this point, an optimal conflict resolution strategy can be obtained. This optimal conflict resolution strategy includes a series of transition actions that need to be performed to transition from the current device topology state diagram to the target node set. Each transition action corresponds to one or more control commands that can be issued, thus forming a complete and lowest-cost collaborative control scheme. Then, the optoelectronic device is controlled according to the optimal conflict resolution strategy.

[0065] On the other hand, embodiments of the present invention also provide a multi-scene optoelectronic display collaborative control system, which uses the above-mentioned multi-scene optoelectronic display collaborative control method, including: The state perception module is used to acquire the state data of the optoelectronic device and user instructions, and to construct the device topology state diagram based on the optoelectronic device state data and deconstruct the user instructions into a target constraint set. The fault recovery module is used to generate a fault recovery plan based on the device topology state diagram and an iterative greedy search strategy, and then repair the optoelectronic device according to the fault recovery plan. The decision module is used to generate the optimal conflict resolution strategy based on the target constraint set and the new device topology state diagram using a graph search algorithm. A control module is provided for controlling the optoelectronic device according to an optimal conflict resolution strategy.

[0066] Finally, it should be noted that the above 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 preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A multi-scene optoelectronic display collaborative control method, characterized in that: include: Acquire the status data of optoelectronic devices and construct a device topology status diagram based on the status data; Based on the device topology state diagram, an iterative greedy search strategy is used to generate a fault recovery plan. Then, the optoelectronic device is repaired according to the fault recovery plan, and a new device topology state diagram is generated. The system acquires user commands, deconstructs them into a set of target constraints, and uses a graph search algorithm to generate an optimal conflict resolution strategy based on the target constraint set and a new device topology state diagram. The system then controls the optoelectronic device according to the optimal conflict resolution strategy.

2. The multi-scene optoelectronic display collaborative control method as described in claim 1, characterized in that: Methods for generating fault recovery plans based on device topology state diagrams and using an iterative greedy search strategy include: S1: Determine the fault node and the root cause node based on the equipment topology state diagram, calculate the initial impact score of the fault node, and generate a candidate action set based on the root cause node and the equipment topology state diagram. The candidate action set includes multiple recovery actions. S2: Initialize the device topology state diagram and create a blank fault recovery action sequence; S3: Randomly select a recovery action from the candidate action set, adjust the device topology state diagram according to the recovery action, and then recalculate the global impact score of the faulty node; S4: If the global impact score of this iteration is lower than the global impact score or the initial impact score of the previous iteration, then add the recovery action to the fault recovery action sequence; otherwise, discard the action and repeat steps S3-S4 until all recovery actions are taken. S5: Repeat steps S2-S4 to obtain multiple different fault recovery action sequences. Then calculate the global recovery degree of each fault recovery action sequence and select the fault recovery action sequence with the highest global recovery degree as the fault recovery plan.

3. The multi-scene optoelectronic display collaborative control method as described in claim 2, characterized in that: Methods for determining the root cause node of a fault based on the device topology state diagram include: A fault propagation graph is generated based on the device topology state diagram. The fault propagation graph includes multiple normal nodes, faulty nodes, and multiple fault-affected edges. Based on the fault impact edges, potential root cause nodes that will affect the faulty nodes are selected from the normal nodes, and then the impact probability of each potential root cause node is calculated based on the fault impact table. The potential root cause node with the highest probability of impact is selected as the root cause node of the failure.

4. The multi-scene optoelectronic display collaborative control method as described in claim 2, characterized in that: If the faulty node cannot be determined based on the device topology status diagram, no fault recovery plan will be generated.

5. The multi-scene optoelectronic display collaborative control method as described in claim 1, characterized in that: Methods for generating optimal conflict resolution strategies using graph search algorithms based on the target constraint set and a new device topology state diagram include: A1: Construct a node state space and an action space based on the device topology state diagram. The node state space includes multiple nodes, and the action space includes multiple transition actions. Construct a corresponding local cost for each transition action. A2: Construct the evaluation function, action set, and global cost function; A3: Use the A* algorithm to extract a transition action from the action space that minimizes the evaluation function, add the extracted transition action to the action sequence, and generate the current node's transition state graph; The evaluation function is: in, UC(n) is the sum of the local costs of the transition actions in the action set from the initial state to the current state n. UC(n) is the number of constraints in the current node's transition state graph that do not satisfy the target constraint set. NAC(n) is the number of nodes in the current node's transition state graph that need to have their states adjusted. A4: Repeat step A3 until the current node's transition state diagram satisfies the objective constraint set and the global cost function is minimized; A5: The action set obtained from the last iteration is used as the optimal conflict resolution strategy.

6. The multi-scene optoelectronic display collaborative control method as described in claim 5, characterized in that: In step A2, if multiple transition actions are equivalent, the transition action that minimizes the evaluation function is retained, and other equivalent transition actions are removed from the action space.

7. A multi-scene optoelectronic display collaborative control system, characterized in that: The method includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements a multi-scene optoelectronic display collaborative control method as described in any one of claims 1-6, comprising: The state awareness module is used to acquire the state data of optoelectronic devices and user commands, and to construct the device topology state diagram based on the optoelectronic device state data and decompose the user commands into a set of target constraints. The fault recovery module is used to generate a fault recovery plan based on the device topology state diagram using an iterative greedy search strategy, then repair the optoelectronic equipment according to the fault recovery plan, and generate a new device topology state diagram. The decision module is used to generate the optimal conflict resolution strategy based on the target constraint set and the new device topology state diagram using a graph search algorithm. A control module is provided for controlling the optoelectronic device according to an optimal conflict resolution strategy.

Images