Unmanned aerial vehicle cluster collaborative formation control method and system

Abstract

The invention discloses an unmanned aerial vehicle cluster collaborative formation control method and system. The method comprises the following steps: acquiring multi-source sensor data in real time, fusing the multi-source sensor data by using an extended Kalman filtering algorithm, and performing state estimation when part of sensor data is missing to obtain state information; based on the state information, interference is detected by monitoring related parameters of communication signals and sensor data, the interference type is identified through spectrum and waveform analysis, and the intensity is evaluated; determining interfered unmanned aerial vehicles and adjacent unmanned aerial vehicles according to a communication neighborhood, performing local formation adjustment on the interfered unmanned aerial vehicles and the adjacent unmanned aerial vehicles by using a distributed control algorithm, and estimating a speed optimization algorithm of the adjacent unmanned aerial vehicles by using Kalman filtering under incomplete information to realize unmanned aerial vehicle cluster collaborative formation control; and establishing an early warning model based on the distance and the speed, and realizing collision avoidance by adopting an optimized artificial potential field method. The anti-interference capability of the unmanned aerial vehicle cluster is improved, the collision probability is reduced, and the formation stability is enhanced.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) control technology, and in particular to a method and system for UAV swarm collaborative formation control. Background Technology

[0002] With the continuous development of technology, drones have been widely used in various fields, including military and civilian applications. Drone swarming and coordinated formation flight, as an emerging application mode, offers numerous advantages such as high efficiency, flexibility, and powerful mission execution capabilities. In the military field, drone swarms can perform various tasks such as reconnaissance, surveillance, and attack, and coordinated formation flight can improve combat efficiency and survivability. In the civilian field, drone swarms can be used for agricultural plant protection, logistics distribution, surveying, and other tasks, greatly improving work efficiency and quality.

[0003] However, numerous challenges arise during the coordinated flight of drone swarms. Among these, electromagnetic interference is a significant issue. In modern society, the electromagnetic environment is increasingly complex, with various electronic devices and communication signals generating electromagnetic interference. When drone swarms are in such an environment, their communication and sensor systems are easily affected.

[0004] From a communication system perspective, drones need to exchange information such as position, speed, and heading via wireless communication to achieve coordinated formation flight. Electromagnetic interference can cause communication signals to be lost, delayed, or erroneous, preventing drones from obtaining information from other drones in a timely and accurate manner. For example, during formation flight, if a drone cannot receive position information from neighboring drones due to communication interference, it may deviate from its original formation position, thus affecting the stability of the entire formation.

[0005] From a sensor system perspective, drones rely on various sensors to perceive their own status and the surrounding environment. Electromagnetic interference can cause deviations in the data output by these sensors, such as inaccurate attitude information measured by the inertial measurement unit (IMU) or errors in distance data measured by lidar. This can lead to errors in the drone's judgment of its own position and the surrounding environment, thereby increasing the risk of formation chaos and collisions.

[0006] Currently, while some technologies and methods exist for UAV swarm control, most are designed under ideal information conditions and have limited capabilities to cope with incomplete information. For example, some traditional swarm control algorithms rely on precise global information; when this information is interfered with or lost, their performance deteriorates sharply. Furthermore, existing technologies are insufficient in suppressing transient electromagnetic interference, often failing to adjust the swarm effectively and promptly when interference occurs, leading to swarm chaos or even collisions. Therefore, developing a UAV swarm cooperative swarm control technology that can effectively suppress interference and reduce the probability of collisions under incomplete information conditions is of significant practical importance. Summary of the Invention

[0007] To address the aforementioned technical problems, this invention proposes a method and system for controlling the coordinated formation of unmanned aerial vehicle (UAV) swarms. Under conditions of incomplete information, this method suppresses transient electromagnetic interference, reduces the probability of collisions in UAV swarm formations, and improves the stability and reliability of UAV swarm coordinated formation flight.

[0008] To achieve the above objectives, the technical solution of the present invention is as follows:

[0009] A method for cooperative formation control of unmanned aerial vehicle (UAV) swarms includes the following steps:

[0010] Step 1: Acquire multi-source sensor data in real time, fuse the multi-source sensor data using the extended Kalman filter algorithm, and perform state estimation when some sensor data is missing to obtain state information;

[0011] Step 2: Based on the status information, interference is detected by monitoring relevant parameters of communication signals and sensor data, and the type and intensity of interference are identified by spectrum and waveform analysis.

[0012] Step 3: Based on the communication neighborhood, determine the interfered UAV and neighboring UAVs. The interfered UAV and neighboring UAVs use a distributed control algorithm to perform local formation adjustment. Under incomplete information, Kalman filtering is used to estimate the speed of neighboring UAVs and optimize the algorithm to realize UAV swarm cooperative formation control.

[0013] Step 4: During the collaborative formation control of the UAV swarm, an early warning model is established based on distance and speed, and an optimized artificial potential field method is used to achieve collision avoidance.

[0014] Preferably, multi-source sensor data is collected through an inertial measurement unit, lidar, and GPS installed on the UAV, and the sampling frequency of the inertial measurement unit, lidar, and GPS decreases sequentially.

[0015] Preferably, state estimation is performed when some sensor data is missing, and the calculation formula is as follows:

[0016] in, This is the state estimate when the GPS signal is lost. yes The predicted state vector at each time step. It is the rate of change of state calculated based on the acceleration and angular velocity measured by the inertial measurement unit.

[0017] Preferably, step 2 includes the following steps:

[0018] Real-time acquisition of monitoring information, including any one of the following: bit error rate of communication signal, signal strength, and rate of change and fluctuation range of sensor data;

[0019] When the monitored information exceeds a preset threshold, electromagnetic interference is determined to have occurred. The type of interference is identified by analyzing the frequency and waveform characteristics of the signal. The signal-to-interference ratio is used to assess the interference intensity of the communication signal, and the relative deviation of the sensor data is used to assess the interference intensity of the sensor data.

[0020] Preferably, step 3 includes the following steps:

[0021] Suppose there are a total of 1 drone, each drone Having a unique number and When detecting drones Electromagnetic interference occurs; define the communication neighborhood. For drones The communication neighborhood is a collection of other neighboring drones that can communicate directly. Determination method: Set up a drone The position is Other drones The position is The communication radius is ,like ,but ,in Denotes the Euclidean norm;

[0022] Set up drones The speed is Each drone Based on the received positions of neighboring drones and speed ( Define the expected relative position. For drones Relative to neighboring drones The expected relative positions of the drones are needed to achieve formation adjustments. The control objective is to make the actual relative position Approaching the desired relative position According to the consensus algorithm, drones The speed update rule is expressed as:

[0023]

[0024] in, To control the gain, Indicates a time step;

[0025] Set up drones The acceleration is In each control cycle Inside, drones The position update formula is:

[0026]

[0027] drones The speed update formula is:

[0028]

[0029] During the local formation adjustment process, the formation status is continuously monitored, and formation error indicators are defined. for:

[0030]

[0031] when Less than the preset error threshold At that time, it was considered that the formation had returned to a stable state.

[0032] Preferably, the method further includes the following step: when the speed of adjacent drones cannot be accurately obtained. At that time, Kalman filtering is used to estimate the speed of adjacent UAVs.

[0033] Preferably, step 4 includes the following steps:

[0034] The relative position vector between the two drones is The relative velocity vector is Introducing prediction time ,predict Distance between the two drones after the time interval :

[0035] Let the safety threshold be ,when At that time, a collision warning signal will be issued.

[0036] Based on the drone's maximum flight speed and minimum safe distance To calculate the prediction time :

[0037] Gravitational potential field To guide the drone toward the target location, a repulsive potential field is used. The formula for enabling a drone to avoid obstacles is as follows: in, It is the repulsion coefficient. It is the range of the repulsive force.

[0038] Assume the reliability of the sensor data is . ,when When ∈ (0, 0.3], it indicates a high degree of information incompleteness, and the gravitational potential field and repulsive potential field are weighted:

[0039] Overall optimal potential field Optimized synergy ,

[0040] The drone based on the optimized combined force Calculate the acceleration = Thus, the adjusted speed is obtained. , where m is the mass of the drone.

[0041] Based on the above, the present invention also discloses a drone swarm cooperative formation control system, comprising:

[0042] The information fusion and estimation module is used to acquire multi-source sensor data in real time, fuse multi-source sensor data using the extended Kalman filter algorithm, and perform state estimation when some sensor data is missing to obtain state information.

[0043] The interference detection and identification module is used to detect interference based on state information by monitoring relevant parameters of communication signals and sensor data, and to identify the type and intensity of interference by using spectrum and waveform analysis.

[0044] The formation adjustment strategy module is used to determine the interfered UAV and neighboring UAVs based on the communication neighborhood. The interfered UAV and neighboring UAVs use a distributed control algorithm to perform local formation adjustment, and use Kalman filtering to estimate the speed of neighboring UAVs under incomplete information to optimize the algorithm, thereby realizing the collaborative formation control of the UAV swarm.

[0045] The collision warning and avoidance module is used to establish a warning model based on distance and speed during the collaborative formation control of UAV swarms, and to achieve collision avoidance using an optimized artificial potential field method.

[0046] Based on the above technical solution, the beneficial effects of this invention are as follows: This invention discloses a method and system for cooperative formation control of UAV swarms. The method involves: acquiring multi-source sensor data in real time; fusing the multi-source sensor data using an extended Kalman filter algorithm; performing state estimation when some sensor data is missing to obtain state information; detecting interference by monitoring communication signals and related parameters of sensor data; identifying the type of interference and assessing its intensity using spectrum and waveform analysis; determining the interfered UAV and neighboring UAVs based on the communication neighborhood; using a distributed control algorithm to perform local formation adjustments for the interfered UAV and neighboring UAVs; and using a Kalman filter to estimate the speed of neighboring UAVs under incomplete information conditions to optimize the algorithm; establishing an early warning model based on distance and speed; and using an optimized artificial potential field method to achieve collision avoidance. This invention provides a UAV swarm cooperative formation control technology under incomplete information conditions to suppress transient electromagnetic interference, reduce the probability of collisions in UAV swarm formations, and improve the stability and reliability of UAV swarm cooperative formation flight. Attached Figure Description

[0047] Figure 1 This is a flowchart illustrating a method for collaborative formation control of a drone swarm in one embodiment;

[0048] Figure 2 This is a schematic block diagram of a drone swarm collaborative formation control system in one embodiment. Detailed Implementation

[0049] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0050] like Figure 1 As shown, this embodiment provides a method for cooperative formation control of unmanned aerial vehicle (UAV) swarms, including the following steps:

[0051] Step 1: Acquire multi-source sensor data in real time, fuse the multi-source sensor data using the extended Kalman filter algorithm, and perform state estimation when some sensor data is missing to obtain state information.

[0052] 1.1 Sensor Selection and Data Acquisition

[0053] In collaborative formation control of UAV swarms under conditions of incomplete information, to improve the accuracy of acquiring information about the UAVs' own status and the surrounding environment, multiple sensors such as inertial measurement units (IMUs), lidar, and GPS are installed on each UAV. IMUs can measure the angular velocity and acceleration of the UAVs, lidar is used to obtain distance information of surrounding obstacles, and GPS provides the global position information of the UAVs.

[0054] Data acquisition from each sensor is performed at a certain sampling frequency. Let the sampling frequency of the IMU be... The sampling frequency of the lidar is The GPS sampling frequency is Normally, This is because the IMU needs to measure the dynamic changes of the drone in real time, while GPS updates relatively infrequently.

[0055] 1.2 Principle of Extended Kalman Filter (EKF) Algorithm

[0056] The Extended Kalman Filter (EKF) algorithm is used to fuse data from different sensors. EKF is a method for state estimation of nonlinear systems. Its basic idea is to linearize the nonlinear system and then apply the Kalman filter framework for state estimation.

[0057] Let the state vector of the UAV be... It contains information such as the drone's position, velocity, and attitude. The system's state equation can be expressed in a nonlinear form: in, It is the system's control input. It is process noise, and follows a Gaussian distribution. , It is the covariance matrix of the process noise.

[0058] The measurement equation is: in, It is the sensor's measured value. It is measurement noise, which follows a Gaussian distribution. , It is the covariance matrix of the measured noise.

[0059] 1.3 EKF Algorithm Steps

[0060] 1) Initialization

[0061] Initialize the state vector of the Kalman filter Covariance Matrix . For the initial state The estimate, This indicates the uncertainty of the initial estimate.

[0062] 2) State prediction

[0063] State prediction is performed based on IMU data. First, the state transition matrix is ​​calculated. It is a state equation For the state vector Jacobian matrix: in, It is the predicted state vector.

[0064] Then, state prediction is performed:

[0065] Covariance matrix prediction:

[0066] 3) Measurement Update

[0067] Upon receiving data from LiDAR and GPS, a measurement update is performed. First, the measurement matrix is ​​calculated. It is the measurement equation. For the state vector Jacobian matrix:

[0068] Calculate Kalman gain :

[0069] Correcting the state vector using measurement data:

[0070] Update the covariance matrix: .

[0071] 1.4 Missing Information Estimation

[0072] If some sensor data is interfered with or lost, state estimation is performed based on the Kalman filter's estimates. For example, when GPS signals are lost, the drone's position is estimated using the IMU's integration results and previous state estimates, combined with the Kalman filter's predictive capabilities. Let... The state estimate when GPS signal is lost can be calculated using the following formula:

[0073] in, It is the rate of change of state calculated based on the acceleration and angular velocity measured by the IMU.

[0074] By installing multiple sensors and using the Extended Kalman Filter (EKF) algorithm to fuse and process the data, the UAV can ensure relatively accurate information acquisition even when some sensor data is interfered with or lost; at the same time, it can monitor the bit error rate of communication signals in real time. Signal strength and sensor data change rate and fluctuation range Furthermore, by using spectrum analysis and waveform feature analysis to identify interference types and assess interference intensity, the system can effectively suppress the impact of transient interference on formation and provide reliable data support for subsequent formation control.

[0075] Step 2: Detect interference by monitoring relevant parameters of communication signals and sensor data, and use spectrum and waveform analysis to identify the type of interference and assess its intensity.

[0076] 2.1 Interference Detection Principle

[0077] In complex electromagnetic environments, the communication signals and sensor data of drones are susceptible to interference. This module monitors the bit error rate of communication signals in real time. Signal strength and the rate of change of sensor data and fluctuation range To determine whether there is electromagnetic interference.

[0078] Communication signal bit error rate It is an important indicator for measuring communication quality. Exceeding the preset threshold This indicates that the communication signal may be interfered with.

[0079] signal strength It can reflect the stability of the communication link, if Abnormal fluctuations could also be a sign of interference. Abnormal signal strength fluctuations are detected by calculating a fluctuation coefficient using a fluctuation algorithm formula and comparing it to a preset fluctuation coefficient threshold. If the coefficient exceeds the threshold, abnormal signal strength fluctuations are considered to have occurred. Alternatively, during the testing period, network signal data from nodes within the communication network can be acquired using a wireless signal strength tester to obtain signal fluctuation values. These values ​​are then used as a basis for identifying suspected nodes. Further analysis of historical data for these suspected nodes over the specified period yields a node warning value. If this warning value exceeds the node warning threshold, abnormal signal strength fluctuations are considered to have occurred.

[0080] For the rate of change of sensor data Defined as the ratio of the difference in sensor data at adjacent moments to the time interval: in, yes Sensor data at any given time, yes Sensor data at any given time, It is a time interval.

[0081] Fluctuation range It can be represented by the standard deviation of sensor data: in, It is the sample size. It is the first Sensor data for each sample, It is the average value of the sample.

[0082] when Exceeding the rate of change threshold or Exceeding the fluctuation range threshold At that time, it was determined that the sensor data was being interfered with.

[0083] 2.2 Interference Type Identification Method

[0084] After detecting electromagnetic interference, it is necessary to further identify the type of interference. This module identifies the type of interference by analyzing the frequency and waveform characteristics of the signal.

[0085] For communication signals, frequency distribution is obtained using spectral analysis. Assume the communication signal is... Its spectrum It can be obtained through Fourier transform:

[0086] By analyzing the spectrum The bandwidth and center frequency of the signal can be used to make a preliminary judgment on the type of interference. For example, narrowband interference has a narrow spectral bandwidth and is usually concentrated around a specific frequency; broadband interference has a wide spectral bandwidth and covers multiple frequency ranges.

[0087] To more accurately identify interference types, waveform feature analysis can also be incorporated. A waveform similarity index for the signal can be defined. This is used to measure the similarity between a signal and a known interference waveform. Assume the known interference waveform is... Then the waveform similarity index It can be calculated using the cross-correlation function: in, It's a time delay. When Exceeding the similarity threshold At that time, the signal is considered to be similar to a known interference waveform, thus identifying the type of interference.

[0088] 2.3 Interference Intensity Assessment

[0089] In addition to identifying the type of interference, it is also necessary to assess the intensity of the interference. This module uses the signal-to-interference ratio (SIR) to evaluate the interference intensity of communication signals, which is defined as the signal power. With interference power The ratio:

[0090] For sensor data, the intensity of interference can be assessed by the degree of deviation in the sensor data. The relative deviation of the sensor data is defined. for: in, These are measurements from sensors that have been interfered with. This is an estimate under undisturbed conditions. Relative deviation. The larger the value, the stronger the interference.

[0091] Through the above methods of interference detection, type identification, and intensity assessment, this module can detect electromagnetic interference in a timely and accurate manner under conditions of incomplete information, and provide an important basis for subsequent formation adjustments and collision avoidance.

[0092] Step 3: Based on the communication neighborhood, determine the interfered UAV and neighboring UAVs. The interfered UAV and neighboring UAVs use a distributed control algorithm to perform local formation adjustment, and use Kalman filtering to estimate the speed of neighboring UAVs under incomplete information to optimize the algorithm.

[0093] 3.1 Determination of the interfered UAV and neighboring UAVs

[0094] Under conditions of incomplete information, when the interference detection and identification module detects transient electromagnetic interference, it needs to accurately determine the affected drones and their neighboring drones. Assume there are a total of [number missing] drones in the drone swarm. 1 drone, each drone Having a unique number and .

[0095] Define communication neighborhood For drones A collection of other drones that can communicate directly. When a drone is detected... When the communication signal bit error rate exceeds a preset threshold or sensor data is abnormal, the drone is determined to be... Interference occurred. At this time, the group of its neighboring drones is... Communication neighborhood It can be determined in the following ways:

[0096] Set up drones The position is Other drones The position is The communication radius is .like ,but ,in This represents the Euclidean norm.

[0097] 3.2 Application of Distributed Control Algorithms

[0098] A distributed control algorithm is employed to allow the interfered UAV and its neighboring UAVs to exchange position and speed information via local communication. Let the UAV... The speed is Each drone Based on the received positions of neighboring drones and speed ( It can autonomously calculate and adjust its position and attitude.

[0099] Define the expected relative position For drones Relative to neighboring drones The desired relative position. To achieve formation adjustments, drones... The control objective is to make the actual relative position Approaching the desired relative position .

[0100] According to the consensus algorithm, drones The speed update rule can be expressed as:

[0101]

[0102] in, To control the gain, This represents the time step. The meaning of this formula is: [Unmanned Aerial Vehicle] Adjust its speed according to the relative positional deviation of adjacent drones so that the entire local formation approaches the desired relative positional relationship.

[0103] 3.3 Local Formation Adjustment Process

[0104] After identifying the affected drone and its neighboring drones, and calculating the adjusted speed using a distributed control algorithm, the drone... Send control commands to the actuators to adjust the local formation.

[0105] Set up drones The acceleration is According to Newton's second law (in For drones quality For use in drones (Resultant force), the actuator generates a corresponding force according to the speed adjustment command. This changes the drone's acceleration. .

[0106] In each control cycle Inside, drones The position update formula is:

[0107]

[0108] The speed update formula is:

[0109]

[0110] During the adjustment process, the formation status is continuously monitored. Formation error indices are defined. for:

[0111]

[0112] when Less than the preset error threshold At that time, it was considered that the formation had returned to a stable state.

[0113] 3.4 Optimization Considering Incomplete Information

[0114] Under conditions of incomplete information, since some sensor data may be interfered with or lost, the above algorithm needs to be optimized. For example, when the speed information of neighboring drones cannot be accurately obtained, Kalman filtering can be used to estimate the speed of neighboring drones.

[0115] Assume adjacent drones The speed estimate is Based on the prediction and update steps of Kalman filtering, it is continuously updated. This is then applied to distributed control algorithms. This allows for effective formation adjustments even under conditions of incomplete information, improving formation stability and resilience.

[0116] Step 4: Establish an early warning model based on distance and velocity, and use an optimized artificial potential field method to achieve collision avoidance.

[0117] 4.1 Establishment of Collision Warning Model

[0118] In collaborative formation flying of UAVs, a collision early warning model based on distance and speed is established to monitor the collision risk between UAVs in real time. Let the UAVs be... and drones The position vectors are respectively and The velocity vectors are respectively and .

[0119] The relative position vector between the two drones is The relative velocity vector is .

[0120] Distance between the two drones It can be calculated using the Euclidean distance formula:

[0121] To more accurately assess collision risk, a prediction time is introduced. ,predict Distance between the two drones after the time interval :

[0122] Let the safety threshold be ,when At that time, a collision warning signal is issued. To determine the appropriate prediction time... It can be based on the maximum flight speed of the drone and minimum safe distance To calculate:

[0123] 4.2 Design of Collision Avoidance Algorithm

[0124] Upon issuing a collision warning signal, the collision avoidance algorithm is activated. This embodiment employs an artificial potential field method combined with an optimization strategy under incomplete information conditions.

[0125] In the artificial potential field method, the UAV is subjected to both gravitational and repulsive potential fields. The gravitational potential field... To guide the drone toward the target location, a repulsive potential field is used. To make the drone avoid obstacles (other drones).

[0126] The gravitational potential field can be expressed as: in, It is the gravitational coefficient. This is the target location for the drone.

[0127] The repulsive potential field can be represented as: in, It is the repulsion coefficient. It is the range of the repulsive force.

[0128] Overall momentum The combined force on the drone .

[0129] Under conditions of incomplete information, since some sensor data may be affected by interference or loss, the artificial potential field method needs to be optimized. Let the reliability of the sensor data be... ( ),when When the value ∈ (0, 0.3], it indicates a high degree of information incompleteness. In this case, the gravitational and repulsive potential fields are weighted:

[0130] Overall optimal potential field Optimized synergy .

[0131] The drone based on the optimized combined force Adjust the flight attitude and speed. Assume the mass of the drone is... According to Newton's second law The acceleration can be calculated. Thus, the adjusted speed is obtained. ,in It is the control cycle.

[0132] Through the above collision warning model and collision avoidance algorithm, it is possible to monitor the collision risk between drones in real time under incomplete information conditions and take effective collision avoidance measures in a timely manner to reduce the probability of collisions in drone swarm formations.

[0133] It should be understood that although the steps in the flowchart above are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowchart above may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0134] like Figure 2 As shown, in one embodiment, a drone swarm cooperative formation control system is provided, comprising:

[0135] The information fusion and estimation module is used to acquire multi-source sensor data in real time, fuse multi-source sensor data using the extended Kalman filter algorithm, and perform state estimation when some sensor data is missing to obtain state information.

[0136] The interference detection and identification module is used to detect interference based on state information by monitoring relevant parameters of communication signals and sensor data, and to identify the type and intensity of interference by using spectrum and waveform analysis.

[0137] The formation adjustment strategy module is used to determine the interfered UAV and neighboring UAVs based on the communication neighborhood. The interfered UAV and neighboring UAVs use a distributed control algorithm to perform local formation adjustment, and use Kalman filtering to estimate the speed of neighboring UAVs under incomplete information to optimize the algorithm, thereby realizing the collaborative formation control of the UAV swarm.

[0138] The collision warning and avoidance module is used to establish a warning model based on distance and speed during the collaborative formation control of UAV swarms, and to achieve collision avoidance using an optimized artificial potential field method.

[0139] The above are merely preferred embodiments of the present application and are not intended to limit the embodiments of the present application. For those skilled in the art, the embodiments of the present application can have various modifications and variations. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the embodiments of the present application should be included within the protection scope of the embodiments of the present application.

Claims

1. A method for cooperative formation control of unmanned aerial vehicle (UAV) swarms, characterized in that, Includes the following steps: Step 1: Acquire multi-source sensor data in real time, fuse the multi-source sensor data using the extended Kalman filter algorithm, and perform state estimation when some sensor data is missing to obtain state information; Step 2: Based on the status information, interference is detected by monitoring relevant parameters of communication signals and sensor data, and the type and intensity of interference are identified by spectrum and waveform analysis. Step 3: Based on the communication neighborhood, determine the interfered UAV and neighboring UAVs. The interfered UAV and neighboring UAVs use a distributed control algorithm to perform local formation adjustment. Under incomplete information, Kalman filtering is used to estimate the speed of neighboring UAVs and optimize the algorithm to realize UAV swarm cooperative formation control. Step 4: During the collaborative formation control of the UAV swarm, an early warning model is established based on distance and speed, and an optimized artificial potential field method is used to achieve collision avoidance.

2. The method for cooperative formation control of unmanned aerial vehicle (UAV) swarms according to claim 1, characterized in that, Data from multiple sensors is collected by the inertial measurement unit, lidar, and GPS installed on the drone, with the sampling frequency decreasing sequentially among the inertial measurement unit, lidar, and GPS.

3. The method for cooperative formation control of unmanned aerial vehicle (UAV) swarms according to claim 2, characterized in that, State estimation is performed when some sensor data is missing. The calculation formula is as follows: in, This is the state estimate when the GPS signal is lost. yes The predicted state vector at each time step. It is the rate of change of state calculated based on the acceleration and angular velocity measured by the inertial measurement unit.

4. The method for cooperative formation control of unmanned aerial vehicle (UAV) swarms according to claim 1, characterized in that, Step 2 includes the following steps: Real-time acquisition of monitoring information, including any one of the following: bit error rate of communication signal, signal strength, and rate of change and fluctuation range of sensor data; When the monitored information exceeds a preset threshold, electromagnetic interference is determined to have occurred. The type of interference is identified by analyzing the frequency and waveform characteristics of the signal. The signal-to-interference ratio is used to assess the interference intensity of the communication signal, and the relative deviation of the sensor data is used to assess the interference intensity of the sensor data.

5. The method for cooperative formation control of unmanned aerial vehicle (UAV) swarms according to claim 4, characterized in that, Step 3 includes the following steps: Suppose there are a total of 1 drone, each drone Having a unique number and When detecting drones Electromagnetic interference occurs; define the communication neighborhood. For drones The communication neighborhood is a collection of other neighboring drones that can communicate directly. Determination method: Set up a drone The position is Other drones The position is The communication radius is ,like ,but ,in Denotes the Euclidean norm; Set up drones The speed is Each drone Based on the received positions of neighboring drones and speed ( Define the expected relative position. For drones Relative to neighboring drones The expected relative positions of the drones are needed to achieve formation adjustments. The control objective is to make the actual relative position Approaching the desired relative position According to the consensus algorithm, drones The speed update rule is expressed as: in, To control the gain, Indicates a time step; Set up drones The acceleration is In each control cycle Inside, drones The position update formula is: drones The speed update formula is: During the local formation adjustment process, the formation status is continuously monitored, and formation error indicators are defined. for: when Less than the preset error threshold At that time, it was considered that the formation had returned to a stable state.

6. The method for cooperative formation control of unmanned aerial vehicle (UAV) swarms according to claim 5, characterized in that, It also includes the following steps: when the speed of adjacent drones cannot be accurately obtained. At that time, Kalman filtering is used to estimate the speed of adjacent UAVs.

7. The method for cooperative formation control of unmanned aerial vehicle (UAV) swarms according to claim 5, characterized in that, Step 4 includes the following steps: The relative position vector between the two drones is The relative velocity vector is Introducing prediction time ,predict Distance between the two drones after the time interval : Let the safety threshold be ,when At that time, a collision warning signal will be issued. Based on the drone's maximum flight speed and minimum safe distance To calculate the prediction time : Gravitational potential field To guide the drone toward the target location, a repulsive potential field is used. The formula for enabling a drone to avoid obstacles is as follows: in, It is the repulsion coefficient. It is the range of the repulsive force. Assume the reliability of the sensor data is . ,when When ∈ (0, 0.3], it indicates a high degree of information incompleteness, and the gravitational potential field and repulsive potential field are weighted: Overall optimal potential field Optimized synergy , The drone based on the optimized combined force Calculate the acceleration = Thus, the adjusted speed is obtained. , where m is the mass of the drone.

8. A drone swarm collaborative formation control system, characterized in that, include: The information fusion and estimation module is used to acquire multi-source sensor data in real time, fuse multi-source sensor data using the extended Kalman filter algorithm, and perform state estimation when some sensor data is missing to obtain state information. The interference detection and identification module is used to detect interference based on state information by monitoring relevant parameters of communication signals and sensor data, and to identify the type and intensity of interference by using spectrum and waveform analysis. The formation adjustment strategy module is used to determine the interfered UAV and neighboring UAVs based on the communication neighborhood. The interfered UAV and neighboring UAVs use a distributed control algorithm to perform local formation adjustment, and use Kalman filtering to estimate the speed of neighboring UAVs under incomplete information to optimize the algorithm, thereby realizing the collaborative formation control of the UAV swarm. The collision warning and avoidance module is used to establish a warning model based on distance and speed during the collaborative formation control of UAV swarms, and to achieve collision avoidance using an optimized artificial potential field method.

Images