A cooperative control method for dynamic interaction of unmanned information groups

Abstract

This invention belongs to the field of intelligent control technology for unmanned vehicles, specifically relating to a collaborative control method for dynamic interaction of unmanned information swarms. This method mainly involves three parts: initial determination of sub-swarm information, feature-based global planning and operation of the swarm, and feature-based local planning and operation of the swarm. First, sub-swarm information is identified, separated, and extracted based on the limited information available to the sub-swarm members. Based on the calculation results, the state of each sub-swarm member within the swarm can be effectively determined under different environments, thereby obtaining the connection state between sub-swarm members. Considering reasonable system information input and effective interaction, the distance between different sub-swarm members is dynamically adjusted with the overall swarm performance as the objective, thus realizing the global and local planning, operation, and control process of the unmanned swarm's overall architecture under information interaction.

Description

Technical Field

[0001] This invention belongs to the field of intelligent control technology for unmanned vehicles, specifically relating to a collaborative control method for dynamic interaction of unmanned information swarms. This collaborative control method mainly involves overall planning and control under the interaction of limited information sets of unmanned swarms in different environments. During the operation of the unmanned swarm, considering the dynamic changes in the swarm architecture and the interaction and influence of information exchange between individual components, the swarm needs to conduct multi-dimensional overall planning and effective control of the operational status of its components based on its characteristics. In this process, both overall performance requirements and achievable goals for individual components must be considered, completing the global and local planning of the swarm under constraints, thereby achieving a scalable and extensible intelligent operation process for unmanned swarms adapted to different environments. Background Technology

[0002] During unmanned swarm operation, the swarm configuration will change accordingly due to changes in the environment, such as... Figure 1 As shown, when the configuration changes, individuals need to make corresponding dynamic adjustments according to the changes in the group in order to adapt to the group performance requirements in complex environments.

[0003] like Figure 1 As shown, under different operating conditions, the configuration of the group exhibits various complex and diverse variations. However, actual group configuration changes are not limited to the types mentioned above and can be even more complex. During this dynamic process, each sub-entity also undergoes corresponding changes based on requirements. In this dynamic process, the interactive information between sub-entities also changes with the environment and configuration. If we consider the information input of the sub-entities as a dynamic information group, and further consider the differences in interaction between the group information group and the individual information groups, then we need to consider a group control process based on the dynamic interaction of the information group.

[0004] In this process, the swarm information collection relies on the information upload and sharing of each sub-entity, while the effective issuance of sub-entity commands depends on effective decision-making based on global information input. When sub-entities are located in different spatial positions, the external data they collect differs, and this difference increases with the distance between individuals. The integrated control of the unmanned swarm depends on the effective input of external environmental data, but it also requires the establishment of a good interconnection mechanism among the sub-entities. If the distance between the sub-entities is too small, the acquired external information has a high degree of overlap, which is not conducive to the effective adaptation of the unmanned swarm in unknown environments. Conversely, if the distance between the sub-entities is too large, the external information received by each sub-entity has no overlap, and the correlation between them is very low, which is not conducive to the swarm's effective control over the sub-entities. Therefore, swarm control based on spatial layout and sub-entity distance fusion becomes a key scientific problem in the operation of unmanned swarms in their location environment. Summary of the Invention

[0005] (a) Technical problems to be solved

[0006] The technical problem to be solved by this invention is: how to provide a collaborative control method for dynamic interaction of unmanned information groups.

[0007] (II) Technical Solution

[0008] To address the aforementioned technical problems, this invention provides a collaborative control method for dynamic interaction of unmanned information swarms, the collaborative control method comprising:

[0009] Step 1: Initialization and determination of sub-entity information group;

[0010] Step 2: Feature-based population global planning operation phase;

[0011] Step 3: Feature-based group local planning operation.

[0012] The sub-body information group initialization determination step in step 1 includes:

[0013] Step 1.1: Complete the process of identifying and separating sub-entity information groups;

[0014] The unmanned platform group is defined as consisting of the leading vehicle A, the middle vehicle B, and the trailing vehicle C, with their corresponding information group sets being S. A S B and S C In actual driving, the trajectories of different vehicles are not always on the same straight line. Therefore, the straight-line distance between the first vehicle A and the middle vehicle B is L. AB The straight-line distance between the middle car B and the rear car C is L. BC The straight-line distance between car A in front and car C behind is L. AC ;

[0015] For the unmanned platform group, the information group S is composed of the leading vehicle A, the middle vehicle B, and the trailing vehicle C. A S B and S C By identifying and separating the information, and calculating the common information set R of the three sub-information groups, we have:

[0016] R = S A ∩S B ∩S C

[0017] Based on this, calculate the local information set R among different sub-groups of information. AB R BC and R AC Then we have:

[0018]

[0019] Finally, calculate the unique sub-entity information set R for each sub-entity information group. A R B and R C Then we have:

[0020]

[0021] In the process of information group computation, the common information set R and the local information set Ri of the information group are... AB R BC R AC The data is related to the distance between the three sub-entities. The closer the distance, the more information data is in the intersection; the farther the distance, the less information data is in the intersection. Therefore, data feature calculation and collaborative control of the information group are carried out based on the distance between the sub-entities.

[0022] Step 1.2: Complete the sub-entity information group extraction and calculation process;

[0023] Information groups S targeting unmanned platform users A S B and S C First, calculate the information sharing rate η. ABC Information discrepancy rate η A|B|C As shown below:

[0024]

[0025] In the formula, M(x) represents the size of the set x.

[0026] Based on the local information set R among different sub-groups AB R BC and R AC It is necessary to calculate the information detachment rate η for each pair of information groups separately. A|B η B|C and η A|C Then we have:

[0027]

[0028] In the planning and operation of unmanned swarms, global planning is prioritized. After the global planning is completed, local planning and operation are carried out based on the characteristics of the swarm.

[0029] The feature-based population global planning operation in step 2 includes:

[0030] Step 2.1: Complete the calculation of global features of the population;

[0031] When the information sharing rate η ABCWhen η = 1, it represents complete information sharing within the group, with no external information input, corresponding to the information difference rate η. A|B|C =0, which indicates that the group has no external information to refer to in an unknown environment, and the group cannot make decisions in this situation;

[0032] Information difference rate and information sharing rate η A|B|C When η = 1, it represents that the group is in a state of complete input of external information, and the corresponding η is... ABC =0 indicates that the sub-entities within the group are excessively separated, and information between them is completely impossible to share. At this point, the individuals in the group lose information interaction, and the group loses its cluster interconnection function.

[0033] Under ideal conditions, the information sharing rate η ABC Information discrepancy rate η A|B|C A dynamic equilibrium needs to be achieved within a certain range to enable real-time input of external information and reasonable sharing of internal information, thereby providing reasonable decision-making references for system subsystems and overall decision-making;

[0034] The information sharing rate η ABC Information discrepancy rate η A|B|C The calculation method for the span Φ of the system is as follows:

[0035]

[0036] The system span Φ fluctuates dynamically during actual operation, and its corresponding dynamic equilibrium range is [ω1, ω2], with upper and lower limits as follows:

[0037]

[0038] Step 2.2: Complete the group global planning and execution process;

[0039] Based on the system span Φ and distance L AB L BC If the value is used for judgment, the following situations may occur:

[0040] (1) If the system span |Φ|∈[ω1, ω2], then the shared information within the system and the unknown information discovered from the outside world in the group in the external environment reach a dynamic equilibrium. At this time, the group is in a normal driving process and there is no need to adjust the group state.

[0041] (2) If the system span If Φ > 0, then the group is too compact in the external environment. If they continue traveling, the group will not have enough information in the unknown environment, and the information overlap between different sub-groups will be high, which is not conducive to the group making sufficient decisions in complex environments. At this time, it is necessary to increase the minimum distance L between different sub-groups. min =min(LAB L BC To achieve optimal results for different information groups;

[0042] Under this operating condition, if L AB ≠L BC minimum distance L min Need to be increased to the target distance L t Then we have:

[0043]

[0044] If L AB =L BC Distance L AB and L BC Need to be increased to the target distance L t Then we have the following calculation equation:

[0045]

[0046] (3) If the system span And Φ<0, then the group is too relaxed in the external environment. If the journey continues, the information of the group will be too scattered in the unknown environment. No sub-group can make a comprehensive judgment through the correlation data, thus making it impossible to make effective decisions in complex environments.

[0047] At this point, it is necessary to reduce the maximum distance L between different sub-body units. max =max(L AB L BC To achieve optimal group results for different information groups; under this condition, the maximum distance L max Need to be reduced to the target distance L t Then we have:

[0048]

[0049] If L AB =L BC Distance L AB and L BC Need to be increased to the target distance L t Then we have the following calculation equation:

[0050]

[0051] The feature-based group local planning operation in step 3 includes:

[0052] Step 3.1: Complete the calculation process of local features of the population;

[0053] Calculate: Information detachment rate η A|B ηB|C and η A|C If the intermediate value ω3 is given, then:

[0054]

[0055] Based on this, according to the information ionization rate η A|B η B|C and η A|C If the following information determination conditions are set, then the following constraints apply:

[0056] Condition 1: (n A|C +η A|B )>η B|C And (η) A|C +η B|C )>η A|B ;

[0057] Condition two:

[0058] Step 3.2: Complete the group local planning operation process;

[0059] Based on the information judgment criteria in step 3.1, a comprehensive evaluation is conducted, resulting in the following situations:

[0060] (1) If both conditions one and two are met under the current working conditions, then the operating architecture and spatial layout of each sub-group are temporarily in a reasonable state, and no dynamic adjustment is required.

[0061] (2) If only condition 2 is met but condition 1 is not met under the current working condition, then within the group, the layout of the front vehicle A, the rear vehicle C and the middle vehicle B is overly coupled. If the driving continues, the information processing and transmission time of the front vehicle A and the middle vehicle B will be short, which will affect the response decision of the rear vehicle C to the real-time working condition.

[0062] At this moment, the distance L between car A in front and car C behind is... AC To address the problem of excessive information coupling, ΔL needs to be increased. Therefore:

[0063]

[0064] (3) If only condition one is met but condition two is not met under the current working conditions, then within the group, the layout of the front vehicle A, the rear vehicle C and the middle vehicle B is excessively decoupled, that is, the degree of separation of the sub-body is large. If the driving continues, the information of the front vehicle A and the middle vehicle B will be at risk of failure for the rear vehicle C. At this time, the rear vehicle C will be lagging behind in following and responding to the group, which is not conducive to the coordinated operation of the group.

[0065] At this moment, the distance L between car A in front and car C behind is... AC To address the problem of excessive information decoupling, ΔL needs to be reduced. Therefore:

[0066]

[0067] (4) If neither condition one nor condition two is satisfied under the current operating conditions, then the coupling state of the front vehicle A, the rear vehicle C, and the middle vehicle B within the group needs to be further determined; at this time, the information detachment rate η needs to be judged. A|C Is it equal to ω2? If η A|C If ≠ω2, then continue to judge η. A|C Is it equal to ω1? If η A|C =ω1, at this time the distance between the sub-body is not adjusted; if η A|C If ≠ω1, then the distance L between car A and car C is... AC The system coupling needs to be reduced to increase its coupling; the current time is T, and the distance between vehicle A and the following vehicle C is L. AC|T At time (T+1), the distance L between vehicle A and the following vehicle C is... AC|T+1 It needs to be adjusted to:

[0068]

[0069] And if η A|C =ω2, then the distance L between the car in front A and the car behind C is... AC The system coupling needs to be increased to reduce the coupling; at time (T+1), the distance L between the preceding vehicle A and the following vehicle C... AC|T+1 It needs to be adjusted to:

[0070]

[0071] (III) Beneficial Effects

[0072] Compared with existing technologies, this invention takes the smallest unmanned platform swarm as its research basis and proposes a collaborative control method for dynamic interaction of unmanned information swarms based on feature extraction and effective fusion of information swarms. This invention identifies, separates, and extracts features from the swarm information. Based on this, during the planning and operation of the unmanned swarm, the swarm state is determined according to the features, and global planning is prioritized. After the global adjustment is completed, local planning and control operations are performed on the features of individual components, thereby realizing the information interaction and collaborative control process of unmanned swarms under the influence of a limited dataset.

[0073] Compared with traditional research methods, this invention mainly focuses on the study of limited datasets of sub-entities within a group. Considering the distribution characteristics of the group under spatial layout, and further considering the correlation between the information differences of sub-entities and the upper-level control of the group, a method based on dynamic fusion and effective judgment of information groups is proposed. Based on this, global and local planning and control processes are carried out during the group operation, thereby realizing a collaborative control method that is compatible with environmental scalability and adaptability and extended reliability. Attached Figure Description

[0074] Figure 1 This is a schematic diagram of the dynamic driving architecture for unmanned swarms.

[0075] Figure 2 A schematic diagram of a collaborative control architecture for dynamic interaction of unmanned information groups. Detailed Implementation

[0076] To make the objectives, contents, and advantages of the present invention clearer, the specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples.

[0077] Under the dynamic operation requirements of unmanned swarms, the dynamic information interaction and organic collaborative control among different sub-units within the swarm become the core issues for the stable operation of the swarm. The purpose of this invention is to provide a collaborative control method for the dynamic interaction of unmanned information swarms, so as to realize the dynamic information interaction process as the spatial scale between sub-units within the swarm changes, thereby completing the overall planning and operation process of the unmanned swarm with the goal of controllable interaction changes of information between different sub-units and overall information optimization in multiple dimensions.

[0078] The collaborative control architecture for dynamic interaction of unmanned information swarms involved in this invention is as follows: Figure 2 As shown in the figure, in the system control architecture, the spatial layout of the unmanned platform swarm changes dynamically in real time. Therefore, it is necessary to separate and extract the information of the sub-units for calculation. Based on the obtained information, global and local planning calculations of the swarm can be performed to achieve the swarm planning process under the optimal layout.

[0079] First, for the unmanned platform group consisting of the leading vehicle A, the middle vehicle B, and the trailing vehicle C, calculate the common information set R and the local information set R' corresponding to the information groups of the three sub-units. AB R BC and R AC and the sub-body information set R A R B and R C Simultaneously, based on the calculations of various information groups S of the unmanned platform group... A S B and S C Calculate the information sharing rate η respectively. ABC Difference rate η A|B|C and information detachment rate η A|B η B|C and η A|C .

[0080] Based on this, global and local planning calculations are performed using the extracted group characteristics. During the global planning process, the information sharing rate η is used as the basis for the calculations. ABC With difference rate ηA|B|C The system span Φ is calculated based on the changes in the system span Φ and the distance value L. AB L BC Then the distance L between the front car A and the middle car B can be determined. AB And the distance L between the middle car B and the rear car C BC Perform global dynamic adjustments.

[0081] During the local planning process, based on the information detachment rate η A|B η B|C and η A|C The intermediate value ω3 and the dynamic equilibrium limits ω1 and ω2 are used to propose corresponding constraints. This allows for the local dynamic adjustment of the straight-line distance L between the preceding vehicle A and the following vehicle C under different operating conditions based on real-time changes in group information. AC .

[0082] Example 1

[0083] This embodiment performs separation and correlation calculations based on limited information of different sub-entities within the swarm, fully considering the dynamic changes in characteristics during swarm operation, and conducts research on swarm collaborative control based on the dynamic interaction of sub-entity information, thereby completing the information interaction and real-time control process of the unmanned swarm. The following description, in conjunction with the accompanying drawings, further illustrates a collaborative control method for dynamic interaction of unmanned information swarms according to the present invention.

[0084] In real-world environments, the operation of multiple unmanned platforms requires the collaborative interaction of their individual components to achieve group functionality in complex terrains. The selection and decision-making regarding group characteristics depend on the fusion and mutual compensation of information from each component. Through the integration and selection of information from different components, organic collaboration is achieved during the operation of the group queue.

[0085] Based on practical application requirements, this paper introduces the method using a minimum unmanned platform swarm as a foundation. In actual operation, it is assumed that the unmanned platform swarm consists of a leading vehicle A, a middle vehicle B, and a trailing vehicle C, with their corresponding information sets denoted as S. A S B and S C In actual driving, the trajectories of different vehicles are not always on the same straight line. Therefore, the straight-line distance between the first vehicle A and the middle vehicle B is L. AB The straight-line distance between the middle car B and the rear car C is L. BC The straight-line distance between car A in front and car C behind is L. AC Therefore, the information exchange process and collaborative operation method of the unmanned platform are realized.

[0086] Step 1: Initialization and determination of sub-entity information group

[0087] During the operation of unmanned swarms, the first step is to identify, separate, and extract features based on the swarm information. Then, based on the calculation of information overlap and feature set, the initial determination of information within the information domain is completed.

[0088] 1.1 Sub-entity information group identification and separation

[0089] For the unmanned platform group, the information group S is composed of the leading vehicle A, the middle vehicle B, and the trailing vehicle C. A S B and S C First, we need to identify and separate the information, that is, calculate the common information set R of the three sub-information groups. Then we have:

[0090] R = S A ∩S B ∩S C

[0091] Based on this, calculate the local information set R among different sub-entity information groups. AB R BC and R AC Then, respectively:

[0092]

[0093] Finally, calculate the unique sub-entity information set R for each sub-entity information group. A R B and R C Then, respectively:

[0094]

[0095] In the process of information group computation, the common information set R and the local information set Ri of the information group are... AB R BC R AC The data is related to the distance between the three sub-entities; the closer the distance, the more information data is in the intersection; the farther the distance, the less information data is in the intersection. Therefore, data feature calculation and collaborative control of the information group are carried out based on the distance between the sub-entities.

[0096] 1.2 Sub-entity Information Cluster Extraction and Calculation

[0097] Based on the subgroup information group identification and separation calculation results, further feature extraction calculation of the information group is carried out. That is, based on the calculation results of different sets, the group features are extracted and calculated. With the numerical results as feature reference, the information differentiation and fusion calculation process under the explicit processing of group information features is completed.

[0098] Information groups S targeting unmanned platform users A S B and SC First, calculate the information sharing rate η. ABC Information discrepancy rate η A|B|C The following is an example:

[0099]

[0100] In the formula, M(x) represents the size of the set x.

[0101] Based on the local information set R among different sub-groups AB R BC and R AC It is necessary to calculate the information detachment rate η for each pair of information groups separately. A|B η B|C and η A|C The calculation method is as follows:

[0102]

[0103] In the planning and operation of unmanned swarms, global planning is prioritized. After the global planning is completed, local planning and operation are carried out based on the characteristics of the swarm.

[0104] Step 2: Feature-based Population Global Planning Operation

[0105] In the global planning and operation of the group, the information sharing rate η is mainly used. ABC With difference rate η A|B|C The changes in distance L are used to dynamically adjust the distance L between the leading vehicle A and the middle vehicle B in real time. AB And the distance L between the middle car B and the rear car C BC By adjusting the distance, the characteristics of the information group are dynamically adjusted to enable the scalability of the unmanned swarm in complex environments, thereby completing the adaptive process of swarm intelligence based on demand-adjustable swarm intelligence.

[0106] 2.1 Calculation of Global Features of the Population

[0107] When η ABC When η = 1, it represents complete sharing of system information within the group, with no external information input, and the corresponding η is... A|B|C =0, indicating that the group has no external information reference in the unknown environment, and the group cannot make decisions in this situation; while η A|B|C When η = 1, it represents that the group is in a state of complete input of external information, and the corresponding η is... ABC A value of η = 0 indicates excessive separation among the subgroups, with no information sharing between them. In this state, individuals within the group lose information interaction, and the group loses its cluster interconnection function. Ideally, the information sharing rate η is 0. ABC With difference rate η A|B|CA dynamic equilibrium needs to be achieved within a certain range to enable real-time input of external information and reasonable sharing of internal information, thereby providing reasonable decision-making references for system subsystems and overall decision-making.

[0108] Assuming an information sharing rate η ABC With difference rate η A|B|C If there is a system span Φ, the corresponding calculation method is as follows:

[0109]

[0110] The system span Φ fluctuates dynamically during actual operation, and its corresponding dynamic equilibrium range is [ω1, ω2], with upper and lower limits as follows:

[0111]

[0112] 2.2 Group Global Planning and Operation

[0113] Based on the calculated system span Φ and distance L AB L BC If the value is used for judgment, the following situations may occur:

[0114] (1) If the system span |Φ|∈[ω1, ω2], then the shared information within the system and the unknown information discovered from the outside world in the group in the external environment reach a dynamic equilibrium. At this time, the group is in a normal driving process and there is no need to adjust the group state.

[0115] (2) If the system span If Φ > 0, then the group is too tightly packed in the external environment. If travel continues, the group will lack sufficient information in the unknown environment, and the information overlap between different sub-groups will be high, which is not conducive to the group making sufficient decisions in complex environments. Therefore, it is necessary to increase the minimum distance L between different sub-groups. min =min(L AB L BC This is to achieve the optimal results for different information groups.

[0116] Under this operating condition, if L AB ≠L BC minimum distance L min Need to be increased to the target distance L t Then we have:

[0117]

[0118] If L AB =L BC Distance L AB and L BC Need to be increased to the target distance L tThen we have the following calculation equation:

[0119]

[0120] (3) If the system span If Φ < 0, then the group is too relaxed in the external environment. If the journey continues, the information of the group will be too scattered in the unknown environment. No individual can make a comprehensive judgment through the correlation data, thus making it impossible to make effective decisions in complex environments.

[0121] At this point, it is necessary to reduce the maximum distance L between different sub-body units. max =max(L AB L BC This is to achieve optimal group results for different information groups. Under this condition, the maximum distance L max Need to be reduced to the target distance L t Then we have:

[0122]

[0123] If L AB =L BC Distance L AB and L BC Need to be increased to the target distance L t Then we have the following calculation equation:

[0124]

[0125] Step 3: Feature-based Population Local Programming Operation

[0126] During the local planning process of a group, the information detachment rate η is mainly used. A|B η B|C and η A|C To dynamically adjust the straight-line distance L between the car in front (A) and the car behind (C) in real time. AC This enables unmanned swarm operations with extended reference local information, thereby completing the swarm autonomous adjustment process based on information sharing.

[0127] 3.1 Calculation of Local Features of the Population

[0128] In actual group operations, the information detachment rate η A|B η B|C and η A|CA dynamic equilibrium needs to be achieved. An abnormal increase in the information detachment rate indicates excessive information overlap between the two corresponding sub-sub ...

[0129] Calculate the free information rate η A|B η B|C and η A|C If the intermediate value ω3 is given, then:

[0130]

[0131] Based on this, according to the information ionization rate η A|B η B|C and η A|C If the following information determination conditions are set, then the following constraints apply:

[0132] Condition 1: (n A|C +η A|B )>η B|C And (η) A|C +η B|C )>η A|B .

[0133] Condition two:

[0134] 3.2 Local Planning and Operation of the Group

[0135] Based on the above information and judgment criteria, a comprehensive evaluation is conducted, resulting in the following situations:

[0136] (1) If both conditions one and two are met under the current working conditions, then the operating architecture and spatial layout of each sub-group are temporarily in a reasonable state, and no dynamic adjustment is required.

[0137] (2) If only condition 2 is met but condition 1 is not met under the current working condition, then within the group, the layout of the front vehicle A, the rear vehicle C and the middle vehicle B is overly coupled. If the driving continues, the information processing and transmission time of the front vehicle A and the middle vehicle B will be short, which will affect the response decision of the rear vehicle C to the real-time working condition.

[0138] At this moment, the distance L between car A in front and car C behind is... AC To address the problem of excessive information coupling, ΔL needs to be increased. Therefore:

[0139]

[0140] (3) If only condition one is met but condition two is not met under the current working conditions, then within the group, the layout of the front vehicle A, the rear vehicle C and the middle vehicle B is excessively decoupled, that is, the degree of separation of the sub-body is large. If the driving continues, the information of the front vehicle A and the middle vehicle B will be at risk of failure for the rear vehicle C. At this time, the rear vehicle C will be lagging behind in following and responding to the group, which is not conducive to the coordinated operation of the group.

[0141] At this moment, the distance L between car A in front and car C behind is... AC To address the problem of excessive information decoupling, ΔL needs to be reduced. Therefore:

[0142]

[0143] (4) If neither condition one nor condition two is satisfied under the current operating conditions, then the coupling state of the front vehicle A, the rear vehicle C, and the middle vehicle B within the group needs to be further determined. At this point, the information detachment rate η needs to be considered. A|C Is it equal to ω2? If η A|C If ≠ω2, then continue to judge η. A|C Is it equal to ω1? If η A|C =ω1, at this time the distance between the sub-body is not adjusted; if η A|C If ≠ω1, then the distance L between car A and car C is... AC This needs to be reduced to avoid increasing system coupling. At time T, the distance between vehicle A and the following vehicle C is L. AC|T At time (T+1), the distance L between vehicle A and the following vehicle C is... AC|T+1 It needs to be adjusted to:

[0144]

[0145] And if η A|C =ω2, then the distance L between the car in front A and the car behind C is... AC The system coupling needs to be increased to reduce its coupling. At time (T+1), the distance L between the preceding vehicle A and the following vehicle C... AC|T+1 It needs to be adjusted to:

[0146]

[0147] In summary, this invention belongs to the field of intelligent control technology for unmanned vehicles, specifically relating to a collaborative control method for dynamic interaction of unmanned information swarms. This method mainly involves three parts: initial determination of sub-swarm information, feature-based global planning and operation of the swarm, and feature-based local planning and operation of the swarm. First, sub-swarm information is identified, separated, and extracted based on the limited information available to the sub-swarm members. Based on the calculation results, the state of each sub-swarm member within the swarm can be effectively determined under different environments, thereby obtaining the connection state between sub-swarm members. Considering reasonable system information input and effective interaction, the distance between different sub-swarm members is dynamically adjusted with the overall swarm performance as the objective, thus realizing the global and local planning, operation, and control process of the unmanned swarm's overall architecture under information interaction.

[0148] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A collaborative control method for dynamic interaction of unmanned information groups, characterized in that, The collaborative control method includes: Step 1: Initialization and determination of sub-entity information group; Step 2: Feature-based population global planning operation phase; Step 3: Feature-based group local planning operation phase; The sub-body information group initialization determination step in step 1 includes: Step 1.1: Complete the process of identifying and separating sub-entity information groups; The unmanned platform group is defined as consisting of the leading vehicle A, the middle vehicle B, and the trailing vehicle C, with their corresponding information group sets being as follows: S A , S B and S C In actual driving, the trajectories of different vehicles are not always on the same straight line. Therefore, the straight-line distance between the first vehicle A and the middle vehicle B is... L AB The straight-line distance between the middle car B and the rear car C is L BC The straight-line distance between car A in front and car C behind is L AC ; The information group for the unmanned platform group consists of the leading vehicle A, the middle vehicle B, and the trailing vehicle C. S A , S B and S C Information is identified and separated, and the common information set of the three sub-groups is calculated. R Then we have: Based on this, calculate the local information set among different sub-entity information groups. R AB , R BC and R AC Then we have: Finally, the unique sub-entity information sets of different sub-entity information groups are calculated. R A , R B and R C Then we have: In the process of information group computation, the common information set of the information group R and local information set R AB , R BC , R AC The data is related to the distance between the three sub-entities. The closer the distance, the more information data is in the intersection; the farther the distance, the less information data is in the intersection. Therefore, data feature calculation and collaborative control of information groups are carried out based on the distance between the sub-entities. The initialization and determination step of the sub-body information group also includes: Step 1.2: Complete the sub-entity information group extraction and calculation process; Various information groups targeting unmanned platform users S A , S B and S C First, calculate the information sharing rate separately. η ABC Information discrepancy rate η A|B|C As shown below: In the formula, M ( x ) represents the computation set x The corresponding size; Based on the local information sets among different sub-body information groups R AB , R BC and R AC It is necessary to calculate the information detachment rate for each pair of information groups separately. η A|B , η B|C and η A|C Then we have: In the planning and operation of unmanned swarms, global planning is prioritized. After the global planning is completed, local planning and operation are carried out based on the characteristics of the swarm.

2. The collaborative control method for dynamic interaction of unmanned information groups as described in claim 1, characterized in that, The feature-based population global planning operation in step 2 includes: Step 2.1: Complete the calculation of global features of the population; When information sharing rate η ABC When the value is 1, it indicates that the system information within the group is completely shared, and there is no external information input, corresponding to the information difference rate. η A|B|C =0, which means that the group has no external information reference in the unknown environment, and the unmanned group cannot make decisions at this time; Information discrepancy rate and information sharing rate η A|B|C When =1, it represents that the group is in a state of complete input of external information, corresponding to η ABC =0, which indicates that the sub-entities within the group are excessively separated and cannot share information with each other at all. At this time, the individuals in the group lose information interaction and the group loses its cluster interconnection function. In an ideal state, the information sharing rate η ABC Information discrepancy rate η A|B|C A dynamic equilibrium needs to be achieved within a certain range to enable real-time input of external information and reasonable sharing of internal information, thereby providing reasonable decision-making references for system subsystems and overall decision-making; The information sharing rate η ABC Information discrepancy rate η A|B|C The calculation method for the span Φ of the system is as follows: The system span Φ fluctuates dynamically during actual operation, and its corresponding dynamic equilibrium range is [ ω 1, ω 2], the upper and lower limits of the range are respectively: 。 3. The collaborative control method for dynamic interaction of unmanned information groups as described in claim 2, characterized in that, The feature-based population global planning operation in step 2 further includes: Step 2.2: Complete the group global planning and execution process; Based on the system span Φ and distance L AB , L BC If the value is used for judgment, the following situations may occur: (1) If the system span |Φ|∈[ ω 1, ω [2] Then, in the external environment, the shared information within the system and the unknown information discovered from the outside reach a dynamic equilibrium. At this time, the group is in a normal driving process and there is no need to adjust the group state. (2) If the system span |Φ| [ ω 1, ω 2] If Φ>0, then the group is too compact in the external environment. If they continue to travel, the group will not have enough information in the unknown environment, and the information overlap between different sub-groups will be high, which is not conducive to the group making sufficient decisions in complex environments. At this time, it is necessary to increase the minimum distance between different sub-groups. L min =min( L AB , L BC To achieve optimal results for different information groups; Under this operating condition, if L AB ≠ L BC minimum distance L min Need to increase to the target distance L t Then we have: if L AB = L BC ,distance L AB and L BC Need to increase to the target distance L t Then we have the following calculation equation: (3) If the system span |Φ| [ ω 1, ω 2] And Φ<0, then the group is too relaxed in the external environment. If it continues to drive, the information of the group will be too scattered in the unknown environment. No sub-group can make a comprehensive judgment through the correlation data, so it cannot make effective decisions in the complex environment. At this point, it is necessary to reduce the maximum distance between different sub-units. L max =max( L AB , L BC To achieve optimal results for different information groups; under this operating condition, the maximum distance L max Need to reduce to the target distance L t Then we have: if L AB = L BC ,distance L AB and L BC Need to increase to the target distance L t Then we have the following calculation equation: 。 4. The collaborative control method for dynamic interaction of unmanned information groups as described in claim 3, characterized in that, The feature-based population local planning operation in step 3 includes: Step 3.1: Complete the calculation process of local features of the population; Calculate: Information Free Rate η A|B , η B|C and η A|C median value ω 3, then we have: Based on this, according to the information detachment rate η A|B , η B|C and η A|C If the following information determination conditions are set, then the following constraints apply: Condition 1: ( η A|C + η A|B )> η B|C and( η A|C + η B|C )> η A|B ; Condition two: .

5. The collaborative control method for dynamic interaction of unmanned information groups as described in claim 4, characterized in that, The feature-based group local planning operation step 3 further includes: Step 3.2: Complete the group local planning operation process; Based on the information judgment criteria in step 3.1, a comprehensive evaluation is conducted, resulting in the following situations: (1) If both conditions one and two are met under the current working conditions, then the operating architecture and spatial layout of each sub-group within the group are temporarily in a reasonable state, and no dynamic adjustment is required.

6. The collaborative control method for dynamic interaction of unmanned information groups as described in claim 5, characterized in that, In step 3.2, a comprehensive evaluation is performed based on the information judgment conditions in step 3.

1. The following situations apply: (2) If only condition 2 is satisfied but condition 1 is not satisfied under the current working condition, then within the group, the layout of the front vehicle A, the rear vehicle C and the middle vehicle B is overly coupled. If the driving continues, the information processing and transmission time of the front vehicle A and the middle vehicle B will be short, which will affect the response decision of the rear vehicle C to the real-time working condition. At this moment, the distance between car A in front and car C behind is... L AC Δ needs to be increased L To address the problem of excessive information coupling, we have: 。 7. The collaborative control method for dynamic interaction of unmanned information groups as described in claim 6, characterized in that, In step 3.2, a comprehensive evaluation is performed based on the information judgment conditions in step 3.

1. The following situations apply: (3) If only condition one is met but condition two is not met under the current working condition, then within the group, the layout of the front vehicle A, the rear vehicle C and the middle vehicle B is excessively decoupled, that is, the degree of separation of the sub-body is large. If the driving continues, the information of the front vehicle A and the middle vehicle B will be at risk of failure for the rear vehicle C. At this time, the rear vehicle C will be lagging behind in following and responding to the group, which is not conducive to the coordinated operation of the group. At this moment, the distance between car A in front and car C behind is... L AC Need to reduce Δ L To address the problem of excessive information decoupling, we have: 。 8. The collaborative control method for dynamic interaction of unmanned information groups as described in claim 7, characterized in that, In step 3.2, a comprehensive evaluation is performed based on the information judgment conditions in step 3.

1. The following situations apply: (4) If neither condition one nor condition two is satisfied under the current operating conditions, then the coupling state of the front vehicle A, the rear vehicle C, and the middle vehicle B within the group needs to be further determined; at this time, the information detachment rate needs to be judged. η A|C and ω Are 2 equal? ​​If so... η A|C ≠ ω 2. Then continue to judge. η A|C and ω 1. Are they equal? ​​If so... η A|C = ω 1. At this time, the distance between the sub-sub ... η A|C ≠ ω 1. What is the distance between car A in front and car C behind? L AC It needs to be reduced to avoid increasing system coupling; currently it is... T At what time is the distance between car A and the car following car C? L AC|T ,exist T At time +1, the distance between car A and the following car C. L AC|T+1 It needs to be adjusted to: And if η A|C = ω 2. What is the distance between car A in front and car C behind? L AC It is necessary to increase the system's coupling to reduce its coupling; T At time +1, the distance between car A in front and car C behind. L AC|T+1 It needs to be adjusted to: 。

Images