阅读说明：本技术 一种面向无人机编队通信的轨迹生成与跟踪方法及系统 (Trajectory generation and tracking method and system for unmanned aerial vehicle formation communication ) 是由 杨志华 任安宁 齐晓晗 于 2020-06-12 设计创作，主要内容包括：本发明提供了一种面向无人机编队通信的轨迹生成与跟踪方法,包括以下步骤：S1、构建无人机通信系统模型,采用中继无人机重新建立起源无人机和目的无人机之间的通信；S2、预测源无人机和目的无人机的轨迹,并实时传送给中继无人机；S3、根据源无人机和目的无人机的预测轨迹,实时规划中继无人机的轨迹。本发明还提供了一种面向无人机编队通信的轨迹生成与跟踪系统。本发明的有益效果是：为在恶劣环境下,低空环境存在阴影和衰落,两通信无人机链路连接断开的情况提供了一种重新建立稳定连接的方法,打破传统通信的源节点和目的节点固定不动的假设。(The invention provides a trajectory generation and tracking method for formation communication of unmanned aerial vehicles, which comprises the following steps: s1, constructing an unmanned aerial vehicle communication system model, and reestablishing communication between the origin unmanned aerial vehicle and the destination unmanned aerial vehicle by adopting the relay unmanned aerial vehicle; s2, predicting the tracks of the source unmanned aerial vehicle and the target unmanned aerial vehicle, and transmitting the tracks to the relay unmanned aerial vehicle in real time; and S3, planning the track of the relay unmanned aerial vehicle in real time according to the predicted tracks of the source unmanned aerial vehicle and the target unmanned aerial vehicle. The invention also provides a track generation and tracking system for unmanned aerial vehicle formation communication. The invention has the beneficial effects that: the method for reestablishing the stable connection is provided for the conditions that shadow and fading exist in the low-altitude environment and the link connection of the two communication unmanned aerial vehicles is disconnected in the severe environment, and the assumption that the source node and the destination node of the traditional communication are fixed is broken through.)

1. A trajectory generation and tracking method for formation communication of unmanned aerial vehicles is characterized by comprising the following steps:

s1, constructing an unmanned aerial vehicle communication system model, dividing a communicated inorganic cluster into a source unmanned aerial vehicle and a target unmanned aerial vehicle, and reestablishing communication between the source unmanned aerial vehicle and the target unmanned aerial vehicle by using a relay unmanned aerial vehicle when the source unmanned aerial vehicle and the target unmanned aerial vehicle cannot communicate with each other, namely no direct communication link exists;

s2, predicting the tracks of the source unmanned aerial vehicle and the target unmanned aerial vehicle, and transmitting the tracks to the relay unmanned aerial vehicle in real time;

and S3, planning the track of the relay unmanned aerial vehicle in real time according to the predicted tracks of the source unmanned aerial vehicle and the target unmanned aerial vehicle.

2. The trajectory generation and tracking method for formation communication of unmanned aerial vehicles according to claim 1, wherein: in step S1, the relay drone reestablishes communication between the originating drone and the destination drone using the DF forwarding scheme.

3. The trajectory generation and tracking method for formation communication of unmanned aerial vehicles according to claim 2, wherein: in step S2, the trajectories of the source drone and the target drone are predicted using a trajectory prediction algorithm based on the extended kalman filter, and the position at the next time is predicted at the current time.

4. The trajectory generation and tracking method for formation communication of unmanned aerial vehicles according to claim 3, wherein: the trajectory prediction algorithm based on the extended Kalman filtering comprises the following steps:

inputting: an initial position value, an initial reported value;

and (3) outputting: predicted trajectory values, deviation values;

1): preprocessing a track;

2): initializing parameters;

3): acquiring the current state and the position of the current moment;

4): circulating for given times;

5): recording the value predicted each time by the extended Kalman filtering algorithm;

6) calculating the deviation between the predicted value and the actual value;

7) ending the cycle;

8) outputting the predicted value and the deviation.

5. The trajectory generation and tracking method for formation communication of unmanned aerial vehicles according to claim 4, wherein: step S3 includes the following substeps:

s31, solving the optimal problem by a PSO algorithm based on a penalty function;

s32, generating a smooth track of the relay unmanned aerial vehicle;

and S33, controlling the relay unmanned aerial vehicle to fly out of the optimal track.

6. The trajectory generation and tracking method for formation communication of unmanned aerial vehicles according to claim 5, wherein: in step S1, assuming that the relay drone only acts within a set time, the total time of the whole motion process is T, and the total time is divided into N time slots, and the coordinates of the source drone node a, the relay drone node R, and the destination drone node B of the nth time slot are (ps) respectively _x[n],ps_y[n],ps_z[n])，(pr_x[n],pr_y[n],pr_z[n])，(pd_x[n],pd_y[n],pd_z[n]) N is 1,2, the distance between the source drone node a and the relay drone node R is set to d_sr[n]The distance between the relay unmanned aerial vehicle node R and the target unmanned aerial vehicle node B is set as d_rd[n]

The height difference between the source unmanned aerial vehicle node A and the relay unmanned aerial vehicle node R is set as hd_sr[n]The height difference between the relay unmanned aerial vehicle node R and the destination unmanned aerial vehicle node B is set as hd_rd[n]，

hd_sr[n]＝|ps_z[n]-pr_z[n]|n＝1,2,...,N (3)

hd_rd[n]＝|pr_z[n]-pd_z[n]|n＝1,2,...,N (4)

Adopting non line of sight (NLOS) transmission, the channel transmission model is as follows:

wherein the content of the first and second substances,

wherein eta_LOS，η_NLOSA, b are constants related to the propagation environment, f is the carrier frequency, c is the speed of light;

the power loss of the nth slot is:

the channel transmission coefficient of the nth time slot is as follows:

sn represents small scale fading, which is independent and same distributed CN (0, 1);

adopting DF to forward the way, according to the channel condition, guaranteeing that the channel transmission rate is optimal, supposing that the transmitting power of the source unmanned aerial vehicle and the relay unmanned aerial vehicle are equal, the maximum communication rates of the source unmanned aerial vehicle and the target unmanned aerial vehicle are respectively:

h_srchannel state matrix, P, for relaying to a destination node_sIn order to transmit the power, the power is,²representing a noise variance; h is_rdChannel state matrix, P, for relaying to a destination node_rIn order to transmit the power, the power is,²representing a noise variance; b is the channel bandwidth;

reasonably constraining the transmission rate, the speed and the turning radius of the unmanned aerial vehicle;

C_sr[n]-C_rd[n]≥0 (10)

x_r(n+1)-x_r(n)≤mV_rmax(11)

C_srChannel transmission rate, C, from the source drone to the relay drone_rdRepresenting the channel transmission rate from the relay unmanned aerial vehicle to the target unmanned aerial vehicle; x is the number of_r(n) represents the location of the relay drone of point n of the discrete point;

m is a proportionality coefficient; v_rmaxThe unit of the maximum flying speed of the relay unmanned aerial vehicle is m/s; r_minIs the minimum turning radius, and g is the gravitational acceleration.

7. The trajectory generation and tracking method for formation communication of unmanned aerial vehicles according to claim 6, wherein: in step S2, based on the trajectory prediction algorithm of the extended kalman filter, performing state estimation on the dynamic behavior of the mobile unmanned aerial vehicle, updating the estimation on the state variable by using the estimation value at the previous time and the observation value at the current time, and further predicting the trajectory position at the next time;

the state is updated according to the following formula:

X(k)＝f_k-1(X(k-1)) (13)

Z(k)＝h_k(X(k)) (14)

X(k|k-1)＝f_k-1(X(k-1|k-1) (15)

P(k|k-1)＝f_k-1P(K-1|K-1)f_k-1 ^T+f_k-1Qf_k-1 ^T(16)

k_g(k)＝P(k|k-1)(h_k)^T/(h_kP(k|k-1)h_k ^T+h_kR_kh_k ^T) (17)

X(k|k)＝X(k|k-1)+k_g(k)[Z(k)-h_kX(k|k-1)](18)

in equations (13) to (18), X (k | k-1) is the result of prediction using the previous state, A is the system parameter, X (k-1| k-1) is the result of the last state optimization, P (k | k-1) is the covariance matrix corresponding to X (k | k-1), P (k-1| k-1) is the covariance matrix corresponding to X (k-1| k-1), Q and R are the white Gaussian noise variance of the system process, k is the white noise variance of the system process_gFor extended Kalman gain, H is a parameter of the measurement system, and P (k | k) is a covariance matrix corresponding to X (k | k).

8. The trajectory generation and tracking method for formation communication of unmanned aerial vehicles according to claim 7, wherein: in step S31, a DF forwarding scheme is adopted to ensure that the channel transmission rate is optimal according to the channel conditions, and assuming that the transmission powers of the source drone and the relay drone are equal, the constructed optimization problem is:

s.t.C_sr[n]-C_rd[n]≥0 (19a)

x_r(n+1)-x_r(n)≤mV_rmax(19b)

h_rdchannel state matrix, P, for relaying to a destination node_rIn order to transmit the power, the power is,²representing the noise variance, reasonably constraining the transmission rate, the speed and the turning radius of the unmanned aerial vehicle, C_srChannel transmission rate, C, from the source drone to the relay drone_rdRepresenting the channel transmission rate, x, from the relaying drone to the destination drone_r(n) represents the location of the relay drone of point n of the discrete point;

m is a proportionality coefficient; v_rmaxThe unit of the maximum flying speed of the relay unmanned aerial vehicle is m/s; r_minIs the minimum turning radius, g is the acceleration of gravity;

in order to obtain the maximum transmission rate, the obtained optimization problem is a non-convex problem, a PSO algorithm with a penalty function is adopted to solve the optimization problem, the constrained optimization problem is converted into an unconstrained optimization problem, an optimal solution is solved, and a fitness function of a particle swarm is constructed according to the established optimization problem:

U represents a penalty factor.

9. The trajectory generation and tracking method for formation communication of unmanned aerial vehicles according to claim 7, wherein: in step S32, the source drone and the destination drone are tracked according to their trajectoriesGenerating M optimal points by the trace prediction, starting from an initial position, obtaining M +1 points in total, calculating polynomial coefficients of M sections of traces to obtain each section of trace, and knowing the time t of each section of trace_iTotal time is T;

l(t)＝α₀+α₁t+α₂t²+α₃t³+α₄t⁴+α₅t⁵+α₆t⁶+α₇t⁷(21)

where l (t) is the expression of the trajectory at time t, α₀,α₁,α₂,...,α_kThe method comprises the steps of obtaining polynomial coefficients for polynomial coefficients, distributing time evenly, obtaining the polynomial coefficients by using M polynomials, obtaining continuous expressions of each section, and constructing an equality constraint equation in order to guarantee the continuity of tracks;

the initial position and the target position of the relay unmanned aerial vehicle are fixed, the motion track of the unmanned aerial vehicle in a period of time is selected, in the period of time, the initial position, the speed and the acceleration of the track are equal to those of the relay unmanned aerial vehicle, and the tail end position of the track is equal to that of the relay unmanned aerial vehicle in the period of time;

l⁽¹⁾[0]＝(vr_x[0]，vr_y[0],vr_z[0]) (24)

l⁽¹⁾[T]＝(vr_x[T],vr_y[T],vr_z[T]) (25)

l⁽²⁾[0]＝(ar_x[0]，ar_y[0],ar_z[0]) (26)

l⁽²⁾[T]＝(ar_x[T],ar_y[T],ar_z[T]) (27)

parameter l [0 ]],l[T],l⁽¹⁾[0],l⁽¹⁾[T],l⁽²⁾[0],l⁽²⁾[T]Representing the initial position, the T-time position, the initial speed, the T-time speed, the initial acceleration and the T-time acceleration of the relay unmanned aerial vehicle in a period of time;

The positions, the speeds and the accelerations of the adjacent intermediate points are continuous, the motion of the unmanned aerial vehicle cannot change suddenly, and the motion track of the unmanned aerial vehicle is a continuously-guided curve;

parameter l_k[tf],l_k+1[ts],Respectively representing the end position of a section k, the start position of a section k +1, representing the end speed of the section k, the start speed of the section k +1, representing the end acceleration of the section k, and the start acceleration of the section k +1, solving polynomial coefficients, finding out the shortest curve between the two points, and constructing an optimization equation as follows:

s.t.

A_eqα＝b_eq(29a)

and solving the parameters to obtain a polynomial expression.

10. The utility model provides a track generation and tracking system towards unmanned aerial vehicle formation communication which characterized in that: comprising a readable storage medium having stored therein execution instructions for, when executed by a processor, implementing the method of any one of claims 1 to 9.

Technical Field

The invention relates to an unmanned aerial vehicle, in particular to a trajectory generation and tracking method and system for formation communication of the unmanned aerial vehicle.

Background

The low-altitude high-speed civil unmanned aerial vehicle plays an important role in many fields such as logistics transportation, evening performance, live competition and the like. The unmanned aerial vehicle inevitably meets complicated communication environment at the in-process of carrying out some tasks, compares in fixed basic station on ground, and unmanned aerial vehicle has high mobility, deploys nimble characteristics, therefore unmanned aerial vehicle ubiquitous in various applications, including unmanned aerial vehicle relay under emergency, follow satellite uninstallation data etc. in the hot spot region.

In wireless communication, relaying is a method for improving system communication throughput and assisting system re-communication in which a direct link is lost due to channel change. Under the influence of node mobility, a common relay base station cannot meet the requirement on the mobility flexibility of the base station, and along with the popularization of unmanned aerial vehicles, the unmanned aerial vehicles are widely used as an air relay auxiliary communication method.

Compared with the traditional static relay, the unmanned aerial vehicle mobile relay has the following advantages:

(1) the system can be deployed quickly, and can be used as a relay to join the system at any time when the communication system is needed. Is suitable for processing unknown and emergency events.

(2) The mobile communication system has high mobility and flexibility, can freely move between a sending end and a receiving end, can adjust the position at any time to achieve the optimal communication performance, and enhances the performance of the communication system.

In recent years, research on relay unmanned aerial vehicles mainly includes unmanned aerial vehicles as auxiliary communication tools of ground base stations, and when the coverage of the ground base stations is limited or limited, the unmanned aerial vehicles are used as relays for transmitting information. The auxiliary unmanned aerial vehicle can be divided into static unmanned aerial vehicle deployment, semi-dynamic unmanned aerial vehicle deployment and dynamic unmanned aerial vehicle deployment according to whether the auxiliary unmanned aerial vehicle is movable. The static unmanned aerial vehicle is deployed to realize that an optimal communication position is selected between a transmitting end and a receiving end for communication, so that the communication requirements of the transmitting end and the receiving end can be met, or the coverage of a communication area is met by deploying the static unmanned aerial vehicle. Semi-dynamic unmanned aerial vehicle deploys that ground user is static, and unmanned aerial vehicle removes, and in practical application, unmanned aerial vehicle need constantly remove in order to satisfy different demands. The dynamic unmanned aerial vehicle deployment is that the positions of ground users and unmanned aerial vehicles are changed, and for some mobile users, mainly the ground hot spot users are required, and in order to ensure the communication quality in the moving process, the deployment of the dynamic unmanned aerial vehicle is required.

In the present phase, research on relay unmanned aerial vehicles mainly focuses on providing aerial assistance for ground base stations, generally fixing the flight height of the unmanned aerial vehicle, and searching the optimal position of communication between two ground base stations needing communication. The relay unmanned aerial vehicle is supplementary to ground basic station, and both ends communication base station is fixed, also provides convenience for the solution of problem.

However, when the unmanned aerial vehicle moving at a high speed flies at a low altitude, shadow fading is generated due to occlusion, and communication interruption can be caused when the environment is complex, namely, the communication of the unmanned aerial vehicle moving at a high speed is interrupted, and link connection cannot be reestablished, so that communication interruption is caused.

Therefore, in the case of link disconnection between two communicating drones, how to provide a method for reestablishing a stable connection breaks through the assumption that the source node and the destination node of the conventional communication are stationary, which is a technical problem to be solved urgently by those skilled in the art.

Disclosure of Invention

In order to solve the problems in the prior art, the invention provides a trajectory generation and tracking method and system for unmanned aerial vehicle formation communication.

The invention provides a trajectory generation and tracking method for formation communication of unmanned aerial vehicles, which comprises the following steps:

s1, constructing an unmanned aerial vehicle communication system model, dividing a communicated inorganic cluster into a source unmanned aerial vehicle and a target unmanned aerial vehicle, and reestablishing communication between the source unmanned aerial vehicle and the target unmanned aerial vehicle by using a relay unmanned aerial vehicle when the source unmanned aerial vehicle and the target unmanned aerial vehicle cannot communicate with each other, namely no direct communication link exists;

s2, predicting the tracks of the source unmanned aerial vehicle and the target unmanned aerial vehicle, and transmitting the tracks to the relay unmanned aerial vehicle in real time;

And S3, planning the track of the relay unmanned aerial vehicle in real time according to the predicted tracks of the source unmanned aerial vehicle and the target unmanned aerial vehicle.

As a further improvement of the present invention, in step S1, the relay drone reestablishes communication between the originating drone and the destination drone, using the DF forwarding scheme.

As a further improvement of the present invention, in step S2, the trajectories of the source drone and the target drone are predicted separately using a trajectory prediction algorithm based on the extended kalman filter, and the position at the next time is predicted at the current time.

As a further improvement of the invention, the trajectory prediction algorithm based on the extended Kalman filtering comprises:

inputting: an initial position value, an initial reported value;

and (3) outputting: predicted trajectory values, deviation values;

1): preprocessing a track;

2): initializing parameters;

3): acquiring the current state and the position of the current moment;

4): circulating for given times;

5): recording the value predicted each time by the extended Kalman filtering algorithm;

6) calculating the deviation between the predicted value and the actual value;

7) ending the cycle;

8) outputting the predicted value and the deviation.

As a further improvement of the present invention, step S3 includes the following sub-steps:

S31, solving the optimal problem by a PSO algorithm based on a penalty function;

s32, generating a smooth track of the relay unmanned aerial vehicle;

and S33, controlling the relay unmanned aerial vehicle to fly out of the optimal track.

The invention also provides a trajectory generation and tracking system for formation communication of unmanned aerial vehicles, which comprises a readable storage medium, wherein execution instructions are stored in the readable storage medium, and when executed by a processor, the execution instructions are used for realizing the method in any one of the above.

The invention has the beneficial effects that: through the scheme, a stable connection reestablishing method is provided for the conditions that shadow and fading exist in a low-altitude environment and the link connection of the two communication unmanned aerial vehicles is disconnected in a severe environment, and the assumption that a source node and a destination node of traditional communication are immovable is broken through.

Drawings

Fig. 1 is an interruption scene diagram of a trajectory generation and tracking method for formation communication of unmanned aerial vehicles according to the present invention.

Fig. 2 is a plot of trajectory prediction and predicted deviation for the direction of source drone X, Y.

Fig. 3 is a plot of trajectory prediction and predicted deviation in the Z direction of the source drone.

Fig. 4 is a diagram of trajectory prediction and predicted deviation of the direction of the destination drone X, Y.

Fig. 5 is a diagram of trajectory prediction and predicted deviation in the Z direction of the target drone.

Fig. 6 is a diagram of relay drone optimal position and maximum transmission rate.

Fig. 7 is a three-dimensional path point diagram of the relay drone.

Fig. 8 is a diagram of the flight trajectory of the relay drone.

Fig. 9 is a diagram comparing the flight trajectory of the relay unmanned aerial vehicle with the planned trajectory position.

Fig. 10 is a plot of the flight speed of the relay drone compared to the planned trajectory speed.

Detailed Description

The invention is further described with reference to the following description and embodiments in conjunction with the accompanying drawings.

A track generation and tracking method for formation communication of unmanned aerial vehicles is characterized in that a relay unmanned aerial vehicle is used for carrying out auxiliary communication, a DF forwarding mode is adopted, under the condition that the unmanned aerial vehicle continuously moves, a PSO algorithm with penalty factors is adopted to plan a relay unmanned aerial vehicle track which enables a separated formation to be reconnected for communication and to obtain the optimal transmission rate, the positions of a source unmanned aerial vehicle and a relay unmanned aerial vehicle are predicted by considering the real-time performance of the motion of the unmanned aerial vehicle, the position of the relay unmanned aerial vehicle is planned in real time, and the real-time performance and the effectiveness of the algorithm are guaranteed. The dynamic constraint of the unmanned aerial vehicle is considered, and the actual flying characteristics of the unmanned aerial vehicle are highlighted. And controlling the relay unmanned aerial vehicle to fly according to the planned smooth track by using an unmanned aerial vehicle control algorithm, wherein the track deviation finally reflects the deviation of the communication speed.

The specific process is as follows:

1. unmanned aerial vehicle communication system model construction

The drone swarm performs a specific task, and due to the blockage of certain obstacles or channel fading, a connected drone is grouped into two parts that cannot communicate, i.e., there is no direct communication link. A relay unmanned aerial vehicle R is adopted, so that the relay unmanned aerial vehicle R helps to establish the connection of the unmanned aerial vehicle cluster again. Considering a two-hop system, as shown in fig. 1, first, the drone cluster is divided into two parts, and it is assumed that each part can select a cluster head node, which is a source node a and a destination node B for information transmission. Secondly, the connection of the whole unmanned aerial vehicle cluster can be realized as long as the source node A and the destination node B can carry out optimal communication. Due to the high-speed mobility of the unmanned aerial vehicle, the communication quality of the unmanned aerial vehicle cluster is guaranteed to be optimal in the moving process.

Assuming that the relay unmanned aerial vehicle only acts in a certain time, the total time of the whole motion process is T, the relay unmanned aerial vehicle is divided into N time slots, and the coordinates of the nth time slot node A, R, B are (ps)_x[n],ps_y[n],ps_z[n])，(pr_x[n],pr_y[n],pr_z[n])，(pd_x[n],pd_y[n],pd_z[n]). N is 1,2, the distance between the source unmanned aerial vehicle node a and the relay unmanned aerial vehicle node R is set as d_sr[n]The distance between the relay unmanned aerial vehicle node R and the target unmanned aerial vehicle node B is set as d _rd[n]。

The height difference between the source unmanned aerial vehicle node A and the relay unmanned aerial vehicle node R is set as hd_sr[n]The height difference between the relay unmanned aerial vehicle node R and the destination unmanned aerial vehicle node B is set as hd_rd[n]，

hd_sr[n]＝|ps_z[n]-pr_z[n]|n＝1,2,...,N (3)

hd_rd[n]＝|pr_z[n]-pd_z[n]|n＝1,2,...,N (4)

The unmanned aerial vehicle flying at low altitude and high speed is considered, the channel condition of the unmanned aerial vehicle flying at low altitude is complex, shelters of building buildings and the like may exist, and influences of shadows, fading and the like are mainly considered. Consider non line-of-sight (NLOS) transmission. The channel transmission model is:

wherein the content of the first and second substances,

wherein eta_LOS，η_NLOSA, b are constants related to the propagation environment, f is the carrier frequency, and c is the speed of light.

The power loss of the nth slot is:

the channel transmission coefficient of the nth time slot is as follows:

sn represents small scale fading and is independent of the same distribution CN (0, 1).

Adopting a DF (decode and forward) forwarding mode, ensuring the optimal channel transmission rate according to the channel condition, and assuming that the transmitting power of a source unmanned aerial vehicle and a relay unmanned aerial vehicle are equal, the maximum communication rates of the source unmanned aerial vehicle and a target unmanned aerial vehicle are respectively as follows:

h_srchannel state matrix, P, for relaying to a destination node_sIn order to transmit the power, the power is,²representing the noise variance. h is_rdChannel state matrix, P, for relaying to a destination node_rIn order to transmit the power, the power is,²representing the noise variance. B is the channel bandwidth.

And carrying out reasonable constraint on the transmission rate, the speed and the turning radius of the unmanned aerial vehicle.

C_sr[n]-C_rd[n]≥0(10)

x_r(n+1)-x_r(n)≤mV_rmax(11)

C_srChannel transmission rate, C, from the source drone to the relay drone_rdRepresenting the channel transmission rate of the relay drone to the destination drone. x is the number of_r(n) represents the location of the relay drone of point n of the discrete point.

And m is a proportionality coefficient. V_rmaxThe unit is m/s for relaying the maximum flying speed of the unmanned aerial vehicle. R_minIs the minimum turning radius, and g is the gravitational acceleration.

2. Source unmanned aerial vehicle and target unmanned aerial vehicle trajectory prediction

When the environment is more complicated, the navigation positioning system of unmanned aerial vehicle self can be destroyed, can't convey the real-time position to relay unmanned aerial vehicle, considers the real-time nature of unmanned aerial vehicle motion, utilizes the extended Kalman filtering algorithm to predict source unmanned aerial vehicle and relay unmanned aerial vehicle's position, and real-time and the validity of algorithm are guaranteed in the position of real-time planning relay unmanned aerial vehicle. And performing state estimation on the dynamic behavior of the mobile unmanned aerial vehicle based on a dynamic trajectory prediction algorithm of the extended Kalman filtering, updating estimation on a state variable by using an estimation value at the previous moment and an observation value at the current moment, and predicting the trajectory position at the next moment.

The state is updated according to the following formula:

X(k)＝f_k-1(X(k-1)) (13)

Z(k)＝h_k(X(k)) (14)

X(k|k-1)＝f_k-1(X(k-1|k-1) (15)

P(k|k-1)＝f_k-1P(K-1|K-1)f_k-1 ^T+f_k-1Qf_k-1 ^T(16)

k_g(k)＝P(k|k-1)(h_k)^T/(h_kP(k|k-1)h_k ^T+h_kR_kh_k ^T) (17)

X(k|k)＝X(k|k-1)+k_g(k)[Z(k)-h_kX(k|k-1)](18)

Where X (k | k-1) is the result of the prediction using the previous state, A is the system parameters, X (k-1| k-1) is the optimal result for the previous state, P (k | k-1) is the covariance matrix corresponding to X (k | k-1), P (k-1| k-1) is the covariance matrix corresponding to X (k-1| k-1), Q and R are the white Gaussian noise variances of the system process, k is the system parameters_gIs an extended kalman gain. H is a parameter of the measurement system, and P (k | k) is a covariance matrix corresponding to X (k | k).

And respectively predicting the tracks of the source unmanned aerial vehicle and the target unmanned aerial vehicle by using an extended Kalman filtering algorithm, and predicting the position of the next moment at the current moment.

Algorithm 1: trajectory prediction algorithm based on extended Kalman filtering

Inputting: initial position value, initial measured value

And (3) outputting: predicted trajectory values, deviation values

1: preprocessing a track;

2: initializing parameters;

3: acquiring the current state and the position of the current moment;

4: circulating for given times;

5: recording the value predicted each time by the extended Kalman prediction method;

6, calculating the deviation between the predicted value and the actual value;

7, ending the circulation;

and 8, outputting the predicted value and the deviation.

3. PSO algorithm based on penalty function for solving optimal problem

After the motion path points of the source unmanned aerial vehicle and the target unmanned aerial vehicle at the next moment are predicted by using an extended Kalman filtering algorithm at the current moment, the motion path points are transmitted to the relay unmanned aerial vehicle in real time, so that the relay unmanned aerial vehicle can calculate the path point at the next moment.

Adopting DF forwarding mode, according to the channel condition, ensuring the optimal channel transmission rate, and assuming that the transmission power of the source unmanned aerial vehicle and the transmission power of the relay unmanned aerial vehicle are equal, the constructed optimization problem is as follows:

s.t.C_sr[n]-C_rd[n]≥0 (19a)

x_r(n+1)-x_r(n)≤mV_rmax(19b)

h_rdchannel state matrix, P, for relaying to a destination node_rIn order to transmit the power, the power is,²representing the noise variance. And carrying out reasonable constraint on the transmission rate, the speed and the turning radius of the unmanned aerial vehicle. C_srChannel transmission rate, C, from the source drone to the relay drone_rdRepresenting the channel transmission rate of the relay drone to the destination drone. x is the number of_r(n) represents the location of the relay drone of point n of the discrete point.

And m is a proportionality coefficient. V_rmaxThe unit is m/s for relaying the maximum flying speed of the unmanned aerial vehicle. R_minIs the minimum turning radius, and g is the gravitational acceleration.

In order to obtain the maximum transmission rate, the obtained Optimization problem is a non-convex problem, a Particle Swarm Optimization (PSO) algorithm with a penalty function is adopted to solve the Optimization problem, the constrained Optimization problem is converted into an unconstrained Optimization problem, and an optimal solution is obtained. And constructing a fitness function of the particle swarm according to the established optimization problem to obtain:

u represents a penalty factor.

4. Relay unmanned aerial vehicle smooth trajectory generation

The relay unmanned aerial vehicle generates a smooth and continuous curve from generated discrete path points, a 7-time minisnap track generation method is adopted to obtain a group of path points, in order to guarantee the effectiveness of the obtained path points, the relay unmanned aerial vehicle is expected to fly through the optimal path points, M optimal points are generated according to the track prediction of a source unmanned aerial vehicle and a target unmanned aerial vehicle, M +1 points are shared from an initial position, the polynomial coefficient of M sections of tracks is calculated, and each section of track is obtained. And knowing the time t of each trace_iThe total time is T.

l(t)＝α₀+α₁t+α₂t²+α₃t³+α₄t⁴+α₅t⁵+α₆t⁶+α₇t⁷(21)

Where l (t) is the expression of the trajectory at time t. Alpha is alpha₀,α₁,α₂,...,α_kThe polynomial coefficient is calculated for the polynomial coefficient and has M polynomials with uniform time distribution, thereby obtaining a continuous expression of each segment. To ensure continuity of the trajectory, an equality constraint is constructedAnd (4) an equation.

The initial position and the target position of the relay unmanned aerial vehicle are fixed, the motion track of the unmanned aerial vehicle in a period of time is selected, in the period of time, the initial position, the speed and the acceleration of the track are equal to the initial position, the speed and the acceleration of the relay unmanned aerial vehicle, and the tail end position of the track is equal to the tail end position, the speed and the acceleration of the relay unmanned aerial vehicle in the period of time.

l⁽¹⁾[0]＝(vr_x[0]，vr_y[0],vr_z[0]) (24)

l⁽¹⁾[T]＝(vr_x[T],vr_y[T],vr_z[T]) (25)

l⁽²⁾[0]＝(ar_x[0]，ar_y[0],ar_z[0]) (26)

l⁽²⁾[T]＝(ar_x[T],ar_y[T],ar_z[T]) (27)

Parameter l [0 ]],l[T],l⁽¹⁾[0],l⁽¹⁾[T],l⁽²⁾[0],l⁽²⁾[T]Represent relay unmanned aerial vehicle initial position in a period of time, T position at a moment, initial velocity, T speed at a moment, initial acceleration, T acceleration at a moment.

The position, the speed and the acceleration of adjacent intermediate point are continuous, and the motion of unmanned aerial vehicle can not take place the sudden change, and its motion trajectory is the curve that can lead in succession.

Parameter l_k[tf],l_k+1[ts],

Respectively represents a k-segment ending position, a k + 1-segment starting position, a k-segment ending speed, a k + 1-segment starting speed, a k-segment ending acceleration and a k + 1-segment starting acceleration. I amThe purpose of the method is to calculate polynomial coefficients, there may be many curves satisfying constraints, we need to find the shortest curve between two points, and construct an optimization equation as follows:

s.t.

A_eqα＝b_eq(29a)

and solving the parameters to obtain a polynomial expression.

5. Control relay unmanned aerial vehicle to fly out of optimal track

In order to ensure the feasibility of the scheme, the characteristics of the unmanned aerial vehicle body and the related angle requirements are combined, and the smaller the error between the actual flying track and the expected flying track is, the better the error is, namely the control errors of the three coordinate directions and the yaw angle at the lower side are 0. Namely:

(a_rx-a_rx,des)+k_d1(v_rx-v_rx,des)+k_p1(p_rx-p_rx,des)＝0 (30)

(a_ry-a_ry,des)+k_d2(v_ry-v_ry,des)+k_p2(p_ry-p_ry,des)＝0 (31)

(a_rz-a_rz,des)+k_d3(v_rz-v_rz,des)+k_p3(p_rz-p_rz,des)＝0 (32)

ψ＝ψ_des(33)

k_p1、k_p2、k_p3respectively representing the proportional control parameters in the directions of the X, Y and Z positions of the coordinate system. Correspondingly, k_d1、k_d2、k_d3Respectively representing differential control parameters in the directions of the X, Y and Z positions of the coordinate system. v. of _rx、v_ry、v_rz、a_rx、a_ry、a_rz、p_rx,p_ry,p_rzRepresenting velocity, acceleration and position in the X, Y, Z directions of the coordinate system, respectively. v. of_x,des,v_y,des,v_z,des、a_x,des,a_y,des,a_z,des、p_x,des,p_y,des,p_z,desRepresenting the desired velocity, acceleration and position in the X, Y, Z directions of the coordinate system, respectively. Psi and psi_desRespectively yaw angle and desired yaw angle.

6. Simulation result

The positions of a source unmanned aerial vehicle and a target unmanned aerial vehicle are predicted, and three coordinate direction trajectory equations of the source unmanned aerial vehicle are selected as follows:

p_sx＝3+10sin(3m)

p_sy＝2+10sin(2m)

p_sz＝10cos(4m)

the three coordinate direction trajectory equations of the target unmanned aerial vehicle are respectively as follows:

p_dx＝7+10sin(3m)

p_dy＝8+10sin(2m)

p_dz＝20+10cos(4m)

the positions of 20 path points of a source unmanned aerial vehicle and a target unmanned aerial vehicle are predicted, the time interval is 1S, the predicted track values and the predicted error results of the source unmanned aerial vehicle in three coordinate axis directions are shown in figures 2 and 3, and the error between the position predicted by using the extended Kalman filtering algorithm and the actual position is very small through comparison of the figures. The predicted track values and the predicted error results of the target unmanned aerial vehicle in the three coordinate axis directions are shown in fig. 4 and 5, and the comparison of the graphs shows that the error between the position predicted by using the extended Kalman filtering algorithm and the actual position is very small.

In the experiment, the fact that large-scale fading exists in the low-altitude environment when the unmanned aerial vehicle flies is assumed, and eta is selected_LOS＝0.1，η_NLOS21, a-5.0188, b-0.3511, and a carrier frequency of 2.4 × 10 ⁹HZ，c＝3×10⁸m/s, the transmitting power of the source unmanned aerial vehicle and the transmitting power of the relay unmanned aerial vehicle are both 20dB, the noise power value used in simulation is-163 dBm, and small-scale fading existing in the environment is independent and identically distributed CN (0, 1). Maximum flight speed v of relay unmanned aerial vehicle_max＝50m/s，g＝9.81m/s²。

The predicted track value is substituted into an optimization function for calculating the optimal position of the relay unmanned aerial vehicle, so that the optimal position point and the maximum speed value in the three coordinate axis directions of the relay unmanned aerial vehicle shown in fig. 6 are obtained, and the maximum speed value can be seen to be 26.3bit/s, and tends to be stable along with the continuous movement of the unmanned aerial vehicle, which shows that the relay unmanned aerial vehicle can stably provide communication service for the unmanned aerial vehicle which continuously moves at a high speed. Fig. 7 is a diagram of the three-dimensional path points of relay drones connected together, which is not in line with the flight characteristics of drones, which require a smooth continuous trajectory.

By adjusting the control parameters of the unmanned aerial vehicle, a trajectory diagram containing the expected trajectory and actually controlling the unmanned aerial vehicle to fly shown in fig. 8 is obtained, and it can be observed that within a fixed flight time of 20S, the planned expected trajectory and the actually flying trajectory have a certain deviation due to the influence of the control parameters or calculation errors. Fig. 9 further shows deviations of the desired trajectory from the actual flight trajectory in various directions of the coordinate axis, corresponding to deviations of the information transfer rate. FIG. 10 is a graph comparing desired speed to actual airspeed. Through fig. 7, 8 and 9, it can be seen that the method for reestablishing the communication link connection by using the relay unmanned aerial vehicle in the environment of communication interruption of the unmanned aerial vehicle, which is provided by the invention, is effective, and the error exhibited by actual flight of the unmanned aerial vehicle is acceptable, thereby proving the feasibility of the method.

The track generation and tracking method and system for unmanned aerial vehicle formation communication provided by the invention have the following advantages:

(1) the method for reestablishing the stable connection is provided for the conditions that shadow and fading exist in the low-altitude environment and the link connection of the two communication unmanned aerial vehicles is disconnected in the severe environment, and the assumption that the source node and the destination node of the traditional communication are fixed is broken through.

(2) The quad-rotor unmanned aerial vehicle has the characteristic of high-speed flight, and the adopted unmanned aerial vehicle position prediction method can reduce the deviation of the position calculation of the relay unmanned aerial vehicle caused by time delay and realize real-time position calculation.

(3) By considering the dynamic characteristics of the unmanned aerial vehicle, the path point of the relay unmanned aerial vehicle calculated by the invention is changed into a smooth and continuous feasible track of the unmanned aerial vehicle, and the unmanned aerial vehicle is controlled to fly out of the planning track by using an unmanned aerial vehicle control method, so that the scheme is more comprehensive and complete. The defect that in the prior art, only data is considered, and the characteristics of the unmanned aerial vehicle are not really considered by taking discrete points as the track of the unmanned aerial vehicle is overcome.

The foregoing is a more detailed description of the invention in connection with specific preferred embodiments and it is not intended that the invention be limited to these specific details. For those skilled in the art to which the invention pertains, several simple deductions or substitutions can be made without departing from the spirit of the invention, and all shall be considered as belonging to the protection scope of the invention.