Multi-task allocation method and device for unmanned transportation equipment, equipment and medium

Abstract

The invention relates to a multi-task allocation method and device for unmanned transportation equipment, equipment and a medium, and the method comprises the steps: constructing a multi-task combination model, and setting related constraint conditions; obtaining an initial population through chaotic mapping; performing rapid non-dominated sorting on the initial population, and based on a dynamic congestion degree sorting strategy, selecting and determining a plurality of parent populations through a champion; for each parent population, generating a new individual through adaptive crossover operation, and then performing mutation operation through Gaussian mutation to obtain a filial population; combining the parent population and the child population, performing rapid non-dominated sorting operation, and screening excellent individuals through a probability distribution function to obtain a new parent population; performing iterative computation based on the new parent population, and gradually optimizing a result to obtain an optimal multi-task allocation scheme; and performing multi-task allocation on the unmanned transportation equipment through the optimal multi-task allocation scheme. The method can improve the transportation efficiency, and balance the task scheduling complexity and the utilization rate of unmanned transportation equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent task allocation, and in particular to a multi-task allocation method, device, equipment and medium for unmanned transport equipment. Background Art

[0002] With the development of automation technology, robots have been gradually introduced in many industries to replace manual labor to perform some simple and repetitive tasks. In this process, unmanned transportation equipment has become an important part of multiple automation systems. On the premise of ensuring the timeliness of task transportation, how to efficiently allocate the tasks of multiple unmanned transportation equipment has become one of the key factors to improve the efficiency of system operation. As the amount of tasks gradually increases, the amount of transportation tasks that need to be completed per unit time is also increasing. Therefore, improving transportation efficiency has become an important bottleneck restricting the development of multiple unmanned transportation equipment systems.

[0003] At present, most unmanned transport equipment systems adopt the "single-time single-task" mode, that is, each unmanned transport equipment only performs one task in a task cycle. Although this mode reduces the complexity of task allocation and unmanned transport equipment scheduling, the overall time and path required to complete all tasks are longer under the same task magnitude. In contrast, the "single-time multi-task" mode can improve the utilization rate of unmanned transport equipment and reduce the number of unmanned transport equipment required under the same number of tasks and time conditions. However, this mode also increases the difficulty of coordination and scheduling between unmanned transport equipment systems, thereby limiting the optimization effect of the scheduling algorithm. Summary of the invention

[0004] Based on this, it is necessary to provide a multi-task allocation method, device, equipment and medium for unmanned transportation equipment that can solve the problems of low transportation efficiency, high task scheduling complexity and low utilization rate of unmanned transportation equipment in response to the above-mentioned technical problems.

[0005] A multi-task allocation method for unmanned transport equipment, the method comprising: Obtain the number of unmanned transport equipment and the types of missions; According to the number of unmanned transport equipment and the types of tasks, a multi-task combination model with the goal of minimizing total cost and maximizing on-time completion is constructed, and relevant constraints are set; Initializing parameters, and obtaining an initial population through chaotic mapping based on the multi-task combination model and constraint conditions; Performing a fast non-dominated sort on the initial population, and determining a number of parent populations through tournament selection based on a dynamic crowding sorting strategy; For each of the parent populations, a new individual is generated through an adaptive crossover operation, and then a mutation operation is performed through a Gaussian mutation to obtain a child population; wherein randomness is introduced into the Gaussian mutation, and mean and standard deviation parameters are adjusted to obtain an improved Gaussian mutation, and the mutation operation is performed through the improved Gaussian mutation; The parent population and the child population are merged to obtain an intermediate population; a fast non-dominated sorting operation is performed on the intermediate population, and excellent individuals are screened through a probability distribution function to obtain a new parent population; Iterative calculations are performed based on the new parent population, and the results are gradually optimized to obtain an optimal multi-task allocation scheme; and multi-task allocation is performed on the unmanned transport equipment through the optimal multi-task allocation scheme.

[0006] A multi-task allocation device for unmanned transport equipment, the device comprising: The data acquisition module is used to obtain the number of unmanned transport equipment and the types of tasks; A model building module, used to build a multi-task combination model with the goal of minimizing total cost and maximizing on-time completion according to the number of unmanned transport equipment and the types of tasks, and set relevant constraints; An initial population acquisition module is used to perform parameter initialization and obtain an initial population through chaotic mapping based on the multi-task combination model and constraint conditions; A parent population acquisition module is used to perform fast non-dominated sorting on the initial population and determine a number of parent populations through tournament selection based on a dynamic congestion sorting strategy; A crossover mutation operation module is used to generate new individuals for each parent population through an adaptive crossover operation, and then perform a mutation operation through a Gaussian mutation to obtain a child population; wherein randomness is introduced into the Gaussian mutation, and the mean and standard deviation parameters are adjusted to obtain an improved Gaussian mutation, and the mutation operation is performed through the improved Gaussian mutation; A new parent population acquisition module is used to merge the parent population and the child population to obtain an intermediate population; perform a fast non-dominated sorting operation on the intermediate population, select excellent individuals through a probability distribution function, and obtain a new parent population; The iterative update module performs iterative calculations based on the new parent population, gradually optimizes the results, and obtains the optimal multi-task allocation plan; and multi-tasks are allocated to the unmanned transport equipment through the optimal multi-task allocation plan.

[0007] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the multi-task allocation method of the unmanned transport equipment when executing the computer program.

[0008] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the multi-task allocation method for unmanned transport equipment.

[0009] The multi-task allocation method, device, equipment and medium of the above-mentioned unmanned transport equipment, the method obtains the number of unmanned transport equipment and the type of task; according to the number of unmanned transport equipment and the type of task, a multi-task combination model with the goal of minimizing total cost and maximizing on-time completion is constructed, and relevant constraints are set; parameter initialization is performed, and based on the multi-task combination model and constraints, an initial population is obtained through chaotic mapping; the initial population is quickly non-dominated sorted, and based on a dynamic congestion sorting strategy, a number of parent populations are determined through tournament selection; for each parent population, an adaptive crossover operation is performed Generate new individuals, and then perform mutation operation through Gaussian mutation to obtain the offspring population; introduce randomness into Gaussian mutation, and adjust the mean and standard deviation parameters to obtain the improved Gaussian mutation, and perform mutation operation through the improved Gaussian mutation; merge the parent population and the offspring population to obtain the intermediate population; perform fast non-dominated sorting operation on the intermediate population, screen excellent individuals through probability distribution function, and obtain a new parent population; perform iterative calculation based on the new parent population, gradually optimize the results, and obtain the optimal multi-task allocation plan; perform multi-task allocation on the unmanned transport equipment through the optimal multi-task allocation plan.

[0010] The method provided by the present invention constructs a multi-task combination model with the goal of minimizing total cost and maximizing on-time completion, determines the priority of tasks and the allocation strategy of vehicles; when obtaining the initial population, the chaotic mapping is used to enhance the correlation of the initial individuals, so as to better cover the search space; when sorting, the dynamic congestion sorting strategy is used to improve the diversity of the solution set, so that the individuals can more flexibly reflect their distribution in the Pareto frontier in the congestion calculation; in screening excellent individuals, the individual selection of each level is performed through the probability distribution function, which can prevent the algorithm from converging in advance. At the same time, in order to ensure the diversity of individuals in the entire population, and also to improve the overall distribution of the Pareto optimal solution and prevent premature maturity, in the crossover mutation operation, new individuals are generated through adaptive crossover operation, and mutation operation is performed through improved Gaussian mutation, so as to increase the diversity of solutions, improve the possibility of global search, improve the overall convergence of the algorithm, and finally obtain the optimal multi-task allocation scheme, and the multi-task allocation of unmanned transportation equipment through the optimal multi-task allocation scheme can improve transportation efficiency, balance the complexity of task scheduling and the utilization rate of unmanned transportation equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0012] Figure 1 A schematic flow chart of a multi-task allocation method for unmanned transport equipment in one embodiment; Figure 2 A schematic diagram of the relationship between system on-time completion and delivery time in an embodiment; Figure 3 A Pareto frontier diagram of task allocation performed by the NSGA-II algorithm in one embodiment; Figure 4 A Pareto frontier schematic diagram of task allocation by the method provided by the present invention in one embodiment; Figure 5 A schematic diagram of a trajectory for path planning using the NSGA-II algorithm in one embodiment; Figure 6 A schematic diagram of a trajectory for path planning by the method provided by the present invention in one embodiment; Figure 7 A structural block diagram of a multi-task allocation device for unmanned transport equipment in one embodiment; Figure 8 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment.

[0013] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0014] 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.

[0015] It should be noted that the technical solutions between the various embodiments of the present invention can be combined with each other, but it must be based on the fact that ordinary technicians in the field can implement it. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0016] The following will describe the implementation of the present invention in detail with reference to the accompanying drawings in the embodiment diagram of the present invention.

[0017] Example 1 The task scheduling problem of multiple unmanned transport equipment is a typical NP problem. The number of task combinations, total quality, and combined path costs are mutually constrained and have strong uncertainty. Within a given time frame, the system generates a transport task sequence according to a certain strategy, aiming to minimize the total distance of the entire task group and maximize the load of the task group while meeting the maximum load limit and taking into account the transportation time. Establishing a multi-task combination model is the primary problem in solving the scheduling of multi-task unmanned transport equipment. Through reasonable task combinations, the system efficiency can be improved and the transportation cost can be reduced.

[0018] Based on this, this embodiment discloses a multi-task allocation method for unmanned transportation equipment, which constructs a multi-task combination model with the goal of minimizing total cost and maximizing on-time completion, and uses it as an objective function to determine the priority of tasks and the allocation strategy of vehicles; when obtaining the initial population, the use of chaotic mapping can enhance the correlation of initial individuals, thereby better covering the search space; when sorting, a dynamic congestion sorting strategy is adopted to improve the diversity of the solution set, so that individuals can more flexibly reflect their distribution in the Pareto frontier in the congestion calculation; in screening excellent individuals, individual selection at each level is performed through probability distribution functions, which can prevent the algorithm from converging prematurely. At the same time, in order to ensure the diversity of individuals in the entire population, and to improve the overall distribution of the Pareto optimal solution and prevent premature maturity, in the crossover mutation operation, new individuals are generated through adaptive crossover operation, and mutation operation is performed through the improved Gaussian mutation to increase the diversity of solutions, improve the possibility of global search, and improve the overall convergence of the algorithm. Finally, the optimal multi-task allocation scheme is obtained. Multi-task allocation is performed on unmanned transport equipment through the optimal multi-task allocation scheme, which solves the problems of low transportation efficiency, high complexity of task scheduling, and low utilization rate of unmanned transport equipment.

[0019] The unmanned transport equipment in the present invention includes but is not limited to automatic guided vehicles (unmanned transport equipment), unmanned delivery robots, unmanned aerial vehicles, unmanned trucks, etc.

[0020] In addition, for the convenience of analysis, the following assumptions are made for the multi-task combination model: (1) All unmanned transport equipment are initially placed in the parking area and can be called upon.

[0021] (2) The unmanned transport equipment has the same speed for empty and loaded transport during transportation, and the straight-line speed is constant. , the turning speed is .

[0022] (3) Ignore the impact of unmanned transport equipment failures, power supply and other factors, and ensure that all unmanned transport equipment operates normally during operation.

[0023] (4) Ignore the volume restrictions of the unmanned transport equipment transport vehicle for cargo and only ensure that the total weight of the task group's cargo does not exceed the maximum load of the unmanned transport equipment.

[0024] (5) The unmanned transport equipment models are the same, and the loading and unloading time at each cargo point is not considered. Only the driving time and waiting time during the task execution are considered.

[0025] (6) Unmanned transport equipment is not charged during operation. After all transport tasks are completed, the unmanned transport equipment is concentrated in the parking area for charging (to match the subsequent unmanned transport equipment scheduling).

[0026] (7) There are no restrictions on the issuance of transport tasks, that is, every unmanned transport equipment can perform the task.

[0027] (8) The transportation task execution process is continuous and no interruption is allowed.

[0028] In summary, referring to Figure 1, a multi-task allocation method for unmanned transport equipment provided in an embodiment of the present invention includes the following steps: Step 201, obtaining the number of unmanned transport equipment and the type of task.

[0029] Step 202, based on the number of unmanned transport equipment and the types of tasks, a multi-task combination model with the goal of minimizing total cost and maximizing on-time completion is constructed, and relevant constraints are set.

[0030] Step 203, initializing parameters, based on the multi-task combination model and constraint conditions, obtaining the initial population through chaotic mapping.

[0031] Step 204 , perform fast non-dominated sorting on the initial population, and determine a number of parent populations through tournament selection based on a dynamic congestion sorting strategy.

[0032] Step 205, for each parent population, generate new individuals through adaptive crossover operation, and then perform mutation operation through Gaussian mutation to obtain a child population; wherein, randomness is introduced into the Gaussian mutation, and the mean and standard deviation parameters are adjusted to obtain an improved Gaussian mutation, and the mutation operation is performed through the improved Gaussian mutation.

[0033] Step 206, merging the parent population and the child population to obtain an intermediate population; performing a fast non-dominated sorting operation on the intermediate population, screening excellent individuals through a probability distribution function, and obtaining a new parent population.

[0034] Step 207, performing iterative calculations based on the new parent population, gradually optimizing the results, and obtaining an optimal multi-task allocation scheme; performing multi-task allocation on the unmanned transport equipment through the optimal multi-task allocation scheme.

[0035] In the specific implementation process of step 201, the application scenario is first determined, the number of unmanned transport equipment and the type of task are obtained. The present invention takes the laboratory multi-unmanned transport equipment as the research object, and its transportation task can be described as: at most Unmanned transport equipment of the same model that can perform transport tasks, with a maximum load capacity of , No. Unmanned transport equipment to carry out The load capacity of each task Indicates. tasks, the total task set is Indicates that the system is based on Task group for generating the available driving time and load capacity of unmanned transport equipment , Each task The information included is ,in represents the horizontal coordinate of the task point, Indicates the vertical coordinate of the task point; Representation Task Acceptable delivery time windows; Representation Task Desired delivery time window; Indicates the weight of the transport task. After the transport task group issues the order, the unmanned transport equipment starts from the current point, completes all transport tasks in sequence, and then returns to the parking area. When all unmanned transport equipment returns to the parking area, it means that all transport tasks in this time period have been completed. According to the laboratory's production plan, transport tasks have fixed time windows, so unmanned transport equipment must meet the time window requirements when performing tasks, and delays due to transport task scheduling are not allowed.

[0036] In the specific implementation process of step 202, the establishment of task allocation is the primary problem to be solved in the scheduling of multi-task unmanned transport equipment. Reasonable task combination can improve system efficiency and reduce transportation costs. Types of tasks , a multi-task combination model is established with the goal of minimizing the total cost of the unmanned transportation equipment system and maximizing the on-time completion rate of the system, and relevant constraints are set. Then, the model is used as the objective function to solve the problem in order to obtain the optimal multi-task allocation plan.

[0037] Among them, the multi-task combination model expression is: (1) (2) (3) (4) In the formula, Indicates the lowest total cost; Indicates the electric energy coefficient per unit load during the driving process of unmanned transport equipment; Indicates the weight loaded by the unmanned transport equipment when performing a mission; It represents the ratio of the straight-line driving time of unmanned transport equipment to the power consumption; It represents the ratio of turning time and power consumption of unmanned transport equipment; Indicates Unmanned transport equipment from the mission To the task The decision variable is used to determine whether the unmanned transport equipment has completed the task After whether to execute the task ; Indicates Unmanned transport equipment from the distribution center to the mission The decision variables are used to determine the Whether the unmanned transport equipment starts from the distribution center to perform the task ; Indicates Unmanned transport equipment from the distribution center to the mission Straight-line driving time; Indicates Unmanned transport equipment from the distribution center to the mission Turning time; Indicates Unmanned transport equipment from the mission To the task Straight-line driving time; Indicates Unmanned transport equipment from the mission To the task Turning time; It represents the ratio of waiting time to power consumption of unmanned transport equipment; Indicates completion of task Required working hours; Indicates completion of task Required working hours; It indicates the highest degree of on-time completion; represents the punctuality function; Indicates Unmanned transport equipment to carry out The load of each task; Representation Task Desired delivery time window; Represents the task in the time window penalty function Acceptable delivery time windows; Indicates the delivery time.

[0038] The constraint expression is: (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) In the formula, Indicates Unmanned transport equipment from the mission To the task The decision variables; Indicates The maximum load capacity of unmanned transport equipment; Indicates the number of unmanned transport equipment; Indicates the type of task; Indicates the task point The horizontal axis of Indicates the task point The horizontal axis of Indicates the task point The vertical coordinate of Indicates the task point The vertical coordinate of Indicates The speed of unmanned transport equipment; Representation Task Actual start time; Representation Task Estimated start time; Indicates Unmanned transport equipment actually arrived at the mission time; Indicates Unmanned transport equipment actually arrived at the mission time; Indicates that the unmanned transport equipment is from the mission point To the mission point The travel time between Indicates Unmanned transport equipment completes mission Required working hours; Indicates that the unmanned transport equipment is at the mission point and mission points The travel time between Indicates Unmanned transport equipment actually arrived at the mission time; Indicates Unmanned transport equipment from the mission point Specifically, formula (1) is the sub-model with the lowest total cost of the system. It can be seen that the total cost of the system is composed of the sum of the power consumption of the unmanned transport equipment in straight-line uniform speed driving, turning power consumption and waiting power consumption. Formula (2) is the sub-model with the highest on-time completion rate of the system. Formula (3) defines the decision variable, which represents the Unmanned transport equipment completed the mission After whether to execute the task .

[0039] Formula (4) is the system punctuality function. In multi-task transportation, each task has an expected transportation time window. Considering the system's requirements for vehicle arrival time and actual delivery times, the system punctuality function is established to comprehensively measure the rationality of task allocation, such as Figure 2 As shown in the figure, it is a schematic diagram of the relationship between the system's on-time completion rate and the delivery time, and its horizontal axis represents the delivery time. , vertical coordinate Represents the system's on-time completion function, from Figure 2 It can be seen from the figure that when the unmanned transport equipment is in the time window If the unmanned transport equipment arrives at the service point within the time window, the system's on-time completion degree is 1; If the unmanned transport equipment arrives outside the time window, the system's on-time completion rate is 0; The system's on-time completion rate will increase as the time difference between the delivery time and the expected earliest delivery time decreases; if the unmanned transport equipment arrives within the time window The system's on-time completion rate will decrease as the difference between the delivery time and the expected latest delivery time increases. Thus, the system's on-time completion rate function expression of formula (4) is obtained.

[0040] Formula (5) represents the vehicle carrying capacity constraint, ensuring The total weight of the unmanned transport equipment assigned to the task group does not exceed its maximum load capacity. Formula (6) indicates that the number of unmanned transport equipment participating in the task does not exceed the maximum number of unmanned transport equipment specified by the system; Formula (7) ensures that the starting point and end point of the distribution route of each vehicle are both distribution centers; Formula (8) requires that the start time of the task must be within the time window of the task. Formulas (9) and (10) indicate that a task group can only be executed by one unmanned transport equipment, and at the same time, one unmanned transport equipment can only execute one task group. Formulas (11) and (12) ensure that all tasks are executed. Formula (13) represents the travel time of the unmanned transport equipment between task points. Formula (14) represents the actual start time of the task point. Formula (15) represents the time when the unmanned transport equipment actually arrives at the task point. Formula (16) ensures that the unmanned transport equipment leaves the task point. Arrival at the mission point Time at the mission point Formula (17) represents the waiting time of the unmanned transport equipment at the task point.

[0041] In the specific implementation process of step 203, the initial population is obtained through chaotic mapping, including: randomly generating initial chaotic variables, and performing iterative calculations based on the chaotic mapping function until the number of generated chaotic sequences is equal to the chromosome length; then transforming the generated chaotic sequence to the search space of the objective function until the task sequence meets the population size to generate the initial population.

[0042] It can be understood that the chaotic map has strong randomness, ergodicity and regularity, and is sensitive to initial values ​​and parameters, which can make the algorithm converge faster. The chaotic map can generate random numbers to replace the initial position of the conventional pseudo-random number generation group. In this embodiment, the Tent chaotic map is used to initialize the population to obtain a population with better uniformity and diversity, and its expression is: (18) In the formula, Indicates individual serial number; Represents the chaotic variable number; represents the chaos parameter, and its value range is (0, 1); in this implementation, , is a chaotic variable.

[0043] Specifically, first initialize the parameters and set the maximum number of iterations , population size , upper and lower limits of crossover mutation rate, etc.

[0044] Then the initial population is obtained through chaotic mapping. The specific steps include: first, randomly generate initial chaotic variables , and iterate according to formula (18). If , then recalculate until the number of generated chaotic sequences is equal to the chromosome length. Then, transform the generated chaotic sequence into the objective function according to formula (19), that is, the search space under the multi-task combination model, and repeat the above operation until the task sequence meets the population size to generate the initial population .

[0045] (19) In the formula, and They represent the lower and upper bounds of the search space respectively.

[0046] In the specific implementation process of step 204, the initial population is subjected to fast non-dominated sorting, the basic idea of ​​which is to determine the dominance relationship of the individual Pareto solution set according to the size relationship of the values ​​obtained by all individuals in the population on the objective function, and then divide the non-dominated levels according to the Pareto dominance relationship.

[0047] The specific steps of fast non-dominated sorting are as follows: Step 211, solution set Each solution in Set two parameters and ;in, Indicates being Dominant solution set, Indicates dominance The number of solutions.

[0048] Step 212, setting ,Will The individuals are stored in the collection middle. Step 213, Individuals in, traverse each solution of ,Will For each solution Minus 1.

[0049] Step 214, let , if it exists , then the corresponding solution is saved in middle.

[0050] Step 215, repeat steps 213 and 214 until all populations All individuals in the middle.

[0051] The dynamic crowding ranking strategy includes: deleting the individual with the largest crowding, updating the crowding of two adjacent individuals; then, according to the change of crowding, deleting the individual with the largest crowding again, updating the crowding of two adjacent individuals; repeating the above deletion and updating process to complete the dynamic crowding ranking. It can be understood that this embodiment takes into account the change of crowding of adjacent individuals in real time, and through the dynamic crowding ranking strategy, the distribution of the solution set can be accurately reflected.

[0052] Tournament selection is a method for selecting parent individuals, which first compares them according to the non-dominated sorting level. If two individuals are in the same level, the parent is determined by comparing their crowding distances, and finally the parent is obtained. Parent population .

[0053] In the specific implementation process of step 205, the chromosome is cross-processed by using an adaptive crossover operation, which can ensure that there are no repeated genes in a chromosome. Specifically, two chromosomes are obtained from the population as parent chromosomes, and two crossover points are randomly selected to determine the exchange area. A corresponding matching relationship is established based on the two groups of genes in the crossover area, and the unexchanged part of each chromosome is traversed. If there is a gene that repeats the exchange part, the repeated gene is replaced with a corresponding gene according to this matching relationship, and the conflict can be eliminated. This is repeated until each chromosome does not contain a repeated gene.

[0054] Furthermore, for each parent population , an adaptive crossover operation is performed based on the crossover rate to generate new individuals, and then a Gaussian mutation operation is performed based on the mutation rate to obtain the offspring population .

[0055] It can be understood that the crossover rate and mutation rate have a great influence on the global convergence of the algorithm and the breakthrough of the local optimum, and the crossover rate and mutation rate have different effects at different evolutionary stages. In the early stage of the algorithm, higher crossover rate and mutation rate help to quickly explore the search space and find possible solutions. This can ensure that the algorithm will not fall into the local optimal solution too early, but can cover a wider solution space; in the later stage of the algorithm, the population quality is relatively good and relatively stable, and the crossover mutation rate should be gradually reduced, which can increase the algorithm's exploration of local areas, help to refine the solutions that have been found and further improve the quality of the solutions, protect excellent genes from being destroyed, and avoid stagnation of algorithm search. Due to the coupling between the objectives in the multi-objective optimization problem, and in order to prevent the algorithm search process from becoming a random search, this embodiment dynamically adjusts the crossover rate and mutation rate according to the number of population iterations and the non-dominated level of the population individuals. Among them, the expression of the crossover rate is: (20) The expression of mutation rate is: (twenty one) In the formula, and Indicates the upper and lower limits of the crossover rate; and Indicates the upper and lower limits of the variation value; represents the mean of the non-dominated ranks of all individuals in the current population; represents the smaller value of the non-dominated ranks of two crossed individuals; Represents the evolutionary generation of the current population; represents the total number of evolutionary generations of the population; Indicates the maximum non-dominated level of the current population.

[0056] When the non-dominated level of the crossover and mutation objects is high, that is, the chromosome quality is poor, the maximum crossover rate and mutation rate are used to promote individual evolution; when the non-dominated level of the crossover and mutation objects is low, that is, the chromosome quality is good, the crossover rate and mutation rate are appropriately reduced according to the population evolution stage. In addition, if the current population has a large number of non-dominated layers, that is, the overall convergence of the population is poor, the crossover rate and mutation rate can be appropriately increased to speed up the population convergence. In particular, if all individuals in the current population do not dominate each other, the maximum crossover rate and mutation rate are used to increase the population diversity. Through dynamic adjustment, the diversity and convergence speed of the population can be effectively balanced, thereby improving the efficiency and quality of the optimization algorithm and helping the optimization algorithm perform better in more complex problems.

[0057] In addition, when performing mutation operations, the improved Gaussian mutation is used for mutation operations. The improved Gaussian mutation introduces randomness, which can effectively balance local search and global search, and improve the convergence and search efficiency of the algorithm. Furthermore, by adjusting the mean and standard deviation parameters, the intensity of the mutation can be controlled, which allows the algorithm to increase diversity when a larger mutation is required, or converge to a closer solution when a smaller mutation is required. It has significant advantages in local search.

[0058] The improved Gaussian variation expression is: (twenty two) In the formula, Indicates that the population Individual The position of a gene in the gene sequence, , ; express The position of the new individual genes generated in the gene sequence after the mutation operation; represents a Gaussian variable with a mean of 0 and a variance of 1; Remaining operation; Indicates a rounding operation.

[0059] In the specific implementation process of step 206, it is proposed to use a probability distribution function to screen excellent individuals. Through the designed probability distribution function, the corresponding retention probability can be assigned to individuals of different non-dominated levels according to the population evolution. In the early stage of the algorithm, due to the poor quality of the population, high-quality individuals obtain a higher retention probability to accelerate the evolution of the population. With the iteration of the algorithm, the number of individuals in the first non-dominated level increases sharply. In order to maintain the diversity of the population, the retention probability of individuals in the first non-dominated level is reduced, while the retention probability of individuals in other non-dominated levels is increased to avoid falling into a local optimal solution.

[0060] The probability distribution function expression is: (twenty three) (twenty four) In the formula, represents the non-dominant level; Indicates The set of individuals retained on the non-dominated level; Indicates The probability that individuals on the non-dominated level are retained; Indicates The set of all individuals in the non-dominated level; Indicates The number of individuals on the non-dominated level; Indicates the population size; Represents the evolutionary generation of the current population; Represents the total number of evolutionary generations of the population.

[0061] It is worth noting that when the probability distribution function is used to select excellent individuals, a global memory bank of size N is also set up. With offspring population When , the global memory is used to store the remaining population size The intermediate population , and through the intermediate population Finally, merge the global memory bank and the last generation population, perform non-dominated sorting and crowding calculation, select the retained offspring, and Filter excellent individuals, and obtain a new parent population .

[0062] In the specific implementation process of step 207, iterative calculation is performed based on the new parent population to gradually optimize the results. When , the algorithm ends and the optimal multi-task allocation scheme is obtained; the unmanned transport equipment is multi-tasked by the optimal multi-task allocation scheme. Otherwise, let , continue with step 205.

[0063] In one embodiment, using A task scheduling simulation experiment is carried out to verify the superiority of the method proposed in the present invention.

[0064] The reagents required by the laboratory are stored in the automatic reagent library. There are 32 test benches in total for the experiments. The multi-task combination model with the goal of minimizing total cost and maximizing on-time completion is applied to the task distribution problem. The relevant parameters of the laboratory are shown in Table 1.

[0065] Table 1 Laboratory related parameters

[0066] The material demand, delivery time window and workstation coordinates of each workstation are shown in Table 2. In addition, the coordinates of the material distribution center are (80, 80). There are 32 workstations here, the demand represents the load required by the task, and the expected time window represents the penalty-free time interval for the material to arrive at the workstation.

[0067] Table 2 Coordinates, time windows and material requirements of each task point

[0068] In order to verify that the method provided by the present invention has a better optimization effect, a simulation comparison experiment was designed based on MATLAB 2020b on the Windows 11 operating system to compare the method provided by the present invention with the traditional NSGA-II algorithm. The relevant parameters of the NSGA-II algorithm are set as follows: and , crossover rate , mutation rate , the crossover rate and mutation rate of the present invention are calculated by formula (20) and formula (21). Figure 3 Schematic diagram of the Pareto frontier for task allocation for the NSGA-II algorithm. Figure 4 The figure is a schematic diagram of the Pareto frontier for task allocation using the method provided by the present invention. By comparison, it can be seen that the Pareto optimal frontier obtained by the method provided by the present invention is better in terms of both system total cost and system punctuality, and the quality of the solution is better than that of the traditional NSGA-II algorithm, which verifies the superiority of the method provided by the present invention.

[0069] Figure 5 This is a schematic diagram of the trajectory for path planning using the NSGA-II algorithm. Figure 6Table 3 is a task distribution path table of each AGV using the NSGA-II algorithm and the method provided by the present invention.

[0070] Table 3. Task delivery path table of AGVS before and after algorithm improvement

[0071] The numbers in Table 3 represent the order of the tasks. All AGVs start from the material center and need to return to the material center after completing all the delivery tasks to wait for the next delivery task. Table 4 shows the specific optimization results and objective function value information of the delivery situation.

[0072] Table 4 Objective function values ​​and comparison before and after algorithm improvement

[0073] As shown in Table 4, compared with the traditional NSGA-II algorithm, the AGV task allocation strategy obtained by the method provided by the present invention reduces the total system cost by 10.02% and increases the system on-time completion rate by 8.4%. This further achieves the optimization goal and improves the overall distribution efficiency of laboratory AGVs.

[0074] Although this embodiment Figure 1 The steps in the process are shown in sequence as indicated by the arrows, but the steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of the steps, and the steps can be executed in other orders. Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0075] Example 2 Based on the multi-task allocation method of unmanned transport equipment in Example 1, this embodiment discloses a multi-task allocation device for unmanned transport equipment, such as Figure 7 As shown, the multi-task allocation device of the unmanned transport equipment includes: a data acquisition module 401, a model construction module 402, an initial population acquisition module 403, a parent population acquisition module 404, a crossover mutation operation module 405, a new parent population acquisition module 406 and an iterative update module 407, wherein: The data acquisition module 401 is used to acquire the number of unmanned transport equipment and the types of tasks.

[0076] The model building module 402 is used to build a multi-task combination model with the goal of minimizing total cost and maximizing on-time completion according to the number of unmanned transport equipment and the types of tasks, and set relevant constraints.

[0077] The initial population acquisition module 403 is used to perform parameter initialization and obtain the initial population through chaotic mapping based on the multi-task combination model and constraint conditions.

[0078] The parent population acquisition module 404 is used to perform fast non-dominated sorting on the initial population, and determine a number of parent populations through tournament selection based on a dynamic congestion sorting strategy.

[0079] The crossover mutation operation module 405 is used to generate new individuals for each parent population through an adaptive crossover operation, and then perform a mutation operation through a Gaussian mutation to obtain a child population; wherein randomness is introduced into the Gaussian mutation, and the mean and standard deviation parameters are adjusted to obtain an improved Gaussian mutation, and the mutation operation is performed through the improved Gaussian mutation.

[0080] The new parent population acquisition module 406 is used to merge the parent population and the child population to obtain an intermediate population; perform a fast non-dominated sorting operation on the intermediate population, select excellent individuals through a probability distribution function, and obtain a new parent population.

[0081] The iterative update module 407 performs iterative calculation based on the new parent population, gradually optimizes the results, and obtains the optimal multi-task allocation scheme; and performs multi-task allocation on the unmanned transport equipment through the optimal multi-task allocation scheme.

[0082] In this embodiment, the specific working process and working principle of the data acquisition module 401, the model construction module 402, the initial population acquisition module 403, the parent population acquisition module 404, the crossover mutation operation module 405, the new parent population acquisition module 406 and the iterative update module 407 are the same as those in the method of embodiment 1, so they are not repeated in this embodiment. Each unit module can be implemented in whole or in part by software, hardware and a combination thereof, and each unit module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above unit modules.

[0083] Example 3 like Figure 8 The terminal device disclosed in this embodiment includes a transmitter, a receiver, a memory, and a processor. The transmitter is used to send instructions and data, the receiver is used to receive instructions and data, the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions stored in the memory to implement the method in the above-mentioned embodiment 1.

[0084] It should be noted that the above memory can be independent or integrated with the processor. When the memory is independently provided, the terminal device further includes a bus for connecting the memory and the processor.

[0085] Example 4 This embodiment discloses a computer-readable storage medium, in which computer-executable instructions are stored. When a processor executes the computer-executable instructions, the method in the above-mentioned embodiment 1 is implemented.

[0086] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0087] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0088] The above-mentioned embodiments only express several implementation modes of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the invention. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the attached claims.

Claims

1. A multi-task allocation method for unmanned transport equipment, characterized in that: The method comprises: Obtain the number of unmanned transport equipment and the types of missions; According to the number of unmanned transport equipment and the types of tasks, a multi-task combination model with the goal of minimizing total cost and maximizing on-time completion is constructed, and relevant constraints are set; Initializing parameters, and obtaining an initial population through chaotic mapping based on the multi-task combination model and constraint conditions; Performing a fast non-dominated sort on the initial population, and determining a number of parent populations through tournament selection based on a dynamic crowding sorting strategy; For each of the parent populations, a new individual is generated through an adaptive crossover operation, and then a mutation operation is performed through a Gaussian mutation to obtain a child population; wherein randomness is introduced into the Gaussian mutation, and mean and standard deviation parameters are adjusted to obtain an improved Gaussian mutation, and the mutation operation is performed through the improved Gaussian mutation; The parent population and the child population are merged to obtain an intermediate population; a fast non-dominated sorting operation is performed on the intermediate population, and excellent individuals are screened through a probability distribution function to obtain a new parent population; Iterative calculations are performed based on the new parent population, and the results are gradually optimized to obtain an optimal multi-task allocation scheme; and multi-task allocation is performed on the unmanned transport equipment through the optimal multi-task allocation scheme.

2. The multi-task allocation method for unmanned transport equipment according to claim 1, characterized in that: A multi-task combination model with the goal of minimizing total cost and maximizing on-time completion is constructed. The expression of the multi-task combination model is: ； ； ； ； In the formula, Indicates the lowest total cost; Indicates the electric energy coefficient per unit load during the driving process of unmanned transport equipment; Indicates the weight loaded by the unmanned transport equipment when performing a mission; It represents the ratio of the straight-line driving time of unmanned transport equipment to the power consumption; It represents the ratio of turning time and power consumption of unmanned transport equipment; Indicates Unmanned transport equipment from the mission To the task The decision variables; Indicates Unmanned transport equipment from the distribution center to the mission The decision variables; Indicates Unmanned transport equipment from the distribution center to the mission Straight-line driving time; Indicates Unmanned transport equipment from the distribution center to the mission Turning time; Indicates Unmanned transport equipment from the mission To the task Straight-line driving time; Indicates Unmanned transport equipment from the mission To the task Turning time; It represents the ratio of waiting time to power consumption of unmanned transport equipment; Indicates completion of task Required working hours; Indicates completion of task Required working hours; It indicates the highest degree of on-time completion; represents the punctuality function; Indicates Unmanned transport equipment to carry out The load of each task; Representation Task Desired delivery time window; Represents the task in the time window penalty function Acceptable delivery time windows; Indicates the delivery time.

3. The multi-task allocation method for unmanned transport equipment according to claim 2, characterized in that: Set relevant constraints, the expression of which is: ； ； ； ； ； ； ； ； ； ； ； ； ； In the formula, Indicates Unmanned transport equipment from the mission To the task The decision variables; Indicates The maximum load capacity of unmanned transport equipment; Indicates the number of unmanned transport equipment; Indicates the type of task; Indicates the task point The horizontal axis of Indicates the task point The horizontal axis of Indicates the task point The vertical coordinate of Indicates the task point The vertical coordinate of Indicates The speed of unmanned transport equipment; Representation Task Actual start time; Representation Task Estimated start time; Indicates Unmanned transport equipment actually arrived at the mission time; Indicates Unmanned transport equipment actually arrived at the mission time; Indicates that the unmanned transport equipment is from the mission point To the mission point The travel time between Indicates Unmanned transport equipment completed the mission Required working hours; Indicates that the unmanned transport equipment is at the mission point and mission points The travel time between Indicates Unmanned transport equipment actually arrived at the mission time; Indicates Unmanned transport equipment from the mission point Waiting time.

4. The multi-task allocation method for unmanned transport equipment according to claim 1, characterized in that: The dynamic congestion ranking strategy includes: Delete the individual with the highest crowding degree and update the crowding degree of the two adjacent individuals; Then, according to the change of crowding, the individual with the largest crowding is deleted again, and the crowding of the two adjacent individuals is updated; Repeat the above deletion and update process to complete the dynamic congestion sorting.

5. The multi-task allocation method for unmanned transport equipment according to claim 1, characterized in that: For each of the parent populations, a new individual is generated by an adaptive crossover operation, and then a mutation operation is performed by Gaussian mutation to obtain a child population, including: For each of the parent populations, an adaptive crossover operation is performed based on the crossover rate to generate new individuals, and then a Gaussian mutation operation is performed based on the mutation rate to obtain a child population; Wherein, the expression of the crossover rate is: ； The expression of the mutation rate is: 1 ； In the formula, and Indicates the upper and lower limits of the crossover rate; and Indicates the upper and lower limits of the variation value; represents the mean of the non-dominated ranks of all individuals in the current population; represents the smaller value of the non-dominated ranks of two crossed individuals; Represents the evolutionary generation of the current population; represents the total number of evolutionary generations of the population; Indicates the maximum non-dominated level of the current population.

6. The multi-task allocation method for unmanned transport equipment according to claim 5, characterized in that: The improved Gaussian mutation expression is: ； In the formula, Indicates that the population Individual The position of a gene in the gene sequence, , ; express The position of the new individual genes generated in the gene sequence after the mutation operation; represents a Gaussian variable with a mean of 0 and a variance of 1; Remaining operation; Indicates a rounding operation.

7. The multi-task allocation method for unmanned transport equipment according to any one of claims 1 to 5, characterized in that: The probability distribution function expression is: ； ； In the formula, represents the non-dominant level; Indicates The set of individuals retained on the non-dominated level; Indicates The probability that individuals on the non-dominated level are retained; Indicates The set of all individuals in the non-dominated level; Indicates The number of individuals on the non-dominated level; Indicates the population size; Represents the evolutionary generation of the current population; Represents the total number of evolutionary generations of the population.

8. A multi-task allocation device for unmanned transport equipment, characterized in that: The device comprises: The data acquisition module is used to obtain the number of unmanned transport equipment and the types of tasks; A model building module, used to build a multi-task combination model with the goal of minimizing total cost and maximizing on-time completion according to the number of unmanned transport equipment and the types of tasks, and set relevant constraints; An initial population acquisition module is used to perform parameter initialization and obtain an initial population through chaotic mapping based on the multi-task combination model and constraint conditions; A parent population acquisition module is used to perform fast non-dominated sorting on the initial population and determine a number of parent populations through tournament selection based on a dynamic congestion sorting strategy; A crossover mutation operation module is used to generate new individuals for each parent population through an adaptive crossover operation, and then perform a mutation operation through a Gaussian mutation to obtain a child population; wherein randomness is introduced into the Gaussian mutation, and the mean and standard deviation parameters are adjusted to obtain an improved Gaussian mutation, and the mutation operation is performed through the improved Gaussian mutation; A new parent population acquisition module is used to merge the parent population and the child population to obtain an intermediate population; perform a fast non-dominated sorting operation on the intermediate population, select excellent individuals through a probability distribution function, and obtain a new parent population; The iterative update module performs iterative calculations based on the new parent population, gradually optimizes the results, and obtains the optimal multi-task allocation plan; and multi-tasks are allocated to the unmanned transport equipment through the optimal multi-task allocation plan.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the multi-task allocation method for the unmanned transportation equipment according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the multi-task allocation method for unmanned transportation equipment described in any one of claims 1 to 7 are implemented.

Images