Dynamic task allocation method for humanoid robot in industrial scene and related equipment

Abstract

The embodiment of the invention provides a humanoid robot dynamic task allocation method in an industrial scene and related equipment, and belongs to the technical field of humanoid robots. The method comprises the steps of obtaining a task load level of a current industrial scene; the task load level comprises a low load level, a medium load level and a high load level; determining a corresponding task allocation strategy based on the task load level; the task allocation strategy comprises an event trigger allocation strategy, an improved Hungary algorithm allocation strategy and a hybrid heuristic algorithm allocation strategy; carrying out task allocation on the humanoid robot based on the determined task allocation strategy; the humanoid robot task allocation comprises task-robot matching, task sequence planning and cooperation scheme generation. According to the method, the distribution strategies can be dynamically switched in different load scenes, the real-time performance and global optimization are both considered, and the utilization rate of the robot and the industrial scene cooperation efficiency are improved.

Description

Technical Field

[0001] This application relates to the field of humanoid robot technology, and in particular to a method and related equipment for dynamic task allocation of humanoid robots in industrial scenarios. Background Technology

[0002] In related technologies, the task allocation schemes for humanoid robots have the following limitations:

[0003] 1. Static allocation mode is the main mode: Most systems use preset task sequences or fixed priority rules, which can lead to some robots being overloaded while others are idle during high loads, resulting in low utilization.

[0004] 2. Poor adaptability of single algorithms: Most related technologies use a single scheduling algorithm, which makes it difficult to balance real-time performance and global optimization.

[0005] 3. Insufficient consideration of collaboration factors: In industrial scenarios, multiple robots frequently collaborate (such as jointly transporting heavy workpieces), resulting in low efficiency and high conflict rate in related technical collaboration.

[0006] 4. Lack of adaptation to humanoid robot characteristics: Humanoid robots in related technologies have high energy consumption and high complexity in multi-joint motion planning, resulting in a mismatch between task allocation schemes and actual execution capabilities.

[0007] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the Invention

[0008] The main objective of this application is to propose a dynamic task allocation method and related equipment for humanoid robots in industrial scenarios. This method can combine the multi-degree-of-freedom and multi-skill characteristics of humanoid robots to dynamically switch allocation strategies under different load scenarios, taking into account both real-time performance and global optimization, thereby improving robot utilization and industrial scenario collaboration efficiency.

[0009] To achieve the above objectives, one aspect of this application proposes a method for dynamic task allocation of humanoid robots in industrial scenarios, the method comprising the following steps:

[0010] Obtain the task load level of the current industrial scenario; the task load level includes low load level, medium load level, and high load level;

[0011] The corresponding task allocation strategy is determined based on the task load level; the task allocation strategy includes an event-triggered allocation strategy, an improved Hungarian algorithm allocation strategy, and a hybrid heuristic algorithm allocation strategy.

[0012] Humanoid robot tasks are assigned based on the determined task allocation strategy; the humanoid robot task allocation includes task-robot matching, task sequence planning, and collaborative scheme generation.

[0013] In some embodiments, obtaining the task load level of the current industrial scenario includes:

[0014] The task load index is calculated by obtaining the number of tasks to be assigned per unit time, the average task complexity coefficient, and the robot idle rate. The formulas used include:

[0015] L=α×N / T+β×C-γ×I;

[0016] Wherein, L is the task load index; N is the number of tasks to be assigned per unit time; T is the time window in hours; C is the average task complexity coefficient, ranging from 1 to 5; I is the robot idle rate, ranging from 0 to 1; α, β, and γ are weighting coefficients and α+β+γ=1.

[0017] The load level is determined based on the task load index: when L < 0.3, it is a low load level; when 0.3 ≤ L ≤ 0.7, it is a medium load level; and when L > 0.7, it is a high load level.

[0018] In some embodiments, determining the corresponding task allocation strategy based on the task load level includes:

[0019] When the task load level is low, the task allocation strategy is determined to be an event-triggered allocation strategy.

[0020] When the task load level is medium load level, the task allocation strategy is determined to be the improved Hungarian algorithm allocation strategy;

[0021] When the task load level is high, the task allocation strategy is determined to be a hybrid heuristic algorithm allocation strategy.

[0022] In some embodiments, the task allocation for the humanoid robot based on the determined task allocation strategy includes:

[0023] When the task allocation strategy is an event-triggered allocation strategy, monitor task triggering events;

[0024] If a new task is detected, the skill matching degree, movement cost from the current location to the task point, and remaining battery percentage of each idle robot are obtained. The overall score for each robot is then calculated using the following formulas:

[0025] S=ω1×M+ω2×(1 / C)+ω3×E;

[0026] Where S is the robot's overall score; ω1, ω2, and ω3 are weighting coefficients; M is the skill matching degree, ranging from 0 to 1; C is the movement cost; and E is the remaining battery percentage, ranging from 0 to 1.

[0027] The new task is assigned to the idle robot with the highest overall score, thus completing the task-robot matching.

[0028] In some embodiments, the task allocation for the humanoid robot based on the determined task allocation strategy includes:

[0029] When the task allocation strategy is the improved Hungarian algorithm allocation strategy, collect all currently unassigned tasks and idle robots;

[0030] Based on the information of the tasks to be assigned and the idle robots, a cost matrix is ​​constructed; the cost matrix includes several matrix elements.

[0031] The formulas used to calculate the elements of the matrix include:

[0032] c ij =λ1×t ij +λ2×e ij +λ3×s ij ;

[0033] Among them, c ij Let be an element of the matrix, representing the total cost of the i-th robot performing the j-th task; t ij The estimated time for the i-th robot to perform the j-th task; e ij For relative energy consumption cost; s ij The cost of skill adaptation; λ1, λ2, and λ3 are weighting coefficients, and λ1+λ2+λ3=1;

[0034] The cost matrix is ​​solved using an improved Hungarian algorithm to obtain the globally optimal task-robot matching scheme.

[0035] If the number of tasks to be assigned is greater than the number of idle robots, the tasks to be assigned are prioritized and the higher-priority tasks are assigned first, thus completing the task sequence planning.

[0036] In some embodiments, the task allocation for the humanoid robot based on the determined task allocation strategy includes:

[0037] When the task allocation strategy is a hybrid heuristic algorithm allocation strategy, the task queue and robot status are obtained; the task queue and robot status include position, battery level, and current task sequence;

[0038] Based on the task queue and robot status, an initial task allocation scheme is generated using a greedy algorithm, and tasks are allocated to the robot with the lowest current load in order of priority; the load is the total estimated execution time of the tasks already assigned to the robot.

[0039] The initial task allocation scheme is optimized based on a genetic algorithm to generate a cooperative scheme; the genetic algorithm includes setting encoding and fitness functions, and performing selection, crossover and mutation operations.

[0040] In some embodiments, the hybrid heuristic algorithm includes:

[0041] Implement a dynamic adjustment mechanism; the dynamic adjustment mechanism includes:

[0042] Monitor new tasks or robot malfunction events;

[0043] When a new task is added or a robot malfunction occurs, the perturbation coefficient is calculated using the following formula:

[0044] δ=ΔT / T0;

[0045] Where δ is the disturbance coefficient; ΔT is the number of new tasks; and T0 is the total number of original tasks.

[0046] When T0 = 0, if ΔT ≥ 1, then δ = 1.0;

[0047] If δ>0.2, trigger re-optimization; otherwise, only make local adjustments to the new task and insert it into the existing sequence.

[0048] To achieve the above objectives, another aspect of this application proposes a dynamic task allocation system for humanoid robots in industrial scenarios, used to implement the method described above. The system includes:

[0049] The first module is used to obtain the task load level of the current industrial scenario; the task load level includes low load level, medium load level and high load level;

[0050] The second module is used to determine the corresponding task allocation strategy based on the task load level; the task allocation strategy includes an event-triggered allocation strategy, an improved Hungarian algorithm allocation strategy, and a hybrid heuristic algorithm allocation strategy.

[0051] The third module is used to allocate tasks to the humanoid robot based on the determined task allocation strategy; the humanoid robot task allocation includes task-robot matching, task sequence planning, and collaborative scheme generation.

[0052] To achieve the above objectives, another aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described above.

[0053] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the methods described above.

[0054] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer program product, including a computer program that, when executed by a processor, implements the aforementioned method.

[0055] The embodiments of this application include at least the following beneficial effects: This application provides a method, system, electronic device, storage medium, and program product for dynamic task allocation of humanoid robots in industrial scenarios. The invention includes: obtaining the task load level of the current industrial scenario; the task load level includes low load level, medium load level, and high load level; determining a corresponding task allocation strategy based on the task load level; the task allocation strategy includes an event-triggered allocation strategy, an improved Hungarian algorithm allocation strategy, and a hybrid heuristic algorithm allocation strategy; allocating humanoid robot tasks based on the determined task allocation strategy; the humanoid robot task allocation includes task-robot matching, task sequence planning, and collaborative scheme generation. This invention can dynamically switch allocation strategies under different load scenarios, balancing real-time performance and global optimization, thereby improving robot utilization and collaborative efficiency in industrial scenarios. Attached Figure Description

[0056] Figure 1 This is a flowchart of the dynamic task allocation method for humanoid robots in industrial scenarios provided in the embodiments of this application;

[0057] Figure 2 This is a data flow diagram of the dynamic task allocation method for humanoid robots in industrial scenarios provided in the embodiments of this application;

[0058] Figure 3 This is a task load level determination logic diagram provided in the embodiments of this application;

[0059] Figure 4 This is a flowchart of the event triggering allocation strategy provided in the embodiments of this application;

[0060] Figure 5 This is a flowchart of the improved Hungarian algorithm allocation strategy provided in the embodiments of this application;

[0061] Figure 6 This is a flowchart of the hybrid heuristic algorithm allocation strategy provided in the embodiments of this application;

[0062] Figure 7 This is a timing diagram of a multi-robot collaborative task provided in an embodiment of this application;

[0063] Figure 8This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0064] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit it. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0065] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0066] Before providing a detailed description of the embodiments of this application, some of the nouns and terms involved in the embodiments of this application will be explained first. The nouns and terms involved in the embodiments of this application are subject to the following interpretations.

[0067] 1) IIoT, Industrial Internet of Things.

[0068] In related technologies, the task allocation schemes for humanoid robots have the following limitations:

[0069] 1. Static allocation mode is the main mode: Most systems use preset task sequences or fixed priority rules, which cannot be dynamically adjusted according to real-time task load and robot status (such as battery level or fault). This results in some robots being overloaded while others are idle during high load, leading to low utilization.

[0070] 2. Poor adaptability of single algorithms: Existing methods mostly use a single scheduling algorithm (such as a pure greedy algorithm or a global optimization algorithm), which makes it difficult to balance real-time performance and global optimization. For example, greedy algorithms have a fast response but are prone to getting trapped in local optima; global optimization algorithms (such as the Hungarian algorithm) have a significantly increased computation time when the number of tasks is large, and cannot meet the real-time requirements of industrial scenarios.

[0071] 3. Insufficient consideration of collaboration factors: In industrial scenarios, multiple robots frequently collaborate (such as jointly transporting heavy workpieces). Existing allocation strategies often neglect the complementarity of skills, positional coordination, and task dependencies between robots, resulting in low collaboration efficiency and high conflict rate.

[0072] 4. Lack of adaptation to humanoid robot characteristics: The unique attributes of humanoid robots, such as high energy consumption, high complexity of multi-joint motion planning, and time-consuming skill module switching, are not fully considered, resulting in a mismatch between task allocation schemes and actual execution capabilities.

[0073] In view of this, this application provides a task allocation method that can adapt to dynamic industrial scenarios, balance real-time performance and global optimization, and is suitable for the characteristics of humanoid robots, aiming to achieve the following objectives:

[0074] 1. To address the dynamic task load in industrial scenarios, a multi-stage task allocation mechanism is constructed, employing adaptive algorithm strategies under different load levels;

[0075] 2. Optimize the cost model for task allocation by combining the humanoid robot's motion characteristics, skill system, and collaboration requirements;

[0076] 3. Improve the real-time performance and global optimization of task allocation to increase robot utilization and overall production efficiency.

[0077] Figure 1 This is an optional flowchart of the dynamic task allocation method for humanoid robots in industrial scenarios provided in this application embodiment. Figure 1 The method may include, but is not limited to, steps S101 to S103.

[0078] Step S101: Obtain the task load level of the current industrial scenario; the task load level includes low load level, medium load level and high load level;

[0079] Step S102: Determine the corresponding task allocation strategy based on the task load level; the task allocation strategy includes event-triggered allocation strategy, improved Hungarian algorithm allocation strategy, and hybrid heuristic algorithm allocation strategy.

[0080] Step S103: Assign tasks to the humanoid robot based on the determined task allocation strategy; the humanoid robot task allocation includes task-robot matching, task sequence planning, and collaborative scheme generation.

[0081] Steps S101 to S103 of this application embodiment, through a dynamic task allocation method for humanoid robots in industrial scenarios, aim to solve the problem that the static allocation mode of existing industrial humanoid robot task scheduling systems is difficult to adapt to complex dynamic scenarios, resulting in low robot utilization and poor collaboration efficiency. To this end, the method of this application includes: obtaining the task load level of the current industrial scenario, including three levels: low load, medium load, and high load; determining the corresponding task allocation strategy based on the task load level, including an event-triggered allocation strategy, an improved Hungarian algorithm allocation strategy, and a hybrid heuristic algorithm allocation strategy; and allocating tasks to the humanoid robot based on the determined task allocation strategy. Through the above implementation, combined with the multi-degree-of-freedom and multi-skill characteristics of humanoid robots, the allocation strategy is dynamically switched under different load scenarios, taking into account both real-time performance and global optimization, thereby improving robot utilization and collaboration efficiency in industrial scenarios.

[0082] In some embodiments, step S101 may include, but is not limited to, steps S111 to S112:

[0083] Step S111: Obtain the number of tasks to be assigned per unit time, the average task complexity coefficient, and the robot idle rate; calculate the task load index using the following formulas:

[0084] L=α×N / T+β×C-γ×I;

[0085] Where L is the task load index; N is the number of tasks to be assigned per unit time; T is the time window in hours; C is the average task complexity coefficient, ranging from 1 to 5; I is the robot idle rate, ranging from 0 to 1; α, β, and γ are weighting coefficients and α+β+γ=1.

[0086] Step S112: Determine the load level based on the task load index: when L<0.3, it is a low load level; when 0.3≤L≤0.7, it is a medium load level; and when L>0.7, it is a high load level.

[0087] In step S111 of some embodiments, a comprehensive mathematical formula is used to combine three key variables (number of tasks, task difficulty, and system idle state) with preset weighting coefficients to calculate the task load index (L). This makes system state assessment measurable and comparable, providing an objective data basis for subsequent automated decision-making and avoiding the arbitrariness of human judgment.

[0088] In step S112 of some embodiments, the continuous load index is discretized into three distinct levels (low, medium, and high), which makes the scheduling strategy highly refined and scenario-based, ensuring that the system can adopt the most appropriate strategy under load conditions.

[0089] In some embodiments, step S102 may include, but is not limited to, steps S201 to S203:

[0090] Step S201: When the task load level is low, determine the task allocation strategy as the event-triggered allocation strategy.

[0091] Step S202: When the task load level is medium load level, the task allocation strategy is determined to be the improved Hungarian algorithm allocation strategy.

[0092] Step S203: When the task load level is high, determine the task allocation strategy as a hybrid heuristic algorithm allocation strategy.

[0093] In steps S201-S203 of some embodiments, an intelligent and flexible adaptive task allocation method is constructed. In the present invention, the most suitable strategy can be switched when facing different load levels.

[0094] In some embodiments, step S103 may include, but is not limited to, steps S301 to S303:

[0095] Step S301: When the task allocation strategy is an event-triggered allocation strategy, monitor the task triggering event;

[0096] Step S302: If a new task is detected, obtain the skill matching degree, movement cost from the current position to the task point, and remaining battery percentage for each idle robot, and calculate the comprehensive score for each robot using the following formulas:

[0097] S=ω1×M+ω2×(1 / C)+ω3×E;

[0098] Where S is the robot's overall score; ω1, ω2, and ω3 are weighting coefficients; M is the skill matching degree, ranging from 0 to 1; C is the movement cost; and E is the remaining battery percentage, ranging from 0 to 1.

[0099] Step S303: Assign the new task to the idle robot with the highest overall score to complete the task-robot matching.

[0100] In steps S301-S303 of some embodiments, in the comprehensive scoring formula (event triggering strategy), ω1+ω2+ω3=1, and the unit of C is meters. This embodiment of the invention avoids the situation where a few robots quickly run out of power due to being constantly selected, while other robots remain idle for extended periods, thus helping to extend the overall working time of the robot fleet.

[0101] In some embodiments, step S103 may include, but is not limited to, steps S304 to S307:

[0102] Step S304: When the task allocation strategy is the improved Hungarian algorithm allocation strategy, collect all currently unassigned tasks and idle robots;

[0103] Step S305: Construct a cost matrix based on the information of the tasks to be assigned and the idle robots; the cost matrix includes several matrix elements.

[0104] The formulas used to calculate matrix elements include:

[0105] c ij =λ1×t ij +λ2×e ij +λ3×s ij ;

[0106] Among them, c ij Let be a matrix element, representing the total cost of the i-th robot performing the j-th task; t ij The estimated time for the i-th robot to perform the j-th task; e ij For relative energy consumption cost; s ij The cost of skill adaptation; λ1, λ2, and λ3 are weighting coefficients, and λ1+λ2+λ3=1;

[0107] Step S306: Solve the cost matrix based on the improved Hungarian algorithm to obtain the globally optimal task-robot matching scheme;

[0108] Step S307: If the number of tasks to be assigned is greater than the number of idle robots, the tasks to be assigned are prioritized and the higher priority tasks are assigned first, thus completing the task sequence planning.

[0109] In steps S304-S305 of some embodiments, the cost matrix M is constructed. m×n Where m is the number of robots, n is the number of tasks, and the matrix element c ij Let c represent the total cost of robot i performing task j, and the total cost formula (improved Hungarian algorithm) is: ij =λ1×t ij +λ2×e ij +λ3×s ij , where t ij e is the estimated time (in seconds) for robot i to perform task j. ij The relative energy cost (ranging from 0 to 1, with higher values ​​indicating higher energy consumption) is s. ij The skill adaptation cost (ranging from 0 to 1, with lower values ​​indicating better adaptation) is λ1 + λ2 + λ3 = 1.

[0110] The relative energy consumption cost e of the embodiments of the present invention ij Value description:

[0111] e ij The relative energy consumption level used to quantify the energy consumption of robot i when performing task j is determined by the "unit task energy consumption ratio" calculation, as follows:

[0112]

[0113] Among them, E ij The actual energy consumption for robot i to perform task j (collected in real time by sensors, unit: kilowatt-hours);

[0114] E j,max The maximum energy consumption value for all robots performing task j (determined in advance through historical task data statistics or simulation tests, serving as a normalization benchmark).

[0115] Value retrieval logic:

[0116] 1. Data source: Relying on the energy consumption sensors of robots in industrial scenarios and historical task databases (such as task execution energy consumption records stored in the MES system);

[0117] 2. Calculation Process: Before each task assignment, the robot's real-time energy consumption data is retrieved and correlated with the historical maximum energy consumption associated with the task type. This data is then normalized using the formula described above to ensure that e... ij ∈[0,1] (a higher value indicates greater relative energy consumption);

[0118] 3. Innovative Connection: Unlike the traditional "fixed energy consumption coefficient", this solution uses dynamic real-time data collection and historical benchmark normalization to adapt to complex situations such as robot aging and load changes in industrial scenarios, thereby improving the accuracy of energy efficiency optimization for task allocation.

[0119] Skill adaptation cost s of this invention embodiment ij Value description:

[0120] s ij The skill matching degree between robot i and task j is used to quantify the skill matching degree, and is determined by the "skill matching degree inverse coefficient" calculation, as follows:

[0121]

[0122] Among them, S ik Let represent the degree of mastery of robot i over skill type k (pre-tested and scored through teaching and testing, with values ​​[0,1], where 1 indicates complete mastery);

[0123] W jk The requirement weight of task j for skill type k (determined through task decomposition analysis, with values ​​[0,1]). );

[0124] n represents the total number of skill types involved in task j (e.g., assembly tasks may involve skills such as "precision operation" and "cooperative movement").

[0125] Value retrieval logic:

[0126] 1. Data sources: Relying on the robot skill archive (storing skill mastery scores for each robot) and the task breakdown system (analyzing task requirements and skill weights);

[0127] 2. Calculation process:

[0128] Before task assignment, retrieve the robot's skill mastery level S from the archive. ik Obtain the skill requirement weight W from the task decomposition system. jk ;

[0129] The "skill matching degree" is calculated by weighted summation, and then s is obtained by subtracting the matching degree from 1. ij , ensure s ij ∈[0,1] (the lower the value, the better the fit).

[0130] In steps S304-S307 of some embodiments, by solving the entire cost matrix, the solution with the lowest total cost that matches all tasks with all robots can be found.

[0131] In some embodiments, step S103 may include, but is not limited to, steps S308 to S310:

[0132] Step S308: When the task allocation strategy is a hybrid heuristic algorithm allocation strategy, obtain the task queue and robot status; the task queue and robot status include position, battery level and current task sequence;

[0133] Step S309: Based on the task queue and robot status, generate an initial task allocation scheme using a greedy algorithm, and allocate tasks to the robot with the lowest current load in order of priority; the load is the total estimated execution time of the tasks already assigned to the robot.

[0134] Step S310: Optimize the initial task allocation scheme based on the genetic algorithm to generate a cooperative scheme; the genetic algorithm includes setting the encoding and fitness function, and performing selection, crossover and mutation operations.

[0135] In steps S308-S310 of some embodiments, a hybrid design of greedy algorithm and genetic algorithm is implemented. First, a feasible and decent initial solution is quickly generated using a greedy algorithm. Then, a genetic algorithm is used to simulate natural evolution. Through the operations of "selection, crossover, and mutation", a global search is performed based on the initial solution. This can help to escape local optima and find better solutions, providing a highly robust, efficient, and intelligent solution for high-load scenarios.

[0136] In step S310 of some embodiments, the initial task allocation scheme is optimized based on a genetic algorithm:

[0137] Encoding: A two-dimensional chromosome structure (R,T) is used, where R is the robot number and T is the task execution order;

[0138] Fitness function (hybrid heuristic algorithm): F = 1 / (T) total +μ×C conflict +ν×E total F represents the fitness value; a higher value indicates a better solution.

[0139] Among them, T total C represents the total task completion time (in seconds). conflict E represents the cost of collaboration conflict (a relative value, ranging from 0 to 10). total Let μ be the total energy consumption (relative value) of all robots, and μ and ν be the penalty coefficients.

[0140] Selection, crossover, and mutation operations: Tournament selection is used, with a crossover probability P. c =0.7, mutation probability P m =0.1;

[0141] Iterate and optimize until the termination condition is met (the termination condition is that the number of iterations reaches a preset threshold of 50 generations or the fitness value improvement rate is less than 1% for 10 consecutive generations), and output the optimal task allocation and sequence scheme.

[0142] The collaboration conflict cost C of this invention embodiment conflict Value description:

[0143] C conflict To quantify the degree of task conflict in multi-robot collaboration, a weighted calculation based on spatiotemporal overlap and motion interference coefficient is used, as shown in the formula:

[0144] C conflict =α·O st +β·I act

[0145] Among them, O st Spatiotemporal overlap (value [0,1]) represents the overlap ratio between the time window of the robot task execution and the working area;

[0146] I act The motion interference coefficient (values ​​[0,1]) represents the physical interference risk of the robot's motion sequence (such as the probability of crossover of the robot arm's motion trajectory);

[0147] α and β are weighting coefficients (satisfying α+β=1, with default α=0.6 and β=0.4, which can be dynamically adjusted according to the scenario).

[0148] 1. Spatiotemporal overlap O st calculate:

[0149]

[0150] in, The time overlap between the execution of tasks i and j for robots (through the task pre-execution time window [t]). i_start ,t i_end ] and [t j_start ,t j_end (Request for submission);

[0151] T task Total task duration (the maximum execution time of all robot tasks);

[0152] The spatial overlap area of ​​the working areas of robots i and j (extracting the spatial intersection of the robot motion envelope through a digital twin model of an industrial scene);

[0153] A work Total working area (the global spatial extent of task execution in the digital twin model);

[0154] Number of robot pairwise combinations (m is the total number of robots participating in the collaboration, to avoid double counting).

[0155] 2. Motion interference coefficient I act calculate:

[0156]

[0157] The number of interference events between robot action sequences i and j (through kinematic simulation, the trajectory intersection and collision risk of the robotic arm and end effector during task execution are simulated, and the number of interference events is counted).

[0158] The total number of frames in the robot's i and j action sequences (based on the temporal resolution of task execution, decomposed into discrete motion frame statistics).

[0159] The value retrieval logic and innovation of this invention embodiment:

[0160] 1. Data source:

[0161] Time window: Pre-allocated execution time based on the robot task scheduling system;

[0162] Spatial area: Extracted in real time through an industrial digital twin platform;

[0163] Motion intervention: Offline / online trajectory simulation based on robot kinematics library.

[0164] 2. Innovation points:

[0165] Unlike the traditional "conflict marking method" (which only determines whether there is a conflict), this invention uses a two-dimensional quantification of spatiotemporal overlap and action interference to accurately characterize the "degree" of collaborative conflict rather than its "presence".

[0166] By integrating digital twin spatial analysis and kinematic simulation, it adapts to the collaborative needs of complex tasks in industrial scenarios (such as multi-robot assembly and collaborative handling) and improves the stability of task allocation.

[0167] In some embodiments, the hybrid heuristic algorithm of step S310 includes:

[0168] Step S311, execute the dynamic adjustment mechanism; the dynamic adjustment mechanism includes:

[0169] Step S312: Monitor for new task additions or robot malfunction events;

[0170] Step S313: When a new task is added or a robot malfunction event occurs, the disturbance coefficient is calculated. The formula used includes:

[0171] δ=ΔT / T0;

[0172] Where δ is the disturbance coefficient; ΔT is the number of new tasks; and T0 is the total number of original tasks.

[0173] When T0 = 0, if ΔT ≥ 1, then δ = 1.0;

[0174] In step S314, if δ>0.2, trigger re-optimization; otherwise, only make local adjustments to the new task and insert it into the existing sequence.

[0175] In steps S311-S314 of some embodiments, the dynamic adjustment mechanism can sense the degree of environmental change and adaptively select the optimal response strategy accordingly, which improves computational efficiency.

[0176] This application also provides a dynamic task allocation system for humanoid robots in industrial scenarios, used to implement the aforementioned method. The system includes:

[0177] The first module is used to obtain the task load level of the current industrial scenario; the task load level includes low load level, medium load level and high load level.

[0178] The second module is used to determine the corresponding task allocation strategy based on the task load level; the task allocation strategy includes event-triggered allocation strategy, improved Hungarian algorithm allocation strategy, and hybrid heuristic algorithm allocation strategy;

[0179] The third module is used to allocate tasks to humanoid robots based on a defined task allocation strategy; humanoid robot task allocation includes task-robot matching, task sequence planning, and collaborative scheme generation.

[0180] As an optional implementation method, refer to Figure 2 The dynamic task allocation method for humanoid robots in industrial scenarios according to embodiments of the present invention includes the following steps:

[0181] 1. Obtain the task load level of the current industrial scenario. The load level is determined based on a combination of task density, complexity, and robot idle rate, and includes three levels: low load, medium load, and high load.

[0182] 2. Determine the corresponding task allocation strategy based on the load level: use an event-triggered allocation strategy for low load, a modified Hungarian algorithm allocation strategy for medium load, and a hybrid heuristic algorithm allocation strategy for high load;

[0183] 3. Based on the determined task allocation strategy, allocate tasks to the humanoid robot and output the task-robot matching scheme and execution sequence.

[0184] Step 1: Reference Figure 3 Determining the task load level

[0185] Obtain the following parameters through sensor networks or production management systems:

[0186] The number of new tasks N per unit of time (unit: tasks / hour);

[0187] Average task complexity coefficient C: Assigned a value based on the task type (e.g., handling, assembly, inspection) (1-5, the higher the value, the more complex the task).

[0188] Robot idle rate I: Number of idle robots / Total number of robots (0-1).

[0189] The load index L is calculated based on the formula L=α×N / T+β×C-γ×I, where T is the time window (default 1 hour), and α, β, and γ are weighting coefficients (which can be adjusted according to the scenario; default α=0.4, β=0.4, γ=0.2). Levels are determined based on the L value:

[0190] Low load level: L<0.3;

[0191] Medium load rating: 0.3 ≤ L ≤ 0.7;

[0192] High load level: L>0.7.

[0193] Step 2: Matching of task allocation strategies

[0194] Low load level (event-triggered allocation strategy): Suitable for scenarios with sparse tasks and sufficient robot resources. When a new task is triggered (such as a material request from a workstation), the comprehensive score of idle robots (skill matching degree, movement cost, and remaining battery power) is calculated in real time, and tasks are quickly allocated, prioritizing response speed.

[0195] Medium Load Level (Improved Hungarian Algorithm Allocation Strategy): Suitable for scenarios with moderate task load and a similar number of robots and tasks. A comprehensive cost matrix (including time, energy consumption, and skill costs) is constructed, and the globally optimal match is solved using an improved Hungarian algorithm, balancing efficiency and optimization.

[0196] High-load level (hybrid heuristic algorithm allocation strategy): Suitable for scenarios with intensive tasks and limited robot resources. It adopts a hybrid strategy of "greedy initial solution + genetic algorithm optimization" to balance computational efficiency and global optimum, while introducing a dynamic adjustment mechanism to cope with sudden changes in tasks.

[0197] Step 3: Reference Figure 4 Execution of task assignment

[0198] Task assignment is performed according to the selected strategy, specifically including:

[0199] Task-robot matching: Determine which robot will perform each task;

[0200] Task sequence planning: Optimize the execution order of multiple tasks assigned to each robot;

[0201] Collaboration scheme generation: For tasks with dependencies, generate robot collaboration timing (such as synchronization action nodes and information exchange timing).

[0202] The beneficial effects of this invention include:

[0203] 1. Strong dynamic adaptability: The strategy is adaptively switched by dividing the load level, which ensures response speed under low load and pursues global optimization under high load, thus solving the adaptability problem of a single algorithm.

[0204] 2. Humanoid robot characteristic adaptation: The cost model incorporates unique parameters such as motion energy consumption and skill switching costs, making the allocation scheme more in line with actual execution capabilities;

[0205] 3. Improved collaboration efficiency: In high-load strategies, task dependencies and robot collaboration factors are considered to reduce collaboration conflicts and improve team work efficiency;

[0206] 4. Balance between real-time performance and optimization: The multi-stage strategy design enables the system to maintain high efficiency in different scenarios. It provides millisecond-level response under low load and finds an approximate optimal solution within an acceptable time under high load through heuristic algorithms.

[0207] The following is a detailed introduction and explanation of the solutions in the embodiments of the present invention, with reference to specific application examples.

[0208] Example 1: Task Load Level Determination Process

[0209] This embodiment details how to determine the task load level based on real-time data:

[0210] 1. Data Acquisition:

[0211] Obtain task data for the most recent hour through an Industrial Internet of Things (IIoT) platform: Total number of tasks N = 30;

[0212] Task complexity coefficient C: material handling task (0.3), assembly task (0.7), inspection task (0.5), average C = 0.5;

[0213] Robot status data: Total number of robots: 10, 3 currently idle, idle rate I = 3 / 10 = 0.3.

[0214] 2. Load index calculation:

[0215] Using the formula L=α×N / T+β×C-γ×I, where α=0.4, β=0.4, γ=0.2, T=1 hour, the number of tasks N / T per unit time is normalized (maximum task volume is set to 50), then N / T=30 / 50=0.6, load index:

[0216] L=0.4×0.6+0.4×0.5-0.2×0.3=0.24+0.2-0.06=0.38.

[0217] 3. Level determination: Since L = 12.14 > 0.7, it is determined to be a high load level, triggering the hybrid heuristic algorithm allocation strategy.

[0218] Example 2: Event Triggering Allocation Strategy under Low Load Level

[0219] When the load level is low (e.g., L = 0.2), an event-triggered allocation strategy is adopted, and the specific steps are as follows:

[0220] 1. Event monitoring: The workstation sensor detects a "Material Replenishment" task request (Task ID: T001), triggering the allocation process;

[0221] 2. Robot Status Acquisition: Obtain the parameters of the currently idle robots (R01, R02, R03), as shown in Table 1:

[0222] Table 1. Parameters of currently idle robots (R01, R02, R03)

[0223] robot Skill matching degree M (material handling) Movement cost C (meters) Remaining power E R01 0.9 (Specialized in data transfer) 50 0.8 R02 0.6 (universal type) 30 0.9 R03 0.7 (Dual-purpose type) 40 0.7

[0224] 3. Overall score calculation:

[0225] The formula S = ω1 × M + ω2 × (1 / C) + ω3 × E is used, where ω1 = 0.5, ω2 = 0.3, and ω3 = 0.2.

[0226] R01: S=0.5×0.9+0.3×(1 / 50)+0.2×0.8=0.45+0.006+0.16=0.616;

[0227] R02: S=0.5×0.6+0.3×(1 / 30)+0.2×0.9=0.3+0.01+0.18=0.49;

[0228] R03: S=0.5×0.7+0.3×(1 / 40)+0.2×0.7=0.35+0.0075+0.14=0.4975.

[0229] 4. Task allocation: R01 has the highest overall score, so T001 is assigned to R01, and motion path instructions are generated.

[0230] Example 3: Reference Figure 5 Improved Hungarian Algorithm Allocation Strategy under Medium Load Levels

[0231] When the load level is medium (e.g., L = 0.5), an improved Hungarian algorithm allocation strategy is adopted, and the specific steps are as follows:

[0232] 1. Tasks and Robot Assembly:

[0233] Tasks to be assigned: T1 (assembly), T2 (inspection), T3 (transportation);

[0234] Idle robots: R1, R2, R3 (quantity matching).

[0235] 2. Cost matrix construction:

[0236] Comprehensive cost c ij =λ1×t ij +λ2×e ij +λ3×s ij Where λ1 = 0.5, λ2 = 0.3, and λ3 = 0.2, the parameters are shown in Table 2:

[0237] Table 2. Overall Cost

[0238]

[0239] The cost matrix is ​​as follows:

[0240]

[0241] 3. Improved Hungarian algorithm for solving:

[0242] Step 1: Subtract the minimum value of each row from the matrix to obtain the simplified matrix;

[0243] Step 2: Cover all zero elements with the fewest lines. If the number of lines equals the order, find the optimal matching.

[0244] Step 3: The optimal matching is found to be R1-T1, R2-T3, R3-T2, and the total minimum cost is 60.28+40.35+55.18=155.81.

[0245] 4. Allocation Results: Tasks are allocated according to the matching scheme described above.

[0246] Example 4: Reference Figure 6 Hybrid heuristic algorithm allocation strategy under high load levels

[0247] When the load level is high (e.g., L = 0.8), a hybrid heuristic algorithm allocation strategy is adopted, and the specific steps are as follows:

[0248] 1. Initial solution generation (greedy algorithm):

[0249] Task queue (priority from high to low): T1, T2, T3, T4, T5;

[0250] Robots: R1 (current load 2), R2 (current load 1), R3 (current load 3);

[0251] Initial allocation: The robots with the lowest current load are allocated according to priority, resulting in R2-T1, R2-T2, R1-T3, R1-T4, and R3-T5.

[0252] 2. Genetic Algorithm Optimization:

[0253] Encoding: Chromosome (R,T) = [(R2,T1),(R2,T2),(R1,T3),(R1,T4),(R3,T5)];

[0254] Fitness function: F = 1 / (T) total +μ×C conflict +ν×E total ), where T total = 650 seconds, C conflict =0 (no conflict), E total =4.2 (relative energy consumption), μ=10, ν=5, then F=1 / (650+10×0+5×4.2)=0.0015;

[0255] Selection: The tournament selection method was used to select the chromosomes with the top 50% fitness.

[0256] Crossover: Perform two-point crossover on the task sequence. For example, the task sequence [T1,T2] of R2 crosses with [T3,T4] of R1 to form [T1,T4] and [T3,T2];

[0257] Mutation: Randomly swap the robot assignments of two tasks, such as mutating R3-T5 to R1-T5;

[0258] Iteration: After 50 iterations, the optimal solution was obtained: R2-T1, R1-T2, R1-T3, R2-T4, R3-T5, with the total completion time reduced to 580 seconds and the fitness improved to 0.0017.

[0259] 3. Dynamic adjustment:

[0260] A new task T6 was added, with a perturbation coefficient δ = 1 / 5 = 0.2, triggering a local adjustment;

[0261] Insert T6 into the task sequence of R3 (currently with the lowest load) to obtain the final solution.

[0262] In the timing planning of multi-robot collaborative tasks, for tasks with dependencies, a robot collaboration timing sequence is generated, including synchronization action nodes and information exchange timings. (Reference) Figure 7 When the assembly task is started and Robot 1 and Robot 2 are ready, they synchronize action nodes and information interaction points, perform collaborative operations, and confirm the completion of the task.

[0263] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method.

[0264] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0265] Please see Figure 8 , Figure 8 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes:

[0266] The processor 801 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application.

[0267] The memory 802 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 802 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 802 and is called and executed by the processor 801 using the methods described in the embodiments of this application.

[0268] The 803 input / output interface is used to implement information input and output.

[0269] The communication interface 804 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0270] Bus 805 transmits information between various components of the device (e.g., processor 801, memory 802, input / output interface 803, and communication interface 804);

[0271] The processor 801, memory 802, input / output interface 803, and communication interface 804 are connected to each other within the device via bus 805.

[0272] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0273] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0274] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0275] It is understood that the content of the above method embodiments is applicable to the embodiments of this program product. The specific functions implemented by the embodiments of this program product are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0276] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0277] The method, system, electronic device, storage medium, and program product for dynamic task allocation of humanoid robots in industrial scenarios provided in this application embodiment combine the multi-degree-of-freedom and multi-skill characteristics of humanoid robots to dynamically switch allocation strategies under different load scenarios, taking into account both real-time performance and global optimization, thereby improving robot utilization and industrial scenario collaboration efficiency.

[0278] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0279] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0280] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0281] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0282] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0283] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0284] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0285] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0286] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0287] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0288] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for dynamic task allocation of humanoid robots in industrial scenarios, characterized in that, The method comprises the following steps: obtaining the task load level of the current industrial scene; the task load level comprises a low load level, a medium load level and a high load level; determining a corresponding task allocation strategy based on the task load level; the task allocation strategy comprises an event-triggered allocation strategy, an improved Hungarian algorithm allocation strategy and a hybrid heuristic algorithm allocation strategy; performing humanoid robot task allocation based on the determined task allocation strategy; the humanoid robot task allocation comprises task-robot matching, task sequence planning and cooperation scheme generation.

2. The method of claim 1, wherein, The task load level of the current industrial scene is obtained, comprising: obtaining the number of tasks to be allocated per unit time, the average complexity coefficient of the tasks and the robot idle rate, calculating the task load index, and using the formula comprising: L = α × N / T + β × C - γ × I; wherein L is the task load index; N is the number of tasks to be allocated per unit time; T is the time window, in hours; C is the average complexity coefficient of the tasks, with a value range of 1 to 5; I is the robot idle rate, with a value range of 0 to 1; α, β, γ are weight coefficients and α + β + γ = 1; determining the load level according to the task load index: when L < 0.3, it is a low load level, when 0.3 ≤ L ≤ 0.7, it is a medium load level, and when L > 0.7, it is a high load level.

3. The method of claim 1, wherein, The corresponding task allocation strategy is determined based on the task load level, comprising: when the task load level is a low load level, the task allocation strategy is determined to be an event-triggered allocation strategy; when the task load level is a medium load level, the task allocation strategy is determined to be an improved Hungarian algorithm allocation strategy; when the task load level is a high load level, the task allocation strategy is determined to be a hybrid heuristic algorithm allocation strategy.

4. The method of claim 1, wherein, The humanoid robot task allocation is performed based on the determined task allocation strategy, comprising: when the task allocation strategy is an event-triggered allocation strategy, monitoring task triggering events; if a new task is detected, obtaining the skill matching degree of each idle robot, the moving cost from the current position to the task point and the remaining power proportion, calculating the comprehensive score of each robot, and using the formula comprising: S = ω1 × M + ω2 × (1 / C) + ω3 × E; wherein S is the comprehensive score of the robot; ω1, ω2, ω3 are weight coefficients; M is the skill matching degree, with a value range of 0 to 1; C is the moving cost; E is the remaining power proportion, with a value range of 0 to 1; assigning the new task to the idle robot with the highest comprehensive score to complete the task-robot matching.

5. The method of claim 1, wherein, The humanoid robot task allocation is performed based on the determined task allocation strategy, comprising: when the task allocation strategy is an improved Hungarian algorithm allocation strategy, collecting all the tasks to be allocated and idle robots at present; constructing a cost matrix according to the information of the tasks to be allocated and the idle robots; the cost matrix comprises a plurality of matrix elements; calculating the matrix elements, and using the formula comprising: c ij = λ1 x t ij + λ2 x e ij + λ3 x s ij ; wherein c ij is the matrix element, representing the comprehensive cost of the ith robot performing the jth task; t ij is the estimated time consumption of the ith robot performing the jth task; e ij is the relative energy consumption cost; s ij is the skill adaptation cost; λ1, λ2, λ3 are weight coefficients, and λ1+λ2+λ3=1; The improved Hungarian algorithm is used to solve the cost matrix to obtain a globally optimal task-robot matching scheme. If the number of the tasks to be allocated is greater than the number of the idle robots, the tasks to be allocated are prioritized, and the tasks with high priority are allocated preferentially to complete task sequence planning.

6. The method of claim 1, wherein, The human-shaped robot task allocation based on the determined task allocation strategy includes: When the task allocation strategy is a mixed heuristic algorithm allocation strategy, a task queue and a robot state are obtained; the task queue and the robot state include position, power, and current task sequence; According to the task queue and the robot state, a greedy algorithm is used to generate an initial task allocation scheme, and tasks are allocated to the robot with the lowest current load in turn according to priority; the load is the total estimated execution time of the tasks allocated to the robot. The initial task allocation scheme is optimized based on a genetic algorithm to generate a cooperation scheme; the genetic algorithm includes setting a code and a fitness function, and performing selection, crossover, and mutation operations.

7. The method of claim 6, wherein, The mixed heuristic algorithm includes: A dynamic adjustment mechanism is performed; the dynamic adjustment mechanism includes: A task addition or robot failure event is monitored; When the task addition or the robot failure event occurs, a disturbance coefficient is calculated, and the formula used includes: δ=ΔT / T0; Wherein, δ is the disturbance coefficient; ΔT is the number of added tasks; T0 is the total number of original tasks; When T0=0, if ΔT≥1, δ=1.0; If δ>0.2, re-optimization is triggered, otherwise, only the newly added task is locally adjusted and inserted into the existing sequence.

8. A dynamic task allocation system for humanoids in industrial scenarios for implementing the method according to any one of claims 1 to 7, characterized in that, The system includes: A first module for obtaining a task load level of a current industrial scene; the task load level includes a low load level, a medium load level, and a high load level; A second module for determining a corresponding task allocation strategy based on the task load level; the task allocation strategy includes an event-triggered allocation strategy, an improved Hungarian algorithm allocation strategy, and a mixed heuristic algorithm allocation strategy; A third module for performing human-shaped robot task allocation based on the determined task allocation strategy; the human-shaped robot task allocation includes task-robot matching, task sequence planning, and cooperation scheme generation.

9. An electronic device, comprising: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the method of any one of claims 1 to 7 when executing the computer program.

10. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the method of any one of claims 1 to 7.

Images