A heterogeneous unmanned cluster cooperative formation control method under communication delay

Abstract

The application discloses a kind of communication time delay under heterogeneous unmanned cluster cooperative formation control method, this method is modeled to the static formation control problem of heterogeneous unmanned cluster, give the dynamic model of heterogeneous unmanned cluster, based on the shortest path network of time delay, displacement difference and speed difference are introduced in consistency protocol, static formation control protocol is proposed, consider the static formation control problem of discrete heterogeneous unmanned cluster containing n unmanned vehicles in two-dimensional plane, aim at designing control law to control unmanned cluster to form the formation with predetermined geometric relationship, and control the speed of two-order individual converges to zero.This method is applicable to the consistency formation control protocol of heterogeneous unmanned cluster, information transmission is optimized by introducing the shortest path network of time delay, to accelerate the formation convergence speed under the premise of not increasing the complexity of physical communication link, to meet the demand of complex task.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of unmanned aerial vehicle control, in particular to a heterogeneous unmanned cluster cooperative formation control method under communication time delay. BACKGROUND

[0002] With the continuous development of information technology and intelligent manufacturing field, the functionality and performance of unmanned vehicle system (unmanned aerial vehicles, unmanned ground vehicles and satellites, etc.) have gradually become complete. However, in the face of increasingly complex and diversified application scenarios and task requirements, single intelligent unmanned vehicle is limited by its inherent hardware and software architecture, and its limitations are more and more prominent when executing large-scale or highly complex tasks. In order to break through this bottleneck, the concept of unmanned cluster cooperative formation is proposed and widely concerned. In view of the fact that the unmanned cluster in the actual application scenario is composed of heterogeneous units, such as different models and types of unmanned aerial vehicles cooperating to perform reconnaissance, combat and other tasks, and the air-ground cooperative combat mode between unmanned aerial vehicles and unmanned vehicles, in-depth study of the formation control algorithm of heterogeneous unmanned cluster has important practical significance for promoting the technological progress and practical application in related fields.

[0003] The core mechanism of the formation control strategy based on the consistency theory of multi-agent system is to build a local information interaction network between unmanned aerial vehicles. Under this strategy, the physical communication link is effectively limited between the unmanned aerial vehicle and its adjacent individuals, and by optimizing the topology of the communication network, the predetermined formation control target is achieved with lower communication cost.

[0004] The core advantage of this method is its high efficiency of communication resource utilization, but due to the physical limitations of communication bandwidth, the time consumption of sensor measurement process, the hysteresis of control instruction processing and execution and other multiple factors, communication time delay has become an important challenge that almost all actual systems need to face. Communication time delay not only may have an adverse effect on the stability of the formation shape, but also may significantly reduce the dynamic response speed of the system, thereby threatening the efficiency and success rate of task execution. Therefore, it is of great significance to effectively reduce the time delay in the field of heterogeneous unmanned cluster formation control algorithm research to ensure the stability and rapidity of the system under high delay environment. SUMMARY

[0005] In view of the above shortcomings in the prior art, the present application provides a heterogeneous unmanned cluster cooperative formation control method under communication time delay.

[0006] In order to achieve the above application purpose, the technical scheme adopted by the present application is:

[0007] A heterogeneous unmanned cluster cooperative formation control method under communication time delay, comprising the following steps:

[0008] S1, regarding the unmanned cluster as an agent system containing n agents, wherein the first m agents have second-order dynamic characteristics, and the last n-m agents have first-order dynamic characteristics;

[0009] S2, constructing a shortest path network communication topology model based on the adjacency matrix and the communication delay matrix of the agent system as weights;

[0010] S3, designing a static formation consistency protocol based on the shortest path network communication topology model;

[0011] S4, introducing the expected relative position and the expected speed into the static formation consistency protocol to obtain a formation control protocol, and controlling the agent system to form a formation with a set geometric relationship based on the designed formation control protocol.

[0012] Further, the agent system in S1 is represented as:

[0013]

[0014] x l ((k+1)T)=x l (kT)+Tu l (kT),l∈{m+1,...,n}

[0015] In the formula, x i is the displacement of the agent i in the first m agents, v i is the speed of the agent i in the first m agents, u i is the control input of the first m agents, T is the sampling period; x l is the displacement of the agent l in the last n-m agents, u l is the control input of the agent l in the last n-m agents, and k is the discrete sampling point.

[0016] Further, the adjacency matrix and the communication delay matrix in S2 are represented as:

[0017] A=[a ij ],τ=[τ ij ]

[0018] In the formula, A is the adjacency matrix, a ij is an element in the adjacency matrix; τ is the communication delay matrix, and τ ij is an element in the communication delay matrix.

[0019] Further, the static formation consistency protocol in S3 is represented as:

[0020]

[0021] In the formula, xi is the displacement of agent i in the first m agents, v i is the velocity of agent i in the first m agents, u i is the control input of the first m agents, T is the sampling period; x l is the displacement of agent l in the last n-m agents, u l is the control input of agent l in the last n-m agents, k is the discrete sampling point; a, b, g are control parameters; a ij is the element in the adjacency matrix of the first m agents, a lj is the element in the adjacency matrix of the last n-m agents, t ij is the element in the communication delay matrix.

[0022] Further, the formation control protocol in S4 is represented as:

[0023]

[0024] wherein, x i is the displacement of agent i in the first m agents, v i is the velocity of agent i in the first m agents, u j is the velocity of agent j in the first m agents, u i is the control input of the first m agents, T is the sampling period; x l is the displacement of agent l in the last n-m agents, u l is the control input of agent l in the last n-m agents, k is the discrete sampling point; a, b, g are control parameters; a ij is the element in the adjacency matrix of the first m agents, a lj is the element in the adjacency matrix of the last n-m agents, t ij is the element in the communication delay matrix, x j * is the formation of the first m agents with a set geometric relationship, x l * is the formation of the last n-m agents with a set geometric relationship.

[0025] Further, the agent system controlled based on the designed formation control protocol in S4 is represented as:

[0026] x i ((k+1)T)=x i (kT)+Tv i (kT)

[0027] v i ((k+1)T)=v i (kT)-gTv i(kT)

[0028]

[0029] In the formula, x i is the displacement of agent i in the first m agents, v i is the speed of agent i in the first m agents, u i is the control input of the first m agents, T is the sampling period; x l is the displacement of agent l in the latter n-m agents, u l is the control input of agent l in the latter n-m agents, k is a discrete sampling point; α, β, γ, η are control parameters; a ij is an element in the adjacency matrix of the first m agents, a lj is an element in the adjacency matrix of the latter n-m agents, τ ij is an element in the communication delay matrix, x j * is the formation of the first m agents with a set geometric relationship, x l * is the formation of the latter-m agents with a set geometric relationship.

[0030] The present application has the following beneficial effects:

[0031] 1. A consensus formation control protocol suitable for a heterogeneous unmanned cluster composed of first-order and second-order discrete individuals is proposed and verified, information transmission is optimized by introducing a shortest path network with delay, the formation convergence speed is accelerated without increasing the complexity of physical communication links, and the complex task requirements are met.

[0032] 2. The research object of the present application is a heterogeneous unmanned cluster composed of first-order and second-order discrete individuals, and the formation control problem under communication delay is considered.

[0033] 3. A consensus formation control protocol suitable for a heterogeneous cluster is proposed, and a shortest path network with delay is innovatively introduced to accelerate the formation convergence speed. BRIEF DESCRIPTION OF DRAWINGS

[0034] Figure 1 It is a flowchart of the cooperative formation control method of the heterogeneous unmanned cluster under communication delay of the present application.

[0035] Figure 2 It is a multi-agent communication topology graph based on a shortest path network with delay of a traditional adjacency network.

[0036] Figure 3 It is a multi-agent communication topology graph based on a traditional adjacency network.

[0037] Figure 4A multi-agent communication topology graph based on a shortest path network with time delay for an embodiment of the present application.

[0038] Figure 5a A heterogeneous cluster formation control simulation result plane trajectory graph based on a shortest path network with time delay for an embodiment of the present application.

[0039] Figure 5b An X-direction velocity convergence graph of a heterogeneous cluster formation control simulation result based on a shortest path network with time delay for an embodiment of the present application.

[0040] Figure 5c A Y-direction velocity convergence graph of a heterogeneous cluster formation control simulation result based on a shortest path network with time delay for an embodiment of the present application.

[0041] Figure 6a A heterogeneous cluster formation control simulation result plane trajectory graph based on an adjacency network for an embodiment of the present application.

[0042] Figure 6b An X-direction velocity convergence graph of a heterogeneous cluster formation control simulation result based on an adjacency network for an embodiment of the present application.

[0043] Figure 6c A Y-direction velocity convergence graph of a heterogeneous cluster formation control simulation result based on an adjacency network for an embodiment of the present application. DETAILED DESCRIPTION

[0044] The specific embodiments of the present application are described below to facilitate the understanding of the present application for those skilled in the art, but it should be clear that the present application is not limited to the scope of the specific embodiments, and for those skilled in the art, it is obvious that various changes are within the spirit and scope of the present application defined and determined by the appended claims, and all the applications utilizing the concept of the present application are within the scope of protection.

[0045] A heterogeneous unmanned cluster cooperative formation control method under communication time delay, as shown in Figure 1 , includes the following steps:

[0046] S1, regarding the unmanned cluster as an agent system containing n agents, the first m agents have second-order dynamic characteristics, and the last n-m agents have first-order dynamic characteristics;

[0047] The unmanned cluster is regarded as a multi-agent system, the first m agents have second-order dynamic characteristics, and the last n-m agents have first-order dynamic characteristics. The actual communication topology model between agents can be described by an adjacency network, and the communication time delay between agents is τ ij >0. Since the heterogeneous system is complex, the input time delay is not considered. The system dynamics equation is as follows:

[0048]

[0049] In the formula, x i Let v be the displacement of agent i among the first m agents. i Let u be the velocity of agent i among the first m agents. i x is the control input for the first m agents, and T is the sampling period; l Let u be the displacement of agent l in the last nm agents. l is the control input for agent l in the last nm agents, and k is the discrete sampling point.

[0050] S2. Construct a network communication topology model based on the shortest path with the shortest delay using the adjacency matrix and communication delay matrix of the intelligent agent system as weights;

[0051] Traditionally, adjacency networks are defined as models used to describe the actual communication relationships between agents. However, considering the prevalent communication latency between agents, and to improve the utilization of information in the communication network and enhance its security and reliability, a shortest path communication network, weighted by communication latency, is established without altering the actual physical communication network. This is achieved by increasing the amount of network information and reallocating communication resources.

[0052] The delay-shortest path network is a network of shortest paths with communication delay as the weight. Its topology model uses a directed graph. It is represented that, where V = {v1, v2, ..., v} n} is a directed graph The set of nodes represents the n intelligent agents participating in the unmanned swarm formation. edge set adjacency matrix and communication latency between intelligent agents The definition is as follows:

[0053] (1) If there is at least one slave node v in the traditional adjacency network j to node v i If there is a directed path, then agent i can receive information from j. The communication delay of each directed path is defined as the sum of the communication delays of each edge along that path. If multiple directed paths exist from agent j to agent i, the directed path with the shortest communication delay is taken as the unique actual path for information transmission between agent j and agent i (if the directed path with the shortest communication delay is not unique, any one of them can be chosen). This directed path is then used... It means that, among which, p ij Each ordered pair of nodes within a network is an edge in the adjacency network, i.e. Thus, the communication delay from agent j to agent i and define the element of the adjacency matrix of

[0054] (2) If there is no directed path from node v j to node v i in the traditional adjacency network, agent i cannot receive information from j,

[0055] According to the definition, the shortest path network communication model corresponding to the traditional adjacency network can be obtained As shown in Figure 2 , the node set of is the same as that of the adjacency network, and the edge set of is increased with the edge (v1, v3) relative to the adjacency network, as shown by the dashed line in the figure. The adjacency matrix and the communication delay matrix of are respectively:

[0056]

[0057] The shortest path network here is not the shortest path network in the traditional sense calculated with the weight of the edge as the weight value, but the shortest path network calculated and generated with the communication delay corresponding to each edge as the weight value. The delay shortest path network only takes the communication delay as the weight value and does not contain the input delay, because the communication delay between agents has the characteristics of transmission and accumulation in the communication link, while the input delay of each agent is independent of each other.

[0058] The connectivity of the delay shortest path network is stronger than that of the adjacency network, but it does not delete or add any actual physical communication link, and the delay shortest path network only increases the communication volume and optimizes the communication path on the basis of the existing physical communication network.

[0059] S3, designing a static formation consensus protocol based on the delay shortest path network communication topology model;

[0060] Based on formula (1), the consensus protocol is:

[0061]

[0062] Wherein, α>0, β>0, γ>0, η>0 all represent control parameters, a ij is the element in the adjacency matrix A of the adjacency network.

[0063] Then the closed-loop heterogeneous system of (1) is:

[0064] ​

[0065] The consensus conditions agreed by the heterogeneous system (3) can be given by the following lemma:

[0066] Lemma 1 Consider the heterogeneous system (3) consisting of n agents, if:

[0067] (i) G contains a directed spanning tree,

[0068] (ii)

[0069] (iii)

[0070] (iv)

[0071] (v)

[0072] are all satisfied, then the system (3) can asymptotically reach consensus, i.e., there is

[0073] S4, the desired relative position and the desired speed are introduced into the static formation consensus protocol to obtain a formation control protocol, and the agent system is controlled to form a formation with a set geometric relationship based on the designed formation control protocol.

[0074] The following consensus protocol based on equation (2) considers the static formation control problem of the heterogeneous system of equation (1). The desired relative position and the desired speed are introduced into the protocol (2) to obtain a formation control protocol:

[0075]

[0076] where x i * represents a formation with a predetermined geometric relationship. Based on (4), we can obtain:

[0077]

[0078] where, is the adjacency matrix of the shortest path delay network model is the element of the adjacency matrix , and represents the minimum communication delay from node j to node i. Substituting (5) into (1) can obtain the mathematical model of the heterogeneous unmanned cluster system:

[0079]

[0080] where T is the sampling period, v *(kT) represents the expected velocity of the entire formation, which can vary with the discrete time series, and

[0081] In order to satisfy the requirement that the system (6) can asymptotically form a formation with a predetermined geometric relationship under the action of the consensus protocol (5), we now study its sufficient conditions and give Theorem 1.

[0082] Theorem 1 Consider a heterogeneous system consisting of n agents (6), if:

[0083] (i) The communication topology G contains a directed spanning tree.

[0084] (ii)

[0085] (iii)

[0086] (iv)

[0087] (v)

[0088] If all conditions are met, then system (6) can asymptotically form a formation with a predetermined geometric relationship, i.e. have

[0089] Proof: For (6), let have:

[0090]

[0091] v i ((k+1)T)=v i (kT)-γTv i (kT)

[0092]

[0093] Easy to know The edge set of a graph contains the edge set of its adjacency network. Therefore, the graph can be obtained from the fact that the adjacency network contains a directed spanning tree. It also contains directed spanning trees. Furthermore, by Lemma 1, we can obtain... have Right now Q.E.D.

[0094] Simulation Experiment

[0095] Consider a heterogeneous cluster system consisting of six UAVs, the dynamics model can be described by equation (1), where agents 1, 2, 3 are second-order agents, and agents 4, 5, 6 are first-order agents. The communication topology model graph based on the adjacency network description between agents is shown in Figure 3 .

[0096] Obviously, G contains a directed spanning tree. To simplify the calculation, the weight of the edge of graph G is taken as 1, the sampling period is taken as T = 0.1 s, and the adjacency matrix A of G and the communication delay between agents are assigned as follows:

[0097]

[0098] After introducing the delay shortest path network, the communication topology of the system can be updated to graph as shown in Figure 4 , at this time, the corresponding adjacency matrix and the communication delay matrix are:

[0099]

[0100] The control parameters are taken as α = β = η = 1, γ = 5, which are verified to satisfy the conditions of theorem 4.1. The desired formation shape is set to be a hexagon: The initial positions of the agents are respectively: x1 = [1, 6] T , x2 = [2, 5] T , x3 = [3, 4] T , x4 = [4, 3] T , x5 = [5, 2] T , x6 = [6, 1] T ; The initial velocities of the second-order agents are respectively: v1 = [0.7, 0.8] T , v2 = [0.5, 0.6] T , v3 = [0.3, 0.4] T .

[0101] To verify the feasibility of the formation control protocol, the effectiveness of the sufficient condition given by theorem 1 is verified by simulation, and the simulation results are shown in Figure 5a , Figure 5b , Figure 5c . From the results, it can be seen that the formation of the first-order and second-order agents converges to the pre-set hexagonal formation, as shown by the wireframe in Figure 5a . The velocity of the second-order agent converges to zero, and the feasibility of the control protocol is verified.

[0102] To verify the improvement of the convergence speed of the heterogeneous cluster formation based on the shortest delay path network, the system (1) under the action of the heterogeneous cluster formation control protocol (4) based on the traditional adjacency network is simulated under the condition that other conditions are unchanged, and the results are as shown in Figure 6a 、 Figure 6b 、 Figure 6c From the comparison Figure 5a 、 Figure 5b 、 Figure 5c With Figure 6a 、 Figure 6b 、 Figure 6c It can be found from the results that although the systems under the two networks can form the expected formation, the formation control algorithm based on the shortest delay path network has a significantly faster formation convergence speed, so the shortest delay communication network and the corresponding control protocol proposed in the application can improve the utilization rate of network information and speed up the speed of the system to achieve formation.

[0103] The application is described with reference to flowcharts and / or block diagrams of the method, device (system) and computer program product according to the embodiments of the application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device that implements the functions specified in the flow Figure 1 One or more flows and / or blocks Figure 1 One or more blocks or blocks.

[0104] These computer program instructions can also be stored in a computer readable memory that can guide the computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer readable memory produce a product including instruction devices, which implement the functions specified in the flow Figure 1 One or more flows and / or blocks Figure 1 One or more blocks or blocks.

[0105] These computer program instructions can also be loaded into a computer or other programmable data processing device, so that a series of operation steps are performed on the computer or other programmable device to produce a computer implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in the flow Figure 1 One or more flows and / or blocks Figure 1 One or more blocks or blocks.

[0106] The principles and implementation manners of the present application are described by using specific examples in the present application, and the above examples are only used for helping to understand the method of the present application and its core idea; meanwhile, for the ordinary skilled in the art, according to the idea of the present application, the specific implementation manners and application ranges can be changed, and the above description should not be understood as the limitation of the present application.

[0107] Those skilled in the art will understand that the examples described herein are for the purpose of understanding the principles of the present application and should be understood as not limiting the scope of protection of the present application. Those skilled in the art can make various other specific modifications and combinations according to the technical inspiration disclosed in the present application without departing from the essence of the present application, and these modifications and combinations are still within the scope of protection of the present application.

Claims

1. A method for cooperative formation control of heterogeneous unmanned swarm under communication time-varying delay, characterized in that, The method comprises the following steps: S1. regarding the unmanned swarm as an agent system comprising a first agent having a second-order dynamic characteristic and a second agent having a first-order dynamic characteristic, wherein the agent system is represented as: ​ wherein is the position of the agent in the front m is the displacement of the agent in the front m is the velocity of the agent in the front is the control input of the agent in the front m is the sampling period; is the position of the agent in the back is the displacement of the agent in the back is the control input of the agent in the back is the discrete sampling point;​​​​​​​ S2, constructing a shortest path network communication topology model based on an adjacency matrix and a communication delay matrix of the agent system; S3, designing a static formation consistency protocol based on the shortest path network communication topology model, wherein the adjacency matrix and the communication delay matrix of the shortest path network are represented as: ， wherein is an adjacency matrix of a shortest path network, is an element in the adjacency matrix; is a communication delay matrix under the condition, is an element in the communication delay matrix; S4, introducing an expected relative position and an expected speed into the static formation consistency protocol to obtain a formation control protocol, and controlling the agent system to form a formation with a set geometric relationship based on the designed formation control protocol, wherein the formation control protocol is represented as: In the formula, For the front m Among the intelligent agents displacement, for Among the intelligent agents displacement, For the front m Among the intelligent agents speed, for Among the intelligent agents speed, For the front m Control input for an intelligent agent The sampling period; For the Empress Among the intelligent agents displacement, For the Empress Among the intelligent agents The control input, These are discrete sampling points; For control parameters; These are the elements in the adjacency matrix of the shortest path network with the first m agents. These are elements in the communication delay matrix of the first m agents. For the Empress Elements in the adjacency matrix of the shortest path network for each agent. For the Empress Elements in the communication delay matrix of the shortest path network for each agent For the front The formation of the agents based on their set geometric relationships. for The formation of the agents based on their set geometric relationships. For the Empress The formation of the agents based on the set geometric relationships.

2. The heterogeneous unmanned swarm cooperative formation control method under communication time-varying delay according to claim 1, characterized in that, The adjacency matrix and the communication delay matrix in the S2 are represented as: ， wherein is an adjacency matrix, is an element in the adjacency matrix; is a communication latency matrix, is an element in the communication latency matrix. 3.The method of claim 1, wherein, The static formation consistency protocol in the S3 is represented as: In the formula, For the front m Among the intelligent agents displacement, For the front m Among the intelligent agents speed, For the front m Control input for an intelligent agent The sampling period; For the future Among the intelligent agents displacement, For the future Among the intelligent agents The control input, These are discrete sampling points; For control parameters; For the front Elements in the adjacency matrix of each agent For the future Elements in the adjacency matrix of each agent All of these are elements in the communication delay matrix, and .

Images