Adaptive intelligent task planning and scheduling system based on convolutional neural network

Abstract

The invention discloses an intelligent task planning and scheduling system, and particularly relates to an adaptive intelligent task planning and scheduling system based on a convolutional neural network, which comprises a data input module, a feature extraction module, a reinforcement learning optimization module, a task planning module and a scheduling execution module, the feature extraction module is connected with the reinforcement learning optimization module, the reinforcement learning optimization module is connected with the task planning module, and the task planning module is connected with the scheduling execution module. By using the strong feature extraction capability of the convolutional neural network, key features can be automatically extracted from complex and changeable task data, and accurate data support is provided for task planning and scheduling. The task planning and scheduling strategy is optimized through the reinforcement learning algorithm, the method can automatically adapt to the dynamically changing environment, and the task execution efficiency and effect are improved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an adaptive intelligent task planning and scheduling system based on convolutional neural networks. Background Art

[0002] With the rapid development and widespread application of information technology, various intelligent systems are playing an increasingly important role in the fields of industry, military, transportation, and medical care. These systems often need to handle complex and changing tasks and perform efficient and accurate planning and scheduling to meet real-time needs in different scenarios. However, traditional task planning and scheduling methods often rely on manually preset rules or models, which are difficult to adapt to dynamically changing environments and uncertain factors, resulting in low planning efficiency and poor scheduling effects.

[0003] Driven by the wave of intelligence, using machine learning technology, especially deep learning, to improve the adaptability and intelligence level of task planning and scheduling systems has become a research hotspot. As one of the representative algorithms in the field of deep learning, convolutional neural networks have achieved remarkable results in image recognition, natural language processing and other fields with their powerful feature extraction capabilities. Applying convolutional neural networks to the field of task planning and scheduling can automatically learn the essential characteristics and scheduling rules of tasks from a large amount of data, providing the system with more accurate and efficient decision support.

[0004] Although relying solely on convolutional neural networks for feature extraction can improve the system's ability to understand and identify tasks, it may still face many challenges in the actual scheduling process, such as how to dynamically adjust the scheduling strategy according to the current environment, how to balance the priority and resource allocation between different tasks, etc. These problems need to be solved by introducing more intelligent optimization algorithms.

[0005] The present invention proposes an adaptive intelligent task planning and scheduling system based on convolutional neural networks, which aims to efficiently extract task features through convolutional neural networks, and combine optimization algorithms such as reinforcement learning to achieve adaptive adjustment of task planning and scheduling, so as to cope with complex and changeable actual application scenarios and improve the overall performance and efficiency of the system. Summary of the invention

[0006] In order to solve the above-mentioned problems in the prior art, the present invention provides an adaptive intelligent task planning and scheduling system based on convolutional neural network to solve the current technical problems.

[0007] The technical solution adopted by the present invention to solve its technical problem is: The present invention provides an adaptive intelligent task planning and scheduling system based on a convolutional neural network, comprising a data input module, a feature extraction module, a reinforcement learning optimization module, a task planning module and a scheduling execution module, wherein the data input module is connected to the feature extraction module, the feature extraction module is connected to the reinforcement learning optimization module, the reinforcement learning optimization module is connected to the task planning module, and the task planning module is connected to the scheduling execution module.

[0008] Preferably, the data input module is used to receive and pre-process task-related data, including task description, resource status, time constraints and other information.

[0009] Preferably, the feature extraction module uses a convolutional neural network (CNN) to extract features from the preprocessed task data, and specifically uses the following formula to calculate the convolution layer: ; in, represents the pixel value in the i-th row and j-th column of the output feature map, Represents the pixel value corresponding to the convolution kernel in the input feature map, is the weight of the convolution kernel, b is the bias term, and ReLU is the activation function.

[0010] Preferably, the reinforcement learning optimization module uses the extracted feature information to optimize the task planning and scheduling strategy through a reinforcement learning algorithm (Q-learning), wherein the objective function of the reinforcement learning algorithm can be expressed as: ; Among them, θ is the policy network parameter, π is the policy, τ represents a trajectory, and r t is the immediate reward obtained at time step t, γ is the discount factor, and T is the number of time steps in which the task is completed.

[0011] Preferably, the task planning module generates a task planning scheme, including task allocation, execution order, resource allocation, etc., according to the optimal strategy output by the reinforcement learning optimization module.

[0012] Preferably, the scheduling execution module dynamically schedules task execution according to the plan generated by the task planning module, and adjusts the scheduling strategy according to real-time feedback to achieve adaptive intelligent task scheduling.

[0013] Preferably, the system also includes a performance evaluation module for evaluating the effectiveness of task planning and scheduling, using evaluation indicators such as task completion time, resource utilization, task success rate, etc.

[0014] To achieve the above object, the technical solution adopted by the present invention includes the following steps: Step S1. Data input and preprocessing: Receive task-related data through the data input module, including but not limited to task description, resource status, time constraints and other information, and then preprocess these data, including data cleaning, format conversion and necessary normalization processing, to ensure the consistency and applicability of the data; Step S2. Feature extraction: The preprocessed task data is sent to the feature extraction module, and a convolutional neural network is used to perform deep feature extraction on the task data. The convolutional neural network can automatically learn and extract high-level abstract features from the task data through multi-layer convolution operations, providing strong support for subsequent task planning and scheduling; Step S3. Reinforcement learning optimization: Based on feature extraction, the task planning and scheduling strategy is optimized using the reinforcement learning optimization module. This module uses the reinforcement learning algorithm (Q-learning) to continuously trial and error and learn, with the goal of maximizing cumulative rewards, and gradually adjusts and optimizes the task planning and scheduling strategy; Step S4. Task planning: Based on the optimal strategy output by the reinforcement learning optimization module, the task planning module generates a specific task planning plan, which includes key contents such as task allocation, execution order, and resource allocation, aiming to ensure that the task can be completed efficiently and orderly; Step S5. Scheduling execution and adaptive adjustment: The scheduling execution module dynamically schedules task execution according to the task planning scheme. During the execution process, the system will make adaptive adjustments through the scheduling execution module according to real-time feedback and environmental changes to cope with uncertain factors and emergencies, ensuring the stability and reliability of task execution; Step S6. Performance evaluation: The system also includes a performance evaluation module, which is used to comprehensively evaluate the effect of task planning and scheduling, and to quantitatively analyze system performance through evaluation indicators such as task completion time, resource utilization, task success rate, etc., to provide a basis for subsequent optimization and improvement; The beneficial effects of the present invention are: Efficient feature extraction: Utilizing the powerful feature extraction capabilities of convolutional neural networks, key features can be automatically extracted from complex and changeable task data, providing accurate data support for task planning and scheduling.

[0015] Intelligent optimization: Through the reinforcement learning algorithm, task planning and scheduling strategies are optimized, which can automatically adapt to the dynamically changing environment and improve the efficiency and effectiveness of task execution.

[0016] Adaptive adjustment: The system can make adaptive adjustments based on real-time feedback and environmental changes to ensure the stability and reliability of task execution and improve the robustness of the system.

[0017] Comprehensive evaluation: The performance evaluation module is used to comprehensively evaluate system performance, provide a scientific basis for subsequent optimization and improvement, and promote continuous upgrading and improvement of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The above aspects and advantages of the present invention will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which: Figure 1 is a schematic diagram of an adaptive intelligent task planning and scheduling system based on a convolutional neural network according to an embodiment of the present invention; Figure 2 It is a flow chart of an adaptive intelligent task planning and scheduling system based on a convolutional neural network according to an embodiment of the present invention. Description of Reference Numerals

[0019] exist Figure 1-Figure 2 Among them, 1-data input module; 2-feature extraction module (including CNN); 3-reinforcement learning optimization module (including Q-learning); 4-task planning module; 5-scheduling execution module; 6-data input and processing; 7-feature extraction; 8-reinforcement learning optimization; 9-task planning; 10-scheduling execution and adaptive adjustment; 11-performance evaluation. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] like Figure 1 As shown, the present invention proposes an innovative adaptive intelligent task planning and scheduling system, which deeply integrates convolutional neural networks and reinforcement learning technology to optimize the task execution efficiency in complex environments.

[0022] The data input module 1 is responsible for receiving and preprocessing task-related data, such as task description, resource status vector (where n is the number of features), as well as time constraints, etc. Preprocessing includes data cleaning, format conversion, and normalization to ensure that the data is suitable for subsequent processing.

[0023] The feature extraction module 2 uses a convolutional neural network, the core of which is to extract high-level features from the input data through multiple layers of convolution kernels and pooling layers. Assuming that the input data is in the form of an image after preprocessing, the output feature map of the convolutional neural network can be expressed as: (1); Among them, Conv represents the convolution operation, W1 is the convolution kernel weight, and b1 is the bias term; Pool represents the pooling operation, which is used to reduce the dimension and extract important features; ReLU is the activation function, which increases nonlinearity. Finally, the extracted features are used for subsequent reinforcement learning optimization.

[0024] In the reinforcement learning optimization module 3, Q-learning learns the optimal strategy by iteratively updating the Q value table. Assuming that the state space is S and the action space is A, the Q function can be expressed as Q(s,a), where s∈S, a∈A. The update of the Q value is based on the following formula (2): (2); Among them, s t and a t are the state and action at time t, r t+1 Is to perform action a t After the state s t Transfer to state s t+1 The instant reward obtained, α is the learning rate, and γ is the discount factor. In the deep Q network, the Q function is approximated by a neural network, the input is the output of the feature extraction module, and the output is the Q value of all possible actions.

[0025] The task planning module 4 selects the action with the highest Q value as the optimal action in the current state according to the Q value table (or Q network prediction) output by the reinforcement learning optimization module, and generates a task planning scheme accordingly. The scheme includes task allocation, execution order and resource scheduling.

[0026] The scheduling execution module 5 executes tasks according to the task planning scheme, and monitors the task execution status and resource usage. If there are environmental changes or abnormal situations, the module will collect new status information and feed it back to the reinforcement learning optimization module to dynamically adjust the strategy.

[0027] In summary, the adaptive intelligent task planning and scheduling system described in the present invention realizes the intelligent, efficient and adaptive execution of tasks by integrating the feature extraction capability of convolutional neural networks, the optimization capability of reinforcement learning and efficient task planning and scheduling strategies. The system can significantly improve the accuracy and efficiency of task planning and scheduling, reduce labor costs, and reduce resource waste. It can be widely used in multiple fields such as manufacturing, logistics, and service industries to meet business needs such as production planning, logistics scheduling, and service management.

[0028] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An adaptive intelligent task planning and scheduling system based on convolutional neural network, characterized in that: It includes a data input module, a feature extraction module, a reinforcement learning optimization module, a task planning module and a scheduling execution module. The output of the data input module is the input of the feature extraction module, the output of the feature extraction module is the input of the reinforcement learning optimization module, the output of the reinforcement learning optimization module is the input of the task planning module, and the output of the task planning module is the input of the scheduling execution module.

2. The adaptive intelligent task planning and scheduling system based on convolutional neural network according to claim 1, characterized in that: The data input module is used to receive and pre-process task-related data, including task description, resource status, and time constraint information.

3. The adaptive intelligent task planning and scheduling system based on convolutional neural network according to claim 1, characterized in that: The feature extraction module uses a convolutional neural network to extract features from the preprocessed task data.

4. The adaptive intelligent task planning and scheduling system based on convolutional neural network according to claim 1, characterized in that: The reinforcement learning optimization module uses the extracted feature information to optimize task planning and scheduling strategies through a reinforcement learning algorithm.

5. The adaptive intelligent task planning and scheduling system based on convolutional neural network according to claim 1, characterized in that: The task planning module generates a task planning scheme according to the optimal strategy output by the reinforcement learning optimization module, including task allocation, execution sequence, and resource allocation.

6. The adaptive intelligent task planning and scheduling system based on convolutional neural network according to claim 1, characterized in that: The scheduling execution module dynamically schedules task execution according to the plan generated by the task planning module, and adjusts the scheduling strategy according to real-time feedback to achieve adaptive intelligent task scheduling.

7. The adaptive intelligent task planning and scheduling system based on convolutional neural network according to claim 1, characterized in that: It includes a performance evaluation module, which is used to evaluate the effectiveness of task planning and scheduling, using evaluation indicators such as task completion time, resource utilization, and task success rate.

Images