Task planning domain software rapid construction method and device based on feature variability

Abstract

The invention provides a task planning field software rapid construction method and device based on feature variability, and relates to the field of software design. The method comprises the following steps: constructing basic function components for realizing different plans; constructing a feature model of the task planning series software, setting task types as multi-choice feature points, setting mode types as single-choice feature points, and setting additional functions as selectable feature points; according to the constructed feature model, generating all effective software variants by calculating cosine similarity of multi-choice, single-choice and selectable feature point combinations; and based on the calling dependency relationship and priority among the components, developing a scheduling algorithm for optimizing the integration sequence of the components, and according to the scheduling algorithm, calculating the importance weight of the specific integration position sequence of each component by adopting an attention mechanism to construct a final application program. According to the scheme, the planning software can be efficiently generated, and meanwhile, the assembly integration sequence is reasonably scheduled, so that high-quality software products meeting personalized requirements of users are delivered in a relatively short time.

Description

Technical Field

[0001] The present invention belongs to the field of software design, and in particular relates to a method and device for quickly constructing software in the task planning field based on feature variability. Background Art

[0002] Mission planning is a very important and widely used field, involving rescue planning, supply planning, drone planning, ride planning, navigation planning, etc. Traditional mission planning software systems are usually customized for a specific type of mission or scenario, lacking flexibility and scalability.

[0003] As the field of mission planning continues to develop, users' demands for planning software are also changing and expanding. Users may need a software that supports multiple mission planning types at the same time, or add new planning methods, optimization algorithms, and auxiliary functions based on the original planning functions. However, it is very difficult to quickly respond to these changing needs using the traditional software development model.

[0004] In the existing technology, some studies have proposed methods based on software product lines and service orientation to improve the reusability and flexibility of software. However, these methods often require a lot of domain analysis and modeling work, and it is still a challenge to quickly build software systems that meet the personalized needs of users. On the other hand, feature modeling technology is widely used to describe and build highly configurable software systems. By building a feature model and selecting features as needed, different software instances can be generated more conveniently. However, the existing feature models mainly focus on the functional characteristics of the software and lack modeling and optimization support for the software construction process.

[0005] Therefore, a new method for rapid software construction is urgently needed, which can efficiently generate various variants of planning software based on feature models and reasonably schedule the order of component integration, so as to deliver high-quality software products that meet users' personalized needs in a shorter time. Summary of the invention

[0006] In view of this, the present invention provides a method for quickly constructing task planning domain software based on feature variability, comprising the following steps:

[0007] Step 1: Build basic functional components to implement different plans;

[0008] Step 2, construct a feature model for the task planning series software, set the task type to a multiple-select feature point, the method type to a single-select feature point, and the additional function to an optional feature point;

[0009] Step 3, based on the constructed feature model, all valid software variants are generated by calculating the cosine similarity of multiple-choice, single-choice and optional feature point combinations;

[0010] Step 4: Based on the calling dependencies and priorities between components, develop a scheduling algorithm for optimizing the order of component integration. The scheduling algorithm uses an attention mechanism to calculate the importance weight of the specific location order of each component integration to build the final application.

[0011] In particular, in step 1, the basic functional components of different plans include: implementing rescue planning, supply planning, drone planning, ride planning and navigation planning.

[0012] In particular, in step 3, all valid software variants are calculated and generated, including:

[0013]

[0014] Among them, P represents the combined feature vector, which is a vector generated based on the combination of multiple-choice, single-choice and optional feature points; C represents the feature model vector, which is a pre-constructed feature model vector representing all variable features of the task planning series software; w P , w C are the weight coefficients of the combined feature vector P and the feature model vector C in each dimension, which are used to adjust the importance of features in different dimensions. They are usually set to non-negative values, and the larger the value, the more important the dimension. i , C i Respectively represent the values ​​of the combined feature vector P and the feature model vector C in the i-th dimension; can be 0 or 1, 1 means the feature exists, 0 means it does not exist;

[0015] N P , N C They are the normalization factors of the P vector and the C vector respectively. The purpose is to normalize the vectors to a similar scale to avoid the influence of length difference on the similarity. The value is the L2 norm of the vector.

[0016] In particular, combined features with cosine similarity values ​​above a preset threshold are considered valid variants.

[0017] In particular, the attention mechanism is used to calculate the importance weights of the specific position order of each component integration to build the final application, including:

[0018]

[0019] Among them, α i,k W represents the attention weight of the kth component at the i-th integration position; l , U l , Γ l is the model learnable parameter; F i,kRepresents the feature vector of the kth component at the i-th integration position; the integration order of each component is determined according to the attention weight.

[0020] In particular, the eigenvector F i,k Contains various property information describing the component.

[0021] In particular, the attribute information includes component type, function, call dependency and priority.

[0022] The present invention also proposes a device for quickly constructing task planning domain software based on feature variability, comprising:

[0023] Basic functional component building module, used to build basic functional components to implement different plans;

[0024] The feature model building module is used to build the feature model of the task planning series software, setting the task type as a multiple-select feature point, the mode type as a single-select feature point, and the additional function as an optional feature point;

[0025] A software variant generation module is used to generate all valid software variants by calculating the cosine similarity of multiple-choice, single-choice and optional feature point combinations according to the constructed feature model;

[0026] The final application building module is used to develop a scheduling algorithm for optimizing the component integration order based on the call dependencies and priorities between components. The scheduling algorithm uses the attention mechanism to calculate the importance weight of the specific location order of each component integration to build the final application.

[0027] The method provided by the present invention has the following beneficial technical effects:

[0028] 1. Rapidly build a task planning software system that meets the personalized needs of users. By building a task planning software feature model based on feature variability, users can flexibly select the required task type, planning method and additional functions according to actual needs, and quickly generate customized software variants, greatly shortening the software development cycle.

[0029] 2. Improve the reusability and scalability of the software, encapsulate different planning functions into reusable components, and generate new software variants through feature model combination. If there are new planning requirements in the future, you only need to develop the corresponding functional components and update the feature models without affecting the reuse of existing components.

[0030] 3. Optimize the software building process, design a scheduling algorithm based on component call dependencies and priorities, and use the attention mechanism to calculate the weight of the integration order of each component to ensure that the final generated application has high performance and reliability.

[0031] 4. Reduce software development and maintenance costs. The feature-based construction method avoids repeated analysis, modeling and coding work, which can greatly save manpower and time costs. At the same time, the maintenance cost of deriving new variants based on existing components is also low.

[0032] 5. Enhance the flexibility and customizability of the software system. The feature model makes the functional and non-functional characteristics of the software explicit. Users can freely combine them according to their needs to achieve highly flexible customization to meet the application needs of different fields and scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 A schematic diagram of a process flow of a method for rapidly constructing task planning domain software based on feature variability proposed by the present invention;

[0034] Figure 2 This is a schematic diagram of a device for quickly constructing task planning domain software based on feature variability proposed by the present invention. DETAILED DESCRIPTION

[0035] The present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0036] The present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0037] The present invention provides a method for quickly constructing task planning domain software based on feature variability, such as Figure 1 As shown, the following steps are included:

[0038] Step 1, construct basic functional components to implement different plans; in step 1, the basic functional components of different plans include: implementing rescue planning, supply planning, drone planning, ride planning and navigation planning.

[0039] Rescue planning includes planning rescue routes, manpower and material resources deployment, etc. based on information such as the accident location, casualties, and on-site environment. Integrate the collaborative operation process of various rescue forces (firefighting, medical, etc.), and consider emergency plans such as traffic control and evacuation. Supply planning includes: planning the types and quantities of supplies according to the needs of disaster relief operations, considering the storage, transportation, distribution routes and time arrangements of supplies, or establishing a tracking and dispatching mechanism for supplies. Drone planning includes: planning the take-off and landing points, routes, and altitudes of drones according to operational needs; integrating drone reconnaissance, surveillance, search and other uses and processing data transmission, command control, power management, etc. Ride planning includes: planning vehicle dispatch according to the starting point, destination and number of personnel evacuated; arranging boarding points, boarding order, and transportation routes; considering vehicle types and passenger capacity, etc. Navigation planning includes: integrating the global positioning system (GPS), geographic information system (GIS), etc.; providing accurate and on-site navigation services for various rescue vehicles and personnel, and dynamically planning detour routes based on real-time road conditions.

[0040] Step 2, construct a feature model of a mission planning series software, set the mission type as a multiple-select feature point, the mode type as a single-select feature point, and the additional function as an optional feature point; the feature model includes: root feature: mission planning series software; the mission type uses multiple-select feature points, including: rescue planning, supply planning, drone planning, ride planning, navigation planning; the mode type uses single-select feature points, including: constraint solving, heuristic algorithm-based, machine learning-based, rule reasoning; additional functions use optional feature points, including: decision support, risk assessment, resource management, information sharing, 2D / 3D visualization, real-time monitoring and report generation.

[0041] Step 3, based on the constructed feature model, all valid software variants are generated by calculating the cosine similarity of multiple-choice, single-choice and optional feature point combinations; in step 3, all valid software variants are calculated and generated, specifically including:

[0042]

[0043] Among them, P represents the combined feature vector, which is a vector generated based on the combination of multiple-choice, single-choice and optional feature points; C represents the feature model vector, which is a pre-constructed feature model vector representing all variable features of the task planning series software; w P , w C are the weight coefficients of the combined feature vector P and the feature model vector C in each dimension, which are used to adjust the importance of features in different dimensions. They are usually set to non-negative values, and the larger the value, the more important the dimension. i , C iRespectively represent the values ​​of the combined feature vector P and the feature model vector C in the i-th dimension; can be 0 or 1, 1 means the feature exists, 0 means it does not exist;

[0044] N P , N C They are normalization factors for P vector and C vector respectively. The purpose is to normalize the vectors to a similar scale to avoid the influence of length difference on similarity. The value is the L2 norm of the vector. The combined features with cosine similarity values ​​higher than the preset threshold are regarded as valid variants.

[0045] Step 4: Based on the call dependencies and priorities between components, develop a scheduling algorithm for optimizing the order of component integration. The scheduling algorithm uses an attention mechanism to calculate the importance weights of the specific position order of each component integration and build the final application. The algorithm uses an attention mechanism to calculate the importance weights of the specific position order of each component integration and build the final application, which specifically includes:

[0046]

[0047] Among them, α i,k W represents the attention weight of the kth component at the i-th integration position; l , U l , Γ l is the model learnable parameter; F i,k The feature vector of the kth component at the i-th integration position is represented; the integration order of each component is determined according to the attention weight. Feature vector F i,k Includes various attribute information describing the component, including component type, function, call dependency and priority.

[0048] The present invention also discloses a device for quickly constructing task planning domain software based on feature variability, such as Figure 2 As shown, including:

[0049] The basic functional component building module is used to build basic functional components for implementing different plans; in the basic functional component building module, the basic functional components for different plans include: implementing rescue planning, supply planning, drone planning, ride planning and navigation planning.

[0050] Rescue planning includes planning rescue routes, manpower and material resources deployment, etc. based on information such as the accident location, casualties, and on-site environment. Integrate the collaborative operation process of various rescue forces (firefighting, medical, etc.), and consider emergency plans such as traffic control and evacuation. Supply planning includes: planning the types and quantities of supplies according to the needs of disaster relief operations, considering the storage, transportation, distribution routes and time arrangements of supplies, or establishing a tracking and dispatching mechanism for supplies. Drone planning includes: planning the take-off and landing points, routes, and altitudes of drones according to operational needs; integrating drone reconnaissance, surveillance, search and other uses and processing data transmission, command control, power management, etc. Ride planning includes: planning vehicle dispatch according to the starting point, destination and number of personnel evacuated; arranging boarding points, boarding order, and transportation routes; considering vehicle types and passenger capacity, etc. Navigation planning includes: integrating the global positioning system (GPS), geographic information system (GIS), etc.; providing accurate and on-site navigation services for various rescue vehicles and personnel, and dynamically planning detour routes based on real-time road conditions.

[0051] A feature model building module is used to build a feature model for a mission planning series software, setting the mission type as a multiple-select feature point, the mode type as a single-select feature point, and the additional function as an optional feature point; the feature model includes: root feature: mission planning series software; the mission type uses multiple-select feature points, including: rescue planning, supply planning, drone planning, ride planning, and navigation planning; the mode type uses single-select feature points, including: constraint solving, heuristic algorithm-based, machine learning-based, and rule reasoning; the additional functions use optional feature points, including: decision support, risk assessment, resource management, information sharing, 2D / 3D visualization, real-time monitoring, and report generation.

[0052] The software variant generation module is used to generate all valid software variants by calculating the cosine similarity of multiple-choice, single-choice and optional feature point combinations according to the constructed feature model; in the software variant generation module, all valid software variants are calculated and generated, specifically including:

[0053]

[0054] Where P represents the combined feature vector, which is a vector generated by combining multiple-choice, single-choice and optional feature points; C represents the feature model vector, which is a pre-constructed feature model vector representing all variable features of the task planning series software; wP, w C are the weight coefficients of the combined feature vector P and the feature model vector C in each dimension, which are used to adjust the importance of features in different dimensions. They are usually set to non-negative values, and the larger the value, the more important the dimension. i , C iRespectively represent the values ​​of the combined feature vector P and the feature model vector C in the i-th dimension; can be 0 or 1, 1 means the feature exists, 0 means it does not exist;

[0055] N P , N C They are normalization factors for P vector and C vector respectively. The purpose is to normalize the vectors to a similar scale to avoid the influence of length difference on similarity. The value is the L2 norm of the vector. The combined features with cosine similarity values ​​higher than the preset threshold are regarded as valid variants.

[0056] The final application building module is used to develop a scheduling algorithm for optimizing the component integration order based on the call dependencies and priorities between components. The scheduling algorithm uses the attention mechanism to calculate the importance weight of the specific location order of each component integration to build the final application.

[0057] The attention mechanism is used to calculate the importance weight of the specific position order of each component integration to build the final application, including:

[0058]

[0059] Among them, α i,k W represents the attention weight of the kth component at the i-th integration position; l , U l , Γ l is the model learnable parameter; F i,k The feature vector of the kth component at the i-th integration position is represented; the integration order of each component is determined according to the attention weight. Feature vector F i,k Includes various attribute information describing the component, including component type, function, call dependency and priority.

[0060] In summary, the above are only preferred embodiments of the present invention and are not intended to limit the protection scope of the present invention. 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.

[0061] It is obvious to those skilled in the art that the embodiments of the present invention are not limited to the details of the above exemplary embodiments, and that the embodiments of the present invention can be implemented in other specific forms without departing from the spirit or basic features of the embodiments of the present invention. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-restrictive, and the scope of the embodiments of the present invention is limited by the attached claims rather than the above description, so it is intended to include all changes that fall within the meaning and scope of the equivalent elements of the claims in the embodiments of the present invention. Any figure mark in the claims should not be regarded as limiting the claims involved. In addition, it is obvious that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units, modules or devices stated in the system, device or terminal claims can also be implemented by the same unit, module or device through software or hardware. The words first, second, etc. are used to indicate names, and do not indicate any particular order.

[0062] Finally, it should be noted that the above implementation modes are only used to illustrate the technical solutions of the embodiments of the present invention and are not intended to limit them. Although the embodiments of the present invention have been described in detail with reference to the above preferred implementation modes, those skilled in the art should understand that the technical solutions of the embodiments of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the embodiments of the present invention.

[0063] In summary, the above are only preferred embodiments of the present invention and are not intended to limit the protection scope of the present invention. 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.

[0064] It is obvious to those skilled in the art that the embodiments of the present invention are not limited to the details of the above exemplary embodiments, and that the embodiments of the present invention can be implemented in other specific forms without departing from the spirit or basic features of the embodiments of the present invention. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-restrictive, and the scope of the embodiments of the present invention is limited by the attached claims rather than the above description, so it is intended to include all changes that fall within the meaning and scope of the equivalent elements of the claims in the embodiments of the present invention. Any figure mark in the claims should not be regarded as limiting the claims involved. In addition, it is obvious that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units, modules or devices stated in the system, device or terminal claims can also be implemented by the same unit, module or device through software or hardware. The words first, second, etc. are used to indicate names, and do not indicate any particular order.

[0065] Finally, it should be noted that the above implementation modes are only used to illustrate the technical solutions of the embodiments of the present invention and are not intended to limit them. Although the embodiments of the present invention have been described in detail with reference to the above preferred implementation modes, those skilled in the art should understand that the technical solutions of the embodiments of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for rapidly constructing task planning domain software based on feature variability, characterized in that: The following steps are involved: Step 1: Build basic functional components to implement different plans; Step 2: Construct a feature model for the task planning software series, set the task type to a multiple-select feature point, the method type to a single-select feature point, and the additional function to an optional feature point; Step 3: Based on the constructed feature model, all valid software variants are generated by calculating the cosine similarity of multiple-choice, single-choice and optional feature point combinations; Step 4: Based on the calling dependencies and priorities between components, a scheduling algorithm is developed to optimize the order of component integration. The scheduling algorithm uses an attention mechanism to calculate the importance weight of the specific location order of each component integration to build the final application.

2. The method for rapidly constructing task planning domain software based on feature variability according to claim 1, characterized in that: In step 1, the basic functional components of different plans include: realizing rescue planning, supply planning, drone planning, ride planning and navigation planning.

3. The method for rapidly constructing task planning domain software based on feature variability according to claim 1, characterized in that: In step 3, all valid software variants are calculated and generated, including: Among them, P represents the combined feature vector, which is a vector generated based on the combination of multiple-choice, single-choice and optional feature points; C represents the feature model vector, which is a pre-constructed feature model vector representing all variable features of the task planning series software; w P , w C are the weight coefficients of the combined feature vector P and the feature model vector C in each dimension, which are used to adjust the importance of features in different dimensions and are set to non-negative values. The larger the value, the more important the dimension. i , C i Respectively represent the values ​​of the combined feature vector P and the feature model vector C in the i-th dimension; can be 0 or 1, 1 means the feature exists, 0 means it does not exist; N P , N C They are the normalization factors of the P vector and the C vector respectively. The purpose is to normalize the vectors to a similar scale to avoid the influence of length difference on the similarity. The value is the L2 norm of the vector.

4. The method for rapidly constructing task planning domain software based on feature variability according to claim 3 is characterized in that: Combined features with cosine similarity values ​​higher than a preset threshold are considered valid variants.

5. The method for rapidly constructing task planning domain software based on feature variability according to claim 4, characterized in that: The attention mechanism is used to calculate the importance weights of the specific position order of each component integration to build the final application, including: Among them, α i,k represents the attention weight of the kth component at the i-th integration position; W l , U l , Γ l is the model learnable parameter; F i,k Represents the feature vector of the kth component at the i-th integration position; the integration order of each component is determined according to the attention weight.

6. The method for rapidly constructing task planning domain software based on feature variability according to claim 5, characterized in that: Eigenvector F i,k Contains various property information describing the component.

7. The method for rapidly constructing task planning domain software based on feature variability according to claim 6, characterized in that: The attribute information includes component type, function, call dependency and priority.

8. A device for rapidly constructing task planning domain software based on feature variability, characterized in that: include: Basic functional component building module, used to build basic functional components to implement different plans; The feature model building module is used to build the feature model of the task planning series software, setting the task type as a multiple-select feature point, the mode type as a single-select feature point, and the additional function as an optional feature point; A software variant generation module is used to generate all valid software variants by calculating the cosine similarity of multiple-choice, single-choice and optional feature point combinations according to the constructed feature model; The final application building module is used to develop a scheduling algorithm for optimizing the component integration order based on the call dependencies and priorities between components. The scheduling algorithm uses the attention mechanism to calculate the importance weight of the specific location order of each component integration to build the final application.

Images