Multi-unmanned aerial vehicle formation control method based on distributed model predictive control

Abstract

The invention discloses a multi-unmanned aerial vehicle formation control method based on distributed model predictive control. The method comprises the following steps: describing a motion law of each unmanned aerial vehicle by adopting a state-space equation; modeling a static obstacle as a geometric body, and setting a minimum safe distance; considering position tracking cost, control energy cost, input change cost and formation cost, and establishing an objective function for each unmanned aerial vehicle; adding a CBF security constraint according to whether the security domain has an obstacle avoidance demand or not; constructing an optimal control problem for each unmanned aerial vehicle; in each control period, the optimal control problem is solved; only the first item in the control sequence is adopted as an actual control instruction; and then the system state is updated based on the state-space equation, and prediction and optimization are carried out again in the next period. According to the invention, the problems of obstacle avoidance and formation keeping when the multi-unmanned aerial vehicle formation passes through the obstacle in a complex environment are effectively solved. The method can be widely applied to the field of unmanned aerial vehicle control.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) control, and more particularly to a multi-UAV formation control method based on distributed model predictive control. Background Technology

[0002] Multi-UAV cooperative swarming refers to the process by which UAVs, through communication and interaction, achieve and maintain a predetermined swarm configuration. Swarming technology has significant application value in disaster relief, environmental monitoring, and other fields. Compared to single-UAV solutions, cooperative swarming reduces runtime, lowers mission failure rates, and enhances multi-tasking capabilities, making it more suitable for complex mission scenarios. In scenarios with obstacle threats, multi-UAV swarms, through interaction between UAVs and with environmental information, achieve obstacle avoidance, thus meeting safety requirements in complex environments.

[0003] Currently, artificial potential field methods are commonly used for obstacle avoidance in drone formation scenarios with obstacle threats. This type of method requires pre-constructing the attractive potential function of the target point and the repulsive potential function of the obstacle, and then using the synthesized potential field to guide the trajectory of each drone. However, this method is highly sensitive to the form and parameters of the potential field, its tuning relies on experience, and it lacks robustness. In complex environments, it is prone to getting trapped in local minima, oscillations, or failing to avoid obstacles, leading to decreased propulsion efficiency and even causing individual drones to deviate from the desired formation. Summary of the Invention

[0004] In view of this, in order to address the technical problem that most existing multi-UAV formation control methods rely on artificial potential field methods for obstacle avoidance, which are prone to getting trapped in local minima in complex environments, leading to low safety, this invention proposes a multi-UAV formation control method based on distributed model predictive control. This method includes the following steps: State-space equations are used to describe the motion of each UAV and to determine the mathematical relationship between the system state variables and the control input. Static obstacles are modeled as geometric shapes, and minimum safe distances between UAVs and between UAVs and obstacles are set as hard constraints for subsequent control strategies. Considering the costs of position tracking, control energy, input variation, and formation, a rolling optimization objective function is established for each UAV. Add CBF safety constraints based on whether the safety domain has obstacle avoidance requirements; For each UAV, construct an optimal control problem constrained by system dynamics, input limitations, and safe distance. Within each control cycle, the optimal control problem is solved by combining the current state with preset constraints to obtain the control input sequence in the future time domain; The first term in the control sequence is used as the actual control command to drive the UAV's movement; then the system state is updated based on the state-space equation, and prediction and optimization are performed again in the next cycle to form a closed-loop control structure.

[0005] Based on the above scheme, this invention provides a multi-UAV formation control method based on distributed model predictive control. By explicitly transforming the minimum safe interval between UAVs and between UAVs and obstacles into CBF inequality constraints, the invariance of the safe feasible region is guaranteed, and the risks of local minima, deadlock, and collisions in strongly constrained scenarios are avoided by traditional distance threshold or artificial potential field methods. At the same time, a constraint triggering mechanism based on the safe detection range is introduced, which activates relevant constraints only when there is a potential threat, reducing redundant calculations in obstacle-free areas and improving overall computational efficiency. In addition, soft constraints for formation are added to the DMPC framework, enabling the formation to adjust its formation more flexibly and maintain cooperative stability during obstacle avoidance. Attached Figure Description

[0006] Figure 1 This is a flowchart illustrating the steps of a multi-UAV formation control method based on distributed model predictive control according to the present invention. Figure 2 It is the flight trajectory of the UAV in the obstacle threat map that changes over time under the formation control method of the present invention; Figure 3 This is a top-down view of the UAV's flight trajectory in the obstacle threat map under the formation control method of this invention; Figure 4 The error curve between the drone formation and the expected formation during flight. Detailed Implementation

[0007] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0008] It should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings. Unless otherwise specified, the embodiments and features described in this application can be combined with each other.

[0009] It should be understood that the terms "system," "apparatus," "unit," and / or "module" used in this application are a method of distinguishing different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0010] As indicated in this application and claims, unless the context clearly indicates otherwise, the words "a," "an," "a," and / or "the" are not specifically singular and may include the plural. Generally, the terms "comprising" and "including" only indicate the inclusion of expressly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements. An element defined by the phrase "comprising an..." does not exclude the presence of other identical elements in the process, method, product, or apparatus that includes the element.

[0011] In the description of the embodiments of this application, "a plurality of" refers to two or more. The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature.

[0012] Furthermore, flowcharts are used in this application to illustrate the operations performed by the system according to embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, the steps can be processed in reverse order or simultaneously. Additionally, other operations can be added to these processes, or one or more steps can be removed from them.

[0013] Reference Figure 1 This is a flowchart illustrating an optional example of the multi-UAV formation control method based on distributed model predictive control proposed in this invention. This method can be applied to computer equipment, and the formation control method proposed in this embodiment may include, but is not limited to, the following steps: Step S1: Define the initial target state of the UAV system.

[0014] Among them, the multi-unmanned aerial vehicle system includes The drones can be deployed arbitrarily in three-dimensional space. Each drone is equipped with an onboard local coordinate system and communication unit. The system employs distributed information exchange, eliminating the need for global information exchange. The communication topology for establishing the target formation is an undirected graph. Determine the neighbor set for each drone.

[0015] Step S2: Establish the kinematic model of the UAV, treat each UAV model as a second-order point mass system, and establish the state-space equation of the system.

[0016] Step S3: Define the sampling period, prediction time domain length, and boundaries of the input and control states for model predictive control; The boundaries include the constraints on the drone's speed and acceleration. , ,in For the first The drone in The speed of time For the first The drone in The acceleration of time.

[0017] Step S4: Perform geometric modeling of static obstacles in the environment, and define the minimum safe distance between the drone and neighboring drones, as well as the minimum safe distance between the drone and obstacles.

[0018] Among them, environmental obstacles are simplified into cylinders or polyhedra. Record parameters such as dimensions and height.

[0019] Step S5: Combine the basic cost term within the prediction domain with the formation cost to establish the objective function for rolling optimization of the UAV.

[0020] Step S6: Define the control barrier function between drones and the control barrier function between drones and obstacles, and add CBF safety constraints based on preset judgment conditions.

[0021] Step S7: Construct the optimal control problem for each drone.

[0022] Step S8: Solve the optimal control problem to obtain the optimal control sequence; Step S9: Execute only the first control variable and keep the rest as rolling predictions. After the state update and information interaction are executed, update the local state, send the necessary prediction endpoints or aggregation information to the neighbors, receive the neighbor information, and refresh the local optimization problem for the next moment.

[0023] Step S10: If the target location is reached or a given time step is reached, the process ends.

[0024] In some feasible embodiments, the state-space equation of step S3 is expressed as follows: Among them, drones The state vector is defined as Control input is defined as ,in, Indicates the sampling period. Represents a position vector. Represents the velocity vector. This represents the acceleration vector.

[0025] In some feasible embodiments, step S5 specifically includes: During the sampling period and prediction step size Next, regarding the first The objective function for rolling optimization is established for the UAV, and its basic cost term consists of the following three parts: (1) Position tracking cost, used to drive the state to converge from the current attitude to the preset reference trajectory. : (2) Control energy cost and suppress control input amplitude: (3) Input change cost, constrain drastic changes in control quantity at adjacent time points to maintain system stability: The above three items are weighted. , , After weighting, the first The basic objective function of a drone A soft constraint on formation consistency is introduced as a formation cost. To maintain the desired formation within the prediction domain, a communication topology graph is used. Relative displacement with respect to expectation Construct the first The cost of swarming drones: in For nodes The set of neighbors. As a soft constraint with weight By incorporating it into the objective function, we obtain the first... The overall objective function of the drone: When a potential collision risk occurs within the prediction domain, the accuracy can be increased as needed. Alternatively, it may trigger corresponding soft constraint weights to maintain the formation while ensuring safety.

[0026] In some feasible embodiments, step S6 specifically includes: Define the control barrier function between drones, drones For any neighbor The control barrier function is When the drone is within the safe detection range Encountering a neighbor's drone Add CBF safety constraints as follows: Define a control barrier function between the drone and the obstacle, and let The drone is the closest distance to the side of the cylinder. For any obstacle The control function is When the drone is within the safe detection range Encountering obstacles Add CBF safety constraints as follows: In some feasible embodiments, step S7, the expression for the optimal control problem is: Indicates drone exist The state of time, that is .

[0027] When there is in the prediction domain or When necessary, introduce corresponding CBF safety constraints and increase their weights. Only apply safety obstacle avoidance constraints when needed to balance safety and path efficiency.

[0028] This invention introduces CBF constraints to ensure that UAVs maintain a safe distance during obstacle avoidance and optimizes formation flight paths. Simultaneously, it incorporates the consistency principle into the DMPC framework as a formation cost, thereby significantly improving formation stability in obstacle environments. This method effectively solves the obstacle avoidance and formation maintenance problems of multi-UAV formations traversing obstacles in complex environments, avoiding the safety distance failure issues that occur with traditional methods in strongly constrained scenarios. The designed controller can quickly restore formation stability, reduce formation errors, and demonstrates superior obstacle avoidance capabilities and higher performance compared to traditional DMPC methods. It also exhibits strong adaptability and scalability, making it suitable for various complex multi-UAV collaborative tasks.

[0029] Based on the above method, this embodiment provides a simulation experiment to verify the effectiveness of the above control method.

[0030] Specifically, this simulation experiment employed a multi-UAV system consisting of five UAVs, arbitrarily deployed in three-dimensional space for simulation verification. Each UAV initially followed... The axes form a straight line arrangement, with the specific positions as follows: , , , , The desired drone formation is a pentagon, and the drones' communication neighbor sets are as follows: , , , , Additionally, select parameters. , , Six cylindrical obstacle threats are also placed in the environment.

[0031] Under the control algorithm of multi-drone formation, different colored lines represent different drones, and the flight trajectories of the five drones change over time as follows: Figure 2 As shown, its top-down perspective is as follows Figure 3 As shown. By Figure 2 and Figure 3 It can be seen that the UAV system was able to achieve obstacle avoidance at every time step throughout the entire flight. Meanwhile, Figure 4 The study demonstrates the deviation between the positions of each UAV formation and the expected formation positions during flight. The results show that when obstacles are present, the system will produce a certain deviation to complete obstacle avoidance. However, when there are no obstacles, the UAV system can quickly recover to the expected formation, thus ensuring the stability of the overall formation. In summary, this algorithm can effectively achieve obstacle avoidance and maintain formation stability under obstacle threats.

[0032] A multi-UAV formation control device based on distributed model predictive control: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements a multi-UAV formation control method based on distributed model predictive control as described above.

[0033] The content of the above method embodiments is applicable to the device embodiments. The specific functions implemented by the device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0034] A storage medium storing processor-executable instructions, which, when executed by a processor, are used to implement a multi-UAV formation control method based on distributed model predictive control as described above.

[0035] The content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0036] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the embodiments described. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of this application.

Claims

1. A multi-UAV formation control method based on distributed model predictive control, characterized in that, Includes the following steps: Define the state-space equations for a multi-UAV system; Geometric modeling of static environmental obstacles and definition of minimum safe distance; Define the objective function for MPC control; Construct an optimal control problem for each drone; By combining preset constraints, the optimal control problem is solved to obtain the optimal control sequence; Based on the optimal control sequence, the first control variable is executed while the rest are kept as rolling predictions. The state is then updated by combining the state space equation, the minimum safe distance, and the UAV's objective function.

2. The multi-UAV formation control method based on distributed model predictive control according to claim 1, characterized in that, The state-space equation of the multi-UAV system is expressed as follows: in, Indicates the first The drone in The position of the time step Indicates the first The drone in The speed of the time step Indicates the sampling period. Indicates the first The drone in The position of the time step Indicates the first The drone in The speed of the time step Indicates the first The drone in Acceleration at any moment.

3. The multi-UAV formation control method based on distributed model predictive control according to claim 2, characterized in that, The objective function of the UAV includes a base cost and a formation cost, wherein: The basic cost is: in, Indicates the basic cost. Indicates the cost of location tracking. This indicates the cost of controlling energy. Indicates the cost of input changes. Indicates the corresponding weight. This represents the total prediction step size, where k represents the time step. Preset reference trajectory, This represents a matrix composed of control cost weights; The formation cost is: in, Indicates the cost of formation. Indicates the first The neighbors of the drone gathered. Indicates the first The drone in The position of the time step Indicates the first The expected position of the drone formation. Indicates the first The expected position of the drone formation.

4. The multi-UAV formation control method based on distributed model predictive control according to claim 1, characterized in that, Also includes: Add CBF safety constraints based on whether the safety domain has obstacle avoidance requirements.

5. The multi-UAV formation control method based on distributed model predictive control according to claim 2, characterized in that, The expression for the optimal control problem is as follows: in, Indicates drone exist The state at that time, This represents the minimum speed constraint value for the drone. This indicates the maximum speed constraint value for the drone. This represents the minimum acceleration constraint value for the drone. This represents the minimum acceleration constraint value for the drone. This indicates a CBF safety constraint.

6. The multi-UAV formation control method based on distributed model predictive control according to claim 4, characterized in that, The CBF security constraints specifically include: When a drone encounters a neighboring drone within the safe detection range, add the following CBF safety constraint: in, Indicates the first A drone in Time step for any neighbor The control barrier function, , This indicates the minimum safe distance between drones. Represented as a hyperparameter; When the drone encounters an obstacle within the safety detection range, the CBF safety constraint is added as follows: in, Indicates the first A drone in Time step for any obstacle The control barrier function, , Indicates the first Drag a drone to any obstacle The closest distance to the side of the cylinder. This indicates the minimum safe distance between the drone and the obstacle.

7. The multi-UAV formation control method based on distributed model predictive control according to claim 1, characterized in that, Also includes: Send necessary prediction endpoints or aggregation information to neighboring drones, receive neighbor information, and refresh the local optimization problem for the next moment; The process ends when the target location is reached or a given time step is reached.

8. A multi-UAV formation control device based on distributed model predictive control, characterized in that, include: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the multi-UAV formation control method based on distributed model predictive control as described in any one of claims 1-7.

Images