Communication perception control integrated design method for unmanned aerial vehicle platform

Abstract

The invention discloses a communication perception control integrated design method for an unmanned aerial vehicle platform, and the method comprises the following steps: enabling an unmanned aerial vehicle to transmit a special radar signal in the process of transmitting a communication signal to a user through transmitting beam forming, so as to improve the communication and perception performances; based on the direction vector of the unmanned aerial vehicle, the total lift force of the unmanned aerial vehicle and the six-degree-of-freedom model, constructing and optimizing a horizontal flight six-degree-of-freedom model of the unmanned aerial vehicle to maximize communication average weight and transmission rate; performing convex optimization processing on the communication and sensing beam forming vectors; performing gradient optimization processing on the track of the unmanned aerial vehicle; according to the method, alternate solution optimization is carried out through the communication sensing beamforming vector and the unmanned aerial vehicle trajectory, a continuous and smooth flight trajectory is obtained, vibration and abrupt change in the flight process of the unmanned aerial vehicle are effectively reduced, and stability and trackability are improved.

Description

Technical Field

[0001] The present invention relates to the field of unmanned aerial vehicle communication technology, and in particular to a communication perception and control integrated design method for an unmanned aerial vehicle platform. Background Art

[0002] As mobile communications gradually move towards the era of intelligent connectivity of all things, future mobile communication systems must not only achieve ultra-high speed, ultra-low latency, and ultra-high reliability, but also have millimeter-level precision perception capabilities to support various intelligent applications such as autonomous driving, traffic monitoring, human activity recognition, and smart homes. To this end, integrated perception and communication have received great attention from academia and industry. Compared with information embedding methods [1] and waveform combination schemes [2], multi-antenna systems provide additional spatial information that can significantly improve perception performance and effectively increase system throughput. Beamforming technology focuses signals in multiple specific directions at the same time, which can ensure high-quality services for multiple users and high-precision perception of multiple targets. However, due to non-line-of-sight signal paths or clutter caused by obstacles and scatterers in the surrounding environment, the multi-input multi-output ISAC of the ground network is prone to serious degradation of interawareness performance.

[0003] Due to the high altitude advantage of UAVs, they are expected to become a promising new type of aerial ISAC platform to overcome the existing limitations. Therefore, UAV ISAC with transmit beamforming function has been widely studied. For example, reference [3] considers the scenario where UAVs perform perception tasks in the target area and provide communication services to multiple users at the same time. The research goal is to maximize the weighted average communication rate under the perception performance constraint. To this end, the two-dimensional trajectory and transmit beamforming vector of the UAV are optimized. Reference [4] designs a new adaptive ISAC mechanism in the UAV-assisted system to avoid over-perception, in which the perception duration does not need to be consistent with the communication duration and can be flexibly configured according to application requirements. Reference [5] discusses two joint optimization schemes under different states of UAVs. On the one hand, a joint design method of communication precoding and UAV flight trajectory is proposed to solve the problem of maximizing the minimum user rate; on the other hand, a joint optimization method of UAV perception position, communication and perception precoding is proposed to solve the problem of maximizing the minimum target detection probability. Reference [6] studies a multi-UAV-assisted ISAC scenario, that is, the UAV detects the target and transmits the corresponding data to the user at the same time. A joint design of UAV trajectory, user association and beamforming is proposed to maximize the sum of weighted bit rates of all ground users while minimizing the demand for radar-aware services.

[0004] Existing research works solve the trajectory optimization problem of UAVs under discrete velocity constraints by modeling UAVs as mass points. They use time discretization techniques to obtain a series of segmented trajectories. However, existing models only focus on the position and velocity of the UAV, while ignoring its rotational motion, internal forces, and torque. This means that the complex dynamic characteristics of the UAV as a rigid body are ignored, which may lead to inaccuracies in trajectory optimization and control strategies. In other words, it is difficult for the UAV controller to accurately follow the planned trajectory in practical applications. The mismatch between the planned trajectory and the actual trajectory may lead to increased communication link delays and acquisition errors, as well as a decrease in the target recognition ability of the perception system, ultimately reducing communication and perception performance.

[0005] [1] A.Hassanien, MGAmin, YDZhang, and F.Ahmad, "Dual-Function Radar-Communications: Information Embedding Using Sidelobe Control and WaveformDiversity," IEEE Transactions on Signal Processing, vol.64, no.8, pp.2168–2181, Apr.2016.

[0006] [2] Q.Li, K.Dai, Y.Zhang, and H.Zhang, "Integrated Waveform for a JointRadar-Communication System With High-Speed ​​Transmission," IEEE WirelessCommunications Letters, vol.8, no.4, pp.1208–1211, Aug.2019.

[0007] [3] ZHLyu, GXZhu, and J.Xu, "Joint Maneuver and Beamforming Design for UAV-Enabled Integrated Sensing and Communication," IEEE Transactions on Wireless Communications, vol.22, no.4, pp.2424–2440, Apr.2023.

[0008] [4]C.Deng,X.Fang,and

[0009] [5] R.Chai,

[0010] [6] R. Zhang, Y. Zhang, R. Tang, H. Zhao, Q. Xiao, and C. Wang, "AJoint UAV Trajectory, User Association, and Beamforming Design Strategy for Multi-UAVAssisted ISAC Systems," IEEE Internet of Things Journal, vol. 11, no. 18, pp. 29360–29374, Sep. 2024. Summary of the invention

[0011] In order to overcome the shortcomings of the prior art, the present invention provides a communication, perception and control integrated design method for UAV platforms, which realizes alternating solution and optimization of communication perception beamforming vectors and UAV trajectories to obtain a continuous and smooth flight trajectory, effectively reducing vibration and mutations during the flight of the UAV and improving stability and traceability.

[0012] In order to achieve the above-mentioned object of the invention, the present invention adopts the following technical solutions:

[0013] The first aspect of the present application provides a communication perception control integrated design method for an unmanned aerial vehicle platform, comprising the following steps:

[0014] S101. When the drone transmits communication signals to the user using transmit beamforming, it also sends dedicated radar signals to improve communication and perception performance.

[0015] S102, based on the UAV direction vector, the total lift of the UAV and the six-degree-of-freedom model, construct and optimize the six-degree-of-freedom model of the UAV horizontal flight to maximize the average weighted communication and transmission rate;

[0016] S103, performing convex optimization processing on the communication and perception beamforming vectors;

[0017] S104: performing gradient optimization processing on the trajectory of the UAV.

[0018] Furthermore, in the process of transmitting communication signals to users by using transmit beamforming, the drone simultaneously transmits dedicated radar signals to improve communication and perception performance, including the following steps:

[0019] Determine the channel vector from the UAV to the mth user based on the transmit array response vector and the channel power gain;

[0020] Determine the signal received at the th user based on the channel vector, the transmitted signal of the drone and the additive white Gaussian noise;

[0021] determining a signal-to-noise ratio based on the channel vector and the transmit beamforming vector, thereby determining a spectral efficiency;

[0022] Based on the transmit beam pattern gain and the perceived distance d(p(t),o j ), determine the perceived performance indicators.

[0023] Furthermore, based on the transmit beam pattern gain and the perceived distance d(p(t),o j ), determine the perceptual performance index, which is expressed as follows:

[0024]

[0025] Among them, G d is the covariance matrix, w m is the transmit beamforming vector of the mth user, d(p(t),o j ) is the distance between the UAV and the jth target at time t, p(t) is the position of the UAV at time t, o j is the position of the jth target.

[0026] Furthermore, based on the UAV direction vector, the total lift of the UAV and the six-degree-of-freedom model, a six-degree-of-freedom model of the UAV horizontal flight is constructed and optimized to maximize the average weighted communication and transmission rate, including the following steps:

[0027] Construct six-degree-of-freedom models based on translational dynamics based on Lagrange-Euler equations and rotational dynamics based on Newton-Euler equations;

[0028] Based on the six-degree-of-freedom model, the direction vector of the UAV and the total lift of the UAV, a six-degree-of-freedom model of the UAV horizontal flight is constructed;

[0029] The optimization is performed by building a constrained optimization library to maximize the communication average weighted and transmission rate.

[0030] Furthermore, based on the six-degree-of-freedom model, the direction vector of the drone and the total lift of the drone, a six-degree-of-freedom model of the drone horizontal flight is constructed. The expression of the six-degree-of-freedom model of the drone horizontal flight is as follows:

[0031]

[0032] Where, γ(t) represents the yaw angle, I xx , I yy and I zz represents the moment of inertia on the x-axis, y-axis and z-axis, l represents the length of the rigid body crossbeam, K dx and K dy Represents the drag coefficient on the x-axis and y-axis, K dmx , K dmy and K dmz represents the damping moment coefficient in the x-axis, y-axis and z-axis, K m Represents the moment coefficient, Λ(t)=-ξ 1 (t)=-ξ 2 (t)=-ξ 3 (t)=-ξ 4 (t) and sign(b) represent the sign of b. For the sake of simplicity, the six-degree-of-freedom model of horizontal flight can be abbreviated as M(p(t), Φ(t))=0.

[0033] Furthermore, based on the drone direction vector, the total lift of the drone and the six-degree-of-freedom model, a six-degree-of-freedom model of the drone horizontal flight is constructed and optimized to maximize the expression of the average weighted communication and transmission rate as shown below:

[0034]

[0035] sC 1 :M(p(t),Φ(t))=0,

[0036]

[0037] C 4 :p(0)=p I ,

[0038] C5 :p(T)=p F ,

[0039]

[0040] Among them, R m (t) represents the spectrum efficiency achievable by the mth user, and the constraint C 1 Represents the six-degree-of-freedom model of the drone, with constraint C 2 represents the gain constraint of the UAV receiving beam pattern, where is the beam pattern gain threshold of the jth target, and the constraint C 3 Indicates the flight speed limit of the drone, where V max represents the maximum flight speed, constraint C 4 and C 5 They represent the initial position and terminal position constraints respectively, and the initial position vector is p I =[x I ,y I ,z u ] · and the terminal position vector is p F =[x F ,y F ,z u ] · , constraint C 6 Indicates the UAV transmit power limit.

[0041] Further, performing convex optimization on the communication and sensing beamforming vectors includes the following steps:

[0042] Based on the communication beamforming vector, user weight, user achievable spectrum efficiency, perception covariance matrix, drone receive beam pattern gain constraint and drone transmit power limit constraint, the optimization problem of communication and perception beamforming vector is constructed;

[0043] Discretize the time interval T into P equal subintervals and p+1 time slots, and decouple the optimization problem at different times;

[0044] The non-convex rank constraints are handled by approximation techniques and semi-definite relaxation methods, and the communication and sensing beamforming vectors are obtained after convex optimization.

[0045] Furthermore, by using the approximation technique and the semi-definite relaxation method to handle the non-convex rank constraints, the expressions of the communication and sensing beamforming vectors after convex optimization are obtained as follows:

[0046]

[0047] Among them, w m is the communication beamforming vector, Gd is the covariance matrix, G d is the covariance matrix, g m The conjugate transpose of m is the channel vector from the drone to the mth user, and p is the position of the drone.

[0048] Furthermore, the gradient optimization process for the UAV trajectory includes the following steps:

[0049] Based on the beam pattern gain constraint, flight speed constraint, UAV trajectory constraint, user weight, and user-achievable spectrum efficiency, the UAV trajectory optimization problem is constructed;

[0050] Convert the UAV trajectory optimization problem into a control problem based on the state space model;

[0051] The continuous-time control vector is discretized using the control parameterization method;

[0052] The exact penalty function method is used to transform the constrained nonlinear programming problem into an unconstrained optimization problem to optimize the UAV trajectory.

[0053] Furthermore, the expression for converting the UAV trajectory optimization problem into a control problem based on the state space model is as follows:

[0054]

[0055] C 6 ':x 1 (T) = x F ,x 2 (T) = y F ,

[0056] in, Considering practical factors, the constraint C' is introduced 2 Limit the maneuverability of the drone. In addition, constrain C' 5 Provides a solution for C 1 The remaining constraints are obtained by reformulating the constraints in the original problem in terms of x(t) and u(t).

[0057] The beneficial effects of the present application are as follows: a scenario is achieved in which a multi-antenna UAV senses a target in a specific area while providing communication services to multiple users, and this scenario has potential application prospects in border surveillance and environmental monitoring. In order to address the impact of dynamic constraints on UAV performance, the optimization problem is decomposed into a beamforming optimization subproblem and a UAV trajectory optimization subproblem, and the problems are solved alternately and iteratively. Different from the trajectory discretization of existing methods, this method parameterizes the control variables based on a state space model, and describes the UAV's state variables (such as position and velocity) as functions of the control variables to achieve a continuous and smooth flight trajectory.

[0058] Compared with existing solutions, the present application can significantly reduce the degradation of communication performance and the violation of perception constraints. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0060] Figure 1 It is a schematic diagram of the steps of a communication, perception and control integrated design method for an unmanned aerial vehicle platform of the present invention;

[0061] Figure 2 It is the planning trajectory diagram obtained by the three schemes of the present invention;

[0062] Figure 3 It is the planning trajectory diagram of P0 and P1 of the present invention;

[0063] Figure 4 is the actual trajectory of P0 and P1 of the present invention;

[0064] Figure 5 is a graph showing how the gain of the receiving beam pattern changes over time under different trajectories of the present invention;

[0065] Figure 6 It is a comparison curve of average data throughput under different receiving beam pattern gain thresholds of the present invention. DETAILED DESCRIPTION

[0066] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0067] The following describes the embodiments of the present invention through specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention.

[0068] Embodiment 1:

[0069] A communication, perception and control integrated design method for UAV platforms includes the following steps:

[0070] S101. When the drone transmits communication signals to the user using transmit beamforming, it also sends dedicated radar signals to improve communication and perception performance.

[0071] Radar perception of potential targets, while providing communication services for multiple single-antenna users. Assume that the drone needs to fly from a predetermined initial position to a final position within a limited time range T∈[0,T]; the drone's position vector is p(t)=[x(t),y(t),z(t)] · ,in,[·] · represents the transpose of the matrix; the flight altitude of the drone is z u ; The set of perceived targets in the region of interest is J∈{1,...,J}, and the position of the jth target is o j =[o x,j ,o y,j ,0] · . Assume that there are M users, the user set is represented by M∈{1,...,M}, and the position of the mth user is p m =[p x,m ,p y,m ,0] · .

[0072] The process of transmitting communication signals to users by using transmit beamforming while simultaneously sending dedicated radar signals to improve communication and perception performance includes the following steps:

[0073] Determine the channel vector from the UAV to the mth user based on the transmit array response vector and the channel power gain;

[0074] Determine the signal received at the th user based on the channel vector, the transmitted signal of the drone and the additive white Gaussian noise;

[0075] determining a signal-to-noise ratio based on the channel vector and the transmit beamforming vector, thereby determining a spectral efficiency;

[0076] Based on the transmit beam pattern gain and the perceived distance d(p(t),o j ), determine the perceived performance indicators.

[0077] When the drone transmits communication signals to the user using transmit beamforming, it also sends dedicated radar signals to improve communication and perception performance. The expression is as follows:

[0078]

[0079] Among them, c(t) is the signal emitted by the UAV at time t, w m is the transmit beamforming vector of the mth user, c m is the communication signal, c 0 It is a special radar signal.

[0080] It should be noted that the drone transmits dedicated radar signals and communication signals to users, where the drone uses transmit beamforming to send communication signals to customers. For example, the drone uses transmit beamforming to send communication signals to the mth user at time t. Although communication signals can be used for perception, the performance of perception will be limited. Therefore, by designing a dedicated radar signal c 0 To further improve communication and perception performance. Setting communication signals is independent, that is, c m (t)~CN(0,1), while the dedicated radar signal has zero mean and covariance matrix in(·) H represents the matrix conjugate transpose, and ± represents positive semidefinite. In addition, the communication signal is uncorrelated with the dedicated radar signal, that is, E(c 0 c m )=0 N×1 .

[0081] The expression for the total transmission power of the drone is as follows:

[0082]

[0083] Where E is the total transmission power of the UAV, ‖·‖ is the vector Frobenius norm, tr(·) is the matrix trajectory, and w m is the transmit beamforming vector of the mth user, G d is the covariance matrix; constrains the total transmission power of the UAV, Pmax is the maximum communication power.

[0084] The channel link between the UAV and the ground user is dominated by the line-of-sight link, and the Doppler effect caused by the UAV's maneuverability is fully compensated at both the user and the target. Therefore, the air-to-ground channel is considered to follow the free space path loss model. The channel power gain from the UAV to the mth user is expressed as β m (p(t),p m ), whose expression is as follows:

[0085]

[0086] Among them, β 0 is the channel power gain at a distance of 1 meter, α is the path loss exponent, p(t)=[x(t),y(t),z u ] · is the position vector of the UAV at time t, is the distance between the UAV and the mth user at time t, and x(t) and y(t) are the components of the UAV on the x-axis and y-axis.

[0087] The transmitting array response vector of the UAV in the direction of the mth user is expressed as b(p(t),p m ), whose expression is as follows:

[0088]

[0089] Where N represents the number of antennas of the uniform linear array installed on the UAV, d = λ / 2 represents the distance between adjacent antenna elements, and λ represents the carrier wavelength. is the departure angle of the signal from the drone to the user.

[0090] The channel vector from the drone to the mth user is denoted as g m (p(t)), which is expressed as follows:

[0091]

[0092] Among them, b(p(t),p m ) is the transmitting array response vector of the UAV in the direction of the mth user, β m (p(t),p m ) is the channel power gain from the UAV to the mth user.

[0093] The signal received at the mth user is denoted as s m (t), whose expression is as follows:

[0094]

[0095] in, represents the additive white Gaussian noise at the mth user.

[0096] The signal-to-noise ratio of the mth user is denoted as φ m (p(t),{w i (t)},G d (t)), whose expression is as follows:

[0097]

[0098] Among them, w i is the transmit beamforming vector of the i-th user, G d is the covariance matrix, p(t) is the position vector of the drone at time t, G d is the covariance matrix, is the noise power of the mth user, g m The conjugate transpose of m is the channel vector from the drone to the mth user.

[0099] The spectral efficiency (data rate per unit bandwidth) achievable by the mth user is expressed as follows:

[0100]

[0101] Among them, the signal-to-noise ratio is φ m (p(t),{w i (t)},G d (t)).

[0102] Consider the radar sensing service provided by the drone. In order to improve the sensing performance, the user's communication signal can be used to estimate the target parameters. The power of the sensing signal directed to the target j is called the transmit beam pattern gain, and its expression is as follows:

[0103]

[0104] Among them, G d is the covariance matrix, w m is the transmit beamforming vector of the mth user, p(t) is the position vector of the drone at time t, b(p(t),o j ) is the channel vector from the UAV to the jth target, p(t) is the position of the UAV at time, o j is the position of the jth target, b H (p(t),o j ) is b(p(t),o j ) is the conjugate transpose of .

[0105] Due to path loss, the beam pattern gain received by the drone depends on d(p(t),o j ), which is used as the perceptual performance indicator and expressed as:

[0106]

[0107] Among them, G d is the covariance matrix, w m is the transmit beamforming vector of the mth user, d(p(t),o j ) is the sensing distance from the UAV to the jth target, p(t) is the position of the UAV at time, o j is the position of the jth target.

[0108] S102, based on the UAV direction vector, the total lift of the UAV and the six-degree-of-freedom model, construct and optimize the six-degree-of-freedom model of the UAV horizontal flight to maximize the average weighted communication and transmission rate;

[0109] The constraint optimization library includes: drone trajectory constraints, flight speed constraints, transmit power limits, and beam pattern gain constraints. The drone is considered a rigid body and is operated by controlling the speed of its propellers. For example, vertical motion is achieved by increasing or decreasing the speed of four propellers simultaneously. The direction vector of the drone is represented by Φ(t) = [ζ(t), η(t), γ(t)] · The lift is proportional to the square of the propeller speed and is expressed as:

[0110]

[0111] Among them, ξ i (t) represents the speed of the i-th propeller, L i (t) represents the lift generated by the i-th propeller, K p represents the lift coefficient.

[0112] Since the drone is flying horizontally, the total lift acting on it can be deduced as:

[0113]

[0114] Where ζ(t) represents the roll angle, η(t) represents the pitch angle, and m a represents the mass of the drone, and g represents the acceleration due to gravity.

[0115] Based on the UAV direction vector, the total lift of the UAV and the six-degree-of-freedom model, the six-degree-of-freedom model of the UAV horizontal flight is constructed and optimized to maximize the communication average weighted and transmission rate, including the following steps:

[0116] Construct six-degree-of-freedom models based on translational dynamics based on Lagrange-Euler equations and rotational dynamics based on Newton-Euler equations;

[0117] Based on the six-degree-of-freedom model, the direction vector of the UAV and the total lift of the UAV, a six-degree-of-freedom model of the UAV horizontal flight is constructed;

[0118] The optimization is performed by building a constrained optimization library to maximize the communication average weighted and transmission rate.

[0119] The six-degree-of-freedom model can be constructed through the translational dynamics of the Lagrange-Euler equation and the rotational dynamics based on the Newton-Euler equation. The expression of the six-degree-of-freedom model of the horizontal flight of the drone is as follows:

[0120]

[0121] Where, γ(t) represents the yaw angle, I xx , I yy and I zz represents the moment of inertia on the x-axis, y-axis and z-axis, l represents the length of the rigid body crossbeam, K dx and K dy Represents the drag coefficient on the x-axis and y-axis, K dmx , K dmy and K dmz represents the damping moment coefficient in the x-axis, y-axis and z-axis, K m Represents the moment coefficient, Λ(t)=-ξ 1 (t)+ξ 2 (t)-ξ 3 (t)+ξ 4 (t) and sign(b) represent the sign of b. For the sake of simplicity, the six-degree-of-freedom model of horizontal flight can be abbreviated as M(p(t), Φ(t))=0.

[0122] Under the premise of considering the UAV dynamics model, perception requirements and transmission power constraints, the average weighted sum of communication rates can be maximized by optimizing the UAV trajectory, communication and perception beamforming vectors. Therefore, based on the UAV direction vector, the total lift of the UAV and the six-degree-of-freedom model, the six-degree-of-freedom model of the UAV horizontal flight is constructed and optimized to maximize the expression of the average weighted sum of communication transmission rates as follows:

[0123]

[0124] sC 1 :M(p(t),Φ(t))=0,

[0125]

[0126] C 4:p(0)=p I ,

[0127] C 5 :p(T)=p F ,

[0128]

[0129] Among them, R m (t) represents the spectrum efficiency achievable by the mth user, and the constraint C 1 Represents the six-degree-of-freedom model of the drone, with constraint C 2 represents the gain constraint of the UAV receiving beam pattern, where is the beam pattern gain threshold of the jth target, and the constraint C 3 Indicates the flight speed limit of the drone, where V max represents the maximum flight speed, constraint C 4 and C 5 They represent the initial position and terminal position constraints respectively, and the initial position vector is p I =[x I ,y I ,z u ] · and the terminal position vector is p F =[x F ,y F ,z u ] · , constraint C 6 Indicates the UAV transmit power limit.

[0130] S103, performing convex optimization processing on the communication and perception beamforming vectors;

[0131] Since there is a strong coupling relationship between the UAV trajectory points and the beamforming vectors, the problem can be solved by adopting an alternating optimization strategy to optimize the communication and perception beamforming vectors.

[0132] The convex optimization of the communication and sensing beamforming vectors consists of the following steps:

[0133] Based on the communication beamforming vector, user weight, user achievable spectrum efficiency, perception covariance matrix, drone receive beam pattern gain constraint and drone transmit power limit constraint, the optimization problem of communication and perception beamforming vector is constructed;

[0134] Discretize the time interval T into P equal subintervals and p+1 time slots, and decouple the optimization problem at different times;

[0135] The non-convex rank constraints are handled by approximation techniques and semi-definite relaxation methods, and the communication and sensing beamforming vectors are obtained after convex optimization.

[0136] Based on the communication beamforming vector, user weight, user achievable spectrum efficiency, perception covariance matrix, drone receiving beam pattern gain constraint and drone transmit power limit constraint, the optimization problem of communication and perception beamforming vector is constructed, and its expression is as follows:

[0137]

[0138] sC 2 ,C 6 .

[0139] Among them, ρ m represents the weight of the mth user, R m represents the spectrum efficiency achievable by the mth user, w m (t) is the communication beamforming vector, G d (t) is the perceptual covariance matrix.

[0140] In order to convert the above subproblem into a tractable form, the time interval T is discretized into P equal subintervals and P+1 time slots. The expression of P+1 time slots is as follows:

[0141] 0 = τ 0 <τ 1 <τ 2 <…<τ P-1 <τ P =T.

[0142] According to the gain constraint of the receiving beam pattern of the UAV and the transmission power limit constraint of the UAV, it can be known that the optimization variables {w m [τ n ]} and G d [τ n ] are independent. Therefore, the problem is decoupled at different times and the problem is equivalently decomposed into P sub-problems. At time τ n The optimization problem at is:

[0143]

[0144] In order to solve the above problem, we define Thus, the constraint rank (W m [τ n ])≤1 and W m [τ n ]±0, and then use the sequential convex approximation technique to approximate the objective function as a concave function, which is expressed as follows:

[0145]

[0146] in and Indicates {W m [τ n ]} and G d [τ n ] is the local point at the kth iteration.

[0147] Therefore, the non-convex rank constraint rank (W) is handled by approximation techniques and semi-definite relaxation methods. m [τ n ])≤1, the optimization subproblem is transformed into:

[0148]

[0149] Obviously, the above problem is convex and can be solved efficiently by using the CVX solution package, and the rank constraint rank(W m [τ n ])≤1 is always guaranteed to exist, so the solution to the original subproblem can be obtained by solving the above problem. and Represents the optimal solution obtained, thereby constructing the optimal and Therefore, by using the approximation technique and the semi-definite relaxation method to handle the non-convex rank constraints, the expressions of the communication and sensing beamforming vectors after convex optimization are obtained as follows:

[0150]

[0151] Among them, w m is the communication beamforming vector, G d is the covariance matrix, G d is the covariance matrix, g m The conjugate transpose of m is the channel vector from the drone to the mth user, and p is the position of the drone.

[0152] S104, performing gradient optimization processing on the trajectory of the UAV;

[0153] Based on the beam pattern gain constraint, flight speed constraint, drone trajectory constraint, user weight, and user-achievable spectrum efficiency, the drone trajectory optimization problem is constructed, and its expression is as follows:

[0154]

[0155] sC 1 -C 5 .

[0156] Among them, ρ m represents the weight of the mth user, R m represents the spectrum efficiency achievable by the mth user, p(t) is the trajectory of the drone, and the constraint C 1 Represents the six-degree-of-freedom model of the drone, with constraint C 2 represents the gain constraint of the UAV receiving beam pattern, where is the beam pattern gain threshold of the jth target, and the constraint C 3 Indicates the flight speed limit of the drone, where V max represents the maximum flight speed, constraint C 4 and C 5 They represent the initial position and terminal position constraints respectively, and the initial position vector is p I =[x I ,y I ,z u ] · and the terminal position vector is p F =[x F ,y F ,z u ] · , constraint C 6 Indicates the UAV transmit power limit.

[0157] In order to obtain a high-quality feasible solution, the above problem can be reformulated as an optimal control problem based on a state-space model. Then, the continuous-time control vector is discretized by adopting a control parameterization method. Furthermore, an exact penalty function method is used to handle continuous state inequality constraints. Based on this, an efficient gradient-based UAV trajectory optimization algorithm is designed.

[0158] Gradient optimization of the UAV trajectory includes the following steps:

[0159] Based on the beam pattern gain constraint, flight speed constraint, UAV trajectory constraint, user weight, and user-achievable spectrum efficiency, the UAV trajectory optimization problem is constructed;

[0160] Convert the UAV trajectory optimization problem into a control problem based on the state space model;

[0161] The continuous-time control vector is discretized using the control parameterization method;

[0162] The exact penalty function method is used to transform the constrained nonlinear programming problem into an unconstrained optimization problem to optimize the UAV trajectory.

[0163] The UAV trajectory optimization problem is transformed into a control problem based on the state space model. Combined with the physical meaning of the variables in the six-degree-of-freedom model, the state vector is defined as:

[0164]

[0165] The control variables are defined as:

[0166]

[0167] Therefore, the control vector is expressed as u(t)=[u 1 (t),u 2 (t),u 3 (t)] · Based on the control vector and state vector, the six-degree-of-freedom model can be simplified as Therefore, the UAV trajectory optimization problem is transformed into an optimal control problem and can be expressed as:

[0168]

[0169] C 6 ':x 1 (T) = x F ,x 2 (T) = y F ,

[0170] in, Considering practical factors, the constraint C' is introduced 2 Limit the maneuverability of the drone. In addition, constrain C' 5 Provides a solution for C 1 The remaining constraints are obtained by reformulating the constraints in the original problem in terms of x(t) and u(t).

[0171] By using the control parameterization method to discretize the continuous-time control vector, the control variable u i The parameterized function of (t), i = 1, 2, 3 is expressed as in is the control variable u i (t) in the time interval [τ n-1 ,τ n ), is defined as follows:

[0172]

[0173] make Constraint C 1 ' can be written as By replacing u(t) with Constraint C' 2 and C 3 ' is rewritten as:

[0174]

[0175] in, is the control variable u i (t) in the time interval [τ n-1 ,τ n ),

[0176] The exact penalty function method is used to transform the constrained nonlinear programming problem into an unconstrained optimization problem to optimize the trajectory of the UAV. In order to handle the countless state inequality constraints C 3 'and C' 4 , the exact penalty function method is used to incorporate them into the objective function, thereby transforming the constrained nonlinear programming problem into an unconstrained optimization problem. The objective function is redefined using the exact penalty function method as follows:

[0177]

[0178] in, For the new variables introduced, The original objective function is expressed as

[0179]

[0180] in, g m The conjugate transpose of g m is the channel vector from the drone to the mth user, w m is the transmit beamforming vector of the mth user, φ m is the signal-to-noise ratio received by the mth user, is the noise power of the mth user.

[0181] is a continuous constraint violation, are the terminal constraint violations, respectively expressed as

[0182]

[0183] Among them, μ>0 represents the penalty parameter, k, θ, ò, O are given constants satisfying κ>0, θ>2, ò>0, O∈(0,1).

[0184] It should be noted that in a given system C 1 'and C' 5 The original problem is transformed into a static nonlinear programming problem with only the box constraint C” 2 and Once the objective function is obtained about and The gradient of , the transformation problem can be solved.

[0185] For example, the present invention considers a situation in two-dimensional space where there is one drone, eight users, and an area of ​​1600m 2 The user's position is p 1 =[370m,400m],p 2 =[380m,345m],p 3 =[420m,300m],p 4 =[470m,275m],p 5 =[530m,275m],p 6 =[580m,300m],p 7 =[620m,345m] and p 8 =[630m,400m]. Consider a matrix sensing area with a length of 80m and a width of 20m, with a total of J = 18 sensing points. In order to prove the superiority of the proposed scheme, the problem in the literature [3] without considering the dynamic model is compared as a benchmark scheme, and this problem is regarded as P0. In addition, under the framework of the present invention, the problem considering the three-degree-of-freedom model is regarded as P1, and the problem studied by the present invention is regarded as P2.

[0186] Solve P0, P1 and P2 to obtain the planned trajectory as Figure 2 As shown in Figure 2, the planning trajectories of the three schemes are very similar. During the flight, the UAV continuously strives to get closer to the user to obtain more communication throughput. However, the UAV cannot reach the user location because it needs to maintain a suitable sensing distance from the sensing area to meet the requirements of the sensing beam pattern gain.

[0187] The actual trajectories of P0 and P1 are obtained by designing a proportional-integral-derivative controller with a six-degree-of-freedom model to track the planned trajectory, as shown in Figure 3 and Figure 4 As shown. The actual trajectory of P0 cannot reach the destination, while the actual trajectory of P1 can reach the destination. This is because the planned trajectory of P0 does not take into account the dynamic characteristics of the UAV, resulting in the control input required for the planned trajectory exceeding the execution capability of the UAV system. Therefore, the planned trajectory of P0 cannot be tracked. Although the actual trajectory of P1 can complete the main task as required, it also cannot accurately follow the planned trajectory because the six-degree-of-freedom model is not considered. This illustrates the importance of considering the six-degree-of-freedom model in the trajectory planning process, and also emphasizes that the planned trajectory of P2 is the actual flight trajectory of the UAV.

[0188] The change of beam pattern gain received by the UAV at the sensing point [500m, 600m] over time, such as Figure 5As shown in Figure 2. In all trajectories, the gain of the receiving beam pattern of the drone first decreases and then increases. This is because as the drone gets closer to the user, the perception distance between the drone and the perception position gradually increases, resulting in a gradual decrease in perception performance. Subsequently, as the drone flies toward the destination, the perception distance gradually decreases, thereby improving the perception performance. In addition, it can be observed that although the beam pattern gain of the planned trajectories of P0, P1, and P2 is always greater than the predetermined threshold Θ th , but the beam pattern gain of the actual trajectory of P0 and P1 is less than Θ th The moments of , indicating that the actual trajectory violates the perception performance constraints at these moments. This is because the actual trajectory of the drone deviates from the planned trajectory, resulting in the actual trajectory points not satisfying the perception performance constraints. Compared with scheme P1, the actual trajectory of P0 deviates greatly from the planned trajectory, so the perception constraint violations of the actual trajectory occur more often.

[0189] Average rate and receiving beam pattern gain threshold Θ under different trajectories th The relationship is as Figure 6 As Θ th As increases, the average rate of each scheme decreases. This is because the UAV needs more transmit power to meet higher perception requirements, so that only less transmit power is used for communication. In addition, it can be observed that the communication performance of the planned trajectories of P0 and P1 is worse than that of the planned trajectory of P2. However, the communication performance of the actual trajectories of P0 and P1 is not as good as their respective planned trajectories, and is also lower than that of the planned trajectory of P2. The performance degradation of P0 and P1 is mainly due to the lack of UAV six-degree-of-freedom dynamics. Therefore, the superiority of the proposed scheme P2 is obvious.

[0190] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0191] The terms "first", "second" and "third" etc. in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in a sequence other than those illustrated or described herein, for example. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0192] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A communication, perception and control integrated design method for UAV platforms, characterized in that: The following steps are involved: S101. When the drone transmits communication signals to the user using transmit beamforming, it also sends dedicated radar signals to improve communication and perception performance. S102, based on the UAV direction vector, the total lift of the UAV and the six-degree-of-freedom model, construct and optimize the six-degree-of-freedom model of the UAV horizontal flight to maximize the average weighted communication and transmission rate; S103, performing convex optimization processing on the communication and perception beamforming vectors; S104: performing gradient optimization processing on the trajectory of the UAV.

2. The communication, perception and control integrated design method for UAV platforms according to claim 1 is characterized in that: The UAV uses transmit beamforming to transmit communication signals to the user, and at the same time sends a dedicated radar signal to improve communication and perception performance, including the following steps: Determine the channel vector from the UAV to the mth user based on the transmit array response vector and the channel power gain; Determine the signal received at the th user based on the channel vector, the transmitted signal of the drone and the additive white Gaussian noise; determining a signal-to-noise ratio based on the channel vector and the transmit beamforming vector, thereby determining a spectral efficiency; Based on the transmit beam pattern gain and the perceived distance d(p(t),o j ), determine the perceived performance indicators.

3. The communication, perception and control integrated design method for UAV platforms according to claim 2 is characterized in that: The transmit beam pattern gain and the sensing distance d(p(t),o j ), determine the perceptual performance index, which is expressed as follows: Among them, G d is the covariance matrix, w m is the transmit beamforming vector of the mth user, d(p(t),o j ) is the distance between the UAV and the jth target at time t, p(t) is the position of the UAV at time t, o j is the position of the jth target.

4. The communication, perception and control integrated design method for UAV platforms according to claim 1 is characterized in that: The method of constructing and optimizing the six-degree-of-freedom model of the drone horizontal flight based on the drone direction vector, the total lift of the drone and the six-degree-of-freedom model to maximize the average weighted communication and transmission rate includes the following steps: Construct six-degree-of-freedom models based on translational dynamics based on Lagrange-Euler equations and rotational dynamics based on Newton-Euler equations; Based on the six-degree-of-freedom model, the direction vector of the UAV and the total lift of the UAV, a six-degree-of-freedom model of the UAV horizontal flight is constructed; The optimization is performed by building a constrained optimization library to maximize the communication average weighted and transmission rate.

5. The communication, perception and control integrated design method for UAV platforms according to claim 4 is characterized in that: Based on the six-degree-of-freedom model, the direction vector of the drone and the total lift of the drone, a six-degree-of-freedom model of the drone horizontal flight is constructed. The expression of the six-degree-of-freedom model of the drone horizontal flight is as follows: Where, γ(t) represents the yaw angle, I xx , I yy and I zz represents the moment of inertia on the x-axis, y-axis and z-axis, l represents the length of the rigid body crossbeam, K dx and K dy Represents the drag coefficient on the x-axis and y-axis, K dmx , K dmy and K dmz represents the damping moment coefficient in the x-axis, y-axis and z-axis, K m represents the moment coefficient, Λ(t)=-ξ1(t)+ξ2(t)-ξ3(t)+ξ4(t) and sign(b) represents the sign of b. For simplicity, the six-degree-of-freedom model of horizontal flight can be abbreviated as M(p(t),Φ(t))=0.

6. The communication, perception and control integrated design method for UAV platforms according to claim 4 is characterized in that: Based on the UAV direction vector, the total lift of the UAV and the six-degree-of-freedom model, the six-degree-of-freedom model of the UAV horizontal flight is constructed and optimized to maximize the expression of the average weighted communication and transmission rate as follows: stC1:M(p(t),Φ(t))=0, C4:p(0)=p I , C5:p(T)=p F , Among them, R m (t) represents the spectrum efficiency achievable by the mth user, constraint C1 represents the six-degree-of-freedom model of the UAV, and constraint C2 represents the gain constraint of the receiving beam pattern of the UAV. is the beam pattern gain threshold of the jth target, and constraint C3 represents the flight speed limit of the UAV, where V max represents the maximum flight speed, constraints C4 and C5 represent the initial position and terminal position constraints respectively, and the initial position vector is p I =[x I ,y I ,z u ] · and the terminal position vector is p F =[x F ,y F ,z u ] · ,Constraint C6 represents the UAV transmission power limit.

7. The communication, perception and control integrated design method for UAV platforms according to claim 1 is characterized in that: The convex optimization of the communication and sensing beamforming vectors comprises the following steps: Based on the communication beamforming vector, user weight, user achievable spectrum efficiency, perception covariance matrix, drone receive beam pattern gain constraint and drone transmit power limit constraint, the optimization problem of communication and perception beamforming vector is constructed; Discretize the time interval T into P equal subintervals and p+1 time slots, and decouple the optimization problem at different times; The non-convex rank constraints are handled by approximation techniques and semi-definite relaxation methods, and the communication and sensing beamforming vectors are obtained after convex optimization.

8. The communication, perception and control integrated design method for UAV platforms according to claim 7 is characterized in that: The non-convex rank constraints are processed by the approximation technique and the semi-positive definite relaxation method, and the expressions of the communication and sensing beamforming vectors after convex optimization are obtained as follows: Among them, w m is the communication beamforming vector, G d is the covariance matrix, G d is the covariance matrix, g m The conjugate transpose of m is the channel vector from the drone to the mth user, and p is the position of the drone.

9. The communication, perception and control integrated design method for UAV platforms according to claim 1 is characterized in that: The gradient optimization process for the UAV trajectory comprises the following steps: Based on the beam pattern gain constraint, flight speed constraint, UAV trajectory constraint, user weight, and user-achievable spectrum efficiency, the UAV trajectory optimization problem is constructed; Convert the UAV trajectory optimization problem into a control problem based on the state space model; The continuous-time control vector is discretized using the control parameterization method; The exact penalty function method is used to transform the constrained nonlinear programming problem into an unconstrained optimization problem to optimize the UAV trajectory.

10. The communication, perception and control integrated design method for UAV platforms according to claim 9 is characterized in that: The expression for converting the UAV trajectory optimization problem into a control problem based on the state space model is as follows: C6′:x1(T)=x F ,x2(T)=y F , in, Considering practical factors, constraint C'2 is introduced to limit the maneuverability of the UAV. In addition, constraint C'5 provides the initial state vector necessary to solve the differential equation in C1'. The remaining constraints are obtained by reformulating the constraints in the original problem in the form of x(t) and u(t).

Images