AR multitask resource allocation optimization method

Abstract

The application relates to resource allocation, in particular to an AR multi-task resource allocation optimization method, which determines the types and total amount of available resources in an AR system, collects task information of resources to be allocated, determines a solution space, a target function and a constraint condition of multi-task resource allocation optimization, constructs a resource allocation optimization model, adopts an adaptive dynamic balance of a loach and a pike optimization algorithm to solve the resource allocation optimization model, obtains an optimal resource allocation scheme, and applies the optimal resource allocation scheme to the AR system to realize efficient allocation of multi-task resources; the technical scheme provided by the application can effectively overcome the defects that the prior art cannot realize reasonable, accurate and efficient allocation of multi-task resources in the AR system.

Description

Technical Field

[0001] This invention relates to resource allocation, and more specifically to an AR multi-task resource allocation optimization method. Background Technology

[0002] With the rapid development of augmented reality (AR) technology, its applications in education, entertainment, industrial design, and many other fields are becoming increasingly widespread. During the operation of an AR system, multiple tasks often need to be processed simultaneously, such as image rendering, 3D modeling, and data interaction, and each task has different requirements for system resources. How to rationally and efficiently allocate limited system resources to these tasks has become crucial for improving the performance and user experience of AR systems.

[0003] Traditional resource allocation methods are mostly based on simple rules or fixed strategies, which are difficult to adjust flexibly according to the actual needs of the task and the dynamic changes of system resources. This can easily lead to unreasonable resource allocation when dealing with complex multi-task scenarios, such as some tasks having excess resources and wasting them, while some tasks suffer from insufficient resources, resulting in performance degradation or even failure to run normally.

[0004] Furthermore, existing optimization algorithms also have certain limitations in solving the AR multi-task resource allocation problem. On the one hand, some methods do not explore the solution space comprehensively enough when determining resource allocation schemes, easily overlooking potential optimal solutions, resulting in a final solution that is not truly optimal and affecting the overall performance of the AR system. On the other hand, some methods lack the ability to explore local optima; even if a region close to the optimal solution is found, it is difficult to further explore the global optimum, making it difficult to achieve ideal resource allocation efficiency. In addition, traditional optimization algorithms are prone to getting trapped in local optima when dealing with multi-task resource allocation, making it difficult to escape and explore a broader solution space, and their insufficient population diversity affects the convergence of the algorithm and the accuracy of the solution.

[0005] Therefore, there is an urgent need for an AR multi-task resource allocation optimization method that can effectively solve the above problems, so as to achieve reasonable, accurate and efficient allocation of multi-task resources in AR systems and improve the overall performance and user experience of AR systems. Summary of the Invention

[0006] (a) Technical problems to be solved In view of the above-mentioned shortcomings of the existing technology, the present invention provides an AR multi-task resource allocation optimization method, which can effectively overcome the defects of the existing technology in that it is difficult to achieve reasonable, accurate and efficient allocation of multi-task resources in AR systems.

[0007] (II) Technical Solution To achieve the above objectives, the present invention provides the following technical solution: An AR multi-task resource allocation optimization method includes the following steps: S1. Determine the types and total amount of resources available in the AR system, and collect task information on resources to be allocated; S2. Determine the solution space, objective function, and constraints for multi-task resource allocation optimization, and construct a resource allocation optimization model; S3. The adaptive dynamic equilibrium cuckoo catfish optimization algorithm is used to solve the resource allocation optimization model to obtain the optimal resource allocation scheme. S4. Apply the optimal resource allocation scheme to the AR system to achieve efficient allocation of resources for multiple tasks; Among them, the adaptive dynamic equilibrium catfish optimization algorithm includes: During the global exploration phase, the behavior of cuckoo catfish cooperating to compress the search area is simulated. By dynamically adjusting the compression intensity, the escape space of the prey is gradually reduced, ensuring a broad exploration of the solution space. In the local development phase, the behavior of cuckoo catfish spiraling around prey in natural space is simulated. The randomness of the chaotic sequence is used to generate perturbations around the global optimal solution, thereby improving the ability to develop the global optimal solution. At the same time, the behavior changes of cuckoo catfish during the predation process are simulated. By dynamically adjusting the ratio of exploration to development, a smooth transition is achieved, improving the accuracy of the solution and enhancing the convergence of the algorithm. By simulating the laws of natural selection and introducing parasitism and death mechanisms to enhance population diversity, the algorithm can be helped to escape local optima and optimize population quality.

[0008] Preferably, in S1, the types and total amount of resources available in the AR system are determined, and task information for resources to be allocated is collected, including: Determine the types and total amount of resources available in the AR system, including calculating the total resource C. total Total storage resources S total And total bandwidth resources B total ; Collect task information for resources to be allocated, including the number of tasks and the resource requirements of each task.

[0009] Preferably, in S2, the solution space, objective function, and constraints for multi-task resource allocation optimization are determined, and a resource allocation optimization model is constructed, including: S21. Determine the solution space for multi-task resource allocation optimization: The position of each individual cuckoo catfish in the population represents a solution vector, the dimension of which is the product of the number of resource types and the number of tasks. Each element in the solution vector represents the allocation amount of the corresponding type of resource to the corresponding task. S22. Determine the objective function for optimizing multi-task resource allocation; S23. Determine the constraints for optimizing multi-task resource allocation; S24. Combining the objective function and constraints of multi-task resource allocation optimization, construct a resource allocation optimization model.

[0010] Preferably, the objective function for optimizing multi-task resource allocation in S22 includes: Objective function F(X) i Taking into account both task completion and resource utilization, task completion is measured by task completion time and quality, while resource utilization reflects the efficiency of resource use. ; Among them, X i Let F(X) be the position of the i-th individual cuckoo catfish, i.e., the i-th solution vector. i X represents the position of the i-th individual cuckoo catfish. i The corresponding fitness value, T(X) i Let X be the i-th solution vector. i The average completion time of all tasks, Q(X) i Let X be the i-th solution vector. i The average quality of completion of all tasks, x i,k Let X be the i-th solution vector. i The total allocation of the k-th resource, R k Let k be the total amount of the k-th resource, where k is the resource type index, and k=1, 2, and 3 represent computing resources, storage resources, and bandwidth resources, respectively. , , All are weighting coefficients, and ; S23 defines the constraints for optimizing multi-task resource allocation, including: For each resource type, the total allocation of all tasks shall not exceed the corresponding total resource amount, that is: .

[0011] Preferably, in S3, the adaptive dynamic equilibrium cuckoo-catfish optimization algorithm is used to solve the resource allocation optimization model to obtain the optimal resource allocation scheme, including: S31. Map each solution vector to the position of an individual cuckoo catfish to form an initial population and initialize the algorithm parameters. S32. Determine whether to enter the global exploration phase or the local development phase based on the algorithm progress. If the algorithm enters the global exploration phase, proceed to S33; if it enters the local development phase, proceed to S34. S33. In the global exploration phase, simulate the behavior of cuckoo catfish cooperating to compress the search area. By dynamically adjusting the compression intensity, gradually reduce the escape space of the prey to ensure extensive exploration of the solution space, and proceed to S35. S34. In the local development stage, the behavior of the cuckoo catfish spiraling around its prey in natural space is simulated. The randomness of the chaotic sequence is used to generate perturbations around the global optimal solution, thereby improving the ability to develop the global optimal solution. At the same time, the behavior changes of the cuckoo catfish during the predation process are simulated. By dynamically adjusting the ratio of exploration to development, a smooth transition is achieved, the accuracy of the solution is improved, and the convergence of the algorithm is enhanced. Proceed to S35. S35. Use the objective function of multi-task resource allocation optimization to evaluate all individual cichlid catfish in the current population, calculate the corresponding fitness value, and record and update the historical best solution. S36. Simulate the laws of natural selection, introduce parasitism and death mechanisms to enhance population diversity, help the algorithm escape local optima, and optimize population quality at the same time; S37. Determine whether the iteration termination condition is met. If the iteration termination condition is not met, return to S32. Otherwise, take the historical best solution as the optimal resource allocation scheme.

[0012] Preferably, in S33, during the global exploration phase, the behavior of cuckoo catfish cooperating to compress the search area is simulated. By dynamically adjusting the compression intensity, the escape space of the prey is gradually reduced, ensuring a broad exploration of the solution space, including: S331. Introducing a multidimensional 0-1 random vector Z to simulate unpredictable random resistance in the space. Its randomness helps the algorithm introduce uncertainties during the search process, avoiding the search process from falling into a fixed pattern. In the position update formula, by controlling the weights of the update part based on the position information of other random individuals and the retention part based on its own position information, the movement mode and range of individuals in the search space are dynamically adjusted, affecting the compression intensity. S332. Introduce a random number r that follows a standard normal distribution to describe the strength of cooperation among populations. In the position update formula, its absolute value... The size of this factor will affect the amplification of other random individual positional differences: a larger value will result in a smaller value. This makes individual position updates more susceptible to the influence of other random individual position differences, enhancing the cooperative search effect among the population; smaller This makes individuals more inclined to fine-tune their positions near themselves, dynamically adjusting the search range and cooperation intensity of the population through random changes, and achieving dynamic compression of the search space in conjunction with the multidimensional 0-1 random vector Z. S333. Simulate the cooperative behavior of cuckoo catfish in compressing the search area. This is achieved using a multidimensional 0-1 random vector Z, random numbers r following a standard normal distribution, and the positions of two randomly selected cuckoo catfish individuals from the current population. , The compression intensity is dynamically adjusted to gradually reduce the escape space of the prey. The following formula is used to update the position of all individual cuckoo catfish in the current population during the global exploration phase: ; in, Let be the position of the i-th individual cuckoo catfish at the (t+1)-th iteration after the position update during the global exploration phase. Let be the position of the i-th individual cuckoo catfish at the t-th iteration. This indicates the locational differences of other random individuals. This represents the update portion based on the location information of other random individuals. This indicates the portion of information retained based on its own location.

[0013] Preferably, in S34, during the local development phase, the behavior of the cuckoo catfish spiraling around its prey in natural space is simulated. The randomness of the chaotic sequence is used to generate perturbations around the global optimum, enhancing the ability to develop the global optimum. Simultaneously, the behavioral changes of the cuckoo catfish during predation are simulated, and a smooth transition is achieved by dynamically adjusting the ratio of exploration to development, improving the accuracy of the solution and enhancing the algorithm's convergence. This includes: S341. Simulate the behavior of cuckoo catfish spiraling around prey in natural space. Utilize the randomness of chaotic sequences to generate perturbations around the global optimum, thereby enhancing the ability to develop the global optimum. The following formula is used to perform preliminary position updates for all cuckoo catfish individuals in the current population during the local development phase: ; in, Let i be the position of the i-th cuckoo catfish individual at the (t+1)-th iteration after the initial position update during the local development phase. This represents the position of the current global optimal solution. Randomly select the location of an individual cuckoo catfish from the current population. Let be the spiral angle that varies for the i-th individual cuckoo catfish, and c be a constant controlling the shape of the spiral. m Given the m-th chaotic sequence value, various chaotic sequences can be constructed by introducing different chaotic mappings; S342, Introducing a dynamic equilibrium factor A transition strategy is employed to facilitate a smooth transition from the global exploration phase to the local development phase: ; in, Let T be the balance factor at the t-th iteration, and T be the maximum number of iterations. S343. Simulating the behavioral changes of cuckoo catfish during predation, using dynamic equilibrium factors. The ratio of exploration to development is dynamically adjusted to achieve a smooth transition from the global exploration phase to the local development phase. The following formula is used to update the position of all individual cichlid catfish in the current population again during the local development phase: ; in, Let be the final position of the i-th cuckoo catfish individual at the (t+1)-th iteration after the position update during the local development phase. Let be the optimal position of the i-th individual cuckoo catfish during the global exploration phase.

[0014] Preferably, S36 simulates the laws of natural selection, introducing parasitism and death mechanisms to enhance population diversity, helping the algorithm escape local optima, and simultaneously optimizing population quality, including: For individual cichlid catfish with low fitness rankings in the current population, a parasitic mechanism is used for optimization: Simulating the parasitic behavior of cuckoo catfish, individuals with lower fitness rankings are guided to learn towards the current global optimum, and their positions are updated using the following formula: ; in, , These represent the positions of the j-th catfish individual with the lowest fitness ranking before and after the update. As a learning factor, .

[0015] Preferably, S36 simulates the laws of natural selection, introducing parasitism and death mechanisms to enhance population diversity, helping the algorithm escape local optima, and simultaneously optimizing population quality, including: For individual cichlid catfish with low fitness rankings in the current population, a mortality mechanism is used for optimization: Set a dynamic probability of death P die (t), for individual cichlid catfish with lower fitness ranking, the dynamic mortality probability P is used. die (t) Remove it from the current population and respawn it near the current global optimum to maintain the population size. The dynamic mortality probability P die (t) is calculated using the following formula: ; Among them, P die (t) represents the probability of death at the t-th iteration, and P0 represents the initial probability of death.

[0016] Preferably, in S4, the optimal resource allocation scheme is applied to the AR system to achieve efficient allocation of multi-task resources, including: Based on the optimal resource allocation scheme, computing resources, storage resources, and bandwidth resources are allocated to each task; The system monitors resource usage in real time during the execution of all tasks. If abnormal resource usage or changes in task requirements are detected, the adaptive dynamic balancing catfish optimization algorithm is restarted to solve the resource allocation optimization model and dynamically update the optimal resource allocation scheme.

[0017] (III) Beneficial Effects Compared with the prior art, the AR multi-task resource allocation optimization method provided by the present invention has the following beneficial effects: 1) Accurately allocate resources and improve resource utilization. By determining the types and total amount of available resources in the AR system, collecting task information on resources to be allocated, and constructing a resource allocation optimization model, an adaptive dynamic balancing cuckoo catfish optimization algorithm is used to solve the problem. In the global exploration phase, the solution space can be explored extensively to avoid missing potential optimal solutions. In the local development phase, the ability to develop the global optimal solution can be improved, and the accuracy of the solution can be increased. The final optimal resource allocation scheme can accurately allocate resources according to different task requirements, effectively avoid resource waste and idleness, and achieve efficient utilization of AR system resources. 2) Optimize algorithm performance, enhance convergence and accuracy. The adaptive dynamic equilibrium cuckoo catfish optimization algorithm achieves extensive exploration of the solution space, improves solution accuracy, and enhances algorithm convergence by simulating different behaviors of cuckoo catfish in the global exploration and local development phases, respectively. At the same time, the introduction of parasitism and death mechanisms enhances population diversity and helps the algorithm escape local optima. These innovative designs enable the algorithm to not only fully explore the solution space but also deeply mine the optimal solution when solving resource allocation optimization models, effectively improving the convergence and accuracy of the algorithm and ensuring that a high-quality optimal resource allocation scheme is obtained quickly. 3) Adapt flexibly to tasks and improve the overall adaptability of the system. In the process of finding the optimal resource allocation scheme, it can flexibly switch between the global exploration stage and the local development stage according to the algorithm progress, adapting to the needs of different stages for solution space exploration and optimal solution development. At the same time, in the face of complex and ever-changing multi-task scenarios in AR systems, it can dynamically adjust the resource allocation strategy according to task information, quickly respond to changes in task requirements, provide suitable resource support for different tasks, and effectively improve the overall adaptability and stability of AR systems in multi-task environments. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0019] Figure 1 This is a schematic diagram of the process of the present invention; Figure 2 This is a schematic diagram illustrating the process of using the adaptive dynamic equilibrium cuckoo catfish optimization algorithm in this invention to solve the resource allocation optimization model and obtain the optimal resource allocation scheme. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0021] The core of this invention lies in: Addressing the complex and ever-changing multi-task scenarios and frequent changes in task requirements that AR systems face, an adaptive dynamic equilibrium catfish optimization algorithm is designed. In this algorithm: During the global exploration phase, the behavior of cuckoo catfish cooperating to compress the search area is simulated. By dynamically adjusting the compression intensity, the escape space of the prey is gradually reduced, ensuring a broad exploration of the solution space. In the local development phase, the behavior of cuckoo catfish spiraling around prey in natural space is simulated. The randomness of the chaotic sequence is used to generate perturbations around the global optimal solution, thereby improving the ability to develop the global optimal solution. At the same time, the behavior changes of cuckoo catfish during the predation process are simulated. By dynamically adjusting the ratio of exploration to development, a smooth transition is achieved, improving the accuracy of the solution and enhancing the convergence of the algorithm. By simulating the laws of natural selection and introducing parasitism and death mechanisms to enhance population diversity, the algorithm can be helped to escape local optima and optimize population quality.

[0022] The following describes the specific process of an AR multi-task resource allocation optimization method provided by this invention, using a concrete example (e.g.) Figure 1 (as shown) and technical effects.

[0023] S1. Determine the types and total amount of resources available in the AR system, and collect task information for resources to be allocated, including: Determine the types and total amount of resources available in the AR system, including calculating the total resource C. total Total storage resources S total And total bandwidth resources B total ; Collect task information for resources to be allocated, including the number of tasks and the resource requirements of each task.

[0024] S2. Determine the solution space, objective function, and constraints for multi-task resource allocation optimization, and construct a resource allocation optimization model, including: S21. Determine the solution space for multi-task resource allocation optimization: The position of each individual cuckoo catfish in the population represents a solution vector, the dimension of which is the product of the number of resource types and the number of tasks. Each element in the solution vector represents the allocation amount of the corresponding type of resource to the corresponding task. S22. Determine the objective function for optimizing multi-task resource allocation; S23. Determine the constraints for optimizing multi-task resource allocation; S24. Combining the objective function and constraints of multi-task resource allocation optimization, construct a resource allocation optimization model.

[0025] Specifically, S22 defines the objective function for optimizing multi-task resource allocation, including: Objective function F(X) i Taking into account both task completion and resource utilization, task completion is measured by task completion time and quality, while resource utilization reflects the efficiency of resource use. ; Among them, X i Let F(X) be the position of the i-th individual cuckoo catfish, i.e., the i-th solution vector. i X represents the position of the i-th individual cuckoo catfish. i The corresponding fitness value, T(X) i Let X be the i-th solution vector. i The average completion time of all tasks, Q(X) i Let X be the i-th solution vector. i The average quality of completion of all tasks, x i,k Let X be the i-th solution vector. i The total allocation of the k-th resource, R k Let k be the total amount of the k-th resource, where k is the resource type index, and k=1, 2, and 3 represent computing resources, storage resources, and bandwidth resources, respectively. , , All are weighting coefficients, and .

[0026] Specifically, S23 defines the constraints for optimizing multi-task resource allocation, including: For each resource type, the total allocation of all tasks shall not exceed the corresponding total resource amount, that is: .

[0027] S3. The adaptive dynamic equilibrium cuckoo-catfish optimization algorithm is used to solve the resource allocation optimization model to obtain the optimal resource allocation scheme, such as... Figure 2 As shown, it includes: S31. Map each solution vector to the position of an individual cuckoo catfish to form an initial population and initialize the algorithm parameters. S32. Determine whether to enter the global exploration phase or the local development phase based on the algorithm progress (e.g., enter the global exploration phase when t < 0.6T; enter the local development phase when t ≥ 0.6T, etc.). If the algorithm enters the global exploration phase, proceed to S33; if it enters the local development phase, proceed to S34. S33. In the global exploration phase, simulate the behavior of cuckoo catfish cooperating to compress the search area. By dynamically adjusting the compression intensity, gradually reduce the escape space of the prey to ensure extensive exploration of the solution space, and proceed to S35. S34. In the local development stage, the behavior of the cuckoo catfish spiraling around its prey in natural space is simulated. The randomness of the chaotic sequence is used to generate perturbations around the global optimal solution, thereby improving the ability to develop the global optimal solution. At the same time, the behavior changes of the cuckoo catfish during the predation process are simulated. By dynamically adjusting the ratio of exploration to development, a smooth transition is achieved, the accuracy of the solution is improved, and the convergence of the algorithm is enhanced. Proceed to S35. S35. Use the objective function of multi-task resource allocation optimization to evaluate all individual cichlid catfish in the current population, calculate the corresponding fitness value, and record and update the historical best solution. S36. Simulate the laws of natural selection, introduce parasitism and death mechanisms to enhance population diversity, help the algorithm escape local optima, and optimize population quality at the same time; S37. Determine whether the iteration termination condition is met. If the iteration termination condition is not met, return to S32. Otherwise, take the historical best solution as the optimal resource allocation scheme.

[0028] Specifically, in S33, during the global exploration phase, the behavior of cuckoo catfish cooperating to compress the search area is simulated. By dynamically adjusting the compression intensity, the escape space of the prey is gradually reduced, ensuring a broad exploration of the solution space, including: S331. Introducing a multidimensional 0-1 random vector Z to simulate unpredictable random resistance in the space. Its randomness helps the algorithm introduce uncertainties during the search process, avoiding the search process from falling into a fixed pattern. In the position update formula, by controlling the weights of the update part based on the position information of other random individuals and the retention part based on its own position information, the movement mode and range of individuals in the search space are dynamically adjusted, affecting the compression intensity. S332. Introduce a random number r that follows a standard normal distribution to describe the strength of cooperation among populations. In the position update formula, its absolute value... The size of this factor will affect the amplification of other random individual positional differences: a larger value will result in a smaller value. This makes individual position updates more susceptible to the influence of other random individual position differences, enhancing the cooperative search effect among the population; smaller This makes individuals more inclined to fine-tune their positions near themselves, dynamically adjusting the search range and cooperation intensity of the population through random changes, and achieving dynamic compression of the search space in conjunction with the multidimensional 0-1 random vector Z. S333. Simulate the cooperative behavior of cuckoo catfish in compressing the search area. This is achieved using a multidimensional 0-1 random vector Z, random numbers r following a standard normal distribution, and the positions of two randomly selected cuckoo catfish individuals from the current population. , The compression intensity is dynamically adjusted to gradually reduce the escape space of the prey. The following formula is used to update the position of all individual cuckoo catfish in the current population during the global exploration phase: ; in, Let be the position of the i-th individual cuckoo catfish at the (t+1)-th iteration after the position update during the global exploration phase. Let be the position of the i-th individual cuckoo catfish at the t-th iteration. This indicates the locational differences of other random individuals. This represents the update portion based on the location information of other random individuals. This indicates the portion of information retained based on its own location.

[0029] Specifically, in S34, during the local development phase, the behavior of the cuckoo catfish spiraling around its prey in natural space is simulated. The randomness of the chaotic sequence is used to generate perturbations around the global optimum, enhancing the ability to develop the global optimum. Simultaneously, the behavioral changes of the cuckoo catfish during predation are simulated, and a smooth transition is achieved by dynamically adjusting the ratio of exploration to development, improving the accuracy of the solution and enhancing the algorithm's convergence. This includes: S341. Simulate the behavior of cuckoo catfish spiraling around prey in natural space. Utilize the randomness of chaotic sequences to generate perturbations around the global optimum, thereby enhancing the ability to develop the global optimum. The following formula is used to perform preliminary position updates for all cuckoo catfish individuals in the current population during the local development phase: ; in, Let i be the position of the i-th cuckoo catfish individual at the (t+1)-th iteration after the initial position update during the local development phase. This represents the position of the current global optimal solution. Randomly select the location of an individual cuckoo catfish from the current population. Let be the spiral angle that varies for the i-th individual cuckoo catfish, and c be a constant controlling the shape of the spiral. m Given the m-th chaotic sequence value, various chaotic sequences can be constructed by introducing different chaotic mappings; S342, Introducing a dynamic equilibrium factor A transition strategy is employed to facilitate a smooth transition from the global exploration phase to the local development phase: ; in, Let T be the balance factor at the t-th iteration, and T be the maximum number of iterations. S343. Simulating the behavioral changes of cuckoo catfish during predation, using dynamic equilibrium factors. The ratio of exploration to development is dynamically adjusted to achieve a smooth transition from the global exploration phase to the local development phase. The following formula is used to update the position of all individual cichlid catfish in the current population again during the local development phase: ; in, Let be the final position of the i-th cuckoo catfish individual at the (t+1)-th iteration after the position update during the local development phase. Let be the optimal position of the i-th individual cuckoo catfish during the global exploration phase.

[0030] Specifically, S36 simulates the laws of natural selection, introduces parasitism and death mechanisms to enhance population diversity, helps the algorithm escape local optima, and optimizes population quality, including: For individual cichlid catfish with low fitness rankings in the current population, optimization can be achieved using parasitism or mortality mechanisms: 1) Parasitic mechanism Simulating the parasitic behavior of cuckoo catfish, individuals with lower fitness rankings are guided to learn towards the current global optimum, and their positions are updated using the following formula: ; in, , These represent the positions of the j-th catfish individual with the lowest fitness ranking before and after the update. As a learning factor, ; 2) Death Mechanism Set a dynamic probability of death P die (t), for individual cichlid catfish with lower fitness ranking, the dynamic mortality probability P is used. die (t) Remove it from the current population and respawn it near the current global optimum to maintain the population size. The dynamic mortality probability P die (t) is calculated using the following formula: ; Among them, P die (t) represents the probability of death at the t-th iteration, and P0 represents the initial probability of death.

[0031] The aforementioned technical solution, the adaptive dynamic equilibrium cuckoo catfish optimization algorithm, achieves extensive exploration of the solution space, improves solution accuracy, and enhances algorithm convergence by simulating different behaviors of cuckoo catfish in the global exploration phase and the local development phase, respectively. At the same time, the introduction of parasitism and death mechanisms enhances population diversity and helps the algorithm escape local optima. These innovative designs enable the algorithm to comprehensively explore the solution space and deeply mine the optimal solution when solving resource allocation optimization models, effectively improving the convergence and accuracy of the algorithm and ensuring that a high-quality optimal resource allocation scheme is obtained quickly.

[0032] S4. Apply the optimal resource allocation scheme to the AR system to achieve efficient allocation of resources for multiple tasks, including: Based on the optimal resource allocation scheme, computing resources, storage resources, and bandwidth resources are allocated to each task; The system monitors resource usage in real time during the execution of all tasks. If abnormal resource usage or changes in task requirements are detected, the adaptive dynamic balancing catfish optimization algorithm is restarted to solve the resource allocation optimization model and dynamically update the optimal resource allocation scheme.

[0033] In this technical solution, the available resource types and total amount in the AR system are determined, task information on resources to be allocated is collected, and a resource allocation optimization model is constructed. An adaptive dynamic balancing cuckoo-catfish optimization algorithm is used to solve the problem. In the global exploration phase, the solution space can be explored extensively to avoid missing potential optimal solutions. In the local development phase, the ability to develop the global optimal solution can be improved, and the accuracy of the solution can be increased. The final optimal resource allocation scheme can accurately allocate resources according to different task requirements, effectively avoid resource waste and idleness, and realize the efficient utilization of AR system resources.

[0034] Furthermore, when faced with complex and ever-changing multi-task scenarios in AR systems, the system can dynamically adjust resource allocation strategies based on task information, quickly respond to changes in task requirements, provide appropriate resource support for different tasks, and effectively improve the overall adaptability and stability of the AR system in multi-task environments.

[0035] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An AR multi-task resource allocation optimization method, characterized in that: Includes the following steps: S1. Determine the types and total amount of resources available in the AR system, and collect task information on resources to be allocated; S2. Determine the solution space, objective function, and constraints for multi-task resource allocation optimization, and construct a resource allocation optimization model; S3. The adaptive dynamic equilibrium cuckoo catfish optimization algorithm is used to solve the resource allocation optimization model to obtain the optimal resource allocation scheme. S4. Apply the optimal resource allocation scheme to the AR system to achieve efficient allocation of resources for multiple tasks; Among them, the adaptive dynamic equilibrium catfish optimization algorithm includes: During the global exploration phase, the behavior of cuckoo catfish cooperating to compress the search area is simulated. By dynamically adjusting the compression intensity, the escape space of the prey is gradually reduced, ensuring a broad exploration of the solution space. In the local development phase, the behavior of cuckoo catfish spiraling around prey in natural space is simulated. The randomness of the chaotic sequence is used to generate perturbations around the global optimal solution, thereby improving the ability to develop the global optimal solution. At the same time, the behavior changes of cuckoo catfish during the predation process are simulated. By dynamically adjusting the ratio of exploration to development, a smooth transition is achieved, improving the accuracy of the solution and enhancing the convergence of the algorithm. By simulating the laws of natural selection and introducing parasitism and death mechanisms to enhance population diversity, the algorithm can be helped to escape local optima and optimize population quality.

2. The AR multi-task resource allocation optimization method according to claim 1, characterized in that: S1 determines the types and total amount of resources available in the AR system and collects task information for resources to be allocated, including: Determine the types and total amount of resources available in the AR system, including calculating the total resource C. total Total storage resources S total And total bandwidth resources B total ; Collect task information for resources to be allocated, including the number of tasks and the resource requirements of each task.

3. The AR multi-task resource allocation optimization method according to claim 1, characterized in that: S2 determines the solution space, objective function, and constraints for multi-task resource allocation optimization, and constructs a resource allocation optimization model, including: S21. Determine the solution space for multi-task resource allocation optimization: The position of each individual cuckoo catfish in the population represents a solution vector, the dimension of which is the product of the number of resource types and the number of tasks. Each element in the solution vector represents the allocation amount of the corresponding type of resource to the corresponding task. S22. Determine the objective function for optimizing multi-task resource allocation; S23. Determine the constraints for optimizing multi-task resource allocation; S24. Combining the objective function and constraints of multi-task resource allocation optimization, construct a resource allocation optimization model.

4. The AR multi-task resource allocation optimization method according to claim 3, characterized in that: S22 defines the objective function for optimizing multi-task resource allocation, including: Objective function F(X) i Taking into account both task completion and resource utilization, task completion is measured by task completion time and quality, while resource utilization reflects the efficiency of resource use. ； Among them, X i Let F(X) be the position of the i-th individual cuckoo catfish, i.e., the i-th solution vector. i X represents the position of the i-th individual cuckoo catfish. i The corresponding fitness value, T(X) i Let X be the i-th solution vector. i The average completion time of all tasks, Q(X) i Let X be the i-th solution vector. i The average quality of completion of all tasks, x i,k Let X be the i-th solution vector. i The total allocation of the k-th resource, R k Let k be the total amount of the k-th resource, where k is the resource type index, and k=1, 2, and 3 represent computing resources, storage resources, and bandwidth resources, respectively. , , All are weighting coefficients, and ; S23 defines the constraints for optimizing multi-task resource allocation, including: For each resource type, the total allocation of all tasks shall not exceed the corresponding total resource amount, that is: 。 5. The AR multi-task resource allocation optimization method according to claim 1, characterized in that: In S3, the adaptive dynamic equilibrium cuckoo-catfish optimization algorithm is used to solve the resource allocation optimization model to obtain the optimal resource allocation scheme, including: S31. Map each solution vector to the position of an individual cuckoo catfish to form an initial population and initialize the algorithm parameters. S32. Determine whether to enter the global exploration phase or the local development phase based on the algorithm progress. If the algorithm enters the global exploration phase, proceed to S33; if it enters the local development phase, proceed to S34. S33. In the global exploration phase, simulate the behavior of cuckoo catfish cooperating to compress the search area. By dynamically adjusting the compression intensity, gradually reduce the escape space of the prey to ensure extensive exploration of the solution space, and proceed to S35. S34. In the local development stage, the behavior of the cuckoo catfish spiraling around its prey in natural space is simulated. The randomness of the chaotic sequence is used to generate perturbations around the global optimal solution, thereby improving the ability to develop the global optimal solution. At the same time, the behavior changes of the cuckoo catfish during the predation process are simulated. By dynamically adjusting the ratio of exploration to development, a smooth transition is achieved, the accuracy of the solution is improved, and the convergence of the algorithm is enhanced. Proceed to S35. S35. Use the objective function of multi-task resource allocation optimization to evaluate all individual cichlid catfish in the current population, calculate the corresponding fitness value, and record and update the historical best solution. S36. Simulate the laws of natural selection, introduce parasitism and death mechanisms to enhance population diversity, help the algorithm escape local optima, and optimize population quality at the same time; S37. Determine whether the iteration termination condition is met. If the iteration termination condition is not met, return to S32. Otherwise, take the historical best solution as the optimal resource allocation scheme.

6. The AR multi-task resource allocation optimization method according to claim 5, characterized in that: In S33, during the global exploration phase, the cooperative behavior of cuckoo catfish compressing the search area is simulated. By dynamically adjusting the compression intensity, the escape space of the prey is gradually reduced, ensuring a broad exploration of the solution space, including: S331. Introducing a multidimensional 0-1 random vector Z to simulate unpredictable random resistance in the space. Its randomness helps the algorithm introduce uncertainties during the search process, avoiding the search process from falling into a fixed pattern. In the position update formula, by controlling the weights of the update part based on the position information of other random individuals and the retention part based on its own position information, the movement mode and range of individuals in the search space are dynamically adjusted, affecting the compression intensity. S332. Introduce a random number r that follows a standard normal distribution to describe the strength of cooperation among populations. In the position update formula, its absolute value... The size of this factor will affect the amplification of other random individual positional differences: a larger value will result in a smaller value. This makes individual position updates more susceptible to the influence of other random individual position differences, enhancing the cooperative search effect among the population; smaller This makes individuals more inclined to fine-tune their positions near themselves, dynamically adjusting the search range and cooperation intensity of the population through random changes, and achieving dynamic compression of the search space in conjunction with the multidimensional 0-1 random vector Z. S333. Simulate the cooperative behavior of cuckoo catfish in compressing the search area. This is achieved using a multidimensional 0-1 random vector Z, random numbers r following a standard normal distribution, and the positions of two randomly selected cuckoo catfish individuals from the current population. , The compression intensity is dynamically adjusted to gradually reduce the escape space of the prey. The following formula is used to update the position of all individual cuckoo catfish in the current population during the global exploration phase: ； in, Let be the position of the i-th individual cuckoo catfish at the (t+1)-th iteration after the position update during the global exploration phase. Let be the position of the i-th individual cuckoo catfish at the t-th iteration. This indicates the locational differences of other random individuals. This represents the update portion based on the location information of other random individuals. This indicates the portion of information retained based on its own location.

7. The AR multi-task resource allocation optimization method according to claim 6, characterized in that: In S34, during the local development phase, the behavior of the cuckoo catfish spiraling around its prey in natural space is simulated. The randomness of the chaotic sequence is used to generate perturbations around the global optimum, enhancing the ability to develop the global optimum. Simultaneously, the behavioral changes of the cuckoo catfish during predation are simulated, and a smooth transition is achieved by dynamically adjusting the ratio of exploration to development, improving solution accuracy and enhancing algorithm convergence. This includes: S341. Simulate the behavior of cuckoo catfish spiraling around prey in natural space. Utilize the randomness of chaotic sequences to generate perturbations around the global optimum, thereby enhancing the ability to develop the global optimum. The following formula is used to perform preliminary position updates for all cuckoo catfish individuals in the current population during the local development phase: ； in, Let i be the position of the i-th cuckoo catfish individual at the (t+1)-th iteration after the initial position update during the local development phase. This represents the position of the current global optimal solution. Randomly select the location of an individual cuckoo catfish from the current population. Let be the spiral angle that varies for the i-th individual cuckoo catfish, and c be a constant controlling the shape of the spiral. m Given the m-th chaotic sequence value, various chaotic sequences can be constructed by introducing different chaotic mappings; S342, Introducing a dynamic equilibrium factor A transition strategy is employed to facilitate a smooth transition from the global exploration phase to the local development phase: ； in, Let T be the balance factor at the t-th iteration, and T be the maximum number of iterations. S343. Simulating the behavioral changes of cuckoo catfish during predation, using dynamic equilibrium factors. The ratio of exploration to development is dynamically adjusted to achieve a smooth transition from the global exploration phase to the local development phase. The following formula is used to update the position of all individual cichlid catfish in the current population again during the local development phase: ； in, Let be the final position of the i-th cuckoo catfish individual at the (t+1)-th iteration after the position update during the local development phase. Let be the optimal position of the i-th individual cuckoo catfish during the global exploration phase.

8. The AR multi-task resource allocation optimization method according to claim 7, characterized in that: S36 simulates the laws of natural selection, introducing parasitism and death mechanisms to enhance population diversity, helping the algorithm escape local optima, and simultaneously optimizing population quality, including: For individual cichlid catfish with low fitness rankings in the current population, a parasitic mechanism is used for optimization: Simulating the parasitic behavior of cuckoo catfish, individuals with lower fitness rankings are guided to learn towards the current global optimum, and their positions are updated using the following formula: ； in, , These represent the positions of the j-th catfish individual with the lowest fitness ranking before and after the update. As a learning factor, .

9. The AR multi-task resource allocation optimization method according to claim 8, characterized in that: S36 simulates the laws of natural selection, introducing parasitism and death mechanisms to enhance population diversity, helping the algorithm escape local optima, and simultaneously optimizing population quality, including: For individual cichlid catfish with low fitness rankings in the current population, a mortality mechanism is used for optimization: Set a dynamic probability of death P die (t), for individual cichlid catfish with lower fitness ranking, the dynamic mortality probability P is used. die (t) Remove it from the current population and respawn it near the current global optimum to maintain the population size. The dynamic mortality probability P die (t) is calculated using the following formula: ； Among them, P die (t) represents the probability of death at the t-th iteration, and P0 represents the initial probability of death.

10. The AR multi-task resource allocation optimization method according to claim 1, characterized in that: S4 applies the optimal resource allocation scheme to the AR system to achieve efficient allocation of resources across multiple tasks, including: Based on the optimal resource allocation scheme, computing resources, storage resources, and bandwidth resources are allocated to each task; The system monitors resource usage in real time during the execution of all tasks. If abnormal resource usage or changes in task requirements are detected, the adaptive dynamic balancing catfish optimization algorithm is restarted to solve the resource allocation optimization model and dynamically update the optimal resource allocation scheme.

Images