阅读说明：本技术 一种云机器人的任务调度方法、装置、设备及存储介质 (Task scheduling method, device, equipment and storage medium of cloud robot ) 是由 张腾 高明 金长新 于 2020-05-29 设计创作，主要内容包括：本申请公开了一种云机器人的任务调度方法,首先调用预设的随机编码函数对与目标调度任务对应的云机器人的目标特征属性进行随机编码,得出多个不同的初始特征编码；然后调用预先设置的遗传算法,根据初始特征编码确定出与目标调度任务适应度最高的目标特征编码；再根据目标特征编码确定出对应的目标特征属性信息,并根据目标特征属性信息调度对应的目标机器人。本方法不仅降低了对人力资源的消耗,提高任务调度的效率,而且可以有效地避免工作人员个人主观因素影响,提高制定的云机器人的任务调度方案的优化性。本申请还公开了一种云机器人的任务调度装置、设备及计算机可读存储介质,均具有上述有益效果。(The application discloses a task scheduling method of a cloud robot, which comprises the steps of calling a preset random coding function to carry out random coding on target characteristic attributes of the cloud robot corresponding to a target scheduling task to obtain a plurality of different initial characteristic codes; then, a preset genetic algorithm is called, and a target feature code with the highest fitness with the target scheduling task is determined according to the initial feature code; and determining corresponding target characteristic attribute information according to the target characteristic codes, and scheduling the corresponding target robot according to the target characteristic attribute information. The method not only reduces the consumption of manpower resources and improves the task scheduling efficiency, but also can effectively avoid the influence of personal subjective factors of workers and improve the optimization of the formulated task scheduling scheme of the cloud robot. The application also discloses a task scheduling device, equipment and a computer readable storage medium of the cloud robot, which have the beneficial effects.)

1. A task scheduling method of a cloud robot is characterized by comprising the following steps:

calling a preset random coding function to carry out random coding on the target characteristic attribute of the cloud robot corresponding to the target scheduling task, and obtaining a plurality of different initial characteristic codes;

calling a preset genetic algorithm, and determining a target feature code with the highest fitness with the target scheduling task according to the initial feature code;

and determining corresponding target characteristic attribute information according to the target characteristic codes, and scheduling the corresponding target robot according to the target characteristic attribute information.

2. The method according to claim 1, wherein the process of calling a preset random encoding function to randomly encode the target feature attributes of the cloud robot corresponding to the target scheduling task to obtain a plurality of different initial feature codes specifically comprises:

determining the target characteristic attribute of the cloud robot corresponding to the target scheduling task;

setting corresponding random coding bit numbers according to the weight of each target characteristic attribute;

and calling a preset random coding function to carry out random coding on each target characteristic attribute according to the number of the random coding bits to obtain a plurality of different initial characteristic codes.

3. The method according to claim 1, wherein the step of invoking a preset genetic algorithm and determining a target feature code with the highest fitness with the target scheduling task according to the initial feature code specifically comprises:

inputting the initial feature codes into a first fitness function, and screening out first intermediate feature codes with fitness meeting a preset condition by using the first fitness function;

performing cross operation and mutation operation on each first intermediate characteristic code to obtain a second intermediate characteristic code;

and screening out the target feature codes with the highest target scheduling task adaptability from the second intermediate feature codes through a second adaptability function under the condition that the first iterative robot population corresponding to the second intermediate feature codes reaches a convergence state.

4. The method according to claim 3, wherein the step of performing the interleaving operation and the mutation operation on each of the first intermediate feature codes to obtain a second intermediate feature code comprises:

performing cross operation on each first intermediate feature code to obtain a cross feature code, and determining a second iterative robot population according to the cross feature code;

judging whether the second iterative robot population reaches a convergence state;

if so, determining the second intermediate feature code according to the cross feature code and the first intermediate feature code;

if not, calling a preset random change algorithm to randomly change the target feature attributes in the first intermediate feature codes and/or the cross feature codes to obtain variant feature codes, and determining the second intermediate feature codes according to the variant feature codes, the cross feature codes and the first intermediate feature codes.

5. The method according to claim 3, wherein the step of performing the interleaving operation and the mutation operation on each of the first intermediate feature codes to obtain a second intermediate feature code comprises:

and performing cross operation and mutation operation on each first intermediate feature code based on the cross probability and the mutation probability to obtain the second intermediate feature code.

6. The method according to claim 3, characterized in that the first fitness function and/or the second fitness function is/are in particular an order-based fitness function.

7. Method according to any one of claims 1 to 6, characterized in that said random encoding function is in particular a real encoding function.

8. A task scheduling apparatus of a cloud robot, comprising:

the encoding module is used for calling a preset random encoding function to carry out random encoding on the target characteristic attribute of the cloud robot corresponding to the target scheduling task so as to obtain a plurality of different initial characteristic codes;

the calculation module is used for calling a preset genetic algorithm and determining a target feature code with the highest fitness with the target scheduling task according to the initial feature code;

and the determining module is used for determining corresponding target characteristic attribute information according to the target characteristic codes and scheduling the corresponding target robot according to the target characteristic attribute information.

9. A task scheduling apparatus of a cloud robot, comprising:

a memory for storing a computer program;

a processor for implementing the steps of the task scheduling method of the cloud robot according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that a computer program is stored thereon, which, when being executed by a processor, implements the steps of the task scheduling method of a cloud robot according to any one of claims 1 to 7.

Technical Field

The invention relates to the field of cloud robots, in particular to a task scheduling method, a task scheduling device, task scheduling equipment and a computer-readable storage medium of a cloud robot.

Background

With the gradual maturity of cloud technology, the clustered cloud robot gradually leaves a new way in the industrial and civil fields. Due to the cloud characteristic of the cloud robot, task scheduling can be performed on the cloud robot uniformly through the management platform. When a plurality of cloud robots cooperate to complete a task, different scheduling schemes can be provided due to the influence of factors such as positions, robot attributes and the like. In the prior art, generally, a worker creates a task scheduling scheme through personal intuitive experience. However, due to the complex realistic situation corresponding to the task, the method in the prior art not only needs to consume a large amount of human resources and has low task scheduling efficiency, but also the task scheduling scheme is easily affected by personal subjective factors, so that the task scheduling scheme has errors or is not optimized.

Therefore, how to perform task scheduling on the cloud robot not only can reduce human resources required to be consumed and improve the efficiency of task scheduling, but also can improve the optimization of task scheduling, which is a technical problem that needs to be solved by technical personnel in the field.

Disclosure of Invention

In view of the above, an object of the present invention is to provide a task scheduling method for a cloud robot, which can reduce human resources required to be consumed, improve task scheduling efficiency, and improve optimization of task scheduling when performing task scheduling on the cloud robot; another object of the present invention is to provide a task scheduling apparatus, a device and a computer readable storage medium for a cloud robot, all of which have the above beneficial effects.

In order to solve the technical problem, the invention provides a task scheduling method of a cloud robot, which comprises the following steps:

calling a preset random coding function to carry out random coding on the target characteristic attribute of the cloud robot corresponding to the target scheduling task, and obtaining a plurality of different initial characteristic codes;

calling a preset genetic algorithm, and determining a target feature code with the highest fitness with the target scheduling task according to the initial feature code;

and determining corresponding target characteristic attribute information according to the target characteristic codes, and scheduling the corresponding target robot according to the target characteristic attribute information.

Preferably, the calling a preset random coding function randomly codes the target feature attributes of the cloud robot corresponding to the target scheduling task to obtain a plurality of different initial feature codes, and the process specifically includes:

determining the target characteristic attribute of the cloud robot corresponding to the target scheduling task;

setting corresponding random coding bit numbers according to the weight of each target characteristic attribute;

and calling a preset random coding function to carry out random coding on each target characteristic attribute according to the number of the random coding bits to obtain a plurality of different initial characteristic codes.

Preferably, the step of calling a preset genetic algorithm and determining a target feature code with the highest fitness with the target scheduling task according to the initial feature code specifically includes:

inputting the initial feature codes into a first fitness function, and screening out first intermediate feature codes with fitness meeting a preset condition by using the first fitness function;

performing cross operation and mutation operation on each first intermediate characteristic code to obtain a second intermediate characteristic code;

and screening out the target feature codes with the highest target scheduling task adaptability from the second intermediate feature codes through a second adaptability function under the condition that the first iterative robot population corresponding to the second intermediate feature codes reaches a convergence state.

Preferably, the process of performing a crossover operation and a mutation operation on each of the first intermediate feature codes to obtain a second intermediate feature code specifically includes:

performing cross operation on each first intermediate feature code to obtain a cross feature code, and determining a second iterative robot population according to the cross feature code;

judging whether the second iterative robot population reaches a convergence state;

if so, determining the second intermediate feature code according to the cross feature code and the first intermediate feature code;

if not, calling a preset random change algorithm to randomly change the target feature attributes in the first intermediate feature codes and/or the cross feature codes to obtain variant feature codes, and determining the second intermediate feature codes according to the variant feature codes, the cross feature codes and the first intermediate feature codes.

Preferably, the process of performing a crossover operation and a mutation operation on each of the first intermediate feature codes to obtain a second intermediate feature code specifically includes:

and performing cross operation and mutation operation on each first intermediate feature code based on the cross probability and the mutation probability to obtain the second intermediate feature code.

Preferably, the first fitness function and/or the second fitness function are/is a sequence-based fitness function.

Preferably, the random encoding function is a real number encoding function.

In order to solve the above technical problem, the present invention further provides a task scheduling apparatus for a cloud robot, including:

the encoding module is used for calling a preset random encoding function to carry out random encoding on the target characteristic attribute of the cloud robot corresponding to the target scheduling task so as to obtain a plurality of different initial characteristic codes;

the calculation module is used for calling a preset genetic algorithm and determining a target feature code with the highest fitness with the target scheduling task according to the initial feature code;

and the determining module is used for determining corresponding target characteristic attribute information according to the target characteristic codes and scheduling the corresponding target robot according to the target characteristic attribute information.

In order to solve the above technical problem, the present invention further provides a task scheduling apparatus for a cloud robot, including:

a memory for storing a computer program;

and the processor is used for realizing the steps of any one of the cloud robot task scheduling methods when executing the computer program.

In order to solve the technical problem, the present invention further provides a computer-readable storage medium, where a computer program is stored, and when the computer program is executed by a processor, the computer program implements the steps of any one of the above task scheduling methods for a cloud robot.

The invention provides a task scheduling method of a cloud robot, which comprises the steps of calling a preset random coding function to carry out random coding on target characteristic attributes of the cloud robot corresponding to a target scheduling task to obtain a plurality of different initial characteristic codes; then, a preset genetic algorithm is called, and a target feature code with the highest fitness with the target scheduling task is determined according to the initial feature code; and determining corresponding target characteristic attribute information according to the target characteristic codes, and scheduling the corresponding target robot according to the target characteristic attribute information. Therefore, the method is characterized in that on the basis of the initial feature codes, global search and optimization are carried out through a genetic algorithm to obtain the target feature codes with the highest fitness with the target scheduling task, and a method for creating a task scheduling scheme through personal visual experience of workers in the prior art is replaced, so that the consumption of manpower resources is reduced, the task scheduling efficiency is improved, the genetic algorithm is calculated based on the initial feature codes corresponding to the target feature attributes, the influence of personal subjective factors of the workers can be effectively avoided, and the optimization of the formulated task scheduling scheme of the cloud robot is improved.

In order to solve the technical problem, the invention further provides a task scheduling device, equipment and a computer readable storage medium of the cloud robot, and the task scheduling device, the equipment and the computer readable storage medium have the beneficial effects.

Drawings

In order to more clearly illustrate the embodiments or technical solutions of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.

Fig. 1 is a flowchart of a task scheduling method of a cloud robot according to an embodiment of the present invention;

fig. 2 is a structural diagram of a task scheduling apparatus of a cloud robot according to an embodiment of the present invention;

fig. 3 is a structural diagram of a task scheduling device of a cloud robot according to an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

The core of the embodiment of the invention is to provide a task scheduling method of a cloud robot, which can reduce the human resources required to be consumed, improve the task scheduling efficiency and improve the optimization of task scheduling when the task scheduling is performed on the cloud robot; another core of the present invention is to provide a task scheduling apparatus, a device and a computer-readable storage medium for a cloud robot, all having the above beneficial effects.

In order that those skilled in the art will better understand the disclosure, the invention will be described in further detail with reference to the accompanying drawings and specific embodiments.

Fig. 1 is a flowchart of a task scheduling method of a cloud robot according to an embodiment of the present invention. As shown in fig. 1, a task scheduling method of a cloud robot includes:

s10: and calling a preset random coding function to randomly code the target characteristic attribute of the cloud robot corresponding to the target scheduling task to obtain a plurality of different initial characteristic codes.

Specifically, in this embodiment, first, the factor to be referred to needs to be determined according to the target scheduling task, that is, the cloud robot target feature attribute corresponding to the target scheduling task is determined. The characteristic attributes can be the type, speed, real-time endurance and the like of a mechanical arm mounted by the robot, and the target characteristic attributes are characteristic attributes related to a target scheduling task.

And then, calling a preset random coding function to randomly code the target characteristic attribute, namely setting different coded data for the target characteristic attribute by using the random coding function to express the cloud robot with different attribute values. For example, for a target attribute feature of speed, coded data 1123 and coded data 1124 correspond to values representing different attributes, i.e., different cloud robots. The random encoding function may be a real number encoding function or a binary encoding function, and the like, which is not limited in this embodiment. In actual operation, a plurality of different types of target characteristic attributes are generally determined according to a target scheduling task, and a plurality of different initial characteristic codes are obtained by randomly encoding each target characteristic attribute. That is, one initial feature code includes coded data corresponding to a plurality of different types of target feature attributes, and one initial feature code represents one type of cloud robot having feature attributes corresponding to the initial feature code; a plurality of different initial signature codes represent an initial population of robots.

S20: calling a preset genetic algorithm, and determining a target feature code with the highest fitness with the target scheduling task according to the initial feature code;

s30: and determining corresponding target characteristic attribute information according to the target characteristic codes, and scheduling the corresponding target robot according to the target characteristic attribute information.

Specifically, a genetic algorithm is preset, after the initial feature code is obtained, the preset genetic algorithm is called, the genetic algorithm is used for calculation according to the initial feature code, and the target feature code with the highest fitness with the target scheduling task is determined. It should be noted that the Genetic Algorithm (Genetic Algorithm) is a calculation model of a biological evolution process for simulating natural selection and a Genetic mechanism of the darwinian biological evolution theory, and is a method for searching an optimal solution by simulating the natural evolution process.

Specifically, after the target feature code is determined, the target feature code is decoded according to the previous coding format, and the decoded result is the target feature attribute information. That is to say, for the target scheduling task, the optimized scheduling scheme is to schedule the target cloud robot having the target feature code, so that the corresponding target feature attribute information is determined according to the target feature code, and the corresponding target robot is scheduled according to the target feature attribute information.

The task scheduling method of the cloud robot provided by the embodiment of the invention comprises the steps of calling a preset random coding function to carry out random coding on target characteristic attributes of the cloud robot corresponding to a target scheduling task, and obtaining a plurality of different initial characteristic codes; then, a preset genetic algorithm is called, and a target feature code with the highest fitness with the target scheduling task is determined according to the initial feature code; and determining corresponding target characteristic attribute information according to the target characteristic codes, and scheduling the corresponding target robot according to the target characteristic attribute information. Therefore, the method is characterized in that on the basis of the initial feature codes, global search and optimization are carried out through a genetic algorithm to obtain the target feature codes with the highest fitness with the target scheduling task, and a method for creating a task scheduling scheme through personal visual experience of workers in the prior art is replaced, so that the consumption of manpower resources is reduced, the task scheduling efficiency is improved, the genetic algorithm is calculated based on the initial feature codes corresponding to the target feature attributes, the influence of personal subjective factors of the workers can be effectively avoided, and the optimization of the formulated task scheduling scheme of the cloud robot is improved.

On the basis of the foregoing embodiment, this embodiment further describes and optimizes the technical solution, and specifically, in this embodiment, a process of calling a preset random coding function to randomly code a target feature attribute of the cloud robot corresponding to a target scheduling task to obtain a plurality of different initial feature codes specifically includes:

determining a target characteristic attribute of the cloud robot corresponding to the target scheduling task;

setting corresponding random coding bit numbers according to the weight of each target characteristic attribute;

and calling a preset random coding function to carry out random coding on each target characteristic attribute according to the number of the random coding bits to obtain a plurality of different initial characteristic codes.

Specifically, in this embodiment, after determining the target characteristic attributes of the cloud robot corresponding to the target scheduling task, the corresponding random encoding bits are further set according to the weight of each target characteristic attribute; generally, the greater the weight of a target feature attribute to a target scheduled task, the more bit random number is used to represent the target feature attribute. And after the random coding bit number corresponding to the target characteristic attribute is determined, calling a preset random coding function to carry out random coding on each target characteristic attribute according to the random coding bit number to obtain a plurality of different initial characteristic codes.

As a preferred implementation, in this embodiment, the random encoding function is specifically a real number encoding function.

The real number encoding means that each target characteristic attribute of the cloud robot is represented by one floating point number in a certain range, and the length of the initial characteristic encoding of the cloud robot is related to the number of the target characteristic attributes.

It can be understood that since binary coding requires frequent encoding and decoding, the amount of computation is large, and only limited discrete lattice can be generated, and additional optimal points may be generated, and there is a hamming cliff problem; the real number encoding can avoid the disadvantages, so in the embodiment, a real number encoding function is preferably used for randomly encoding the target characteristic attribute to obtain the initial characteristic encoding. Relatively speaking, the real number coding has the advantages of high precision, large search range, natural and intuitive expression and the like compared with the binary coding, and can overcome the defects of difficulty in solving the high precision problem, inconvenience in reflecting the specific knowledge of the required problem and the like caused by the characteristics of the binary coding.

As can be seen, in the present embodiment, the corresponding random encoding bit number is further set according to the weight of each target feature attribute; and each target characteristic attribute is randomly coded according to the number of the random coding bits through a real number coding function to obtain a plurality of different initial characteristic codes, so that the calculated amount can be reduced, the accuracy of determining the target characteristic codes according to the initial characteristic codes subsequently can be improved, and the optimization of a task scheduling scheme can be further improved.

On the basis of the foregoing embodiment, this embodiment further describes and optimizes the technical solution, and specifically, in this embodiment, a process of calling a preset genetic algorithm and determining a target feature code with the highest fitness with respect to a target scheduling task according to an initial feature code specifically includes:

inputting the initial feature codes into a first fitness function, and screening out first intermediate feature codes with fitness meeting a preset condition by using the first fitness function;

performing cross operation and mutation operation on each first intermediate characteristic code to obtain a second intermediate characteristic code;

and under the condition that the first iterative robot population corresponding to the second intermediate feature codes reaches a convergence state, screening out the target feature codes with the highest adaptability to the target scheduling task from the second intermediate feature codes through a second adaptability function.

Specifically, genetic algorithms include selection operations, crossover operations, and mutation operations. In this embodiment, after the initial feature codes are determined, the initial feature codes are input to a first fitness function to obtain a first fitness value corresponding to each initial feature code, where each first fitness value is a fitness of the cloud robot corresponding to each initial feature code to the target scheduling task; then, according to a preset first screening method, the fitness value with the fitness meeting the preset condition is screened out according to each first fitness value, and the corresponding first intermediate feature code is determined according to the screening result.

It should be noted that, the preset first screening method may be to arrange the first fitness values in order from large to small, and select an initial feature code corresponding to the first fitness value of TOP-N, to obtain a first intermediate feature code; or a first fitness threshold value can be preset, and the initial feature code with the first fitness value exceeding the first fitness threshold value is set as the first intermediate feature code. Of course, in actual operation, other screening methods may also be used to determine the first intermediate feature code, which is not limited in this embodiment.

Specifically, after the first intermediate feature codes are determined, the first intermediate feature codes are subjected to cross operation and mutation operation to obtain second intermediate feature codes. It can be understood that, in this embodiment, the performing of the crossover operation generally corresponds to the gene segments of the cloud robot, that is, the encoded data corresponding to the one or more target characteristic attributes; the mutation operation generally corresponds to a single genetic factor of the cloud robot, namely encoded data corresponding to a target characteristic attribute.

Specifically, after the second intermediate feature codes are obtained, whether a first iteration robot population corresponding to the second intermediate feature codes reaches an end condition of the genetic algorithm is judged, specifically, whether the first iteration robot population reaches a convergence state is judged, if yes, the current first iteration robot population converges to a stable state is shown, the second intermediate feature codes are input into a second fitness function, second fitness values corresponding to the second intermediate feature codes are obtained, then, a target fitness value with the highest fitness to the target scheduling task is screened according to a preset second screening method according to the second fitness values, and a corresponding target feature code is determined according to the screened target fitness values. Specifically, the determining whether the first iterative robot population reaches the end condition of the genetic algorithm specifically means determining whether the maximum fitness value in the first iterative robot population approaches a fixed value.

It should be noted that the fitness function is a function for evaluating the fitness of the cloud robot corresponding to the initial feature code/the cloud robot corresponding to the second intermediate feature code to the target scheduling task. The embodiment does not limit the specific type of the fitness function, and can be selected and changed according to specific situations. Specifically, fitness functions can be roughly divided into two categories: linear stretching fitness function and dynamic adjustment nonlinear fitness function.

As a preferred embodiment, the first fitness function and/or the second fitness function is/are embodied as an order-based fitness function.

Specifically, in the present embodiment, it is preferable to use an order-based fitness function of the linear-stretching fitness functions as the first fitness function and/or the second fitness function. That is, the first fitness function and the second fitness function may each use an order-based fitness function. It should be noted that the greatest advantage of the order-based fitness function is that the probability of the cloud robot being selected is independent of the specific value of the objective function, and is only dependent on the order. The construction method of the sequence-based fitness function comprises the following steps:

firstly, sequencing all cloud robots in the current population according to the quality of objective function values, setting a parameter beta belonging to (0,1), and setting a sequence-based fitness function as follows:

eval(X′_i)＝β*(1-β)^i-1i＝1,2,3,...,m；

wherein, X'_iAn ith cloud robot representing a current population; eval (X'_i) Indicating the sequential fitness of the ith cloud robot of the current population, and β indicating the sequential fitness parameter.

As a preferred embodiment, the process of performing a crossover operation and a mutation operation on each first intermediate feature code to obtain a second intermediate feature code specifically includes:

and carrying out cross operation and mutation operation on each first intermediate feature code based on the cross probability and the mutation probability to obtain a second intermediate feature code.

For cross operation and mutation operation, in the initial stage of evolution, the superior cloud robots in the robot population are almost in a state of no change, and the superior individuals (cloud robots) are not necessarily global superior, which easily leads to the evolution towards local superiority. Therefore, the embodiment improves the adaptive genetic algorithm, so that the cross probability and the variation probability of the cloud robots with large fitness values in the robot population are not zero. Wherein, the improved cross probability calculation formula is as follows:

the improved mutation probability is calculated by the following formula:

wherein f is_maxRepresenting the maximum fitness value in the current population; f. of_aveThe average fitness value of the current population is obtained; f' is a larger fitness value of the two cloud robots to be crossed; f is the adaptability value of the cloud robot to be mutated; p_c1Is the maximum cross probability; p_m1Is the maximum mutation probability.

Therefore, after the cross probability and the mutation probability are improved, the cross probability and the mutation probability of the individuals with excellent performance in the robot population are correspondingly improved, and the individuals with excellent performance cannot be in a state of being in a stagnation state, so that the genetic algorithm can jump out of the local optimal solution to obtain a global optimal solution, namely, the optimization of the task scheduling scheme can be further improved.

As a preferred embodiment, the process of performing a crossover operation and a mutation operation on each first intermediate feature code to obtain a second intermediate feature code specifically includes:

performing cross operation on each first intermediate feature code to obtain a cross feature code, and determining a second iterative robot population according to the cross feature code;

judging whether the second iterative robot population reaches a convergence state;

if yes, determining a second intermediate characteristic code according to the cross characteristic code and the first intermediate characteristic code;

if not, calling a preset random change algorithm to randomly change the target feature attributes in the first intermediate feature codes and/or the cross feature codes to obtain variant feature codes, and determining second intermediate feature codes according to the variant feature codes, the cross feature codes and the first intermediate feature codes.

Specifically, in this embodiment, after the first intermediate feature codes are obtained, the first intermediate feature codes are subjected to a cross operation to obtain cross feature codes; specifically, the cross operation may be performed on the encoded data corresponding to one or more target characteristic attributes; the target characteristic attributes are combined without generating new encoded data of the target characteristic attributes. Each cross feature code and the first intermediate feature code are the second iterative robot population; and then judging whether the second iterative robot population reaches a convergence state. It should be noted that, specifically, determining whether the second iterative robot population reaches the convergence state is to determine whether an average fitness value corresponding to each corresponding cross feature code and each corresponding first intermediate feature code in the second iterative robot population approaches a fixed value, where entering a convergence period is referred to as entering the convergence state, that is, reaching the convergence state when the fitness approaches convergence. The method can be specifically judged by the fitness cycle ratio growth rate of the second iterative robot population.

If the second iterative robot population reaches a convergence state, the current second iterative robot population is converged to a stable state, so that the cloud robot corresponding to the cross feature code and the first intermediate feature code is used as the first iterative robot population, and the code of each cloud robot in the population is called as a second intermediate feature code; if the second iterative robot population does not reach the convergence state, the coded data of the single genetic factor (single target characteristic attribute) of the cloud robot in the first iterative robot population needs to be randomly changed, specifically, a preset random change algorithm is called to change the coded data. For example, assuming that the target feature attribute of the cloud robot speed is defined as 8-bit data encoding, the encoded data of the speed of a certain cloud robot is: 73618191, by randomly changing several bits of the coded data, such as transforming 73618191 to 73638161, a new gene is generated, namely a variant feature code is generated, and then the robot is used as a first iteration robot population according to the variant feature code, the cross feature code and the first intermediate feature code, wherein the code of each cloud robot in the population is called as a second intermediate feature code.

Therefore, the target feature code with the highest target scheduling task adaptability is determined according to the genetic algorithm of the embodiment, the operation mode is convenient and fast, and the target feature code can be accurately determined.

The above detailed description is given for the embodiment of the task scheduling method for the cloud robot, and the present invention further provides a task scheduling apparatus, a device, and a computer-readable storage medium for the cloud robot corresponding to the method.

Fig. 2 is a structural diagram of a task scheduling device of a cloud robot according to an embodiment of the present invention, and as shown in fig. 2, the task scheduling device of the cloud robot includes:

the encoding module 21 is configured to call a preset random encoding function to randomly encode a target feature attribute of the cloud robot corresponding to the target scheduling task to obtain a plurality of different initial feature codes;

the calculation module 22 is used for calling a preset genetic algorithm and determining a target feature code with the highest fitness with the target scheduling task according to the initial feature code;

and the determining module 23 is configured to determine corresponding target feature attribute information according to the target feature codes, and schedule the corresponding target robot according to the target feature attribute information.

The task scheduling device of the cloud robot provided by the embodiment of the invention has the beneficial effects of the task scheduling method of the cloud robot.

As a preferred embodiment, the encoding module specifically includes:

the determining unit is used for determining the target characteristic attribute of the cloud robot corresponding to the target scheduling task;

the setting unit is used for setting corresponding random coding bit numbers according to the weight of each target characteristic attribute;

and the coding unit is used for calling a preset random coding function to carry out random coding on each target characteristic attribute according to the number of the random coding bits so as to obtain a plurality of different initial characteristic codes.

As a preferred embodiment, the calculation module specifically includes:

the first input unit is used for inputting the initial feature codes into a first fitness function and screening out first intermediate feature codes with fitness meeting a preset condition by using the first fitness function;

the operation unit is used for carrying out cross operation and mutation operation on each first intermediate characteristic code to obtain a second intermediate characteristic code;

and the screening unit is used for screening the target feature codes with the highest adaptability to the target scheduling task from the second intermediate feature codes through the second adaptability function under the condition that the first iterative robot population corresponding to the second intermediate feature codes reaches the convergence state.

As a preferred embodiment, the operation unit specifically includes:

the first operation subunit is used for performing cross operation on each first intermediate feature code to obtain a cross feature code, and determining a second iterative robot population according to the cross feature code;

the judging subunit is used for judging whether the second iterative robot population reaches a convergence state; if yes, the first execution subunit is called, if no, the second execution subunit is called,

the first execution subunit is used for determining a second intermediate characteristic code according to the cross characteristic code and the first intermediate characteristic code;

and the second execution subunit is used for calling a preset random change algorithm to randomly change the target feature attribute in the first intermediate feature code and/or the cross feature code to obtain a variation feature code, and determining a second intermediate feature code according to the variation feature code, the cross feature code and the first intermediate feature code.

Fig. 3 is a structural diagram of a task scheduling device of a cloud robot according to an embodiment of the present invention, and as shown in fig. 3, the task scheduling device of the cloud robot includes:

a memory 31 for storing a computer program;

and a processor 32, configured to implement the steps of the task scheduling method for the cloud robot when executing the computer program.

The task scheduling device of the cloud robot provided by the embodiment of the invention has the beneficial effects of the task scheduling method of the cloud robot.

In order to solve the technical problem, the present invention further provides a computer-readable storage medium, on which a computer program is stored, and the computer program, when executed by a processor, implements the steps of the task scheduling method for the cloud robot.

The computer-readable storage medium provided by the embodiment of the invention has the beneficial effects of the task scheduling method of the cloud robot.

The task scheduling method, device, equipment and computer readable storage medium of the cloud robot provided by the invention are described in detail above. The principles and embodiments of the present invention are explained herein using specific examples, which are set forth only to help understand the method and its core ideas of the present invention. It should be noted that, for those skilled in the art, it is possible to make various improvements and modifications to the present invention without departing from the principle of the present invention, and those improvements and modifications also fall within the scope of the claims of the present invention.

The embodiments are described in a progressive manner in the specification, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.

Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.