A comprehensive robot dispatching system integrating multiple brands

Abstract

The present invention discloses a robot integrated scheduling system integrating multiple brands, and relates to the technical field of robot integrated scheduling, comprising a task acquisition module, which is used for a scheduling system to obtain a list of tasks to be executed from a production management system; a robot division module, which is used to generate a plurality of groups of task switching time interval vectors; a task allocation module, which is used to allocate tasks to robots based on a plurality of groups of task switching time interval vectors and a task list; a data acquisition module, which is used to obtain multi-robot efficiency data and system resource data when robots perform tasks; a task switching time interval evaluation model is constructed based on machine learning; and a screening module, which is used to perform data analysis to obtain screened task switching time interval vectors, which are applied to the integrated scheduling of robots of multiple brands, and solve the problem that the setting of task switching time intervals in multi-brand robot systems lacks flexibility, and easily leads to an imbalance between resource utilization and task connection efficiency in the scheduling system.

Description

Technical Field

[0001] The present invention relates to the technical field of robot integrated scheduling, and more specifically, to a robot integrated scheduling system integrating multiple brands. Background Art

[0002] Under the current automation trend of manufacturing and logistics industries, robot systems of different brands are widely used in production lines, warehousing, logistics sorting and other fields. However, due to the differences in control systems and protocols of robots of various brands, scheduling and collaboration are often limited.

[0003] The communication protocols and control interfaces used by robots of different brands vary greatly, such as TCP / IP, EtherCAT, CAN and other bus protocols as well as proprietary control protocols. In order to achieve the interconnection of robots of multiple brands, it is usually necessary to develop a universal protocol converter or communication gateway to perform two-way conversion of data formats and instructions.

[0004] Since different brands of robots have different operating systems and operating environments (such as ROS, Yasawa Motoman, ABB RAPID, etc.), a heterogeneous system integration platform is needed to aggregate the status information, real-time data and sensor data of all robots.

[0005] Data fusion and information synchronization: In a heterogeneous system, the status information reported by each robot may be out of sync or have errors. Data fusion technologies, such as timestamp synchronization and data calibration, can ensure data consistency for robots of different brands in the same scheduling system.

[0006] For example, the invention patent with announcement number CN116638528B announces a multi-robot joint scheduling method for a flexible manufacturing system, which dynamically adjusts the key parameters of the genetic algorithm and controls the iterative search process of the genetic algorithm, thereby improving the algorithm's solution efficiency and realizing efficient joint scheduling of the workshop. A three-layer encoding method matching the machine sequence, process sequence and AGV sequence is adopted; first, the algorithm is initialized, the fitness values ​​of all individuals in the current population are calculated, and the best individual currently searched is recorded; the state data of the current generation population is calculated, and the reward value R is obtained by comparing it with the state data of the previous generation, and the Q-Table is updated to discretize the state data of the current generation; the action value a is selected according to the discrete state value, the row element of the Q-Table and the ε-greedy strategy, and the corresponding genetic algorithm parameter adjustment action is executed; the selection operator, crossover operator and mutation operator of the genetic algorithm are executed according to the adjusted parameters to form a new population; if the population iteration reaches the maximum number of iterations, the best individual and fitness value are output.

[0007] For example, the invention patent with announcement number: CN113256050B announces a robot ring storage system and AGV hybrid scheduling method and system, including the following steps, belonging to the field of production material distribution technology, including: the AGV scheduling equipment receives the transfer task sent by the logistics area workstation and sends it to the production execution master control machine; the production execution master control machine identifies the transfer task to generate a material temporary storage task and sends it to the robot ring storage system. Without changing the overall large process links, the present invention designs a set of logic suitable for the linkage of handling equipment and sorting equipment, so that the handling equipment and picking equipment can be automatically connected with each other in operation, solving the gap in this field in the current market. Moreover, it has high adaptability and does not disrupt the operation scheduling of the original equipment; it is highly flexible and can be compatible with smart devices of different brands and the same scene operation; it reduces manual work and realizes the autonomous linkage of smart devices throughout the entire operation process.

[0008] The above disclosed technical solutions have at least the following technical problems: the task switching time interval is crucial to the coordination and efficiency of multi-brand robot systems; the task switching time interval refers to the minimum time interval set by the scheduling system when arranging different brands of robots to switch tasks; different brands of robots have differences in execution speed, system response time and task preparation time. An appropriate task switching time interval can effectively connect the robot system and avoid affecting the overall workflow due to frequent or too slow task switching.

[0009] If the task switching interval is set too high, each robot will be in a waiting state for a long time after completing a task, resulting in a waste of resources. This will not only reduce the utilization rate of the robot, but also reduce the output efficiency of the entire system, especially in an environment with a tight production rhythm, which may cause delays and affect the continuity of the process.

[0010] When the task switching interval is set too low, the scheduling system will switch robot tasks too frequently, causing various brands of robots to have response lags or action conflicts during task execution due to insufficient preparation. At the same time, frequent task switching may accelerate the wear and tear of equipment, shorten the service life of equipment, and easily cause interference or collision risks during task execution, affecting system safety and stability.

[0011] The prior art lacks flexibility in setting the task switching time interval in a multi-brand robot system, which easily leads to an imbalance between resource utilization and task connection efficiency in the scheduling system, thereby reducing overall production efficiency and equipment life. In view of the above problems, the present invention proposes a solution. Summary of the invention

[0012] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an integrated scheduling system for robots integrating multiple brands. By analyzing the scheduling effects of multi-brand robots under different task switching time intervals, it solves the problem that there is a lack of flexibility in setting the task switching time interval in the multi-brand robot system, which easily leads to an imbalance between resource utilization and task connection efficiency in the scheduling system, thereby reducing the overall production efficiency and equipment life.

[0013] To achieve the above object, the present invention provides the following technical solutions:

[0014] A robot integrated scheduling system integrating multiple brands includes a task acquisition module, a robot division module, a task allocation module, a data acquisition module, an effect evaluation module and a screening module, and there are connections between the modules; the task acquisition module is used for the scheduling system to obtain a list of tasks to be executed from a production management system, and the list of tasks to be executed is sorted according to the priority, required resources and execution requirements of the tasks; the robot division module is used to divide the multi-brand robots into gradients based on performance, and generate several groups of task switching time interval vectors based on computer random numbers according to the task amount of the task list and the gradient number of robots; the task allocation module is used to detect the status of robots of each brand in real time, and screen out several robots that meet the requirements. A robot that meets task requirements and assigns tasks to the robot based on several groups of task switching time interval vectors and task lists, wherein the robot status includes position, power, and idle status; a data acquisition module is used to obtain multi-robot efficiency data and system resource data when the robot performs tasks; an effect evaluation module is used to build a task switching time interval evaluation model based on machine learning according to the multi-robot efficiency data and system resource data; a screening module is used to build a time interval evaluation effect display model according to the output of the task switching time interval evaluation model and the corresponding task switching time interval vector, and perform data analysis to obtain the screened task switching time interval vector, which is applied to the comprehensive scheduling of robots of multiple brands.

[0015] In a preferred embodiment, the method for dividing multi-brand robots into gradients based on performance is as follows: the system collects performance data of robots of each brand through sensors and internal data interfaces, and the performance data includes load capacity, speed, operation accuracy, battery life, and adaptability to the working environment; the performance data collected by all robots are standardized; the number of clusters is automatically determined according to the density clustering of the performance data based on the DBSCAN algorithm, and the number of gradients that need to be classified and the number of robots for each gradient are obtained; the task volume of the task list is quantified by the required task time; while ensuring that the task volume remains unchanged, a number of groups of task switching time interval vectors are randomly generated by computer random numbers, each group of task switching time interval vectors includes the same number of task switching time intervals as the number of gradients that need to be classified, and the task switching time intervals are respectively applied to robots of corresponding gradients.

[0016] In a preferred embodiment, the method of assigning tasks to robots based on several groups of task switching time interval vectors and task lists is specifically as follows: matching the generated task switching time interval vectors with robots of different gradients, with the time intervals in each group of vectors corresponding to robots of the same gradient; assigning tasks to qualified robots in sequence according to the sorted task list, and performing task switching according to the set task switching time interval.

[0017] In a preferred embodiment, the multi-robot efficiency data includes a multi-task concurrent efficiency fluctuation coefficient; the specific method for obtaining the multi-task concurrent efficiency fluctuation coefficient is as follows: obtain the total battery capacity of the robot group; record the response time of the robot group after each task is received, and calculate the average response time; use the ratio of the total battery capacity to the average response time as the task processing capability value of the robot group; calculate the task processing efficiency factor based on the preset task execution efficiency and recovery efficiency; calculate the multi-task concurrent efficiency fluctuation coefficient based on the efficiency factor and the task processing capability value.

[0018] In a preferred embodiment, the system resource data includes a system resource utilization impact coefficient and a system stability impact coefficient; the specific method for obtaining the system resource utilization impact coefficient is as follows: obtain the robot system resource ratio data, the robot system resource ratio data includes processing power, storage capacity and network bandwidth, and calculate the average value of the system resource load excluding task resources when the robot performs a task according to the robot system resource ratio data; calculate the robot average resource scheduling efficiency according to a preset average resource scheduling efficiency calculation formula; calculate the robot system resource average load efficiency according to the preset system resource average load efficiency calculation formula combined with the average value of the system resource load; calculate the system resource utilization impact coefficient according to the ratio of the robot average resource scheduling efficiency and the robot system resource average load efficiency.

[0019] In a preferred embodiment, the specific method for obtaining the system stability influence coefficient is as follows: obtain several task response times in the robot task scheduling process; within the same task time interval, record the response time standard deviation and average response time of each task switching; calculate the task response time variation coefficient based on the standard deviation and average response time of the task response time; calculate the task scheduling stability influence coefficient based on the task response time variation coefficient based on a preset task scheduling stability influence coefficient calculation formula.

[0020] In a preferred embodiment, the time interval evaluation effect display model is constructed based on the output of the task switching time interval evaluation model and the corresponding task switching time interval vector, specifically: obtaining the task switching time interval evaluation coefficients corresponding to different task switching time interval vectors from the task switching time interval evaluation model; setting the X-axis to the task switching time interval vector, and the Y-axis to the corresponding task switching time interval evaluation coefficient, establishing a rectangular coordinate system, and displaying the evaluation effects under different time intervals; marking the task switching time interval evaluation coefficients corresponding to each group of task switching time interval vectors on the rectangular coordinate system to form a series of marked points, and obtaining the evaluation effect display model of the task switching time interval by connecting adjacent marked points.

[0021] In a preferred embodiment, the data analysis is performed to obtain a screened task switching time interval vector, which is applied to the comprehensive scheduling of robots of multiple brands, specifically: the maximum value of the task switching time interval evaluation coefficient corresponding to different time interval vectors is obtained from the task switching time interval evaluation effect display model; the difference between two adjacent maximum values ​​is calculated based on the second-order difference method, and then the changing speed of the difference is calculated to obtain the curvature information; the task switching time interval vector corresponding to the Y-axis maximum value with the largest changing curvature is determined as the screened task switching time interval vector, and applied to the comprehensive scheduling of robots of multiple brands.

[0022] The technical effects and advantages of the present invention's integrated robot dispatching system integrating multiple brands are as follows:

[0023] 1. The present invention uses the task acquisition module to sort tasks according to priority, resource requirements and execution requirements to ensure that key tasks are processed first; the robot division module uses performance data to gradient divide robots and reasonably allocate task loads; the task allocation module performs real-time task allocation based on robot status and task switching time interval vector to improve scheduling flexibility and efficiency; the data acquisition module optimizes the task switching time interval through the analysis of multi-robot efficiency data and system resource data, thereby improving multi-task concurrent processing efficiency, resource utilization and system stability, extending equipment service life and reducing maintenance costs. The overall technical solution can effectively improve the production efficiency and stability of the robot scheduling system, solve the resource utilization and task connection problems in the collaborative operation of multi-brand robots, and has a high practical application value.

[0024] 2. The present invention generates a task switching time interval evaluation coefficient by analyzing factors such as multi-task concurrency efficiency, system resource utilization and stability, which helps to accurately evaluate the optimal task switching timing. Furthermore, based on the output of the evaluation model, the optimal task switching time interval is screened out, thereby improving the efficiency and stability of the comprehensive scheduling of multi-brand robots and maximizing the utilization of system resources, which has significant technical advantages. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a schematic diagram of the structure of a multi-brand robot integrated scheduling system of the present invention. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0027] Embodiment 1, Figure 1 The present invention provides a robot integrated scheduling system integrating multiple brands, including a task acquisition module, a robot division module, a task allocation module, a data acquisition module, an effect evaluation module and a screening module, and there are connections between the modules.

[0028] S1, a task acquisition module, is used for the scheduling system to acquire a list of tasks to be executed from the production management system, wherein the list of tasks to be executed is sorted according to the priority, required resources and execution requirements of the tasks.

[0029] It should be noted that the task priority determines the order in which tasks are processed. Tasks with high priority need to be executed first to ensure that critical tasks are processed in a timely manner. Task priority can be evaluated based on the following factors:

[0030] Production planning: Based on production scheduling or delivery requirements, temporary urgent tasks or production bottleneck tasks should be performed first. Customer needs: External orders or high-value customer orders should be given higher priority.

[0031] Task urgency: Prioritize tasks based on their time requirements (e.g., a task must be completed by a certain time).

[0032] Task priorities are sorted from high to low. High-priority tasks enter the task queue first, and low-priority tasks can be assigned when there are idle resources.

[0033] It should be noted that in the multi-brand robot integrated scheduling system, the order of task execution is determined by priority, resource requirements and execution requirements. Tasks with higher priority will be scheduled first to ensure that critical tasks are processed in a timely manner; tasks with greater resource requirements will be allocated first according to available resources to avoid production bottlenecks caused by insufficient resources; tasks with more complex execution requirements, such as those that require high coordination or specific equipment, will be arranged in the task list for priority execution to ensure that they can be successfully completed within the appropriate time window.

[0034] For the specific task sorting process, there are many analysis methods in the prior art, which will not be described here.

[0035] S2, the robot division module, is used to divide multi-brand robots into gradient divisions based on performance, and to generate several groups of task switching time interval vectors based on computer random numbers according to the task volume of the task list and the gradient number of the robot.

[0036] The method is used to grade the multi-brand robots based on their performance, specifically:

[0037] The system collects performance data of robots of various brands through sensors and internal data interfaces, including load capacity, speed, operation accuracy, battery life, and adaptability to the working environment;

[0038] Standardize the performance data collected by all robots;

[0039] The performance data is clustered according to the density of the DBSCAN algorithm to automatically determine the number of clusters, and the number of gradients that need to be classified and the number of robots in each gradient are obtained;

[0040] Quantify the amount of tasks in the task list by the required task time;

[0041] Under the condition that the task amount remains unchanged, several groups of task switching time interval vectors are randomly generated by computer random numbers, each group of task switching time interval vectors includes the same number of task switching time intervals as the number of gradients that need to be classified, and the task switching time intervals are respectively applied to robots of corresponding gradients.

[0042] It should be noted that, while ensuring that the amount of tasks remains unchanged, a number of task switching time interval vectors are randomly generated by using computer random numbers, which means:

[0043] The total task amount of the task list (for example, the total time or total resources required for all tasks) is used as a constraint condition for generating the task switching time interval.

[0044] Set different task switching time interval vectors and ensure that each set of vectors does not affect the total task volume when applied to the robot gradient. For example, the time interval of each set of vectors can fluctuate within a certain range, so that the frequency of task switching between robots changes, but the total task volume can still be allocated;

[0045] Use random numbers to generate different time interval combinations so that each group of intervals is slightly different. This allows you to test the impact of different intervals on the efficiency of the robot.

[0046] S3, task assignment module, is used to detect the status of robots of various brands in real time, screen out several robots that meet the task requirements, and assign tasks to the robots based on several groups of task switching time interval vectors and task lists. The robot status includes position, power, and idle status.

[0047] The task assignment to the robot based on several groups of task switching time interval vectors and task lists is specifically as follows:

[0048] Match the generated task switching time interval vectors with robots of different gradients, and the time intervals in each set of vectors correspond to robots of the same gradient;

[0049] The tasks are assigned to the qualified robots in sequence according to the sorted task list, and the task switching is performed according to the set task switching time interval.

[0050] S4, a data acquisition module, is used to obtain multi-robot efficiency data and system resource data when the robots perform tasks.

[0051] The multi-robot efficiency data includes a multi-task concurrent efficiency fluctuation coefficient.

[0052] The system resource data includes a system resource utilization influence coefficient and a system stability influence coefficient.

[0053] The multi-task concurrent efficiency fluctuation coefficient is used to measure the efficiency stability of multi-brand robot systems when processing multi-task concurrency. This coefficient reflects the balance between resource allocation and task connection of the scheduling system by monitoring the concurrent processing performance of the system at different task switching time intervals. A reasonable concurrent efficiency fluctuation coefficient can help optimize the task scheduling strategy, improve the system's task throughput and processing efficiency, avoid resource waste or task accumulation caused by improper switching frequency, and ensure that the scheduling system has good resource utilization and task completion speed in high-load scenarios.

[0054] Analyzing the multi-task concurrent efficiency fluctuation coefficient is helpful for evaluating the scheduling effect under different task switching time intervals, thereby solving the problem of lack of flexibility in setting the task switching time interval in multi-brand robot systems, which easily leads to an imbalance between resource utilization and task connection efficiency in the scheduling system, thereby reducing the overall production efficiency and equipment life. It has the following advantages:

[0055] Optimize resource utilization: The multi-task concurrent efficiency fluctuation coefficient helps identify the level of resource utilization at different time intervals. By selecting a more reasonable time interval, it effectively reduces the waste caused by idle or excessive resource occupation, and achieves the best allocation of system resources.

[0056] Improve task connection efficiency: By evaluating the impact of the switching time interval on task connection, this coefficient can help the scheduling system find a smoother task switching solution, thereby reducing the waiting and response delays caused by task switching and improving the overall task execution consistency of the system.

[0057] Enhance system flexibility: In a multi-brand robot system, flexible setting of task switching time intervals helps to cope with different task loads and priority requirements, avoid overly fixed scheduling strategies that lead to rigid system scheduling capabilities, and thus improve multi-brand collaboration efficiency.

[0058] Extending equipment life: Appropriate task switching time intervals can avoid equipment wear caused by frequent switching or waste of resources. In addition, continuous monitoring of the multi-task concurrent efficiency fluctuation coefficient keeps the equipment within a healthy load range, thereby extending its service life and reducing maintenance costs.

[0059] By analyzing the fluctuation coefficient of multi-task concurrent efficiency, the task switching time interval can be dynamically adjusted to balance resource utilization and task connection efficiency, ensuring that the system has a long equipment life cycle and low operation and maintenance costs while running efficiently.

[0060] The specific method for obtaining the multi-task concurrent efficiency fluctuation coefficient is as follows:

[0061] Get the total battery capacity of the robot group;

[0062] Record the response time of the robot group after receiving each task, and calculate the average response time;

[0063] The ratio of the total battery capacity to the average response time is used as the task processing capability value of the robot group;

[0064] According to the preset task execution efficiency and recovery efficiency, the task processing efficiency factor is calculated;

[0065] Calculate the multi-task concurrent efficiency fluctuation coefficient based on the efficiency factor and task processing capability value;

[0066] The specific calculation formula of the multi-task concurrent efficiency fluctuation coefficient is as follows:

[0067]

[0068] In the formula, is the multi-task concurrent efficiency fluctuation coefficient, is the number of responses recorded, is the jth response time, For task execution efficiency, To restore efficiency, is the battery capacity of the ith group, is the average battery capacity for the ith time, and n is the total number of times the battery capacity is obtained.

[0069] The system resource utilization impact coefficient is used to measure the impact of the task switching interval on resource utilization in a multi-brand robot system. A reasonable task switching interval can ensure that the robot has appropriate recovery time between tasks, thereby optimizing resource utilization; an interval that is too short may cause the robot to be put into the next task without fully recovering, increasing the risk of equipment fatigue and failure; an interval that is too long may cause the robot to be idle, reducing the effective utilization of resources. By monitoring this coefficient, the impact of different task switching intervals on robot utilization and task success rate can be evaluated, helping the system to strike a balance between resource utilization and task efficiency, thereby improving overall production efficiency and extending equipment life.

[0070] Analyzing the impact coefficient of system resource utilization can evaluate the scheduling effect under different task switching time intervals, thereby solving the problem of lack of flexibility in setting the task switching time interval in multi-brand robot systems, which easily leads to an imbalance between resource utilization and task connection efficiency in the scheduling system, thereby reducing the overall production efficiency and equipment life. It has the following advantages:

[0071] Improve concurrent task processing efficiency: This coefficient can reflect the fluctuation of the system's processing capacity under multi-task concurrent state. By selecting the optimal task switching interval, the system's concurrent task processing capacity can be improved, task accumulation can be reduced, and smooth task connection can be ensured.

[0072] Optimize resource utilization: The multi-task concurrent efficiency fluctuation coefficient helps to avoid the problem of resource utilization imbalance caused by unreasonable task switching time interval settings. Reasonable interval settings can balance the resource consumption of robots of different brands under different tasks and maximize the use of system resources.

[0073] Extend equipment service life: Reasonable task switching time interval helps reduce robot fatigue loss under high-frequency task switching, thereby extending the service life of the equipment and improving overall production efficiency.

[0074] Improve system stability: By evaluating the fluctuation coefficient of multi-task concurrent efficiency, we can better adjust the task switching interval and reduce the fluctuation of the system under high concurrency conditions, so that the system can still maintain stable operation during peak load.

[0075] The specific method for obtaining the system resource utilization influence coefficient is as follows:

[0076] Obtain robot system resource ratio data, wherein the robot system resource ratio data includes processing capacity, storage capacity and network bandwidth, and calculate the average value of the system resource load of the robot except for task resources when performing a task according to the robot system resource ratio data;

[0077] The average resource scheduling efficiency of the robot is calculated according to the preset average resource scheduling efficiency calculation formula; the average load efficiency of the robot system resources is calculated according to the preset system resource average load efficiency calculation formula combined with the average value of the system resource load;

[0078] The system resource utilization impact coefficient is calculated based on the ratio of the robot's average resource scheduling efficiency and the robot system resource average load efficiency.

[0079] The calculation formula for the robot average resource scheduling efficiency is specifically:

[0080]

[0081] The calculation formula for the average load efficiency of the robot system resources is specifically:

[0082]

[0083] The specific calculation formula of the system resource utilization impact coefficient is as follows:

[0084]

[0085] In the formula, is the average resource scheduling efficiency, is the average load efficiency of system resources, is the average value of system resource load, is the number of robots that execute tasks in the scheduling system when the i-th robot is connected, is the system resource utilization impact coefficient, The network bandwidth of the ith robot, is the weighted average of the processing power and storage capacity of the ith robot, is the core resource requirement of the robot's i-th task, is the total resource requirement of the robot for the i-th task.

[0086] It should be noted that It reflects the resource consumption of the robot that exceeds the core task requirements when completing the i-th task; represents the minimum amount of resources required by the robot to complete the core task function when performing the i-th task; it only includes the resource requirements of the core operation part of the task, and does not include possible system overhead or redundant resources. This part of resources is usually a necessary condition for the successful execution of the task; It represents the minimum amount of resources required by the robot to complete the core task function when executing the i-th task. It only includes the resource requirements of the core operation part of the task, and does not include possible system overhead or redundant resources. This part of resources is usually a necessary condition for the successful execution of the task.

[0087] The system stability impact coefficient is used to measure the change in system stability during the robot's task switching process. The setting of the task switching time interval directly affects the robot's recovery time. If the interval is too short, the robot may be put into the next task without completing sufficient preparation, which will cause system instability and increase the risk of task failure. On the contrary, if the interval is too long, the robot will be idle for a long time, wasting valuable resources and may cause task delays. By evaluating the system stability impact coefficient, the task switching time interval can be reasonably set to optimize the robot resource utilization, improve the task execution success rate, avoid the system's stability degradation caused by excessive switching or idleness, and thus improve the overall production efficiency and equipment life.

[0088] Analyzing the system stability impact coefficient is helpful for evaluating the scheduling effect under different task switching time intervals, thereby solving the problem of lack of flexibility in setting the task switching time interval in multi-brand robot systems, which easily leads to an imbalance between resource utilization and task connection efficiency in the scheduling system, thereby reducing the overall production efficiency and equipment life. It has the following advantages:

[0089] Improve system stability: By properly setting the task switching time interval, you can avoid the robot resources not being fully restored due to too short an interval, thereby reducing system instability and improving the robot's work efficiency and the success rate of task execution.

[0090] Optimize resource utilization: Appropriate time intervals can reduce robot idle time, improve robot utilization, ensure that resources are fully utilized, and avoid idle time waste caused by too long task switching intervals.

[0091] Enhance task connection efficiency: Reasonable time intervals help to smoothly transition between tasks and avoid improper task connection due to too short or too long time intervals, thereby improving overall scheduling efficiency and production line smoothness.

[0092] Extend equipment life: Reasonable task switching time intervals can help reduce excessive load and frequent switching of the robot, thereby reducing equipment wear and failure rate and extending the service life of the equipment.

[0093] Improve overall production efficiency: By optimizing the time interval, ensure that the robot can perform tasks in the best condition, avoid invalid switching and task delays, and ultimately improve the overall production efficiency and stability of task execution.

[0094] By analyzing the coefficients affecting system stability, the task switching time interval can be accurately adjusted to achieve efficient scheduling of multi-brand robot systems, balance resource utilization and task connection efficiency, and reduce the negative impact of system imbalance.

[0095] The specific method for obtaining the system stability influence coefficient is as follows:

[0096] Obtain the response time of several tasks in the robot task scheduling process;

[0097] In the same task time interval, record the standard deviation and average response time of each task switch;

[0098] According to the standard deviation of task response time and the average response time, the coefficient of variation of task response time is calculated;

[0099] The task response time variation coefficient is calculated based on the preset task scheduling stability influence coefficient calculation formula to calculate the task scheduling stability influence coefficient.

[0100] The specific calculation formula of the task response time variation coefficient is as follows:

[0101]

[0102] The specific calculation formula of the system stability influence coefficient is as follows:

[0103]

[0104] In the formula, is the coefficient of variation of task response time, is the task response time at the jth moment, is the total amount of data obtained during the robot task scheduling process, and j is the time label; is the system stability influence coefficient.

[0105] This embodiment uses the task acquisition module to sort tasks according to priority, resource requirements and execution requirements to ensure that key tasks are processed first; the robot division module uses performance data to gradient divide robots and reasonably allocate task loads; the task allocation module performs real-time task allocation based on robot status and task switching time interval vector to improve scheduling flexibility and efficiency; the data acquisition module optimizes the task switching time interval through the analysis of multi-robot efficiency data and system resource data, thereby improving multi-task concurrent processing efficiency, resource utilization and system stability, extending equipment service life and reducing maintenance costs. The overall technical solution can effectively improve the production efficiency and stability of the robot scheduling system, solve the resource utilization and task connection problems in the collaborative operation of multi-brand robots, and has a high practical application value.

[0106] Embodiment 2, S5, effect evaluation module, is used to construct a task switching time interval evaluation model based on machine learning according to multi-robot efficiency data and system resource data.

[0107] According to the multi-robot efficiency data and system resource data, a task switching time interval evaluation model is built based on machine learning, specifically:

[0108] The obtained multi-task concurrent efficiency fluctuation coefficient, system resource utilization influence coefficient and system stability influence coefficient are used to establish a task switching time interval evaluation model to generate a task switching time interval evaluation coefficient;

[0109] The calculation formula of the task switching time interval evaluation coefficient is as follows:

[0110]

[0111] In the formula, is the task switching time interval evaluation coefficient, It is the preset proportional coefficient of the multi-task concurrent efficiency fluctuation coefficient. is the preset proportional coefficient of the system resource utilization impact coefficient, is the preset proportional coefficient of the system stability influence coefficient, is the multi-task concurrent efficiency fluctuation coefficient, is the system resource utilization impact coefficient, is the system stability influence coefficient.

[0112] S6, a screening module, is used to construct a time interval evaluation effect display model based on the output of the task switching time interval evaluation model and the corresponding task switching time interval vector, and perform data analysis to obtain the screened task switching time interval vector, which is applied to the comprehensive scheduling of robots of multiple brands.

[0113] The time interval evaluation effect display model is constructed according to the output of the task switching time interval evaluation model and the corresponding task switching time interval vector, specifically:

[0114] Obtaining task switching time interval evaluation coefficients corresponding to different task switching time interval vectors from a task switching time interval evaluation model;

[0115] Set the X-axis as the task switching time interval vector, and the Y-axis as the corresponding task switching time interval evaluation coefficient, and establish a rectangular coordinate system to show the evaluation results under different time intervals;

[0116] The task switching time interval evaluation coefficient corresponding to each group of task switching time interval vectors is marked on the rectangular coordinate system to form a series of marking points, and the evaluation effect display model of the task switching time interval is obtained by connecting adjacent marking points.

[0117] The data analysis is performed to obtain the screened task switching time interval vector, which is applied to the comprehensive scheduling of robots of multiple brands, specifically: obtaining the maximum value of the task switching time interval evaluation coefficient corresponding to different time interval vectors from the task switching time interval evaluation effect display model;

[0118] The difference between two adjacent maximum values ​​is calculated based on the second-order difference method, and then the changing speed of the difference is calculated to obtain the curvature information;

[0119] The task switching time interval vector corresponding to the maximum value of the Y-axis with the largest change curvature is determined as the screened task switching time interval vector and applied to the comprehensive scheduling of robots of multiple brands.

[0120] This embodiment generates a task switching time interval evaluation coefficient by analyzing factors such as multi-task concurrency efficiency, system resource utilization and stability, helping to accurately evaluate the best task switching timing. Furthermore, based on the output of the evaluation model, the optimal task switching time interval is screened out, thereby improving the efficiency and stability of the comprehensive scheduling of multi-brand robots and maximizing the utilization of system resources, which has significant technical advantages.

[0121] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0122] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0123] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0124] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0125] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0126] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A robot integrated dispatching system integrating multiple brands, characterized in that: It includes task acquisition module, robot division module, task allocation module, data acquisition module, effect evaluation module and screening module, and there are connections between the modules; A task acquisition module is used for the scheduling system to acquire a list of tasks to be executed from the production management system, wherein the list of tasks to be executed is sorted according to the priority, required resources and execution requirements of the tasks; The robot division module is used to divide multi-brand robots into gradient divisions based on performance, and to generate several groups of task switching time interval vectors based on computer random numbers according to the task volume of the task list and the gradient number of the robots; The task assignment module is used to detect the status of robots of various brands in real time, select several robots that meet the task requirements, and assign tasks to the robots based on several groups of task switching time interval vectors and task lists. The robot status includes position, power, and idle status. A data acquisition module is used to obtain multi-robot efficiency data and system resource data when the robots perform tasks; The multi-robot efficiency data includes a multi-task concurrent efficiency fluctuation coefficient; The system resource data includes a system resource utilization influence coefficient and a system stability influence coefficient; An effect evaluation module is used to establish a task switching time interval evaluation model based on the obtained multi-task concurrent efficiency fluctuation coefficient, system resource utilization influence coefficient and system stability influence coefficient, and generate a task switching time interval evaluation coefficient; The multi-task concurrent efficiency fluctuation coefficient is used to measure the efficiency stability of multi-brand robot systems when processing multi-task concurrently; The system resource utilization impact coefficient is used to measure the impact of the task switching time interval on resource utilization in multi-brand robot systems; The system stability impact coefficient is used to measure the change of system stability during the robot's task switching process; The screening module is used to construct a time interval evaluation effect display model based on the task switching time interval evaluation coefficient and the corresponding task switching time interval vector, and perform data analysis to obtain the screened task switching time interval vector, which is applied to the comprehensive scheduling of robots of multiple brands.

2. The multi-brand robot integrated dispatching system according to claim 1 is characterized in that: The method is used to grade the multi-brand robots based on their performance, specifically: The system collects performance data of robots of various brands through sensors and internal data interfaces, including load capacity, speed, operation accuracy, battery life, and adaptability to the working environment; Standardize the performance data collected by all robots; The performance data is clustered according to the density of the DBSCAN algorithm to automatically determine the number of clusters, and the number of gradients that need to be classified and the number of robots in each gradient are obtained; Quantify the amount of tasks in the task list by the required task time; Under the condition that the task amount remains unchanged, several groups of task switching time interval vectors are randomly generated by computer random numbers, each group of task switching time interval vectors includes the same number of task switching time intervals as the number of gradients that need to be classified, and the task switching time intervals are respectively applied to robots of corresponding gradients.

3. The multi-brand robot integrated dispatching system according to claim 2 is characterized in that: The task assignment to the robot based on several groups of task switching time interval vectors and task lists is specifically as follows: Match the generated task switching time interval vectors with robots of different gradients, and the time intervals in each set of vectors correspond to robots of the same gradient; Tasks are assigned to qualified robots in sequence according to the sorted task list, and task switching is performed according to the set task switching time interval.

4. The multi-brand robot integrated dispatching system according to claim 3 is characterized in that: The multi-robot efficiency data includes a multi-task concurrent efficiency fluctuation coefficient; the specific method for obtaining the multi-task concurrent efficiency fluctuation coefficient is as follows: Get the total battery capacity of the robot group; Record the response time of the robot group after receiving each task, and calculate the average response time; The ratio of the total battery capacity to the average response time is used as the task processing capability value of the robot group; According to the preset task execution efficiency and recovery efficiency, the task processing efficiency factor is calculated; The multi-task concurrent efficiency fluctuation coefficient is calculated based on the efficiency factor and task processing capability value.

5. The multi-brand robot integrated dispatching system according to claim 4 is characterized in that: The system resource data includes a system resource utilization influence coefficient and a system stability influence coefficient; the specific method for obtaining the system resource utilization influence coefficient is as follows: Obtain robot system resource ratio data, wherein the robot system resource ratio data includes processing capacity, storage capacity and network bandwidth, and calculate the average value of the system resource load of the robot except for task resources when performing a task according to the robot system resource ratio data; Calculate the average resource scheduling efficiency of the robot according to the preset average resource scheduling efficiency calculation formula; The average load efficiency of the robot system resources is calculated according to the preset system resource average load efficiency calculation formula combined with the average value of the system resource load; The system resource utilization impact coefficient is calculated based on the ratio of the robot's average resource scheduling efficiency and the robot system resource average load efficiency; The calculation formula for the robot average resource scheduling efficiency is specifically: ； The calculation formula for the average load efficiency of robot system resources is as follows: In the formula, is the average resource scheduling efficiency, The network bandwidth of the ith robot, is the weighted average of the processing power and storage capacity of the ith robot, is the core resource requirement of the robot's i-th task, is the total resource requirement of the robot for the i-th task, is the average load efficiency of system resources, is the average value of system resource load, The number of robots that execute tasks in the scheduling system when the i-th robot is connected.

6. The multi-brand integrated robot dispatching system according to claim 5 is characterized in that: The specific method for obtaining the system stability influence coefficient is as follows: Obtain the response time of several tasks in the robot task scheduling process; In the same task time interval, record the standard deviation and average response time of each task switch; According to the standard deviation of task response time and the average response time, the coefficient of variation of task response time is calculated; The task response time variation coefficient is calculated based on a preset system stability influence coefficient calculation formula to calculate the system stability influence coefficient; The specific calculation formula of the system stability influence coefficient is as follows: In the formula, is the coefficient of variation of task response time, is the system stability influence coefficient.

7. The multi-brand robot integrated dispatching system according to claim 6 is characterized in that: The task switching time interval evaluation coefficient and the corresponding task switching time interval vector are used to construct a time interval evaluation effect display model, specifically: Obtaining task switching time interval evaluation coefficients corresponding to different task switching time interval vectors from a task switching time interval evaluation model; Set the X-axis as the task switching time interval vector, and the Y-axis as the corresponding task switching time interval evaluation coefficient, establish a rectangular coordinate system, and show the evaluation results under different time intervals; The task switching time interval evaluation coefficient corresponding to each group of task switching time interval vectors is marked on the rectangular coordinate system to form a series of marking points, and the evaluation effect display model of the task switching time interval is obtained by connecting adjacent marking points.

8. The multi-brand robot integrated dispatching system according to claim 7 is characterized in that: The data analysis is performed to obtain the screened task switching time interval vector, which is applied to the comprehensive scheduling of robots of multiple brands, specifically: obtaining the maximum value of the task switching time interval evaluation coefficient corresponding to different time interval vectors from the task switching time interval evaluation effect display model; The difference between two adjacent maximum values ​​is calculated based on the second-order difference method, and then the changing speed of the difference is calculated to obtain the curvature information; The task switching time interval vector corresponding to the maximum value of the Y-axis with the largest change curvature is determined as the screened task switching time interval vector and applied to the comprehensive scheduling of robots of multiple brands.

9. The multi-brand integrated robot dispatching system according to claim 8 is characterized in that: The calculation formula of the task switching time interval evaluation coefficient is as follows: In the formula, is the task switching time interval evaluation coefficient, It is the preset proportional coefficient of the multi-task concurrent efficiency fluctuation coefficient. is the preset proportional coefficient of the system resource utilization impact coefficient, is the preset proportional coefficient of the system stability influence coefficient, is the multi-task concurrent efficiency fluctuation coefficient, is the system resource utilization impact coefficient, is the system stability influence coefficient.

10. The multi-brand integrated robot dispatching system according to claim 9 is characterized in that: The specific calculation formula of the system resource utilization impact coefficient is as follows: In the formula, is the system resource utilization impact coefficient.

Images