A collaborative positioning method for low-cost three-leader unmanned aerial vehicle formation based on fully connected networks

Abstract

Aiming at the low-cost unmanned aerial vehicle formation collaborative positioning technology, which is not yet fully studied, the present invention proposes a heterogeneous formation collaborative positioning technology based on inertial group measurement information based on data link ranging and network communication. The method first constructs the data link communication rules of the three-lead low-cost heterogeneous formation, and on this basis, a nonlinear equation group is constructed through a fully connected communication structure and the relative distance between the unmanned aerial vehicles measured by the data link. Then, the search range is determined according to the inertial group measurement information, and the particle swarm algorithm is combined to realize collaborative positioning within the tolerance of positioning error to realize online correction of the inertial group measurement position error. Through the centralized communication structure and undirected fully connected data link ranging, the present invention can solve the formation member position information of the high-performance unmanned aerial vehicles in the formation through the optimization method and transmit and update it through the network. The method can be used in conjunction with a variety of formation control or collaborative guidance methods.

Description

Technical Field

[0001] The invention relates to a three-leader low-cost unmanned aerial vehicle formation collaborative positioning method based on a fully connected network. Background Art

[0002] The navigation system is the lifeline of low-cost unmanned aerial vehicles, and the homing guidance system is an important reliance for unmanned aerial vehicles to achieve terminal guidance penetration and hit. The seeker, as an important component of the homing guidance system, is even more important. Generally speaking, the cost of the seeker accounts for as much as 30% to 40% of the cost of the unmanned aerial vehicle. Therefore, many studies have explored solutions to reduce costs based on the navigation and guidance system. At present, the guidance of low-cost unmanned aerial vehicles without seekers mainly adopts external guidance solutions such as ground station assistance. However, the ground station is expensive and easy to be exposed. Once damaged, the cost is extremely high. In addition, the front-end battlefield environment is complex, and ground station communications and GPS navigation are the main interference targets of the enemy.

[0003] On the one hand, the complex system confrontation in the modern battlefield has put forward higher performance requirements for the guidance system. While integrating, optimizing and upgrading the technology, the value of the navigation and guidance system has also risen steadily. On the other hand, the offensive and defensive game has grown and declined, and the denied or degraded combat environment will greatly affect the accuracy of the navigation and guidance system, and the seeker is also very easy to lose the target due to interference. Therefore, the single-soldier combat of high-performance unmanned aerial vehicles is restricted by their own positioning, and the threat level has dropped sharply. Therefore, studying low-cost unmanned aerial vehicle positioning can reduce the cost of the navigation and guidance system, which has great engineering value.

[0004] With the continuous development of network communication technology, compared with the strong threat of high-performance unmanned aerial vehicles, unmanned aerial vehicle formations can give full play to the scale advantage through data link communication. Furthermore, combined with low-cost unmanned aerial vehicles, the cost advantage can be fully improved with the help of unmanned aerial vehicle formations. However, the problem of self-positioning of low-cost unmanned aerial vehicle formations also arises. Although traditional inertial navigation systems have the advantages of low cost and strong autonomy, the drift errors generated over time can easily cause off-target. Therefore, how to combine inertial navigation information and data link communication to achieve collaborative positioning needs to be solved urgently.

[0005] At present, there are few reports on the theoretical research results of the coordinated positioning of heterogeneous unmanned aerial vehicle formations themselves, and how to combine data link ranging to achieve cooperative positioning of non-seeker formations urgently needs a theoretical breakthrough. Therefore, the present invention proposes a three-leader low-cost unmanned aerial vehicle formation coordinated positioning method based on a fully connected network for high- and low-configuration heterogeneous unmanned aerial vehicle formations. Summary of the invention

[0006] Aiming at the low-cost UAV formation collaborative positioning technology, which is not well studied at present, this paper proposes a heterogeneous formation collaborative positioning technology based on inertial group measurement information based on data link ranging and network communication. This method first constructs the data link communication rules of the three-leader low-cost heterogeneous formation, and then constructs a nonlinear equation group based on the full-connected communication structure and the relative distance between the UAVs measured by the data link. Then, the search range is determined according to the inertial group measurement information, and the particle swarm algorithm is combined to realize collaborative positioning within the tolerance of positioning error to realize online correction of the inertial group measurement position error.

[0007] The present invention is described in detail according to the following design steps.

[0008] Step 1: Acquisition of information on heterogeneous formations of three-leader heterogeneous UAVs

[0009] The high-performance leader unmanned aerial vehicle is denoted by m k , k = 1, 2, 3, the corresponding coordinate position is known, denoted as (x k， y k , z k ).

[0010] The low performance unmanned aerial vehicle is denoted as f i , i=4, 5, ..., n (n≥6) corresponding coordinate positions are unknown, denoted as (x fi ,y fi , z fi ).

[0011] The maximum drift error of the low-cost inertial component is recorded as m e , the low-cost unmanned aerial vehicle measurement position information is known, denoted as

[0012] Step 2: Construct a three-leader heterogeneous UAV formation communication network

[0013] First, select a leader drone As a central node, it collects existing information through network communication to perform collaborative positioning and solve and send the solution results.

[0014] Secondly, the leader-follower relationship between the three leader UAVs and the low-cost UAVs is constructed based on the fully connected communication network. There are three low-performance UAVs among the low-performance UAVs. 1 , f 2 , f 3 The three low-performance UAVs are connected to each other and communicate with each other. 1 , f 2 , f 3 The connection is fully formed, as shown in the schematic diagram Figure 1As shown in Figure 1, each unmanned aerial vehicle in the fully connected infrastructure can achieve data link ranging. The data link communication ranging information is known, and the low-cost unmanned aerial vehicle f in the heterogeneous formation i and high-performance leader unmanned aerial vehicle m k The distance between them is denoted as R ik Where i≠k, the distance between low-cost drones is R ij =R ji , that is,

[0015]

[0016] Furthermore, by constructing an additional structure, when the number of low-cost drones in the formation n>6, the low-performance drone f i (i>6) Need to communicate with the drone selected as the central node There is two-way communication between them, and at the same time there is two-way communication between them and any three different drones in the formation, as shown in the schematic diagram Figure 2 As shown. With low-cost unmanned aerial vehicles in heterogeneous formations i (i>6) The collection of drone numbers for communication is denoted as C i ={c 1 , c i2 , c i3 , c i4}, i>6, where c i2 , c i3 , c i4 ∈{1,...,n}-{c 1 The data link communication ranging information is known and can be expressed as R ij , i>6, j=c 1 , c i2 , c i3 , c i4 .

[0017] Step 3: Establish the cooperative positioning equations for the three-leader heterogeneous UAV formation

[0018] Combining the data link ranging information and the relative position equation, let x = [x fi ,y fi , z fi ], i = 4, 5, ..., n (n ≥ 6), the relative position equations of leader-slave collaborative positioning are established as follows

[0019]

[0020] Where x is the solution set of collaborative positioning. The first 12 equations are the basic structural equations composed of three high-performance aircraft and three low-performance aircraft based on a fully connected network. The subsequent equations are given according to the actual communication conditions of the low-performance aircraft.

[0021] Remember q is the qth (q=1, 2, ..., 4(n-3)) row vector norm of G(x), that is,

[0022]

[0023] Thus, the optimized cost function Defined as

[0024]

[0025] Step 4: Collaborative positioning solution based on inertial group measurement information

[0026] The central node of the centralized structure is used to solve the position information of the unmanned aerial vehicle, and the particle swarm optimization method is selected to perform a small-range search based on the inertial group measurement position information to realize the solution of the nonlinear equation group of collaborative positioning of the unmanned aerial vehicle formation. The leader unmanned aerial vehicle solves the collaborative positioning nonlinear equation group to obtain the location of the low-cost unmanned aerial vehicle and transmits it to each unmanned aerial vehicle through the communication network, thereby realizing low-cost collaborative positioning.

[0027] In view of the nonlinear optimization problem described above, the present invention intends to solve it by using a particle swarm optimization algorithm suitable for such problems.

[0028] First, initialize the particle swarm.

[0029] Measure position information based on the inertial group and the maximum drift error of the inertial group m e Determine the scope of your search.

[0030] Set the initialization parameters population size S, number of iterations p = 1, 2, ..., Maxiter, particle speed v, maximum particle speed V max , learning factor c 1 , c 2 , inertia weight w, weight upper and lower bounds w max , w min And stop precision e R , e r .

[0031] Secondly, calculate the fitness of each particle

[0032] The best individual particle history is recorded as pbest, which is the best position that the particle has ever experienced. For each particle, its fitness value is compared with the fitness value of the best historical position it has experienced, and the current fitness value is changed to the best historical fitness value.

[0033] The best position experienced by the population is recorded as gbest, that is, the global historical best. For each particle, its historical best fitness value is compared with the fitness value of the best historical position experienced in the entire population or in the neighborhood. If it is better, the historical best fitness value is used as the current global extreme value.

[0034] Then, update the particle position and velocity and generate a random number rand 1 and rand 2

[0035]

[0036] Then determine whether the stopping condition is met. If the maximum number of iterations is reached or the following formula is satisfied,

[0037]

[0038] Otherwise, the iteration process is repeated.

[0039] The present invention utilizes data link ranging to propose a three-leader low-cost unmanned aerial vehicle formation collaborative positioning method based on a fully connected network structure. Through a centralized communication structure and undirected fully connected data link ranging, the high-performance unmanned aerial vehicles in the formation can calculate the position information of the formation members through an optimization method and transmit and update it through the network.

[0040] The collaborative positioning method designed in the present invention can be used in conjunction with a variety of formation control or collaborative guidance methods. The solution speed and accuracy of the nonlinear collaborative positioning equation can be improved by selecting a suitable algorithm. Similar methods have not been reported yet and are original theoretical innovations. Additional advantages can be obtained through the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] To facilitate understanding and description, a schematic diagram of the framework construction is given and explained in conjunction with the implementation method.

[0042] Figure 1 It is the basic structure for leader-slave data link communication of three-leader heterogeneous unmanned aerial vehicle formation.

[0043] Figure 2 It is a schematic diagram of data link communication of a three-leader heterogeneous unmanned aerial vehicle formation.

[0044] Figure 3 It is the real-time position status of a heterogeneous formation of three leaders - 4 low-cost unmanned aerial vehicles.

[0045] Figure 4 It is the position projection of the collaborative positioning solution result on the XY plane.

[0046] Figure 5 It is the position projection of the collaborative positioning solution result on the XZ plane.

[0047] Figure 6 It is the fitness curve of collaborative positioning of three-leader heterogeneous UAV formation. DETAILED DESCRIPTION

[0048] The present invention proposes a three-leader low-cost unmanned aerial vehicle formation collaborative positioning method based on a fully connected network. First, a formation communication network is constructed based on the formation state information of the three-leader heterogeneous formation, and then a collaborative positioning nonlinear equation group is constructed and solved in combination with data link ranging to obtain the position of the unmanned aerial vehicle for subsequent formation collaborative guidance. The embodiments of the present invention are described in detail below. The embodiments are exemplary and intended to be used to explain the present invention, but cannot be understood as limiting the present invention.

[0049] This example realizes the collaborative positioning of a 3-leader-4 low-cost UAV heterogeneous UAV formation by using the collaborative positioning method designed according to the method of the present invention.

[0050] The precise position information of the low-cost unmanned aerial vehicle to be estimated in this example is given below:

[0051] Low-cost unmanned aerial vehicle 4 Current real position (1200m, 1200m, 221m);

[0052] Low-cost unmanned aerial vehicle 5 Current real position (2100m, 1200m, 1012m);

[0053] Low-cost unmanned aerial vehicle 6 Current real position (1600m, 2100m, 1123m);

[0054] Low-cost unmanned aerial vehicle 7 Current real location (2000m, 1500m, 1828m).

[0055] Combining the leader UAV information, low-cost UAV inertial navigation measurement information and data link ranging aircraft spacing, the 3-leader-4 low-cost UAV heterogeneous UAV formation collaborative positioning method given by the present invention is verified as follows:

[0056] Step 1: Acquisition of information on heterogeneous formations of three-leader heterogeneous UAVs

[0057] The high-tech leader is equipped with a high-precision navigation system, and its own position can be accurately known as follows:

[0058] Team Leader 1 Current location (2000m, 4000m, 3500m);

[0059] Team Leader2 Current location (3000m, 4500m, 3000m);

[0060] Team Leader 3 Current location (4000m,3400m,4000m).

[0061] Low-performance, low-cost unmanned aerial vehicles are equipped with low-precision inertial navigation systems, and the measured position has an inertial group drift error m e <200m:

[0062] Low-cost unmanned aerial vehicle f4 current measurement position (1219.7m, 1235.6m, 2238.3m)

[0063] Low-cost unmanned aerial vehicle 5 Current measurement location (2085.7m, 1234.7m, 1018.4m)

[0064] Low-cost unmanned aerial vehicle 6 Current measurement location (1632.4m, 2168.9m, 1195.0m)

[0065] Low-cost unmanned aerial vehicle 7 Current measurement location (2000.0m, 1516.3m, 1829.2m)

[0066] Step 2: Construct the network topology structure of the three-leader UAV formation

[0067] Select UAV No. 1 as the central node to collect existing information through network communication, perform collaborative positioning solution and send the solution results.

[0068] Build a leader-follower relationship, and all low-cost UAVs are connected to the three leaders. On this basis, a fully connected network of low-cost UAVs is built. The relative position relationship between the three high-performance UAVs and the three low-cost UAVs is obtained through data link ranging as follows

[0069] Low-cost unmanned aerial vehicle and leader m 1 Data link ranging information R 41 =3179.91m,R 51 =3746.79m,R 61 =3069.24m.

[0070] Low-cost unmanned aerial vehicle and leader m 2 Data link ranging information R 42 =3838.73m,R 52 =3956.64m,R 62 =3353.15m.

[0071] Low-cost unmanned aerial vehicle and leader m 3 Data link ranging information R 43 =3980.82m,R S3 =4168.36m,R 63 =3965.78m.

[0072] Data link ranging information between fully connected low-cost unmanned aerial vehicles 54 =1506.79m,R 64 =1475.34m,R 65 =1036.26m.

[0073] On this basis, it is determined that low-cost UAV No. 7 needs to communicate with UAV No. 1 in a two-way manner. At the same time, UAV No. 7 communicates with UAV Nos. 2, 3, and 5, that is, there is a data link ranging R 71 =3007.56m,R 72 =3372.54m,R 73 =3510.78m,R 75 =874.18m.

[0074] Step 3: Establish the cooperative positioning equations for the three-leader heterogeneous UAV formation

[0075] x=[x f4 ,y f4 , z f4 , x f5 ,y f5 , z f5 , x f6 ,y f6 , z f6 , x f7 ,y f7 , z f7 ] The collaborative positioning equations are established as follows

[0076]

[0077] And according to the co-location equations, And optimize the cost function expression.

[0078] Step 4: Collaborative positioning solution based on inertial group measurement information

[0079] First, initialize the parameters.

[0080] Set the parameter population S = 200, the maximum particle speed V max =150, the upper limit of the number of iterations Maxiter = 500, the learning factor c 1 =c 2 =1.8, weight upper and lower bounds w max=0.8, w min =0.3.

[0081] Next, the fitness of each particle is calculated.

[0082] Subsequently, the particle positions and velocities are updated.

[0083] Then determine whether the stopping condition is met. If the maximum number of iterations is reached or the minimum limit e is met R =5,e r =1, then it ends, otherwise repeat the iteration process.

[0084] The number of iterations is 156. The co-location results are as follows:

[0085] Low-cost unmanned aerial vehicle 4 The collaborative positioning solved the position (1198.53m, 1200.78m, 2221.76m), the measurement error was 43.99m, and the position error was 1.832m.

[0086] Low-cost unmanned aerial vehicle 5 The collaborative positioning solved the position (2099.40m, 1198.56m, 1013.94m), the measurement error was 38.95m, and the position error was 2.486m.

[0087] Low-cost unmanned aerial vehicle 6 The collaborative positioning solved the position (1599.89m, 2099.91m, 1123.08m), with a measurement error of 104.84m and a position error of 0.163m.

[0088] Low-cost unmanned aerial vehicle 7 The collaborative positioning solved the position (2000.20m, 1499.61m, 1828.62m), the measurement error was 16.69m, and the position error was 0.761m.

[0089] According to the centralized communication structure, the selected communication link transmits the settlement result to each unmanned aerial vehicle in the unmanned aerial vehicle formation to update the position information.

[0090] Figure 3 It is a real-time position map of the heterogeneous formation of three leading-4 low-cost unmanned aerial vehicles. Figure 4 The position projection of the collaborative positioning solution result on the XY plane. Figure 5 It is the position projection of the collaborative positioning solution result on the XZ plane. Figure 6 It is the fitness curve of collaborative positioning of three-leader heterogeneous UAV formation. Figure 4 and Figure 5In the method, the search circle is drawn with the IMU measurement position as the center and the maximum value of the IMU error drift error as the radius, and the low-cost UAV position is searched in each circle. The search is performed based on the IMU measurement position information and the IMU error range combined with the particle swarm algorithm. When the error meets the set limit, the search is stopped to obtain the collaborative positioning solution position information. The result meets the stop condition, the positioning accuracy is less than 3m, and the collaborative positioning effect is good. The position information provided by this method can realize the collaborative positioning of low-performance UAV formations based on the inertial navigation system through network communication, providing a good information basis for formation collaborative operations.

[0091] This implementation case is an explanation and illustration of the technical solution of the present invention, and is only used as an example to verify the effectiveness of the collaborative positioning method of a three-leader unmanned aerial vehicle formation, and cannot be used to limit the scope of protection of the rights. For those skilled in the art, on the basis of the collaborative positioning method of a three-leader high- and low-configuration unmanned aerial vehicle formation of this application, several changes, modifications, substitutions, variations and improvements can be made according to the needs of engineering practice. Any changes made only partially according to the claims and description of the present invention should still fall within the scope of protection of the present invention.

Claims

1. A three-leader low-cost unmanned aerial vehicle formation collaborative positioning method based on a fully connected network, characterized in that: The steps include: Step 1: Obtaining information about the heterogeneous formation of three-leader heterogeneous UAVs: The high-performance leader unmanned aerial vehicle is denoted by m k , k = 1, 2, 3, the corresponding coordinate position is known, denoted as (x k ,y k , z k ); The low performance unmanned aerial vehicle is denoted as f i , i=4,5,...,n,n≥6, the corresponding coordinate position is unknown, recorded as (x fi ,y fi , z fi ); The maximum drift error of the low-cost inertial component is recorded as m e , the low-cost unmanned aerial vehicle measurement position information is known, denoted as Step 2: Construct a three-leader heterogeneous UAV formation communication network: First, select a leader drone As a central node, it collects existing information through network communication to perform collaborative positioning and send the solution results, c1∈{k|k=1,2,3}; Secondly, the leader-follower relationship between the three leader UAVs and the low-cost UAVs is constructed based on the fully connected communication network. There are three low-performance UAVs f1, f2, and f3 in the low-performance UAVs, which form communication relationships with the three leader UAVs respectively, and the three low-performance UAVs f1, f2, and f3 are fully connected, that is, data link ranging can be achieved between each UAV in the fully connected infrastructure; the data link communication ranging information is known, and the low-cost UAVs f1, f2, and f3 in the heterogeneous formation are connected. i and high-performance leader unmanned aerial vehicle m k The distance between them is denoted as R ik , where i≠k, the distance between low-cost drones is R ij =R ji , that is, Furthermore, by constructing an additional structure, when the number of low-cost drones in the formation n>6, the low-performance drone f i Need to communicate with the drone selected as the central node There is two-way communication between them, and there is two-way communication between them and any three different drones in the formation at the same time; the collection of drone numbers that communicate with the low-cost unmanned aerial vehicle fi in the heterogeneous formation is recorded as C i ={c1,c i2 , c i3 , c i4 }, i>6, where c i2 , c i3 , c i4 ∈{1,...,n}-{c1} data link communication ranging information is known and can be expressed as R ij , i>6, j=c1, c i2 , c i3 , c i4 ; Step 3: Establish the collaborative positioning equations of the three-leader heterogeneous UAV formation: Combining the data link ranging information and the relative position equation, let x = [x fi ,y fi , z fi ], i = 4, 5, ..., n, n ≥ 6, the relative position equations of leader-slave collaborative positioning are established as follows Where x is the solution set of collaborative positioning. The first 12 equations are the basic structural equations composed of 3 high-performance aircraft and 3 low-performance aircraft based on a fully connected network. The subsequent equations are given according to the actual communication conditions of the low-performance aircraft. Remember q is the qth row vector norm of G(x), q = 1, 2, ..., 4(n-3), that is Thus, the optimized cost function Defined as Step 4: Cooperative positioning solution based on inertial group measurement information, using the central node of the centralized structure to solve the position information of the unmanned aerial vehicle, and selecting the particle swarm optimization method to perform range search based on the inertial group measurement position information to achieve the solution of the nonlinear equation group of cooperative positioning of the unmanned aerial vehicle formation. The range search is based on the inertial group measurement position information. and the maximum drift error of the inertial group m e Determine the search range; the leader UAV solves the collaborative positioning nonlinear equations to obtain the location of the low-cost UAV and transmits it to each UAV through the communication network, thereby achieving low-cost collaborative positioning.

2. A method for collaborative positioning of three-leader low-cost unmanned aerial vehicle formations based on a fully connected network as claimed in claim 1, characterized in that: The collaborative positioning solution based on the inertial group measurement information in step 4 is solved by using the particle swarm optimization algorithm. The specific process is as follows: First, initialize the particle swarm; Measure position information based on the inertial group and the maximum drift error of the inertial group m e Determine the scope of the search; Set the initialization parameters population size S, number of iterations p = 1, 2, ..., Maxiter, particle speed v, maximum particle speed V max , learning factors c1, c2, inertia weight w, weight upper and lower bounds w max , w min And stop precision e R , e r ; Secondly, calculate the fitness of each particle; The best individual particle history is recorded as pbest, which is the best position the particle has ever experienced. For each particle, its fitness value is compared with the fitness value of the best historical position it has experienced, and the current fitness value is changed to the best historical fitness value. The best position experienced by the population is recorded as gbest, that is, the global historical best. For each particle, its historical best fitness value is compared with the fitness value of the best historical position experienced in the entire population or in the neighborhood. If it is better, the historical best fitness value is taken as the current global extreme value. Then, update the particle position and velocity, and generate random numbers rand1 and rand2 Then determine whether the stopping condition is met. If the maximum number of iterations is reached or the following formula is satisfied, Otherwise, the iteration process is repeated.

Images