A multi-robot task scheduling method, system, device and storage medium

Abstract

The present invention discloses a multi-robot task scheduling method, system, device and storage medium, which relate to the field of artificial intelligence technology, including: defining the roles of robots and collecting environmental information in real time to build a dynamic environment model; identifying the current tasks to be executed according to the dynamic environment model, evaluating the urgency, complexity and resource requirements of each task, and forming a priority list; decomposing the high-priority tasks in the priority list into subtasks, specifying the roles of robots and task allocation schemes; using a biologically inspired self-organizing algorithm, the robots exchange information according to their own states and task requirements, and dynamically adjust the task allocation scheme; the robots exchange information according to their own states and task requirements, and dynamically adjust the task allocation scheme, thereby realizing the flexibility and adaptability of task allocation, being able to respond to changes in the environment and state in a timely manner, and avoiding task delays due to insufficient resources.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a multi-robot task scheduling method, system, device and storage medium. Background Art

[0002] In recent years, with the rapid development of automation technology and robotics technology, multi-robot systems have been widely used in many fields such as industrial production, logistics warehousing, and mining. Especially in the field of mining, the introduction of multi-robot systems has significantly improved operating efficiency and safety. As an advanced multi-robot system, the multi-joint arm shed robot for mining uses a multi-degree-of-freedom robotic arm to achieve efficient handling and installation of U-beams, significantly reducing the labor intensity of workers and improving the efficiency and safety of complex shed construction. However, although the multi-joint arm shed robot for mining performs well in single-machine operation, it still faces many challenges in multi-robot collaborative operation. Traditional multi-robot task scheduling methods mainly rely on preset rules and fixed task allocation strategies, lacking adaptability and flexibility to dynamic environmental changes. This static task scheduling method is difficult to achieve efficient and real-time task allocation and execution in the face of complex and changeable mine environments, resulting in problems such as low resource utilization and long task response time.

[0003] There are several shortcomings in the application of existing multi-robot task scheduling methods in mine environments. First, existing task scheduling methods are usually based on static environmental models and cannot perceive and adapt to dynamic changes in mine environments in real time, such as sudden obstacles and equipment failures. This results in the need for frequent adjustments to task allocation schemes during actual execution, which increases the complexity and unpredictability of the system. Secondly, existing task scheduling methods are relatively single in terms of task priority evaluation, usually only considering the urgency of the task, while ignoring the complexity of the task and resource requirements. This single evaluation method may lead to waste of resources and inefficient task execution. In addition, existing task scheduling methods lack an effective information exchange mechanism in multi-robot collaborative operations, and cannot achieve real-time communication and dynamic adjustment between robots, thereby limiting the overall performance of the system. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a multi-robot task scheduling method, system, device and storage medium to solve the problems of low task scheduling efficiency, insufficient information sharing and weak collaboration ability of multi-robot systems in dynamic environments.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides a multi-robot task scheduling method, which includes:

[0008] Define the role of the robot and collect environmental information in real time to build a dynamic environment model;

[0009] Identify the tasks to be performed currently according to the dynamic environment model, evaluate the urgency, complexity and resource requirements of each task, and form a priority list;

[0010] Decompose the high-priority tasks in the priority list into subtasks, and specify the roles of the robots and the task allocation scheme;

[0011] Using a biologically inspired self-organizing algorithm, the robot exchanges information based on its own status and task requirements, and dynamically adjusts the task allocation plan;

[0012] The robots execute subtasks according to the adjusted task scheduling plan, output the execution status update and task completion status of each robot, and form a real-time task execution report;

[0013] Operators monitor robot execution through a visual interface, provide feedback and adjust task assignments.

[0014] As a preferred solution of the multi-robot task scheduling method of the present invention, wherein: the role definition includes determining the robot type and defining the role attributes;

[0015] The environmental information includes regional layout, obstacle locations, task distribution and resource status;

[0016] The construction of the dynamic environment model comprises the following steps:

[0017] Classify the environmental information collected in real time and establish a database;

[0018] Using SLAM technology, the environmental information collected in real time is converted into a dynamic map, and obstacles and task locations are marked in the dynamic map to form a dynamic environmental model.

[0019] As a preferred solution of the multi-robot task scheduling method of the present invention, wherein: identifying the current task to be executed according to the dynamic environment model, evaluating the urgency, complexity and resource requirements of each task, and forming a priority list includes the following steps:

[0020] Extract all current task information to be executed from the dynamic environment model;

[0021] According to the nature of the task, classify the task and identify the type of task;

[0022] Analyze the deadlines of tasks and the impact on production, and assign an urgency score to each task;

[0023] Assign a complexity score to each task based on its technical requirements and expected operations;

[0024] Analyze the resources required for each task and give a resource score;

[0025] Calculate the overall priority of each task based on urgency, complexity, and resources;

[0026] The quick sort algorithm is used to sort the tasks and generate a priority list.

[0027] As a preferred solution of the multi-robot task scheduling method of the present invention, wherein: breaking down the high-priority tasks in the priority list into subtasks, specifying the roles of the robots and the task allocation scheme comprises the following steps:

[0028] Check the priority list and select the highest priority task;

[0029] Extract detailed information of selected tasks;

[0030] Determine dismantling standards;

[0031] According to the decomposition criteria, split the task into subtasks and identify the specific content of each subtask;

[0032] Conduct a preliminary assessment of each subtask to determine complexity and resource requirements;

[0033] After completing the task disassembly, access the robot module library to obtain information about available modules;

[0034] Using the matching algorithm, the matching scores of the robot modules are calculated to determine the robot modules required for each subtask;

[0035] Select the robot module with the highest matching score for each subtask and record the assignment;

[0036] Integrate all subtasks and their corresponding robot modules into a comprehensive module allocation plan.

[0037] As a preferred solution of the multi-robot task scheduling method of the present invention, wherein: a biologically inspired self-organizing algorithm is adopted, the robots exchange information according to their own states and task requirements, and the task allocation scheme is dynamically adjusted, including the following steps:

[0038] Extract the current status of each subtask from the comprehensive module allocation scheme and collect the status information of each robot;

[0039] Set the number of virtual ants participating in the scheduling, the pheromone importance factor, the heuristic factor and the pheromone volatilization rate;

[0040] Based on the tasks in the current task pool, as well as the priority and resource requirements of each task, each ant randomly selects a subtask to be executed and evaluates its own status;

[0041] Select a path based on the current pheromone concentration and the matching degree of resources and priorities;

[0042] After each task is completed, the effectiveness of the chosen path is evaluated;

[0043] Based on the evaluation results, the pheromone concentration is updated;

[0044] Set the load and energy thresholds. When the robot's load and energy are lower than the thresholds, start the dynamic adjustment process.

[0045] Based on the urgency and resource requirements of the task, part of the task is transferred to other available robots, and the task status information is updated.

[0046] As a preferred solution of the multi-robot task scheduling method of the present invention, wherein: the robots execute subtasks according to the adjusted task scheduling scheme, output the execution status update and task completion status of each robot, and form a real-time task execution report, including the following steps:

[0047] After the dynamically adjusted task allocation plan is generated, the central controller sends a task allocation confirmation instruction to each robot;

[0048] After receiving the confirmation instruction, the robot confirms it and starts executing according to the assigned subtask;

[0049] During execution, the robot monitors its own status in real time and regularly uploads its own status and task progress data to the central controller;

[0050] The central controller analyzes the uploaded data in real time and takes immediate action when an abnormality is detected;

[0051] The central controller integrates all collected robot status and task execution data to form execution feedback information and compile it into a task execution report.

[0052] In a second aspect, the present invention provides a multi-robot task scheduling system, comprising:

[0053] The dynamic environment model building module defines the role of the robot and collects environmental information in real time to build a dynamic environment model;

[0054] A task identification and evaluation module identifies the tasks to be performed according to the dynamic environment model, evaluates the urgency, complexity and resource requirements of each task, and forms a priority list;

[0055] A task decomposition and allocation module decomposes the high-priority tasks in the priority list into subtasks, specifies the roles of the robots and the task allocation scheme;

[0056] The self-organizing scheduling algorithm module uses a biologically inspired self-organizing algorithm. The robot exchanges information based on its own status and task requirements and dynamically adjusts the task allocation plan;

[0057] Execution status update module: the robot executes subtasks according to the adjusted task scheduling plan, outputs the execution status update and task completion status of each robot, and forms a real-time task execution report;

[0058] Monitoring and feedback module: The operator monitors the robot's execution through a visual interface, provides feedback and adjusts task allocation.

[0059] In a third aspect, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the multi-robot task scheduling method described in the first aspect of the present invention is implemented.

[0060] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the multi-robot task scheduling method described in the first aspect of the present invention is implemented.

[0061] The beneficial effects of the present invention are as follows: by defining the roles of robots and collecting environmental information in real time, a dynamic environmental model is constructed, thereby achieving effective classification of the capabilities of each robot and comprehensive grasp of the environmental status; identifying tasks to be executed according to the dynamic environmental model and evaluating their urgency, complexity and resource requirements to form a priority list, thereby achieving comprehensive analysis and effective sorting of tasks; breaking down high-priority tasks in the priority list into subtasks, and specifying the roles of robots and task allocation plans, thereby achieving decomposition and refinement of complex tasks; by adopting a bio-inspired self-organizing algorithm, robots exchange information according to their own status and task requirements, and dynamically adjust the task allocation plan, thereby achieving flexibility and adaptability in task allocation, being able to respond to changes in the environment and status in a timely manner, and avoiding task delays due to insufficient resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0063] Figure 1 This is a flow chart of the multi-robot task scheduling method in Example 1.

[0064] Figure 2 This is a flow chart of the multi-robot task scheduling system in Example 1. DETAILED DESCRIPTION

[0065] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0066] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0067] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0068] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and provides a multi-robot task scheduling method, comprising the following steps:

[0069] S1. Define the role of the robot and collect environmental information in real time to build a dynamic environment model.

[0070] S1.1. Role definition includes determining the robot type and defining role attributes.

[0071] Furthermore, according to the characteristics of the mining multi-joint arm shed robot, it is classified as a processing robot because it is not only responsible for carrying U-shaped beams, but also involves the precise installation of top beams and auxiliary beams.

[0072] The role attributes are defined as follows:

[0073] Load Capacity: Sufficient strength to lift U-beams weighing several tons.

[0074] Flexibility: The multi-degree-of-freedom robotic arm allows the robot to operate flexibly in tight spaces and to navigate obstacles in the lanes.

[0075] Precision: Ability to accurately place U-beams at predetermined locations, ensuring safety and stability of the roadway.

[0076] Remote control: Supports remote control operation, which can be controlled away from dangerous areas to improve safety.

[0077] S1.2. Environmental information includes area layout, obstacle location, task distribution and resource status.

[0078] Furthermore, the specific steps for collecting environmental information in real time are as follows:

[0079] Sensor layout: LiDAR, infrared sensors and cameras are installed in the mine to monitor the layout of the tunnels, the location of obstacles and the working status of the robots.

[0080] Data collection: The following data is obtained in real time through sensors:

[0081] Area layout: Get map information of mine tunnels to understand their direction, width and height.

[0082] Obstacle location: Monitor dynamic and static obstacles in the lane to ensure the robot can pass safely.

[0083] Task distribution: Identify the locations where U-beams need to be erected and prioritize the tasks.

[0084] Resource Status: Monitor the inventory of U-beams to ensure there is enough material for construction.

[0085] S1.3, building a dynamic environment model includes the following steps:

[0086] The environmental information collected in real time is classified and a database is established. The environmental information collected in real time is converted into a dynamic map using SLAM (simultaneous localization and mapping) technology, and obstacles and task locations are marked in the dynamic map to form a dynamic environmental model.

[0087] S2. Identify the tasks to be performed based on the dynamic environment model, evaluate the urgency, complexity and resource requirements of each task, and form a priority list.

[0088] S2.1. Extract all current task information to be executed from the dynamic environment model.

[0089] Specifically, the task information includes task type, location, expected completion time, and resource requirements to identify the type of task.

[0090] Furthermore, the task types include: transporting U-beams, installing top beams, installing side beams, etc.

[0091] Task location: the specific location of the tunnel that needs to be constructed.

[0092] Expected Completion Time: Determine the expected completion time for each task based on the construction plan.

[0093] Resource requirements: including the number of U-beams, the working status of the robot, etc.

[0094] S2.2. Classify tasks according to their nature, such as urgent tasks, routine tasks, and postponed tasks, and identify the types of tasks.

[0095] Furthermore, the classification criteria for classifying tasks include the following aspects:

[0096] Urgency: Whether the task needs to be processed immediately.

[0097] Scope of impact: The extent to which a task affects production or other tasks.

[0098] Time sensitivity: How close the task's deadline is to the present time.

[0099] Tasks are divided into the following categories:

[0100] Urgent tasks: such as top beam installation, because this is related to the safety and stability of the tunnel.

[0101] Routine tasks: such as the handling of U-beams, need to be completed on time but not immediately.

[0102] Postponable tasks: such as routine inspections, can be performed after other tasks are completed.

[0103] S2.3. Analyze the deadline of the task and the impact on production, and assign an urgency score to each task.

[0104] Going further, the specific steps for assigning an urgency score to each task are as follows:

[0105] Access the current task list through the internal interface, extract the deadline, start time and current time of each task; verify the extracted data to ensure that the time format is correct and the data is complete; calculate the urgency score of each task based on the extracted time information, the expression is:

[0106] ;

[0107] in, Indicates The urgency rating of each task, Represents the task index, Indicates the deadline of the task. Indicates the current time. Indicates the start time of the task.

[0108] S2.4. Assign a complexity score to each task based on its technical requirements and expected operations.

[0109] Furthermore, the specific steps for evaluating the complexity of each task are as follows:

[0110] Mining engineering experts were invited to score each task using a scale of 1 to 10, where 1 represents a very simple task (such as basic handling tasks) and 10 represents an extremely complex task (such as precision machining involving multiple steps). After the expert evaluation, the scores of all experts were collected and the average was calculated as the final complexity score.

[0111] S2.5. Analyze the resources required for each task (such as time, materials, equipment) and give a resource score.

[0112] Furthermore, the specific steps for analyzing the resource requirements of each task and deriving the resource score are as follows:

[0113] List all the resources required for each task, including time, materials, equipment, and manpower;

[0114] Automatically generate a resource list based on the task description and compare it with the currently available resources;

[0115] Resources are scored based on how well their availability matches the task requirements, with scores ranging from 1 to 5, with 1 indicating a very poor match and 5 indicating a perfect match.

[0116] S2.6. Calculate the comprehensive priority of each task based on urgency, complexity and resources; sort the tasks using the quick sort algorithm to generate a priority list.

[0117] S3. Break down the high-priority tasks in the priority list into subtasks, specify the robot's role and task allocation plan.

[0118] S3.1. Check the priority list and select the highest priority task.

[0119] Furthermore, if there are multiple tasks with the same highest priority, they will be further filtered based on their deadlines.

[0120] S3.2. Extract detailed information of the selected task.

[0121] Specifically, the detailed information of the task includes task description, resource requirements, estimated execution time, etc., to facilitate subsequent decomposition and allocation.

[0122] Going further, task description: clarify the specific content of the task, such as moving U-beams, installing top beams, etc.

[0123] Resource requirements: List the time, materials, equipment, etc. required.

[0124] Estimated execution time: Estimate the completion time based on the complexity of the task.

[0125] S3.3. Determine the dismantling standards.

[0126] Specifically, the cracking standards include the task's operational steps, time requirements, resource requirements, etc.

[0127] Furthermore, the operation steps: break down the task into specific steps, such as grabbing the U-beam, moving it to the designated location, installing and fixing it, etc.

[0128] Time requirement: The estimated time to complete each step.

[0129] Resource requirements: The resources required for each step.

[0130] S3.4. According to the decomposition criteria, split the task into subtasks and identify the specific content of each subtask; conduct a preliminary assessment of each subtask to determine the complexity and resource requirements; after completing the task decomposition, access the robot module library to obtain information about available modules.

[0131] Specifically, the information of available modules includes functions, performance parameters and current status (such as availability, charging status, etc.).

[0132] S3.5. Use the matching algorithm to calculate the matching score of the robot module and determine the robot module required for each subtask; select the robot module with the highest matching score for each subtask and record the allocation; integrate all subtasks and their corresponding robot modules into a comprehensive module allocation plan.

[0133] Furthermore, the matching algorithm is used to calculate the matching score of the robot module:

[0134] ;

[0135] in, is the matching score between the 𝑖th subtask and the 𝑗th robot module, 𝑖 is the subtask index, 𝑗 is the robot module index, is the functional matching degree of the 𝑗th robot module, is the resource matching degree between the 𝑖th subtask and the 𝑗th robot module, is the complexity score of the 𝑖th subtask.

[0136] S4. Using a biologically inspired self-organizing algorithm, the robot exchanges information based on its own status and task requirements and dynamically adjusts the task allocation plan.

[0137] S4.1. Extract the current status of each subtask from the comprehensive module allocation scheme.

[0138] Specifically, the current status includes priority, assigned robot module, expected completion time, etc.

[0139] S4.2. Collect the status information of each robot.

[0140] Specifically, the status information includes current location, load status, energy level, etc.;

[0141] S4.3 Determine the key parameters of the ant colony algorithm.

[0142] Specifically, the key parameters include:

[0143] Number of ants: The number of virtual ants participating in the scheduling.

[0144] Pheromone importance factor: the weight that affects pheromone on path selection.

[0145] Heuristic factors: Influence the importance of resources and priorities.

[0146] Pheromone volatilization rate: reflects the decay rate of pheromones over time.

[0147] S4.4. Each ant randomly selects a subtask and builds a path based on its status (such as load, location) and task requirements (such as priority, resource usage); at the beginning of each iteration, a path is selected based on the current pheromone concentration and resource matching degree.

[0148] Furthermore, the expression for path selection is:

[0149] ;

[0150] in, Is the ant from the subtask Transfer to Robotics Module The probability of , which indicates the tendency of ants to choose a path. The higher the value, the greater the probability of choosing the path. This probability is used to guide ants to choose appropriate robot modules during task allocation. It's a task With robot module The pheromone concentration between reflects the quality of the past path. The higher the concentration, the better the performance of the path in the previous execution. Ants tend to choose such a path. It's a task With robot module The degree of matching between resources and priorities is usually combined with the requirements of the subtask and the capabilities of the robot module. The higher the value, the stronger the adaptability of the robot module to the subtask. It provides additional information to help ants consider the specific requirements of the current task when choosing a path. is a set of optional next subtasks or robot modules, representing all possible paths that the current ant can choose. is the index in the collection of optional next subtask or robot module, is the importance factor of pheromone, which controls the influence of pheromone concentration on path selection. A higher 𝛼 value means that ants are more likely to rely on pheromones when choosing a path, while a lower value means that ants will rely more on the matching degree of resources and priorities. is the importance factor of the heuristic information, controlling the influence of the matching degree of resources and priorities on path selection. A high value means that ants place more importance on resource matching and priority, while a low value means that the influence of pheromones is greater. is the pheromone concentration between task 𝑖 and robot module 𝑘, It is the degree of match between the requirements of task 𝑖 and the capabilities of robot module 𝑘.

[0151] S4.5. After each task is completed, evaluate the effectiveness of the selected path.

[0152] Specifically, effectiveness evaluation includes task completion time, resource consumption, and robot load;

[0153] S4.6. Based on the evaluation results, update the pheromone concentration, enhance the paths with good performance, and reduce the pheromone concentration of the paths with poor performance.

[0154] Furthermore, the updated expression of pheromone concentration is:

[0155] ;

[0156] in, It is the pheromone volatilization rate, which controls the decay rate of pheromones. A value of 0 means that the pheromone will evaporate faster, reducing the attractiveness of the path, preventing the path from being overly concentrated on certain choices, and maintaining the diversity of the search. It's on mission With robot module Increased pheromone amount.

[0157] S4.7. Set the load and energy thresholds. When the load and energy of the robot are lower than the thresholds, start the dynamic adjustment process. According to the urgency and resource requirements of the task, transfer part of the task to other available robots, and update the task status information.

[0158] S5. The robot executes the subtask according to the adjusted task scheduling plan, outputs the execution status update and task completion status of each robot, and forms a real-time task execution report, including the following steps:

[0159] After the dynamically adjusted task allocation plan is generated, the central controller sends a task allocation confirmation instruction to each robot; the robot confirms the confirmation instruction after receiving it and starts to execute according to the assigned subtask; during the execution process, the robot monitors its own status in real time and regularly uploads its own status and task progress data to the central controller; the central controller analyzes the uploaded data in real time and takes immediate measures when an abnormal situation is detected; the central controller integrates all the collected robot status and task execution data to form execution feedback information and compiles it into a task execution report.

[0160] This embodiment also provides a multi-robot task scheduling system, including:

[0161] The dynamic environment model construction module defines the role of the robot and collects environmental information in real time to build a dynamic environment model; the task identification and evaluation module identifies the current task to be executed according to the dynamic environment model, evaluates the urgency, complexity and resource requirements of each task, and forms a priority list; the task decomposition and allocation module decomposes the high-priority tasks in the priority list into subtasks, specifies the role of the robot and the task allocation plan; the self-organizing scheduling algorithm module adopts a biologically inspired self-organizing algorithm, and the robot exchanges information according to its own status and task requirements, and dynamically adjusts the task allocation plan; the execution status update module, the robot executes subtasks according to the adjusted task scheduling plan, outputs the execution status update and task completion status of each robot, and forms a real-time task execution report; the monitoring and feedback module, the operator monitors the robot execution through a visual interface, provides feedback and adjusts task allocation.

[0162] This embodiment also provides a computer device, which is suitable for the case of a multi-robot task scheduling method, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the multi-robot task scheduling method proposed in the above embodiment.

[0163] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0164] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for implementing multi-robot task scheduling proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0165] In summary, the present invention achieves effective classification of each robot's capabilities and comprehensive grasp of the environmental status by: defining the roles of the robots and collecting environmental information in real time, and constructing a dynamic environmental model; identifying the tasks to be executed according to the dynamic environmental model and evaluating their urgency, complexity and resource requirements to form a priority list, thereby achieving comprehensive analysis and effective sorting of tasks; breaking down high-priority tasks in the priority list into subtasks, and specifying the roles of the robots and task allocation plans, thereby achieving the decomposition and refinement of complex tasks; by adopting a bio-inspired self-organizing algorithm, the robots exchange information according to their own status and task requirements, and dynamically adjust the task allocation plan, thereby achieving flexibility and adaptability in task allocation, being able to respond to changes in the environment and status in a timely manner, and avoiding task delays due to insufficient resources.

[0166] Example 2, referring to Table 1, is the second example of the present invention. To further verify the technical solution of the present invention, experimental simulation data of a multi-robot task scheduling method is provided.

[0167] The experiment is set in an intelligent warehouse environment with an area of ​​500 square meters and multiple areas and obstacles. The experimental tasks include the handling and sorting of goods. The experiment is divided into two stages: a comparison between the prior art scheduling method and the scheduling method of the present invention.

[0168] In the first stage, the static task allocation strategy of the prior art was adopted, and the 10 robots were evenly assigned tasks, and the task completion time, resource consumption and load of each robot were recorded. In the second stage, the method of the present invention was used, and the robots were assigned different roles (such as handling robots, sorting robots), and environmental information was collected in real time to build a dynamic environment model, and the SLAM technology was used to generate an environmental map.

[0169] The details are shown in Table 1 below:

[0170] Table 1 Experimental data record table

[0171]

[0172] Through the analysis of the above table data, it can be clearly seen that the multi-robot task scheduling method of the present invention has advantages over the prior art.

[0173] First, in terms of task completion time, the average completion time of the handling task of the prior art is 42 minutes, while the handling task of the present invention only takes 30 minutes, an increase of 28.6%. The sorting task also shows obvious advantages. The prior art takes 48 minutes, while the present invention only takes 25 minutes, an efficiency increase of 47.9%. This shows that the present invention can respond to task requirements more quickly and reduce task execution time in a dynamic environment.

[0174] Secondly, in terms of resource consumption, the handling and sorting tasks of the prior art consume 11 kWh and 13 kWh respectively, while the corresponding values ​​of the present invention are 8 kWh and 7 kWh, which are reduced by 27.3% and 46.2% respectively. This shows that the present invention optimizes resource allocation and significantly reduces energy consumption by collecting environmental information and role definition in real time.

[0175] In terms of load status and success rate, the average load status of the robot of the present invention is 55%, while that of the prior art is 76%. This means that the pressure on the robot when performing tasks is significantly reduced, thereby reducing the failure rate and increasing the success rate of tasks. Specifically, the success rate of the prior art is 80%, while the success rate of the present invention is as high as 92%, an increase of 15%.

[0176] Finally, in terms of feedback time, the average feedback time of the prior art is 6 minutes, while the present invention only takes 4 minutes. The fast feedback mechanism enables the system to adjust task allocation in real time, improving the flexibility and response speed of the overall scheduling.

[0177] In summary, through comparison of data, it can be seen that during the implementation process of the present invention, not only the efficiency of task execution is improved and resource consumption is reduced, but also the robot load state is optimized and the task success rate is improved, which fully verifies the practical application value and feasibility of the multi-robot task scheduling method.

[0178] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A multi-robot task scheduling method, characterized in that: include, Define the role of the robot and collect environmental information in real time to build a dynamic environment model; Identify the tasks to be performed currently according to the dynamic environment model, evaluate the urgency, complexity and resource requirements of each task, and form a priority list; Decompose the high-priority tasks in the priority list into subtasks, and specify the roles of the robots and the task allocation scheme; Using a biologically inspired self-organizing algorithm, the robot exchanges information based on its own status and task requirements, and dynamically adjusts the task allocation plan; The robots execute subtasks according to the adjusted task scheduling plan, output the execution status update and task completion status of each robot, and form a real-time task execution report; The role definition includes determining the robot type and defining role attributes; The environmental information includes regional layout, obstacle locations, task distribution and resource status; The construction of the dynamic environment model comprises the following steps: Classify the environmental information collected in real time and establish a database; Using SLAM technology, the real-time collected environmental information is converted into a dynamic map, and obstacles and task locations are marked in the dynamic map to form a dynamic environmental model; Identifying the tasks to be performed according to the dynamic environment model, evaluating the urgency, complexity and resource requirements of each task, and forming a priority list includes the following steps: Extract all current task information to be executed from the dynamic environment model; According to the nature of the task, classify the task and identify the type of task; Analyze the deadlines of tasks and the impact on production, and assign an urgency score to each task; Assign a complexity score to each task based on its technical requirements and expected operations; Analyze the resources required for each task and give a resource score; Calculate the overall priority of each task based on urgency, complexity, and resources; Use the quick sort algorithm to sort the tasks and generate a priority list; Decomposing the high priority tasks in the priority list into subtasks and specifying the roles and task allocation scheme of the robots comprises the following steps: Check the priority list and select the highest priority task; Extract detailed information of selected tasks; Determine dismantling standards; According to the decomposition criteria, split the task into subtasks and identify the specific content of each subtask; Conduct a preliminary assessment of each subtask to determine complexity and resource requirements; After completing the task disassembly, access the robot module library to obtain information about available modules; Using the matching algorithm, the matching scores of the robot modules are calculated to determine the robot modules required for each subtask; Select the robot module with the highest matching score for each subtask and record the assignment; Integrate all subtasks and their corresponding robot modules into a comprehensive module allocation plan; Using a biologically inspired self-organizing algorithm, the robot exchanges information based on its own state and task requirements, and dynamically adjusts the task allocation scheme, including the following steps: Extract the current status of each subtask from the comprehensive module allocation scheme and collect the status information of each robot; Set the number of virtual ants participating in the scheduling, the pheromone importance factor, the heuristic factor and the pheromone volatilization rate; Based on the tasks in the current task pool, as well as the priority and resource requirements of each task, each ant randomly selects a subtask to be executed and evaluates its own status; Select a path based on the current pheromone concentration and the matching degree of resources and priorities; After each task is completed, the effectiveness of the chosen path is evaluated; Based on the evaluation results, the pheromone concentration is updated; Set the load and energy thresholds. When the robot's load and energy are lower than the thresholds, start the dynamic adjustment process. Based on the urgency and resource requirements of the task, part of the task is transferred to other available robots, and the task status information is updated.

2. The multi-robot task scheduling method according to claim 1, characterized in that: The robot executes the subtask according to the adjusted task scheduling plan, outputs the execution status update and task completion status of each robot, and forms a real-time task execution report, including the following steps: After the dynamically adjusted task allocation plan is generated, the central controller sends a task allocation confirmation instruction to each robot; After receiving the confirmation instruction, the robot confirms it and starts executing according to the assigned subtask; During execution, the robot monitors its own status in real time and regularly uploads its own status and task progress data to the central controller; The central controller analyzes the uploaded data in real time and takes immediate action when an abnormality is detected; The central controller integrates all collected robot status and task execution data to form execution feedback information and compile it into a task execution report.

3. A multi-robot task scheduling system, based on the multi-robot task scheduling method according to any one of claims 1 to 2, characterized in that: include, The dynamic environment model building module defines the role of the robot and collects environmental information in real time to build a dynamic environment model; A task identification and evaluation module identifies the tasks to be performed according to the dynamic environment model, evaluates the urgency, complexity and resource requirements of each task, and forms a priority list; A task decomposition and allocation module decomposes the high-priority tasks in the priority list into subtasks, specifies the roles of the robots and the task allocation scheme; The self-organizing scheduling algorithm module uses a biologically inspired self-organizing algorithm. The robot exchanges information based on its own status and task requirements and dynamically adjusts the task allocation plan; Execution status update module: the robot executes subtasks according to the adjusted task scheduling plan, outputs the execution status update and task completion status of each robot, and forms a real-time task execution report; Monitoring and feedback module: The operator monitors the robot's execution through a visual interface, provides feedback and adjusts task allocation.

4. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the multi-robot task scheduling method described in any one of claims 1 to 2 are implemented.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the multi-robot task scheduling method described in any one of claims 1 to 2 are implemented.

Images