CN112596549B - Multi-UAV formation control method, device and medium based on continuous convex rule - Google Patents

Abstract

The invention discloses a multi-UAV formation control method, device and medium based on a continuous convex rule. The method includes: determining a global path planning of the UAV formation; the global path planning is used to determine the UAV formation The action path and formation formation from the starting point to the end point; according to the linearized and discretized UAV dynamics constraints, the collision avoidance constraints of the UAV formation are converted into convex constraints; the unmanned aerial vehicle is determined according to the convex constraints The local path planning of the UAV formation; the local path planning is used to determine the action trajectory of each UAV in the UAV formation; real-time path tracking control is performed on the UAV formation through a preset model. The invention can decide the corresponding formation according to the specific environment, can ensure the expansibility and stability of the formation quantity, prevent formation deadlock, and can be widely used in the technical field of unmanned aerial vehicle control.

Description

Multi-UAV formation control method, device and medium based on continuous convex rule

technical field

The invention relates to the technical field of unmanned aerial vehicle control, in particular to a method, device and medium for controlling a formation of multiple unmanned aerial vehicles based on continuous convex rules.

Background technique

Swarm drones can be used to perform a variety of tasks, such as surveillance, inspection, and automation of factories. In these situations, drones may be required for formation navigation, such as maintaining communication networks, collaboratively manipulating objects, or surveying areas. Mobile formation has many advantages over conventional systems, for example, it can reduce system cost, improve system robustness and efficiency, while providing system redundancy, reconfiguration capability and structural flexibility. In many applications, swarm UAVs are required to avoid collision with each other and avoid collisions with obstacles in the process of reaching the target point, while maintaining the required formation formation, so the accuracy and robustness of the formation control method and the expansion of Sexual requirements are high. In addition, the formation control method needs to take into account the kinematic constraints of UAVs, the coordination between UAVs, and the uncertain interference of the environment.

At present, the relatively mature and common formation control algorithms include the leader-wingman method, the behavior-based method, and the virtual structure method.

Lead-wingman method. At least one drone takes the role of leader, and the rest are designated as followers. The follower keeps track of the leader's position with some prescribed offset, and the leader keeps track of the trajectory it wants. The characteristics of this control strategy are based on a preset formation structure. The wingman is adjusted by tracking the speed, yaw angle and altitude of the leader to maintain the formation.

Based on conduct law. A kind of formation control method that simulates biological reactive behavior mechanism. In the process of multi-UAV formation flight, the behavior response of each UAV to its sensor input information may have four situations: collision avoidance, obstacle avoidance, target acquisition and formation keeping. The biggest feature of this method is to use the average weight of behavioral response control to determine which behavioral response method should be adopted by each UAV in the formation.

Virtual structure method. In the virtual structure approach, the entire formation is treated as a single structure. In the virtual structure approach, control is divided into three steps: first, the desired dynamics of the virtual structure are defined, second, the motion of the virtual structure is transformed into the period motion of each UAV, and finally, the dynamics of each spacecraft are derived tracking control. The virtual structure method controls the formation by sharing the state information of the virtual structure of the formation.

However, the lead-wingman approach has no explicit feedback on the formation, for example, the lead may move too fast for subsequent drones to track. And if there is a problem with the lead plane, the entire system will collapse. Another weakness is that the poor robustness is easily affected by external interference, because under the influence of external or other interference, the error will be propagated backward and amplified step by step.

The main disadvantage of the behavior-based approach is that the group behavior cannot be clearly defined, and the group behavior is called "emergence", which results in the inability to achieve precise formation hold. Another disadvantage is that behavioral methods are difficult to analyze mathematically and often do not guarantee formation characteristics (such as stability).

The virtual structure method regards the formation of the UAV as a rigid virtual structure. During the flight movement of the UAV in formation, a single UAV can be regarded as a fixed position fixed on the virtual structure. As a result, when the formation passes through a narrow area, it cannot pass through the area by stretching or switching formations. Moreover, the virtual mechanism method requires high communication quality, so it is easily restricted by communication.

SUMMARY OF THE INVENTION

In view of this, the embodiments of the present invention provide a multi-UAV formation control method, device and medium based on continuous convex rules with high stability, which can improve the scalability of the number of UAV formations.

One aspect of the present invention provides a multi-UAV formation control method based on continuous convex rules, comprising:

Determine the global path planning of the UAV formation; the global path planning is used to determine the action path and formation formation of the UAV formation from the starting point to the end point;

Transform the collision avoidance constraints of the UAV formation into convex constraints according to the linearized and discretized UAV dynamics constraints;

Determine the local path planning of the UAV formation according to the convex constraint; the local path planning is used to determine the action trajectory of each UAV in the UAV formation;

Real-time path tracking control is performed on the UAV formation through a preset model.

Preferably, the determining of the global path planning of the UAV formation includes:

determining a plurality of drone positions of the drone formation, and determining the outer vertices of the center of rotation of the drone formation;

determining the minimum distance between any pair of UAVs in the UAV formation;

Determine the initial configuration of the UAV formation by isomorphic transformation according to the UAV position, the outer vertex and the minimum distance;

According to the initial configuration of the UAV formation, the target configuration of the UAV formation is determined by a global path planner.

Preferably, according to the initial configuration of the UAV formation, the target configuration of the UAV formation is determined by a global path planner, including:

Build a list of polyhedrons, including all drones in the drone formation in the list of polyhedrons;

Initialize the initial configuration and target configuration of the UAV formation;

Initialize the polyhedron list through the convex region;

draw random samples in the flight space of the drone formation;

Determine whether the random sample is in the obstacle or in the polyhedron list, if so, exclude the random sample; otherwise, perform the following steps:

calculating an unobstructed convex polyhedron of the random sample by an iterative algorithm;

calculating the inscribed ellipsoid of the barrier-free convex polyhedron;

According to the inscribed ellipsoid, the action path of the UAV formation from the starting point to the ending point is determined.

Preferably, in the step of converting the collision avoidance constraint of the UAV formation into a convex constraint according to the linearized and discretized UAV dynamics constraints, the UAV dynamics constraints are:

代表第j只无人机的位置矢量；u_j代表第j只无人机的控制矢量；B＝[0_3×3 I_3×3]，t表示时间；in,

represents the position vector of the jth UAV; u _j represents the control vector of the jth UAV; B=[0 _3×3 I _3×3 ], t represents the time;

The linearized expression is:

表示标称轨迹，即上一次迭代生成的轨迹；

in,

represents the nominal trajectory, that is, the trajectory generated by the previous iteration;

Preferably, the determining the local path planning of the UAV formation according to the convex constraint includes:

Generate initial trajectories for each drone without considering collision avoidance;

According to the initial trajectory, iteratively solve the optimal trajectory of the UAV;

According to the optimal trajectory of each UAV, the optimal trajectory of all UAVs in the UAV formation is determined.

The embodiment of the present invention also provides a multi-UAV formation control device based on continuous convex rules, including:

The global planning module is used to determine the global path planning of the UAV formation; the global path planning is used to determine the action path and formation formation of the UAV formation from the starting point to the end point;

a transformation module, configured to transform the collision avoidance constraint of the UAV formation into a convex constraint according to the linearized and discretized UAV dynamics constraints;

a local planning module, configured to determine the local path planning of the UAV formation according to the convex constraint; the local path planning is used to determine the action trajectory of each UAV in the UAV formation;

A tracking control module is used to perform real-time path tracking control on the UAV formation through a preset model.

Preferably, the global planning module includes:

a first determining unit, configured to determine the positions of a plurality of UAVs of the UAV formation, and determine the outer vertex of the rotation center of the UAV formation;

a second determining unit, configured to determine the minimum distance between any pair of UAVs in the UAV formation;

an isomorphic transformation unit, configured to determine the initial configuration of the UAV formation by isomorphic transformation according to the UAV position, the external vertex and the minimum distance;

The third determining unit is configured to determine the target configuration of the UAV formation through a global path planner according to the initial configuration of the UAV formation.

Preferably, the third determining unit includes:

constructing a subunit for establishing a list of polyhedrons, including all drones in the drone formation in the list of polyhedrons;

a first initialization subunit, used for initializing the initial configuration and target configuration of the UAV formation;

a second initialization subunit, configured to initialize the polyhedron list through the convex region;

a random sampling subunit, used for sampling random samples in the flight space of the drone formation;

A judging subunit, used for judging whether the random sample is in the obstacle or in the polyhedron list, if yes, then the random sample is excluded; otherwise, the following steps are performed:

a first calculation subunit, configured to calculate the barrier-free convex polyhedron of the random sample through an iterative algorithm;

a second calculation subunit for calculating the inscribed ellipsoid of the barrier-free convex polyhedron;

A determination subunit, configured to determine the action path of the UAV formation from the start point to the end point according to the inscribed ellipsoid.

The embodiment of the present invention also provides an electronic device, including a processor and a memory;

the memory is used to store programs;

The processor executes the program to implement the method as described above.

An embodiment of the present invention further provides a computer-readable storage medium, where the storage medium stores a program, and the program is executed by a processor to implement the foregoing method.

The embodiment of the present invention also discloses a computer program product or computer program, where the computer program product or computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. The computer instructions can be read from the computer-readable storage medium by a processor of the computer device, and the processor executes the computer instructions to cause the computer device to perform the foregoing method.

The embodiment of the present invention firstly determines the global path planning of the UAV formation; then converts the collision avoidance constraints of the UAV formation into convex constraints according to the linearization and discretization UAV dynamics constraints; and then according to the convex constraints Constraints determine the local path planning of the UAV formation; the local path planning is used to determine the action trajectory of each UAV in the UAV formation; finally, the UAV formation is carried out through a preset model. Real-time path tracking control. The invention can decide the corresponding formation according to the specific environment, and can ensure the expansibility and stability of the formation quantity, and prevent formation deadlock.

Description of drawings

In order to illustrate the technical solutions in the embodiments of the present application more clearly, the following briefly introduces the drawings that are used in the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can also be obtained from these drawings without creative effort.

FIG. 1 is a flowchart of steps of a method for controlling a formation of multiple UAVs based on a continuous convex rule provided by an embodiment of the present invention.

Detailed ways

In order to make the purpose, technical solutions and advantages of the present application more clearly understood, the present application will be described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, but not to limit the present application.

In view of the problems existing in the prior art, the present invention proposes a scalable and efficient multi-UAV formation control method based on continuous convex planning, which can be used to solve the problem of group UAVs adjusting the formation according to a specific environment and avoiding UAVs. Go to the target location without colliding with each other and avoiding the collision of the drone with obstacles. The invention firstly calculates a set of safe intermediate formations from the starting point to the ending point for the group UAV through global path planning. Then, on the basis of linearizing and discretizing UAV dynamics constraints and converting collision avoidance constraints into convex constraints, local path planning is carried out through continuous convex programming, and combined with model predictive control, the path following of group UAVs is realized in real time. . The technology belongs to the field of swarm UAV formation control, and more specifically, relates to a distributed adaptive formation control method for swarm UAVs based on continuous convex programming.

The present invention is completely based on distributed collaborative planning, and there is no so-called lead plane. When a single UAV (or lead plane) has a problem, it will not lead to the collapse of the entire system, and has stronger robustness. It will not cause a situation similar to the gradual expansion of the error of the leader-wingman method. At the same time, the invention has a clear definition for the formation, and can decide the corresponding formation according to the specific environment, and can ensure the (quantity) scalability and stability of the formation, and prevent formation deadlock. The purpose of the present invention is to achieve the above-mentioned effects

As shown in Figure 1, the method of the present invention comprises the following steps:

Determine the global path planning of the UAV formation; the global path planning is used to determine the action path and formation formation of the UAV formation from the starting point to the end point;

Transform the collision avoidance constraints of the UAV formation into convex constraints according to the linearized and discretized UAV dynamics constraints;

Determine the local path planning of the UAV formation according to the convex constraint; the local path planning is used to determine the action trajectory of each UAV in the UAV formation;

Real-time path tracking control is performed on the UAV formation through a preset model.

Preferably, the determining of the global path planning of the UAV formation includes:

determining a plurality of drone positions of the drone formation, and determining the outer vertices of the center of rotation of the drone formation;

determining the minimum distance between any pair of UAVs in the UAV formation;

Determine the initial configuration of the UAV formation by isomorphic transformation according to the UAV position, the outer vertex and the minimum distance;

According to the initial configuration of the UAV formation, the target configuration of the UAV formation is determined by a global path planner.

Preferably, according to the initial configuration of the UAV formation, the target configuration of the UAV formation is determined by a global path planner, including:

Build a list of polyhedrons, including all drones in the drone formation in the list of polyhedrons;

Initialize the initial configuration and target configuration of the UAV formation;

Initialize the polyhedron list through the convex region;

draw random samples in the flight space of the drone formation;

Determine whether the random sample is in the obstacle or in the polyhedron list, if so, exclude the random sample; otherwise, perform the following steps:

calculating an unobstructed convex polyhedron of the random sample by an iterative algorithm;

calculating the inscribed ellipsoid of the barrier-free convex polyhedron;

According to the inscribed ellipsoid, the action path of the UAV formation from the starting point to the ending point is determined.

Preferably, in the step of converting the collision avoidance constraint of the UAV formation into a convex constraint according to the linearized and discretized UAV dynamics constraints, the UAV dynamics constraints are:

代表第j只无人机的位置矢量；u_j代表第j只无人机的控制矢量；B＝[0_3×3 I_3×3]，t表示时间；in,

represents the position vector of the jth UAV; u _j represents the control vector of the jth UAV; B=[0 _3×3 I _3×3 ], t represents the time;

The linearized expression is:

表示标称轨迹，即上一次迭代生成的轨迹；

in,

represents the nominal trajectory, that is, the trajectory generated by the previous iteration;

Preferably, the determining the local path planning of the UAV formation according to the convex constraint includes:

Generate initial trajectories for each drone without considering collision avoidance;

According to the initial trajectory, iteratively solve the optimal trajectory of the UAV;

According to the optimal trajectory of each UAV, the optimal trajectory of all UAVs in the UAV formation is determined.

The implementation process of the method of the present invention is specifically described below:

The invention combines a sampling-based graph search algorithm to sample and connect convex regions in free space, and combines sampling and nonlinear optimization to find a safe global path. The invention proposes a scalable and efficient multi-UAV formation control method based on continuous convex programming. On the basis of linearizing and discretizing UAV dynamics constraints and converting collision avoidance constraints into convex constraints, the continuous Convex planning performs local path planning, and at the same time combines model predictive control to realize the path following of group UAVs in real time. The invention can be used to solve the problem that a group of UAVs adjusts the formation according to a specific environment and avoids collision between UAVs and avoids collision between UAVs and obstacles to go to the target position.

The specific implementation steps are as follows:

(1), global path planning:

和一系列相对于队形的旋转中心(通常是质心)的外部顶点

给定，其中n_f表示定义队形f的外部顶点的数量。这些顶点表示无人机在编队中的位置的凸包，从而降低了具有多个无人机的编队的复杂度。进一步用d_f表示编队f中任意给定一对无人机之间的最小距离。然后通过同构变换定义一个队形，包括大小s∈R₊，平移t∈R³和由单元四元数q∈SO(3)描述的旋转R(q)，其共轭由

表示。通过这种形式定义，无人机队形可完全由配置z＝[t,s,q]∈R³×R₊×SO(3)定义。给定配置z和队形f，可通过以下方式计算所得队形的无人机位置和外部顶点：First define the formation. Each formation f consists of a series (n) of UAV positions

and a series of outer vertices relative to the formation's center of rotation (usually the center of mass)

given, where n _f represents the number of outer vertices that define the formation f. These vertices represent the convex hull of the drone's position in the formation, reducing the complexity of a formation with multiple drones. We further use d _f to denote the minimum distance between any given pair of UAVs in the formation f. A formation is then defined by isomorphic transformation, including size s ∈ R ₊ , translation t ∈ R ³ and rotation R(q) described by the unit quaternion q ∈ SO(3), whose conjugate is given by

express. With this formal definition, the UAV formation can be completely defined by the configuration z=[t,s,q]∈R ³ ×R ₊ ×SO(3). Given a configuration z and a formation f, the drone positions and outer vertices of the resulting formation can be calculated by:

where the rotation matrix in SO(3) is given by the quaternion operation as follows:

For formation f and configuration z, the embodiment of the present invention uses the following formula to express the external vertex set:

In the description of the method, the embodiment of the present invention relies on this definition to form a formation, but the method is general and can be applied to other definitions, which is not limited in the present invention.

Given an initial configuration (z _s ) and a target configuration (z _g ) of the swarm drones, the global path planner computes a feasible path and intermediate formations connecting them. This is achieved by combining sampling-based methods with constrained nonlinear optimization, where the idea is to sample in a low-dimensional space (the workspace) and then let the optimizer compute the remaining degrees of freedom. In particular, embodiments of the present invention create a graph in which each node is a feasible formation and contains an initial configuration and a target configuration. A feasible edge between two nodes or formations is a convex region in free space that contains both formations. A feasible edge provides a way to transition between two nodes in the graph.

Then the global path planning algorithm is introduced. Consider the barrier-free area F, the starting position s of the formation centroid and its target position g∈R ³ . In this algorithm, the embodiments of the present invention describe the proposed calculation method, and the goal is to obtain a configuration sequence T={z _s ,...,z that enables the swarm of UAVs to navigate from the initial configuration z _s to the target configuration g _{s g} } is composed of the global path. The embodiment of the present invention establishes a list of existing polyhedrons P, and the swarm drones can be completely contained in the polyhedron. The configuration sequence is initialized with initial z _s and target g _s configurations for centroids s and g. Similarly, initialize the list of polyhedra with convex regions _Ps and _Pz , both containing the initial and target configurations, respectively. The method works by drawing random samples in the work area (for the UAV's flight space ^R3 ). The sampling method is similar to Rapid Search Random Tree (RRT). Each random sample ^p∈R3 is excluded if it is within an obstacle or within one of the multiple points in the list P. Otherwise perform the following steps:

1. Through a fast iterative method, a large unobstructed convex polyhedron P _{p is calculated by an iterative algorithm at the sample p} .

2. Through semi-definite programming, calculate the largest ellipsoid ep inscribed in the convex polyhedron _{P p} , if the ellipsoid can be stored under the condition that all UAVs keep a safe distance d _f and maintain the required formation For all drones, the configuration z _w is stored in the configuration sequence T. Here w represents the center of the formation, that is, the center corresponding to the largest ellipsoid _ep .

3. Continue sampling and repeat the above steps until the first feasible path is found, until the entire space F is explored or until the maximum time limit is reached.

4. Return a configuration sequence T corresponding to a feasible path with the shortest (Euclidean) distance.

(2), local path planning

If each configuration z _w corresponding to the sequence obtained by global path planning is given a sequence number in sequence, that is, T={z ₁ , z ₂ ,...,z _m }, then local path planning is to solve how to get from z _{i+ 1} to _zi questions. The method of continuous convex programming is used here. Continuous convex programming is an iterative method for solving convex approximations of non-convex problems, which has the following advantages:

1. Continuous convex programming uses multiple iterations to ensure that the convex approximation of non-convex constraints is accurate, resulting in more fuel-efficient trajectories.

2. Continuous convex programming algorithms can be written using free software such as CVX [30, 31] to convert convex programs into semidefinite programs (SDP) or second-order cone programs (SOCP).

First, linearize and discretize the dynamic constraints of the drone. The dynamic constraints of the UAV are as follows:

是第j只无人机的位置矢量，u_j是第j只无人机的控制矢量。in,

is the position vector of the jth drone, and u _j is the control vector of the jth drone.

In order to rewrite the dynamical constraints into constraints that can be used for convex programming problems, the equation must first be linearized:

The next step in converting dynamic constraints into constraints that can be used in convex programming is to convert dynamic constraints into constraints that can be used in convex programming. To do this, discretizing the problem using the zero-order hold method, we get:

x _j [k+1]=A _j [k]x _j [k]+B _j [k]u _j [k]+z _j [k],k=k ₀ ,...,T-1,j =1,...,N

where x _j [k]=x _j (t _k ), u _j [k]=u _j (t _k ), and:

The final step in transforming the UAV swarm reconstruction into a convex programming problem is to transform the collision avoidance constraints into convex constraints. The collision avoidance constraints are as follows:

Where G=[I _3×3 0 _3×3 ], R _col is the minimum distance allowed between two UAVs.

Since the collision avoidance constraint in its current form is concave, the best convex approximation will be an affine constraint. It can be converted into the following formula:

The proof is as follows:

和

是x_i，x_j先前迭代的标称轨迹，φ是两个向量之间的角度。假定这些标称值是已知的，并且不是优化中的变量。因此，新的避碰约束是仿射的，该形式可用于凸规划问题。in,

and

are the nominal trajectories of the previous iterations of x _i , x _j , and φ is the angle between the two vectors. These nominal values are assumed to be known and not variables in the optimization. Therefore, the new collision avoidance constraint is affine, a form that can be used for convex programming problems.

The goal of optimal group reconstruction is to minimize the L1 norm of the control input. The L1 norm of the control input is equal to the total fuel used during transmission. Therefore, in this embodiment of the present invention, group reconstruction may be defined as follows:

x _j [k+1]=A _j [k]x _j [k]+B _j [k]u _j [k]+z _j [k],k=k ₀ ,...,T-1,j =1,...,N

||u _j [k]|| _∞ ≤U _max k=0,...,T-1,j=1,...,N

x _j [0]=x _j,0 ,x _j [T]=x _j,f j=1,...,N

另外，标称轨迹应接近实际状态轨迹，以使近似误差最小。为了确保标称向量是实际状态向量的良好估计，使用连续凸规划。SCP是一种迭代方法，它解决了一个非凸问题的凸逼近问题，并在下一次迭代中使用该解决方案使该问题凸化，即

其中w是连续凸规划的迭代次数。重复此过程，直到轨迹序列根据以下条件收敛：The approximation used to optimize the trajectory as a convex program requires the nominal trajectory of each drone

Additionally, the nominal trajectory should be close to the actual state trajectory to minimize approximation errors. To ensure that the nominal vector is a good estimate of the actual state vector, continuous convex programming is used. SCP is an iterative method that solves a convex approximation of a non-convex problem and uses that solution in the next iteration to make the problem convex, i.e.

where w is the number of iterations of the continuous convex programming. This process is repeated until the trajectory sequence converges according to:

where ε _scp is the threshold.

To enforce collision avoidance constraints, each UAV communicates its own nominal trajectory to its neighbor UAVs.

The steps of continuous convex programming are as follows. First, initial trajectories are generated for each agent without considering collision avoidance. The iterative process then begins with each drone solving its optimal trajectory. Next, each drone stores the current trajectory as the nominal trajectory for the next iteration and passes this trajectory to its neighboring drones. Finally, this iteration is repeated until the trajectory converges and the drone does not collide.

Since the convex optimization problem can be solved efficiently, the running time is now on the order of one or two time steps, so model predictive control can be used to go to the Update future assignments and control commands, enabling continuous convex programming. Model predictive control can provide some robustness against disturbances and allow for distribution and disconnection of communication networks

To sum up, compared with the prior art, the present invention realizes the adaptive cooperative formation of group UAVs by combining global path planning and local path planning, and the global planning and local planning are decoupled.

The sampling of the global path planning is the sampling rising to the formation, and combined with the idea of quickly exploring the random tree, multiple sampling points (formations) from the starting point to the end point can be quickly found. And each formation can be adaptively transformed according to the environment, resulting in different scaled formations.

Based on the linearization and discretization of dynamic constraints and convex collision avoidance constraints, local path planning constructs a distributed convex programming problem with group reconstruction, and solves it by continuous convex programming. At the same time, the results of model predictive control can realize the formation process in real time and improve the anti-interference performance.

The embodiment of the present invention also provides a multi-UAV formation control device based on continuous convex rules, including:

The global planning module is used to determine the global path planning of the UAV formation; the global path planning is used to determine the action path and formation formation of the UAV formation from the starting point to the end point;

a transformation module, configured to transform the collision avoidance constraint of the UAV formation into a convex constraint according to the linearized and discretized UAV dynamics constraints;

a local planning module, configured to determine the local path planning of the UAV formation according to the convex constraint; the local path planning is used to determine the action trajectory of each UAV in the UAV formation;

A tracking control module is used to perform real-time path tracking control on the UAV formation through a preset model.

Preferably, the global planning module includes:

a first determining unit, configured to determine the positions of a plurality of UAVs of the UAV formation, and determine the outer vertex of the rotation center of the UAV formation;

a second determining unit, configured to determine the minimum distance between any pair of UAVs in the UAV formation;

an isomorphic transformation unit, configured to determine the initial configuration of the UAV formation by isomorphic transformation according to the UAV position, the external vertex and the minimum distance;

The third determining unit is configured to determine the target configuration of the UAV formation through a global path planner according to the initial configuration of the UAV formation.

Preferably, the third determining unit includes:

constructing a subunit for establishing a list of polyhedrons, including all drones in the drone formation in the list of polyhedrons;

a first initialization subunit, used for initializing the initial configuration and target configuration of the UAV formation;

a second initialization subunit, configured to initialize the polyhedron list through the convex region;

a random sampling subunit, used for sampling random samples in the flight space of the drone formation;

A judging subunit, used for judging whether the random sample is in the obstacle or in the polyhedron list, if yes, then the random sample is excluded; otherwise, the following steps are performed:

a first calculation subunit, configured to calculate the barrier-free convex polyhedron of the random sample through an iterative algorithm;

a second calculation subunit for calculating the inscribed ellipsoid of the barrier-free convex polyhedron;

A determination subunit, configured to determine the action path of the UAV formation from the start point to the end point according to the inscribed ellipsoid.

The embodiment of the present invention also provides an electronic device, including a processor and a memory;

the memory is used to store programs;

The processor executes the program to implement the method described in FIG. 1 .

An embodiment of the present invention also provides a computer-readable storage medium, where the storage medium stores a program, and the program is executed by a processor to implement the method described in FIG. 1 .

The embodiment of the present invention also discloses a computer program product or computer program, where the computer program product or computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. The computer instructions can be read from the computer-readable storage medium by a processor of the computer device, and the processor executes the computer instructions to cause the computer device to perform the foregoing method.

In some alternative implementations, the functions/operations noted in the block diagrams may occur out of the order noted in the operational diagrams. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or the blocks may sometimes be executed in the reverse order, depending upon the functionality/operations involved. Furthermore, the embodiments presented and described in the flowcharts of the present invention are provided by way of example in order to provide a more comprehensive understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of the various operations are altered and in which sub-operations described as part of larger operations are performed independently.

Furthermore, while the invention is described in the context of functional modules, it is to be understood that, unless stated to the contrary, one or more of the described functions and/or features may be integrated in a single physical device and/or or software modules, or one or more functions and/or features may be implemented in separate physical devices or software modules. It will also be appreciated that a detailed discussion of the actual implementation of each module is not necessary to understand the present invention. Rather, given the attributes, functions, and internal relationships of the various functional modules in the apparatus disclosed herein, the actual implementation of the modules will be within the routine skill of the engineer. Accordingly, those skilled in the art, using ordinary skill, can implement the invention as set forth in the claims without undue experimentation. It is also to be understood that the specific concepts disclosed are illustrative only and are not intended to limit the scope of the invention, which is to be determined by the appended claims along with their full scope of equivalents.

The functions, if implemented in the form of software functional units and sold or used as independent products, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product in essence, or the part that contributes to the prior art or the part of the technical solution. The computer software product is stored in a storage medium, including Several instructions are used to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, Read-Only Memory (ROM, Read-Only Memory), Random Access Memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program codes .

The logic and/or steps represented in flowcharts or otherwise described herein, for example, may be considered an ordered listing of executable instructions for implementing the logical functions, may be embodied in any computer-readable medium, For use with, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other system that can fetch instructions from and execute instructions from an instruction execution system, apparatus, or apparatus) or equipment. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport the program for use by or in connection with an instruction execution system, apparatus, or apparatus.

More specific examples (non-exhaustive list) of computer readable media include the following: electrical connections with one or more wiring (electronic devices), portable computer disk cartridges (magnetic devices), random access memory (RAM), Read Only Memory (ROM), Erasable Editable Read Only Memory (EPROM or Flash Memory), Fiber Optic Devices, and Portable Compact Disc Read Only Memory (CDROM). In addition, the computer readable medium may even be paper or other suitable medium on which the program may be printed, as the paper or other medium may be optically scanned, for example, followed by editing, interpretation, or other suitable medium as necessary process to obtain the program electronically and then store it in computer memory.

It should be understood that various parts of the present invention may be implemented in hardware, software, firmware or a combination thereof. In the above-described embodiments, various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented by any one or a combination of the following techniques known in the art: Discrete logic circuits, application specific integrated circuits with suitable combinational logic gates, Programmable Gate Arrays (PGA), Field Programmable Gate Arrays (FPGA), etc.

In the description of this specification, description with reference to the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples", etc., mean specific features described in connection with the embodiment or example , structure, material or feature is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

Although embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, The scope of the invention is defined by the claims and their equivalents.

The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the described embodiments, and those skilled in the art can also make various equivalent deformations or replacements on the premise of not violating the spirit of the present invention, These equivalent modifications or substitutions are all included within the scope defined by the claims of the present application.

Claims (9)

1. A multi-unmanned aerial vehicle formation control method based on a continuous convex rule is characterized by comprising the following steps:

determining global path planning of unmanned aerial vehicle formation; the global path planning is used for determining an action path from a starting point to an end point of the unmanned aerial vehicle formation and a formation;

converting collision avoidance constraints of the unmanned aerial vehicle formation into convex constraints according to linearized and discretized unmanned aerial vehicle dynamics constraints;

determining a local path plan of the unmanned aerial vehicle formation according to the convex constraint; the local path plan is used for determining the action track of each unmanned aerial vehicle in the unmanned aerial vehicle formation;

performing real-time path tracking control on the unmanned aerial vehicle formation through a preset model;

according to the linearized and discretized unmanned aerial vehicle dynamics constraint, converting the collision avoidance constraint of the unmanned aerial vehicle formation into a convex constraint, wherein the unmanned aerial vehicle dynamics constraint is as follows:

wherein,

l∈Rⁿa position vector representing a jth drone; u. of_jA control vector representing a jth drone; b is ═ 0_3×3 I_3×3]T represents time;

the expression of the linearization is:

wherein,

representing a nominal track, namely a track generated in the last iteration;

2. the method for controlling formation of multiple unmanned aerial vehicles based on the continuous convex rule according to claim 1, wherein the determining the global path plan of the formation of the unmanned aerial vehicles comprises:

determining a plurality of drone positions for the formation of drones, and determining an outer vertex of a center of rotation for the formation of drones;

determining the minimum distance between any pair of unmanned aerial vehicles in the unmanned aerial vehicle formation;

determining the initial configuration of the formation of the unmanned aerial vehicles through isomorphic transformation according to the positions of the unmanned aerial vehicles, the external vertexes and the minimum distance;

and determining the target configuration of the unmanned aerial vehicle formation through a global path planner according to the initial configuration of the unmanned aerial vehicle formation.

3. The method for controlling formation of multiple drones based on the continuous convex rule according to claim 2, wherein the determining, by a global path planner, the target configuration of the formation of drones according to the initial configuration of the formation of drones comprises:

establishing a polyhedron list, and including all unmanned aerial vehicles in the unmanned aerial vehicle formation in the polyhedron list;

initializing the initial configuration and the target configuration of the unmanned aerial vehicle formation;

initializing the polyhedron list through a convex area;

extracting random samples from the flight space of the formation of unmanned aerial vehicles;

judging whether the random sample is in the barrier or in the polyhedron list, if so, excluding the random sample; otherwise, the following steps are executed:

calculating an obstacle-free convex polyhedron of the random samples through an iterative algorithm;

calculating an inscribed ellipsoid of the barrier-free convex polyhedron;

and determining an action path from a starting point to an end point of the unmanned aerial vehicle formation according to the inscribed ellipsoid.

4. The method for controlling formation of multiple drones based on the continuous convex rule according to claim 1, wherein the determining the local path plan of the formation of drones according to the convex constraint comprises:

generating an initial trajectory for each drone without regard to avoiding collisions;

according to the initial track, iteratively solving the optimal track of the unmanned aerial vehicle;

and determining the optimal tracks of all the unmanned aerial vehicles in the unmanned aerial vehicle formation according to the optimal track of each unmanned aerial vehicle.

5. The utility model provides a many unmanned aerial vehicle formation controlling means based on protruding rule in succession which characterized in that includes:

the global planning module is used for determining global path planning of unmanned aerial vehicle formation; the global path planning is used for determining an action path from a starting point to an end point of the unmanned aerial vehicle formation and a formation;

the conversion module is used for converting collision avoidance constraints of the unmanned aerial vehicle formation into convex constraints according to the linearized and discretized unmanned aerial vehicle dynamics constraints;

the local planning module is used for determining the local path planning of the unmanned aerial vehicle formation according to the convex constraint; the local path plan is used for determining the action track of each unmanned aerial vehicle in the unmanned aerial vehicle formation;

the tracking control module is used for carrying out real-time path tracking control on the unmanned aerial vehicle formation through a preset model;

in the conversion module, the unmanned aerial vehicle dynamics constraint is as follows:

wherein,

l∈Rⁿa position vector representing a jth drone; u. of_jA control vector representing a jth drone; b is ═ 0_3×3 I_3×3]T represents time;

the expression of the linearization is:

wherein,

representing a nominal track, namely a track generated in the last iteration;

6. the sequential convex rule-based multi-UAV formation control device according to claim 5, wherein the global planning module comprises:

a first determination unit for determining a plurality of drone positions of the formation of drones and determining an outer vertex of a center of rotation of the formation of drones;

a second determining unit, configured to determine a minimum distance between any pair of drones in the formation of drones;

the isomorphic transformation unit is used for determining the initial configuration of the formation of the unmanned aerial vehicles through isomorphic transformation according to the positions of the unmanned aerial vehicles, the external vertexes and the minimum distance;

and the third determining unit is used for determining the target configuration of the formation of the unmanned aerial vehicles through a global path planner according to the initial configuration of the formation of the unmanned aerial vehicles.

7. The sequential convex rule-based multi-UAV formation control device according to claim 6, wherein the third determination unit comprises:

the building subunit is used for building a polyhedron list and containing all the unmanned aerial vehicles in the unmanned aerial vehicle formation in the polyhedron list;

the first initialization subunit is used for initializing the initial configuration and the target configuration of the formation of the unmanned aerial vehicles;

the second initialization subunit is used for initializing the polyhedron list through a convex region;

a random sampling subunit, configured to extract random samples in the flight space of the formation of unmanned aerial vehicles;

the judgment subunit is used for judging whether the random sample is in the barrier or in the polyhedron list, and if so, the random sample is excluded; otherwise, the following steps are executed:

a first calculation subunit, configured to calculate an obstacle-free convex polyhedron of the random samples by an iterative algorithm;

the second calculating subunit is used for calculating an inscribed ellipsoid of the barrier-free convex polyhedron;

and the determining subunit is used for determining an action path from the starting point to the end point of the formation of the unmanned aerial vehicles according to the inscribed ellipsoid.

8. An electronic device comprising a processor and a memory;

the memory is used for storing programs;

the processor executing the program realizes the method according to any one of claims 1-4.

9. A computer-readable storage medium, characterized in that the storage medium stores a program, which is executed by a processor to implement the method according to any one of claims 1-4.

Images