An intelligent monitoring system and method for a robot ground station

Abstract

The present invention discloses an intelligent monitoring system and method for a robot ground station, which relates to the technical field of robot monitoring. By defining the strategy space of each robot, generating a target index for each strategy through an objective function, generating a utility index for each strategy using a utility function, combining the target index and the utility index to generate a preference coefficient for each strategy, in the cooperation set, the robot calculates the total cooperation benefit according to the preference coefficient, and after optimizing the revenue distribution through an allocation method, resource sharing is carried out. In the non-cooperation set, after solving the Nash equilibrium allocation strategy, the robot selects the optimal strategy according to the preference coefficient and performs task scheduling based on the selected optimal strategy. The monitoring system adopts a globally optimized strategy, and the robot shares resources in the cooperation set and independently selects the optimal strategy in the non-cooperation set, ultimately realizing the optimization of global tasks and resources.

Description

Technical Field

[0001] The present invention relates to the technical field of robot monitoring, and particularly to an intelligent monitoring system and method for a robot ground station. Background Art

[0002] A monitoring system is a system that provides all-round monitoring and intelligent analysis for a robot operation platform, aiming to achieve real-time monitoring, status evaluation, fault warning, and intelligent maintenance of robot devices. This system is mainly used in a ground control station, and communicates with the robot through a wireless or wired network to collect various status information during the operation of the robot.

[0003] The prior art has the following deficiencies:

[0004] 1. Traditional task scheduling and resource management methods are usually only locally optimized. When planning paths, different robots may choose conflicting paths, or delays may occur during task execution due to uneven resource allocation, and it is impossible to optimize from a global perspective, resulting in low overall system efficiency.

[0005] 2. Existing monitoring systems use static rules or simple priority decision-making mechanisms for task allocation, lacking the ability to adapt to complex environmental changes. When facing a dynamic environment, the system has poor flexibility and cannot adjust strategies in real time.

[0006] Based on this, the present invention proposes an intelligent monitoring system and method for a robot ground station, adopting a global optimization strategy. Robots share resources in the cooperation set and independently select the optimal strategy in the non-cooperation set, ultimately achieving the optimization of global tasks and resources. Summary of the Invention

[0007] The purpose of the present invention is to provide an intelligent monitoring system and method for a robot ground station to solve the deficiencies in the background art.

[0008] To achieve the above purpose, the present invention provides the following technical solution: An intelligent monitoring method for a robot ground station, the monitoring method comprising the following steps:

[0009] The monitoring system obtains all robot information in the ground station based on the management platform, defines the strategy space of each robot, classifies all robots with interaction into the cooperation set, and classifies all robots without interaction into the non-cooperation set.

[0010] After the robot selects the strategy space, a target index is generated for each strategy through the objective function, a utility index is generated for each strategy using the utility function, and a preference coefficient is generated for each strategy by combining the target index and the utility index.

[0011] In the cooperation set, the robot calculates the total cooperation benefit according to the preference coefficient, optimizes the revenue distribution through the allocation method, and then conducts resource sharing. In the non-cooperation set, after solving the Nash equilibrium allocation strategy, the robot selects the optimal strategy according to the preference coefficient, performs task scheduling based on the selected optimal strategy, and conducts real-time monitoring and control during the operation of all robots.

[0012] In a preferred embodiment, combining the target index and the utility index to generate a preference coefficient for each strategy includes the following steps:

[0013] After obtaining the target index and the utility index of the strategy, calculate the preference coefficient of the strategy, and the expression is: , where select is the preference coefficient, is the target index, is the utility index, , are the adjustment coefficients of the target index and the utility index respectively, and , are both greater than 0.

[0014] In a preferred embodiment, after the robot selects the strategy space, generate the target index for each strategy through the objective function, including the following steps:

[0015] Obtain the historical target data of the strategy. The historical target data includes the performance factor and the resource factor. Normalize the performance factor and the resource factor so that the value ranges of the performance factor and the resource factor are mapped to between [0, 1]. Obtain the normalized value of the performance factor and the normalized value of the resource factor, and subtract the normalized value of the resource factor from the normalized value of the performance factor to obtain the target index.

[0016] In a preferred embodiment, use the utility function to generate the utility index for each strategy, including the following steps:

[0017] Obtain the historical utility data of the strategy. The historical utility data includes the completion factor and the conflict factor. Normalize the completion factor and the conflict factor so that the value ranges of the completion factor and the conflict factor are mapped to between [0, 1]. Obtain the normalized value of the completion factor and the normalized value of the conflict factor, and sum the normalized value of the factor and the normalized value of the conflict factor to obtain the utility index.

[0018] In a preferred embodiment, in the cooperation set, the robot calculates the total cooperation benefit according to the preference coefficient, optimizes the revenue distribution through the allocation method, and then conducts resource sharing, including the following steps:

[0019] In the cooperation set, obtain the preference coefficient and operating cost of the acquisition strategy. After normalizing the operating cost, divide the preference coefficient by the normalized value of the operating cost to obtain the total cooperation benefit. After calculating the total cooperation benefit, use the average distribution method to evenly distribute the total cooperation benefit to each robot in the cooperation set. The expression is: , where is the distribution income of the robot, is the number of robots in the cooperation set, is the total cooperation benefit;

[0020] The robot performs resource sharing and coordinated task execution in cooperation according to the optimized distribution income, and adjusts the execution strategy according to the priority of the task.

[0021] In a preferred embodiment, in the non-cooperation set, after solving the Nash equilibrium configuration strategy, the robot selects the optimal strategy according to the preference coefficient, including the following steps:

[0022] The condition of Nash equilibrium is: for any robot i, after the strategies of other robots are selected, robot i cannot improve its own utility by changing its own strategy, that is:

[0023] , where represents the preference coefficient of robot under the corresponding strategy, represents the strategy selected by robot , represents the alternative strategy of robot ; that is, the strategies selected by other robots;

[0024] Through the greedy algorithm, find the optimal strategy combination of each robot when the strategies of other robots are fixed, and obtain the Nash equilibrium strategy;

[0025] After solving the Nash equilibrium, each robot will select the optimal strategy according to the preference coefficient of its own strategy. After sorting all strategies from large to small according to the preference coefficient, select the first strategy in the sorting as the optimal strategy. After the robot determines the optimal strategy, it performs the task according to the optimal strategy.

[0026] In a preferred embodiment, the calculation logic of the completion factor is: obtain the actual task completion time of the robot after selecting the strategy, divide the actual task completion time by the scheduled task completion time to obtain the task completion efficiency, obtain the number of tasks completed by the robot after selecting the strategy, divide the number of tasks completed by the monitoring duration to obtain the task completion rate, and add the task completion rate to the task completion efficiency to obtain the completion factor;

[0027] The calculation logic of the conflict factor is as follows: After obtaining the selection strategy, the number of path conflicts and the number of shared resource conflicts of the robot during the task execution are obtained. The conflict amplitude is obtained by adding the number of robot path conflicts and the number of shared resource conflicts. The conflict influence coefficient is obtained by dividing the conflict amplitude by the total number of conflicts. After obtaining the actual position and the expected position of each monitoring point on the moving path of the robot, the navigation path deviation at the monitoring point is calculated. After obtaining the navigation path deviations at all monitoring points, the navigation deviation coefficient is calculated. The expression is: , where NDI is the navigation deviation coefficient, is the navigation path deviation at the th monitoring point,

[0028] The conflict factor is obtained by summing the conflict influence coefficient and the navigation deviation coefficient. , where is the performance factor, is the number of task coverage nodes of the strategy, is the total number of nodes in the ground station;

[0029] The calculation expression of the resource factor is:

[0030] , where is the resource factor, represents the cumulative time period, represents the power consumption rate of the robot at time , represents the occupancy rate of the shared tool of the robot at time .

[0031] In a preferred embodiment, the calculation expression of the navigation path deviation is: After obtaining the actual position and the expected position of each monitoring point on the moving path of the robot, the navigation path deviation at the monitoring point is calculated. The expression is: , where is the navigation path deviation at the th monitoring point, , are the actual positions, , are the expected positions.

[0032] A robot ground station intelligent monitoring system includes a set partitioning module, a coefficient calculation module, and a task allocation and scheduling module;

[0033] Set Partitioning Module: Obtain all robot information in the ground station based on the management platform, define the strategy space for each robot, classify all robots with interaction into the cooperation set, and classify all robots without interaction into the non - cooperation set;

[0034] Coefficient Calculation Module: After the robot selects the strategy space, generate a target index for each strategy through the objective function, generate a utility index for each strategy using the utility function, and combine the target index and the utility index to generate a preference coefficient for each strategy;

[0035] Task Allocation and Scheduling Module: In the cooperation set, the robot calculates the total cooperation benefit according to the preference coefficient, optimizes the income distribution through the allocation method and then conducts resource sharing. In the non - cooperation set, after solving the Nash equilibrium allocation strategy, the robot selects the optimal strategy according to the preference coefficient, and conducts task scheduling based on the selected optimal strategy, and conducts real - time monitoring and control during the operation of all robots.

[0036] In the above technical solution, the technical effects and advantages provided by the present invention:

[0037] The present invention defines the strategy space for each robot, generates a target index for each strategy through the objective function, generates a utility index for each strategy using the utility function, combines the target index and the utility index to generate a preference coefficient for each strategy. In the cooperation set, the robot calculates the total cooperation benefit according to the preference coefficient, optimizes the income distribution through the allocation method and then conducts resource sharing. In the non - cooperation set, after solving the Nash equilibrium allocation strategy, the robot selects the optimal strategy according to the preference coefficient, and conducts task scheduling based on the selected optimal strategy. The monitoring system adopts a globally optimized strategy. The robots share resources in the cooperation set and independently select the optimal strategy in the non - cooperation set, ultimately achieving the optimization of global tasks and resources. Brief Description of the Drawings

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required to be used in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can also be obtained according to these drawings.

[0039] Figure 1 It is the method flow chart of the present invention. Detailed Embodiments

[0040] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0041] Embodiment 1: Refer to Figure 1 As shown, in this embodiment, a method for intelligent monitoring of a robot ground station is provided. The monitoring method includes the following steps:

[0042] The monitoring system obtains all robot information in the ground station based on the management platform, defines the strategy space for each robot (the strategy space includes task selection, path planning, energy management, resource allocation, etc.), divides all robots with interaction into a cooperation set, and divides all robots without interaction into a non - cooperation set. After the robot selects the strategy space, a target index is generated for each strategy through the objective function, a utility index is generated for each strategy using the utility function, and a preference coefficient is generated for each strategy by combining the target index and the utility index. In the cooperation set, the robot calculates the total cooperation benefit according to the preference coefficient, optimizes the income distribution through the allocation method, and then conducts resource sharing. In the non - cooperation set, after solving the Nash equilibrium allocation strategy, the robot selects the optimal strategy according to the preference coefficient, and performs task scheduling based on the selected optimal strategy, and real - time monitoring and control are carried out during the operation of all robots.

[0043] In this application, by defining the strategy space for each robot, a target index is generated for each strategy through the objective function, a utility index is generated for each strategy using the utility function, and a preference coefficient is generated for each strategy by combining the target index and the utility index. In the cooperation set, the robot calculates the total cooperation benefit according to the preference coefficient, optimizes the income distribution through the allocation method, and then conducts resource sharing. In the non - cooperation set, after solving the Nash equilibrium allocation strategy, the robot selects the optimal strategy according to the preference coefficient, and performs task scheduling based on the selected optimal strategy. The monitoring system adopts a global optimization strategy. The robot realizes the optimization of global tasks and resources by sharing resources in the cooperation set and independently selecting the optimal strategy in the non - cooperation set.

[0044] Embodiment 2: The monitoring system obtains all robot information in the ground station based on the management platform, and defines the strategy space for each robot (the strategy space includes task selection, path planning, energy management, resource allocation, etc.), including the following steps:

[0045] The monitoring system obtains the status information of robots in all ground stations in real time through a data interface, including the current position, task status, remaining battery power, path planning, task assignment, etc. of each robot, providing basic data for subsequent policy decisions to ensure that all robots can share status information under the same platform;

[0046] For each robot, define its optional policy space, which specifically includes the following:

[0047] The list of tasks that each robot can choose to execute. Tasks may include transportation, inspection, maintenance, etc., and decisions are made based on factors such as task priority, time requirements, resource consumption, etc.

[0048] When each robot completes a task, it needs to select a path. The path planning strategy takes into account factors such as the environmental map, obstacles, shortest path, collision avoidance, traffic congestion, etc.

[0049] The robot needs to select whether to return to the charging station or perform energy management based on its remaining battery level and the location of charging facilities. This policy space also includes power consumption optimization, charging strategies, etc.

[0050] Resource allocation: Robots may share resources (such as tools, equipment, etc.) among multiple tasks and need to allocate these resources reasonably. Resource allocation strategies may include priority management, time management, etc.

[0051] Thus, ensure that each robot can make decisions from multiple dimensions (tasks, paths, battery power, resources) to meet the requirements of complex tasks.

[0052] Classify all robots with interaction into the cooperation set and all robots without interaction into the non - cooperation set, including the following steps:

[0053] First, define what kind of interaction relationship will make a robot be classified into the cooperation set. The interaction relationship can be the following types:

[0054] Resource sharing: Multiple robots share certain resources, such as charging stations, item handling tools, sensors, etc.

[0055] Task collaboration: Robots need to collaborate to complete tasks, such as multiple robots carrying items together, patrolling areas, repairing equipment, etc.

[0056] Information sharing: Robots need to exchange data or information to form a collaboration network, such as collaborative path planning, sharing sensor data, etc.

[0057] Space sharing: Robots move in the same or adjacent physical spaces, and path conflicts, collaborative obstacle avoidance, etc. may occur.

[0058] By clarifying the interaction relationships, determine which robots need to collaborate and which robots can work independently.

[0059] After the robot selection strategy space, generate a target index for each strategy through the objective function, including the following steps:

[0060] Operation content: The management platform needs to collect the status information of each robot in the ground station in real time, including but not limited to:

[0061] Current task: The task that the robot is executing, such as transportation, inspection, repair, etc.

[0062] Resource usage: The resources used by the robot, such as batteries, charging stations, tools, etc.

[0063] Path planning: Whether the current path planning of the robot conflicts with other robots.

[0064] Task progress: The progress of each robot in the task.

[0065] Status of other robots: Whether other robots are performing tasks in the same area, whether there is path overlap or resource sharing with this robot.

[0066] By collecting the task and status information of each robot in real time, establish an interaction relationship network between robots to provide data support for subsequent partitioning and decision-making.

[0067] Operation content: Based on the interaction relationships between robots (such as resource sharing, collaborative tasks, information exchange, etc.), construct an interaction matrix. Each element of this matrix represents the interaction relationship between two robots:

[0068] The matrix element value is 1: indicating that there is an interaction relationship between two robots and they need to collaborate or share resources. The matrix element value is 0: indicating that two robots do not need to interact and can perform tasks independently.

[0069] Example: Suppose there are four robots R1, R2, R3, and R4. The interaction matrix may be as shown in Table 1:

[0070] R1 R2 R3 R4 R1 0 1 0 0 R2 1 0 1 0 R3 0 1 0 1 R4 0 0 1 0

[0071] Table 1

[0072] In Table 1, R1 interacts with R2, R2 interacts with R3, R3 interacts with R4, etc. The interaction matrix visually shows the interaction relationships between each robot, providing a basis for subsequent partitioning of the cooperation set and non-cooperation set.

[0073] Based on the interaction matrix, through connected component analysis, robots with interactions are divided into a cooperation set, and robots without interactions are included in the non - cooperation set.

[0074] Cooperation set: If two robots form a connected sub - graph through direct or indirect interaction relationships (such as being indirectly connected through other robots), these robots are included in the same cooperation set.

[0075] Non - cooperation set: Robots that do not form interaction relationships with other robots will be included in the non - cooperation set.

[0076] Example: In the above interaction matrix, R1 and R2 have interactions, R2 and R3 have interactions, and R3 and R4 have interactions. Through connected component analysis, it is concluded that robots R1, R2, R3, and R4 form a connected sub - graph. Therefore, these robots belong to the same cooperation set. If there is a robot R5, which has no interaction with R1, R2, R3, and R4, then R5 will be included in the non - cooperation set.

[0077] Goal: Through connectivity analysis, reasonably divide robots into cooperation sets and non - cooperation sets to ensure that there is a certain cooperation relationship among the robots within each set, while the robots outside each set can work independently.

[0078] After the robot selects the strategy space, generate a target index for each strategy through the objective function, including the following steps:

[0079] Obtain the historical target data of the strategy. The historical target data includes a performance factor and a resource factor. Normalize the performance factor and the resource factor so that the value ranges of the performance factor and the resource factor are mapped to between [0, 1]. Obtain the normalized value of the performance factor and the normalized value of the resource factor. Subtract the normalized value of the resource factor from the normalized value of the performance factor to obtain the target index. The larger the target index, the better the overall performance of the strategy.

[0080] The calculation expression of the performance factor is: , where is the performance factor, is the number of task - covered nodes of the strategy, is the total number of nodes in the ground station;

[0081] The larger the performance factor, the greater the task coverage of the robot choosing this strategy in the ground station, that is, the higher the selection priority of this strategy. The reasons can be explained from the following aspects:

[0082] Task coverage refers to the scope or quantity covered by a robot when performing tasks in a ground station. A strategy with a high performance factor usually means that the strategy can efficiently complete more tasks, cover more areas, or execute more subtasks. An increase in the performance factor generally means that the robot performs better when executing tasks and can more comprehensively meet the task requirements. This enables the robot to give full play to its capabilities when choosing this strategy, thereby increasing the task completion rate, reducing idle time, or unnecessary repetitive tasks.

[0083] When designing the task performance factor, factors such as task completion rate, time efficiency, and accuracy are usually incorporated. A high performance factor indicates that the robot performs excellently in these key indicators, that is, it can complete more tasks efficiently and accurately in a shorter time. For example, a high task completion rate indicates that the strategy can complete more tasks, high time efficiency indicates that the tasks are completed quickly, and high accuracy indicates that there are no errors during task execution. These all directly affect the task coverage and reflect the superiority of the strategy.

[0084] If the performance factor is large, it means that the strategy can cover a wider range of tasks and optimize the efficiency of task execution. In a multi-robot system, tasks are usually allocated according to multiple factors such as the capabilities of the robots, task requirements, and resource allocation. Selecting a strategy with a high performance factor can enable the robots to participate in task execution more efficiently, avoid resource waste and task overlap, and improve the overall task coverage.

[0085] Strategies with high performance factors usually involve better spatio-temporal optimization and resource management. For example, when executing tasks, the robot can efficiently plan paths, reasonably arrange the task order, and ensure maximum task coverage and minimum resource consumption. This optimized strategy not only improves the task coverage but also enhances the overall efficiency of the system. Therefore, a high performance factor not only reflects the quality of the robot's task completion but also serves as the basis for its selection priority.

[0086] By setting the performance factor as the key decision-making basis, the ground station can preferentially select those strategies that can effectively execute tasks, reduce conflicts, and improve efficiency according to the performance of the robot strategies. Through the task coverage indicator, when multiple robots work together, it can ensure that they can reasonably allocate tasks, avoid resource competition and task vacancies. The higher the performance factor of a strategy, the more it means that it can effectively improve the task execution ability of the ground station, so it has a higher selection priority.

[0087] The strategy with high performance factors means greater task completion efficiency, quality, and coverage, enabling the robot to complete more tasks in the ground station with higher resource utilization efficiency. Therefore, choosing such a strategy helps improve the overall system performance and task completion rate. This strategy can better allocate tasks, reduce conflicts, and improve the effective use of resources, so its priority should be higher.

[0088] The calculation expression of the resource factor is: , where is the resource factor, represents the cumulative time period, represents the power consumption rate of the robot at time moment, represents the occupancy rate of the shared tool by the robot at time moment;

[0089] The larger the resource factor, it indicates that within the cumulative time period, the robot consumes more energy and occupies the shared tool for a longer time during the task process using this strategy. As a result, the priority of choosing this strategy is lower. The reasons can be explained from the following aspects:

[0090] When the robot executes tasks, energy is a limited and precious resource. When the resource factor of a strategy is large, it means that the robot will consume more energy during the execution of this strategy. This not only increases the cost of task execution but also may lead to the exhaustion of the robot's energy, affecting the execution of subsequent tasks. Therefore, choosing a strategy with high energy consumption means that the robot will face a higher risk of energy consumption during task execution, which reduces the priority of this strategy. A strategy with high energy consumption usually means that the robot is less efficient when performing the same task, perhaps because the path planning is not optimized or the execution process is not efficient, resulting in more energy waste. For example, a task takes a long time to complete, or the robot uses unnecessary additional resources (such as running at high speed, using too many sensors, etc.), thus exacerbating the energy consumption.

[0091] Strategies with high resource consumption can affect the sustainability of a robot's task execution. For example, under a strategy with high energy consumption, the robot may interrupt the task due to battery depletion before the task is completed, resulting in task interruption or failure. In this way, the task completion rate is low and the efficiency is low, making this strategy no longer be preferentially selected. If a strategy involves a long occupation time of shared tools, it means that this strategy may occupy or compete for resources of other robots or tasks during task execution. This not only affects the overall execution efficiency of the task, but may also lead to bottlenecks in system resources, thus reducing the overall effectiveness of the system. Conflict of shared resources: In a multi-robot system, multiple robots may need to use the same tool or resource (such as sensors, workbenches, robotic arms, etc.). If a certain strategy results in a long occupation time of this tool, it may prevent other robots from using this tool in a timely manner, thus triggering resource conflicts and even causing task delays or failures. A strategy with a long occupation time of shared tools usually means that the task scheduling is not efficient enough and may cause bottlenecks in resource scheduling. A robot that occupies shared resources for a long time may hinder the task execution of other robots, reduce the overall collaborative effectiveness, and ultimately affect the task coverage and completion rate.

[0092] When the resource factor is large, it means that the cumulative effect of resource consumption during the task process is more significant. For a robot performing a long-term task, if it continuously selects a strategy with high resource consumption, the following problems will occur:

[0093] As the task execution time increases, a strategy with high resource consumption may cause the robot's battery to deplete quickly, unable to support the completion of long-term or multi-stage tasks. This may lead to task execution interruption, or the robot may not be able to return to the charging station in time for replenishment, thus affecting the entire task chain. When a robot depends on more energy and shared tools, its independence and flexibility are restricted. For example, if a robot is overly dependent on a certain shared tool (such as a specific sensor, device, etc.), it may be restricted to certain specific areas or task types during the task and unable to flexibly adjust its task execution strategy.

[0094] A strategy with a high resource factor usually reflects a less efficient execution process, and there may be room for optimization in such a strategy. A larger resource factor means that the robot does not fully optimize its resource usage when performing tasks. For example, the task path planning may be unreasonable, resulting in redundant movements and energy consumption; the task sequence is not optimal, leading to unreasonable use of shared tools, etc. Selecting a strategy with a larger resource factor actually means that the sustainability of task execution is poor, and the long-term occupation of resources will cause the system performance to gradually decline. In a multi-robot system, in order to improve the overall task execution efficiency and reduce resource waste, it is necessary to preferentially select those strategies that can efficiently utilize resources (such as energy, time, shared tools, etc.). Strategies with a smaller resource factor can usually better balance task completion and resource usage, enabling the robot to continuously and efficiently execute tasks without frequent charging or waiting for the release of resources.

[0095] Therefore, if a strategy has a large resource factor, it means that it will consume more resources (especially energy and shared tools) when performing tasks, reducing the task execution efficiency and increasing resource conflicts and risks. This makes the priority of this strategy should be reduced.

[0096] Generating a utility index for each strategy using a utility function includes the following steps:

[0097] Obtain the historical utility data of the strategy. The historical utility data includes the completion factor and the conflict factor. Normalize the completion factor and the conflict factor so that the value ranges of the completion factor and the conflict factor are mapped to between [0, 1]. Obtain the normalized value of the completion factor and the normalized value of the conflict factor. Sum the factor normalized value and the conflict factor normalized value to obtain the utility index. The larger the utility index, the higher the overall benefit of the strategy.

[0098] The calculation logic of the completion factor is as follows: Obtain the actual task completion time of the robot after selecting the strategy. Divide the actual task completion time by the scheduled task completion time to obtain the task completion efficiency. Obtain the number of tasks completed by the robot after selecting the strategy. Divide the number of tasks completed by the monitoring duration to obtain the task completion rate. Add the task completion rate and the task completion efficiency to obtain the completion factor;

[0099] The larger the completion factor, the higher the task completion rate and task completion efficiency after selecting this strategy, that is, the higher the selection priority of this strategy. Specifically:

[0100] The larger the completion factor, it means that the task completion rate and task completion efficiency of the robot after selecting this strategy are higher, and thus the selection priority of this strategy is higher. The reasons can be explained from the following aspects:

[0101] The task completion rate reflects the proportion of tasks that the robot can complete on time and with high quality when performing tasks. The larger the completion factor, the more tasks the robot can complete in a shorter time and with high task quality. This usually means that the strategy has strong efficiency and reliability during task execution, thus giving the strategy a higher priority in task allocation. A high completion factor indicates that the robot is more efficient and reliable when performing tasks, can complete more tasks or a higher proportion of tasks within a given time. A larger completion factor means a lower task failure rate for the strategy. The robot is less likely to encounter mid-course failures or task interruptions, improving the overall task completion rate.

[0102] Task completion efficiency includes not only the proportion of tasks completed, but also the speed of task completion and resource utilization efficiency. A strategy with a large completion factor usually means that the robot can complete tasks in a shorter time and consume fewer resources during task execution. High-efficiency task execution improves the overall system performance, saving time and resources. A strategy with high task completion efficiency can ensure that tasks are completed as soon as possible, thus improving the overall scheduling efficiency and reducing task waiting time. A large task completion factor usually also means more efficient resource utilization. For example, optimized path planning, reasonable energy management, fair task allocation, etc., all of which can reduce unnecessary resource waste.

[0103] In a multi-robot cooperation system, a strategy with a larger completion factor can effectively improve the overall task execution efficiency and reduce conflicts between tasks. For a task scheduling system, preferentially selecting a strategy with a large completion factor helps to ensure the maximization of the overall system performance. When robots select tasks, they will preferentially choose those strategies that can complete tasks efficiently, save time and resources. A strategy with a high completion factor usually can avoid task duplication and conflicts. For example, the robot can complete tasks in a shorter time, enabling other robots to be allocated new tasks in a timely manner, avoiding scheduling conflicts caused by task delays. When robots select more efficient strategies, the allocation of resources (such as energy, computing power, shared tools, etc.) will also become more reasonable. Resources can be fully utilized without being wasted under some inefficient strategies.

[0104] A strategy with a high completion factor can not only demonstrate good execution performance in the current task but also accumulate advantages in the long-term task execution process. When a robot selects a strategy with a high completion factor, it can accumulate more task success experiences, providing more data support for future task scheduling and optimization. A high completion factor also indicates a relatively high reliability of the strategy. The robot can successfully complete tasks through this strategy, reducing the time for backtracking, adjustment, and repair caused by errors or failures. A high completion factor enables each task in the task chain to be smoothly connected, allowing the robot to effectively complete a series of tasks without being affected by the failure or delay of individual tasks in the entire task sequence.

[0105] In a multi-robot system, tasks usually have clear goals and time limits. The larger the completion factor, the more it means that the strategy selected by the robot can help achieve these goals and meet the system's expectations. The successful completion rate and high efficiency of tasks are in line with the system goals, improving the overall performance of the system. A strategy with a large completion factor means a higher success rate of tasks, and the results of task execution are closer to expectations. This is because this strategy can effectively avoid task failures caused by reasons such as incorrect path planning, improper task scheduling, or unreasonable resource allocation.

[0106] The larger the completion factor, the more it indicates that the robot can execute tasks with a higher completion rate and efficiency. A high completion factor usually reflects a higher success rate of task completion, a faster task execution speed, and more efficient resource utilization. Due to these advantages, a strategy with a high completion factor can improve the execution effect of tasks and the overall efficiency of the system. Therefore, it should be given a higher priority in task scheduling and resource allocation.

[0107] The calculation logic of the conflict factor is as follows: Obtain the number of path conflicts and the number of shared resource conflicts (including shared tools, shared charging, etc., that is, the number of times the robot and other robots need to share tools simultaneously and the number of times they charge through the shared charging station simultaneously) during the task execution of the robot after selecting the strategy. Add the number of robot path conflicts and the number of shared resource conflicts to obtain the conflict amplitude. Divide the conflict amplitude by the total number of conflicts (obtained by summing the conflict amplitudes of all strategies) to get the conflict impact coefficient. After obtaining the actual position and expected position of each monitoring point on the moving path of the robot, calculate the navigation path deviation at the monitoring point. The expression is: , where is the navigation path deviation at the th monitoring point, , are the actual positions, , are the expected positions. After obtaining the navigation path deviations at all monitoring points, calculate the navigation deviation coefficient. The expression is: , where NDI is the navigation deviation coefficient, is the navigation path deviation at the th monitoring point, is the number of monitoring points on the moving path. The conflict factor is obtained by summing the conflict influence coefficient and the navigation deviation coefficient.

[0108] The larger the conflict factor, the more frequent the conflicts between the robot and other robots during the task execution after selecting this strategy, and the easier the robot is to deviate from the expected moving path during the task execution. That is, the lower the selection priority of this strategy. Specifically:

[0109] The larger the conflict factor (Conflict Factor), the higher the frequency of conflicts between the robot and other robots during the task execution, and the easier it is to deviate from the expected moving path during the task execution. Therefore, the lower the selection priority of this strategy. The reasons can be explained from the following aspects:

[0110] A large conflict factor means that the robot has a high frequency of conflicts with other robots or the environment during the task execution. Conflicts can lead to interruptions, delays or resource competition during the task execution, thus reducing the overall execution efficiency of the task. The following are the negative impacts brought by conflicts:

[0111] Conflicts may cause the task to be paused or restarted. For example, when two robots arrive at the same location simultaneously, they need to pause or be rescheduled, thus prolonging the execution time of the task. Frequent conflicts will increase the uncertainty of the task execution time. The robot needs to handle conflicts frequently, which may affect the smooth progress of the entire task and reduce the predictability and reliability of task completion. Therefore, the strategy with a large conflict factor usually shows frequent task interruptions and rescheduling during the execution process, affecting the smooth completion of the task, thus reducing the priority of this strategy.

[0112] In a multi-robot system, conflicts not only affect the task execution but also may lead to waste of resources (such as energy, time, shared tools, etc.). For example:

[0113] Multiple robots may need to share the same resource (such as a workbench, sensor, execution arm, etc.). When conflicts occur frequently, the allocation and utilization efficiency of the resource are reduced, which may cause the robots to wait and be rescheduled frequently, thus wasting time and energy. Frequent conflicts may force the robot to take unnecessary avoidance actions or re-plan the path, which will consume more energy. Especially during the obstacle avoidance process, the robot may need to change the path, increasing the additional driving distance and causing waste of energy. These wastes not only affect the efficiency of the task but also lead to a decline in the overall effectiveness of the system. Therefore, the strategy with a large conflict factor requires more resources to make up for the losses caused by conflicts, making the priority of this strategy lower.

[0114] A large conflict factor also means that the robot is more likely to deviate from the expected navigation path. Path deviation usually increases the complexity and uncertainty of task execution, leading to the following problems:

[0115] After the robot deviates from the path, more calculations and adjustments are usually required to recalibrate its position. This not only increases the time for task execution but also may cause the path planning to no longer be optimal. The robot deviating from the path may take a longer path, resulting in an extended task completion time, thus affecting the overall task scheduling. Especially when multiple tasks are carried out alternately, the delay will affect the scheduling of other tasks. Frequent path deviations increase the difficulty of task execution and lead to a decrease in the speed and efficiency of the robot when performing tasks. Therefore, the priority of this strategy should be relatively low.

[0116] In a multi-robot system, task scheduling is usually optimized according to factors such as the current positions of the robots, task requirements, and available resources. A strategy with a large conflict factor means that the scheduling of the robot during task execution is unstable, increasing the complexity of task scheduling. The following are the scheduling problems caused by conflicts:

[0117] Due to frequent conflicts, the system needs to frequently adjust the task allocation and path planning of the robots, increasing the computational complexity of the scheduling system. If conflicts occur frequently, it may cause the robots to be unable to share resources efficiently, increasing the resource competition during the scheduling process and reducing the overall efficiency of the system. Therefore, a strategy with a large conflict factor usually causes an additional burden on the scheduling system, making the priority of this strategy relatively low.

[0118] Frequent conflicts and path deviations not only affect the execution effect of the current task but also have a negative impact on the long-term performance of the system. For example:

[0119] A strategy with frequent conflicts may lead to continuous interruption and rescheduling of tasks, and the efficiency of the robot in performing tasks is always at a low level, affecting the productivity and performance of the overall system. The instability brought by conflicts makes the robots unable to maintain efficient cooperation in the long term, but instead increases the complexity and uncertainty of the system, affecting the collaborative effect of the system in a multi-task and multi-robot environment. Therefore, a strategy with a large conflict factor will cause the overall performance of the system to decline, reduce the predictability and reliability of task completion, and further weaken the priority of this strategy.

[0120] The larger the conflict factor, the higher the frequency of conflicts between the robot and other robots or resources during task execution, resulting in unstable task execution, waste of resources, and an increase in path deviations. A strategy with frequent conflicts not only affects the efficiency and reliability of task execution but also increases the scheduling complexity of the system. Therefore, the priority of this strategy should be relatively low because it will reduce the overall performance of the system, increase resource consumption, and waste time.

[0121] Generate a preference coefficient for each strategy by combining the target index and the utility index, including the following steps:

[0122] After obtaining the target index and the utility index of the strategy, calculate the preference coefficient of the strategy. The expression is: , where select is the preference coefficient, is the target index, is the utility index, and are the adjustment coefficients of the target index and the utility index respectively, and and are both greater than 0.

[0123] In the cooperation set, the robot calculates the total cooperation benefit according to the preference coefficient, and optimizes the income distribution through the distribution method and then conducts resource sharing, including the following steps:

[0124] In the cooperation set, obtain the preference coefficient and the operating cost of the strategy. After normalizing the operating cost, divide the preference coefficient by the normalized value of the operating cost to obtain the total cooperation benefit. After calculating the total cooperation benefit, adopt the average distribution method to evenly distribute the total cooperation benefit to each robot in the cooperation set. The expression is: , where is the allocated income of the robot, is the number of robots in the cooperation set, is the total cooperation benefit.

[0125] The robot conducts resource sharing and coordinated task execution in the cooperation according to the optimized allocated income. The purpose of resource sharing and task scheduling is to ensure the efficient completion of tasks, avoid resource conflicts, and adjust the execution strategy according to the priority of tasks, including the following steps:

[0126] Resource sharing: According to the allocated income and the preference coefficient, the robot reasonably shares limited resources such as batteries, computing resources, sensors, etc. during the task execution process.

[0127] Task scheduling: The robot conducts task scheduling according to the optimized strategy preference coefficient and income. Task scheduling needs to consider the synergy effect between robots to ensure the avoidance of resource conflicts and task conflicts during the task execution process.

[0128] Coordination and cooperation: When executing tasks, the robot ensures the smooth completion of tasks by sharing resources and coordinating task progress. At the same time, the robot adjusts the strategy according to the real-time task requirements to optimize the efficiency of task execution.

[0129] In the non - cooperative set, after solving the Nash equilibrium allocation strategy, the robot selects the optimal strategy according to the preference coefficient and performs task scheduling based on the selected optimal strategy, including the following steps:

[0130] In the non - cooperative set, the interaction between robots is based on independent strategy selection. Therefore, it is necessary to solve the Nash equilibrium to determine the optimal strategy for each robot. The Nash equilibrium means that in a task, among the strategy combinations selected by all participants, no participant can obtain higher benefits by unilaterally changing their own strategy. The steps for solving the Nash equilibrium are as follows:

[0131] Each robot has a set of optional strategies, and the strategy space includes task selection, path planning, energy management, etc. The combination of all strategies forms the strategy space of the game.

[0132] All robots will choose a strategy that maximizes their own utility, while considering the strategies of other robots. Solving the Nash equilibrium means finding a strategy combination under which no robot can obtain greater utility by changing its strategy alone. Specifically, the condition for the Nash equilibrium is: for any robot \(i\), after the strategies of other robots are selected, robot \(i\) cannot improve its own utility by changing its strategy, that is:

[0133] , where represents for all robots The meaning of is a universal quantification symbol, used to indicate that the following expression applies to all robots The objects or elements represented by represents robot The preference coefficient of robot under the corresponding strategy represents robot The strategy selected by robot represents robot The alternative strategy of robot That is, the strategies selected by other robots.

[0134] It means that in the Nash equilibrium, for each participant (robot or player), when the strategies of other participants are fixed, choosing the current strategy will not be worse than choosing any other strategy. In other words, in the equilibrium state, no participant has the incentive to unilaterally change their strategy, because if they change their strategy, their utility will not increase and may even decrease. Therefore, the equilibrium is stable, and the strategy choices of all individuals are optimal in the equilibrium state.

[0135] Then, through the greedy algorithm, find the optimal strategy combination for each robot when the strategies of other robots are fixed, and obtain the Nash equilibrium strategy;

[0136] The steps of the greedy algorithm are as follows: In each round, the robot selects the current local optimal strategy without considering the future decisions or strategy adjustments of other robots. For example, the robot will select the optimal strategy based on the current environment and the strategies of other robots, even if this choice may cause the subsequent choice to no longer be optimal. Repeat the greedy algorithm until the strategy selected by each robot no longer changes, or the change is very small. At this point, the algorithm has reached convergence and finally obtains a strategy combination, namely the Nash equilibrium strategy;

[0137] After solving the Nash equilibrium, each robot will select the optimal strategy based on the optimization coefficient of its own strategy. After sorting all strategies from large to small according to the optimization coefficient, the strategy ranked first will be selected as the optimal strategy. After the robot determines the optimal strategy, it will perform the task according to the optimal strategy.

[0138] After selecting the optimal strategy, the robots will be scheduled according to the requirements of the task. Task scheduling is the process of ensuring that the robots execute the tasks according to the optimal strategy. The scheduling process needs to consider the priority of the task, the availability of resources, and the coordination between robots.

[0139] Real-time monitoring and control of all robot operations, including the following steps:

[0140] Based on real-time data, monitor various status parameters of the robot, such as:

[0141] Task status (executing, completed, failed);

[0142] Energy status (power remaining, energy consumption);

[0143] Motion state (speed, path deviation, acceleration);

[0144] Environmental status (obstacle detection, environmental changes);

[0145] Goal: Gain real-time visibility into the robot’s health, task progress, and environment interactions to make timely decisions.

[0146] Track the progress of tasks, evaluate whether the current tasks are proceeding as planned, and adjust task parameters or plans based on real-time data. Ensure that tasks are completed on time and promptly detect problems such as task delays and execution failures. Monitor the robot status in real time and use data analysis techniques (such as machine learning and rule detection) to detect anomalies. Once a failure or anomaly occurs (such as low battery, task interruption, collision, etc.), immediately issue an alarm and initiate emergency handling procedures. Minimize robot failure time and ensure the stability of the robot in task execution.

[0147] Generate and adjust control instructions based on real-time monitoring data and task progress. This includes: adjusting the energy management of the control movement path (such as charging or replacing the battery), adjusting task priorities or reallocating tasks, and optimizing the robot's operation strategy to ensure that the robot executes tasks as efficiently and stably as possible. If multiple robots are performing tasks together, conduct real-time communication and collaboration between the robots. This includes sharing task information, coordinating movement paths, adjusting task priorities, etc. Ensure the smooth cooperation of multiple robots, avoid conflicts and resource conflicts, and achieve collaborative task completion.

[0148] Monitor the energy consumption of each robot, calculate the remaining battery power, and dynamically adjust the robot's energy management strategy according to task requirements. If the battery power is insufficient, operations such as automatically requesting charging or battery replacement can be performed. Ensure that the robot has sufficient energy during task execution and improve energy utilization efficiency. Adjust the robot's path planning and movement strategy according to real-time environmental information. For example, avoid new obstacles and dynamically plan paths according to environmental changes. Ensure that the robot can adapt to dynamic environmental changes and effectively avoid obstacles. Record and store the robot's operation data in real time, including task execution, energy consumption, fault conditions, etc. Generate real-time reports and display them to the operator through the management platform. Provide real-time monitoring data to help the operator make decisions and provide basic data for subsequent optimization and analysis.

[0149] Optimize the robot's behavior through algorithms based on the real-time data and feedback information of the robot's operation. For example, optimize the strategy by analyzing historical data and adjust task allocation, path planning, etc. Through continuous feedback and optimization, improve the robot's autonomy and execution efficiency. If a serious fault, emergency, or safety hazard is detected, trigger the alarm mechanism and initiate the emergency handling process, such as safe shutdown, remote control, etc. Ensure the safety of the robot and the surrounding environment.

[0150] The real-time monitoring and control system can comprehensively monitor, dynamically adjust, and optimize the state of the robot during operation to ensure that the robot can execute tasks efficiently and safely. At the same time, when a fault or abnormality occurs, take timely measures to handle it. The realization of these steps depends on the combination of multiple technologies, including data acquisition, sensor networks, control algorithms, artificial intelligence, etc.

[0151] Embodiment 3: An intelligent monitoring system for a robot ground station according to this embodiment includes a set partitioning module, a coefficient calculation module, and a task allocation and scheduling module;

[0152] Set Partitioning Module: Based on the management platform, obtain all the robot information in the ground station, and define the policy space for each robot (the policy space includes task selection, path planning, energy management, resource allocation, etc.). All robots with interaction are grouped into a cooperation set, and all robots without interaction are grouped into a non - cooperation set. The set partitioning result is sent to the coefficient calculation module and the task assignment and scheduling module;

[0153] Coefficient Calculation Module: After the robot selects the policy space, generate a target index for each policy through the objective function, generate a utility index for each policy using the utility function, and combine the target index and the utility index to generate a preference coefficient for each policy. The preference coefficient is sent to the task assignment and scheduling module;

[0154] Task Assignment and Scheduling Module: In the cooperation set, the robots calculate the total cooperation benefit according to the preference coefficient, and after optimizing the income distribution through the allocation method, they conduct resource sharing. In the non - cooperation set, after solving the Nash equilibrium allocation strategy, the robots select the optimal strategy according to the preference coefficient, and perform task scheduling based on the selected optimal strategy, and conduct real - time monitoring and control during the operation of all robots.

[0155] The above formulas are all dimensionless and take their numerical values for calculation. The formula is obtained by collecting a large amount of data for software simulation to get a formula closest to the real situation. The preset parameters in the formula are set by those skilled in the art according to the actual situation.

[0156] It should be understood that the term “and / or” in this article is merely a description of the association relationship between associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, both A and B exist simultaneously, and B exists alone. Here, A and B can be singular or plural. In addition, the character “ / ” in this article generally represents an “or” relationship between the associated objects before and after, but it may also represent an “and / or” relationship, which can be specifically understood by referring to the context before and after.

[0157] It should be understood that in various embodiments of this application, the magnitudes of the sequence numbers of the above - mentioned processes do not mean the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of this application.

[0158] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. A professional technician can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application. Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0159] As described above, the above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed in this application can easily think of changes or substitutions, and all should be covered by the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

Claims

1. An intelligent monitoring method for a robot ground station, characterized in that: The monitoring method includes the following steps: The monitoring system obtains all robot information in the ground station based on the management platform, defines the policy space for each robot, classifies all robots with interaction into the cooperation set, and classifies all robots without interaction into the non - cooperation set; After the robot selects the strategy space, a target index is generated for each strategy through the objective function, and a utility index is generated for each strategy using the utility function. The preferred coefficient is generated by combining the target index and the utility index, and the expression is: , where select is the preferred coefficient, is the target index, is the utility index, , are the adjustment coefficients of the target index and the utility index respectively, and , are both greater than 0; In the cooperation set, the robots calculate the total cooperation benefit according to the preference coefficient, optimize the benefit distribution through the distribution method and then conduct resource sharing. In the non - cooperation set, after solving the Nash equilibrium allocation strategy, the robots select the optimal strategy according to the preference coefficient, perform task scheduling based on the selected optimal strategy, and conduct real - time monitoring and control during the operation of all robots.

2. The intelligent monitoring method for a robot ground station according to claim 1, wherein: After the robots select the policy space, a target index is generated for each policy through the objective function, including the following steps: obtaining the target index by subtracting the normalized value of the resource factor from the normalized value of the performance factor.

3. The intelligent monitoring method of a robot ground station according to claim 2, characterized in that: The acquisition logic of the normalized value of the performance factor and the normalized value of the resource factor is as follows: Obtain the historical target data of the policy, where the historical target data includes the performance factor and the resource factor, perform normalization processing on the performance factor and the resource factor, map the value ranges of the performance factor and the resource factor to between [0, 1], and obtain the normalized value of the performance factor and the normalized value of the resource factor.

4. The intelligent monitoring method for a robot ground station according to claim 3, characterized in that: Generate a utility index for each policy using the utility function, including the following steps: Obtain the historical utility data of the policy, where the historical utility data includes the completion factor and the conflict factor, perform normalization processing on the completion factor and the conflict factor, map the value ranges of the completion factor and the conflict factor to between [0, 1], obtain the normalized value of the completion factor and the normalized value of the conflict factor, and sum the normalized value of the factor and the normalized value of the conflict factor to obtain the utility index.

5. The intelligent monitoring method for a robot ground station according to claim 2, wherein: In the cooperation set, the robots calculate the total cooperation benefit according to the preference coefficient, optimize the benefit distribution through the distribution method and then conduct resource sharing, including the following steps: In the cooperation set, obtain the optimization coefficient of the acquisition strategy and the operating cost. After normalizing the operating cost, divide the optimization coefficient by the normalized value of the operating cost to obtain the total cooperation benefit. After calculating the total cooperation benefit, adopt the average distribution method to evenly distribute the total cooperation benefit to each robot in the cooperation set. The expression is: , where is the distribution income of the robot, is the number of robots in the cooperation set, is the total cooperation benefit; The robots conduct resource sharing and coordinate task execution in cooperation according to the optimized distribution benefit, and adjust the execution strategy according to the priority of the task.

6. The intelligent monitoring method of a robot ground station according to claim 2, characterized in that: In the non - cooperation set, after solving the Nash equilibrium allocation strategy, the robots select the optimal strategy according to the preference coefficient, including the following steps: The condition of Nash equilibrium is that for any robot i, given the strategies selected by other robots, robot i cannot improve its utility by changing its own strategy, that is: , where represents the preference coefficient of the robot under the corresponding strategy, represents the strategy selected by the robot, represents the alternative strategy of the robot, that is, the strategy selected by other robots; Through the greedy algorithm, find the optimal strategy combination for each robot when the strategies of other robots are fixed to obtain the Nash equilibrium strategy; After solving the Nash equilibrium, each robot will select the optimal strategy according to the preference coefficient of its own strategy. After sorting all strategies from large to small according to the preference coefficient, select the first - ranked strategy as the optimal strategy. After the robot determines the optimal strategy, perform the task according to the optimal strategy.

7. A method for intelligent monitoring of a robot ground station according to claim 4, characterized in that: The calculation logic of the completion factor is as follows: Obtain the actual task completion time of the robot after obtaining the selection strategy, divide the actual task completion time by the scheduled task completion time to obtain the task completion efficiency, obtain the number of tasks completed by the robot after obtaining the selection strategy, divide the number of tasks completed by the monitoring duration to obtain the task completion rate, and add the task completion rate to the task completion efficiency to obtain the completion factor; The calculation logic of the conflict factor is as follows: After obtaining the selection strategy, the number of path conflicts and the number of shared resource conflicts of the robot during the task execution are obtained. The conflict amplitude is obtained by adding the number of robot path conflicts and the number of shared resource conflicts. The conflict influence coefficient is obtained by dividing the conflict amplitude by the total number of conflicts. After obtaining the actual position and the expected position of each monitoring point on the moving path of the robot, the navigation path deviation at the monitoring point is calculated. After obtaining the navigation path deviations at all monitoring points, the navigation deviation coefficient is calculated. The expression is: , where NDI is the navigation deviation coefficient, is the navigation path deviation at the th monitoring point, is the number of monitoring points on the moving path. The conflict factor is obtained by summing the conflict influence coefficient and the navigation deviation coefficient.

8. The intelligent monitoring method of a robot ground station according to claim 3, characterized in that: The calculation expression of the performance factor is as follows: , where is the performance factor, is the number of task coverage nodes of the strategy, is the total number of all nodes in the ground station; The calculation expression of the resource factor is: , where is the resource factor, represents the cumulative time period, represents the power consumption rate of the robot at time , represents the occupancy rate of the shared tool by the robot at time .

9. The intelligent monitoring method for a robot ground station according to claim 7, wherein: The calculation expression for the navigation path deviation is as follows: After obtaining the actual position and the expected position of each monitoring point on the moving path of the robot, calculate the navigation path deviation at the monitoring point, and the expression is: , where is the navigation path deviation at the th monitoring point, , are the actual positions, , are the expected positions.

10. A robot ground station intelligent monitoring system for implementing the monitoring method according to any one of claims 1-9, characterized in that: It includes a set partitioning module, a coefficient calculation module, and a task assignment and scheduling module; Set partitioning module: Based on the management platform, obtain all the robot information in the ground station, define the strategy space of each robot, divide all the robots with interaction into a cooperation set, and divide all the robots without interaction into a non-cooperation set; Coefficient calculation module: After the robot selects the strategy space, generate a target index for each strategy through the objective function, generate a utility index for each strategy using the utility function, and combine the target index and the utility index to generate a preference coefficient for each strategy; Task assignment and scheduling module: In the cooperation set, the robot calculates the total cooperation benefit according to the preference coefficient, optimizes the revenue distribution through the allocation method and then conducts resource sharing. In the non-cooperation set, after solving the Nash equilibrium configuration strategy, the robot selects the optimal strategy according to the preference coefficient, and conducts task scheduling based on the selected optimal strategy, and conducts real-time monitoring and control during the operation of all robots.

Images