Four-rotor unmanned aerial vehicle formation coordination type collision avoidance attitude control method and device

Abstract

The invention provides a four-rotor unmanned aerial vehicle formation coordination type collision avoidance attitude control method and device, and the method comprises the steps: constructing a distributed control frame which comprises a four-rotor unmanned aerial vehicle collision avoidance attitude controller and a particle swarm optimization controller which work cooperatively; a kinematic state model of the four-rotor unmanned aerial vehicle is established through a collision avoidance attitude controller, an obstacle avoidance shift function is designed based on a collision cone model to dynamically adjust a reference trajectory, a fault node is isolated and a formation communication topology is maintained when a fault unmanned aerial vehicle is detected, and a cost function of model prediction control is constructed. Solving an optimal control instruction meeting the kinematics constraint, the input constraint and the stability constraint; a formation tracking precision index and an energy consumption index are defined through a particle swarm optimization controller, and control weight parameters are optimized in real time. According to the invention, the problem that the unmanned aerial vehicle formation obstacle avoidance efficiency and formation tracking are not coordinated can be solved, the calculation complexity is reduced, and the safety is improved; the influence of the fault unmanned aerial vehicle on the formation is eliminated, and the unmanned aerial vehicle energy consumption and formation tracking precision are balanced.

Description

Technical Field

[0001] The present invention belongs to the field of multi-quadrotor UAV formation, specifically, relates to a distributed control technology for multiple quadrotor UAVs. A control framework with two controllers working together is designed, including a quadrotor UAV collision avoidance attitude controller and a PSO optimization controller. Background Art

[0002] Coordinated collision avoidance attitude control for multi-UAV swarms in complex environments is a key research topic in the current field of UAV technology. With the widespread application of UAVs in environmental monitoring, air transportation, building construction, target tracking, and other fields, multi-UAV swarms, with their distributed architecture and collective collaborative operation capabilities, have significantly improved mission execution efficiency and adaptability to complex environments. On the one hand, optimized obstacle avoidance methods based on potential functions increase the algorithm's computational complexity, resulting in low obstacle avoidance efficiency and thus affecting formation tracking. On the other hand, the lack of a solution to address UAV malfunctions can lead to the failure of the entire formation mission. Furthermore, selecting appropriate weights to ensure that the UAVs maintain good tracking performance with low energy consumption is also a challenge.

[0003] Based on this, the present invention provides a coordinated collision avoidance attitude control method for a four-rotor UAV formation, which solves the problems of low obstacle avoidance efficiency and inability to cope with fault conditions of UAVs, and coordinates the tracking capability and energy consumption of the UAV formation. Summary of the Invention

[0004] The present invention proposes a coordinated collision avoidance attitude control strategy for a formation of four-rotor UAVs, which incorporates multiple schemes based on model predictive control and realizes distributed collaborative obstacle avoidance and formation tracking of multiple UAVs.

[0005] The technical solution of the present invention provides a coordinated collision avoidance attitude control method for a quadrotor UAV formation, constructing a distributed control framework, including a collaborative quadrotor UAV collision avoidance attitude controller and a particle swarm optimization controller; The collision avoidance gesture controller performs the following operations: Establish the kinematic state model of the quadrotor drone; Design an obstacle avoidance displacement function based on the collision cone model and dynamically adjust the reference trajectory according to the distance to the obstacle; When a faulty UAV is detected, the adjacency matrix is ​​reconstructed to isolate the faulty node and maintain the formation communication topology; Combining the formation tracking error, obstacle avoidance function, and inter-UAV collision avoidance function, a cost function of model predictive control is constructed to solve the optimal control instructions that satisfy kinematic constraints, input constraints, and stability constraints. The particle swarm optimization controller performs the following operations: Define formation tracking accuracy and energy consumption indicators, and construct a comprehensive evaluation fitness function; The particle swarm optimization algorithm is used to optimize the control weight parameters in real time and output them to the collision avoidance attitude controller to balance tracking accuracy and energy consumption.

[0006] Furthermore, the dynamic adjustment logic of the obstacle avoidance shift function includes: When the distance between the obstacle and the drone is within the collision warning range and meets the obstacle avoidance direction conditions, the obstacle avoidance trajectory is output; When the distance is equal to the preset safety distance, the obstacle avoidance safety trajectory is output; When the distance exceeds the obstacle avoidance detection range or the obstacle avoidance conditions are not met, the original reference trajectory is maintained.

[0007] Moreover, according to the fault identification identifier in the adjacency matrix, the faulty UAV node is automatically isolated and the node data is eliminated in the formation error calculation. Moreover, the calculation of the formation tracking error includes fusing the relative positions, obstacle avoidance displacement values ​​and reference trajectory deviations between UAVs, and introducing the weighted communication adjacency relationship of the fault state identifier.

[0008] Moreover, the design of the collision avoidance function includes generating an exponential collision penalty term based on the Euclidean distance between drones and a safety distance threshold, wherein the closer the distance is to the safety threshold, the higher the penalty intensity.

[0009] Moreover, the cost function satisfies the stability constraints based on the Lyapunov function, the infinity norm of the control input does not exceed the maximum allowed value, and the initial consistency constraints of the predicted state and the actual state.

[0010] Moreover, the execution of the particle swarm optimization algorithm includes taking the sum of squares of formation errors as a tracking accuracy indicator and the sum of squares of control inputs as an energy consumption indicator, and weighting the two to form a fitness function; and solving the optimal weight coefficient by iteratively updating the particle position and velocity.

[0011] On the other hand, the present invention provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the coordinated collision avoidance attitude control method for a formation of four-rotor drones as described above is implemented.

[0012] On the other hand, the present invention provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned coordinated collision avoidance attitude control method for a formation of four-rotor drones.

[0013] On the other hand, the present invention provides a computer program product, including a computer program, which, when executed by a processor, implements the above-mentioned coordinated collision avoidance attitude control method for a four-rotor drone formation.

[0014] The present invention provides a new solution for coordinated collision avoidance attitude control of a formation of quadrotor drones. For multiple quadrotor drones, a multi-drone distributed control method is adopted, and a control framework with two controllers working together is designed, including a quadrotor drone collision avoidance attitude controller and a PSO optimization controller. Specifically, the quadrotor drone collision avoidance attitude controller incorporates a collision cone model and an adaptive fault-tolerant mechanism into the model predictive controller. The obstacle avoidance method based on the collision cone can solve the problem of inconsistency between the obstacle avoidance efficiency of drone formations and formation tracking, reduce computational complexity, and improve safety. The adaptive fault-tolerant control mechanism can promptly eliminate the impact of faulty drones on the formation. The PSO optimization controller can balance the drone energy consumption and formation tracking accuracy. The effectiveness of the solution was verified by simulating multiple quadrotor drones using the physical simulation platform Gazebo.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1) The collision cone-based model predictive control method proposed in this paper can quickly detect potential collisions between drones and obstacles through geometric analysis, making obstacle avoidance decisions in a short time, reducing computing time and resource consumption, and achieving efficient and accurate path tracking.

[0016] 2) The proposed fault handling method creates an adjacency matrix of faulty agents, which improves the robustness of the system and ensures the stability of the formation and the continuity of the mission.

[0017] 3) The proposed particle swarm optimization (PSO) controller can calculate the appropriate weights according to the formation error and energy consumption, which can improve the formation tracking performance and control the energy consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 Schematic diagram of the coordinated collision avoidance attitude control framework for a formation of four-rotor drones proposed in an embodiment of the present invention.

[0019] Figure 2 Schematic diagram of the Gazebo physical simulation platform according to an embodiment of the present invention.

[0020] Figure 3 This is a diagram of the multi-rotor communication topology structure of an embodiment of the present invention.

[0021] Figure 4 This is a trajectory diagram of a formation using an obstacle avoidance method based on collision cones according to an embodiment of the present invention.

[0022] Figure 5It is a trajectory diagram of a formation using the obstacle avoidance method based on potential function according to an embodiment of the present invention.

[0023] Figure 6 Schematic diagram of formation error in the XYZ directions during obstacle avoidance of the UAV UAV2 according to an embodiment of the present invention.

[0024] Figure 7 2 is a schematic diagram showing a comparison of trajectories of the unmanned aerial vehicle (UAV) UAV2 during obstacle avoidance according to an embodiment of the present invention.

[0025] Figure 8 3 is a comparison diagram of the UAV formation with and without fault handling according to an embodiment of the present invention.

[0026] Figure 9 This is an embodiment of the present invention Schematic diagram of the UAV formation error in the XY direction when =1.33.

[0027] Figure 10 This is an embodiment of the present invention Schematic diagram of the UAV formation error in the XY direction when =0.51. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is described in detail below with reference to the accompanying drawings and implementation examples. It should be understood that the specific implementation examples described herein are only used to explain the present invention and are not intended to limit the invention.

[0029] An embodiment of the present invention provides a coordinated collision avoidance attitude control method for a quadrotor UAV formation. A control framework is designed that collaborates with two controllers, including a quadrotor UAV collision avoidance attitude controller and a PSO optimization controller. The collision cone model is integrated into the model predictive control algorithm through an obstacle avoidance shift function to achieve formation tracking and obstacle avoidance. An adjacency matrix based on faulty agents is created, and a new communication topology is constructed to ensure the progress of the formation mission. The particle swarm optimization (PSO) algorithm is used to optimize the weight parameters, so that the UAV formation can improve the formation tracking performance while taking energy consumption into consideration.

[0030] 1. Quadrotor UAV Collision Avoidance Attitude Controller See also Figure 1 The embodiment provides a coordinated collision avoidance attitude control method for a quadrotor UAV formation, which implements a quadrotor UAV collision avoidance attitude controller. The collision avoidance attitude controller performs the following operations: Establish the kinematic state model of the quadrotor drone; Design an obstacle avoidance displacement function based on the collision cone model and dynamically adjust the reference trajectory according to the distance to the obstacle; When a faulty UAV is detected, the adjacency matrix is ​​reconstructed to isolate the faulty node and maintain the formation communication topology; Combining the formation tracking error, obstacle avoidance function and inter-UAV collision avoidance function, a cost function of model predictive control is constructed to solve the optimal control instructions that meet the kinematic constraints, input constraints and stability constraints.

[0031] In the embodiment, the specific implementation steps of the controller are further provided as follows: Step 1: Establish the kinematic state model of the quadrotor drone: Example Considerations A drone is used to extract the position, speed and related parameters of the drone during movement, and the first The state space expression of a UAV.

[0032] A further implementation of the embodiment includes the following sub-steps: Step 1.1: Construct a kinematic model based on position, velocity, and other parameters during motion:

[0033] Among them, the drone number , Represents the position vector (horizontal, longitudinal, vertical) and yaw angle, represents the four-dimensional real space, It represents the speed of the UAV in the world coordinate system; is the yaw angle of the drone The translation matrix from the body-fixed frame to the earth-fixed frame can be expressed as:

[0034] Velocity vector ,in , and is the linear velocity of the drone in the fuselage, is the heading angular velocity; Indicates drone The time derivative of the velocity vector describes the change of the UAV's velocity in three-dimensional space; and represents the response characteristic matrix of the UAV to the control input, and , , in represents the construction of a diagonal matrix, and is with and Related gain and delay , and its control input can be expressed as ,in is the forward speed instruction, is the lateral speed command, is the upward speed command, and is the yaw velocity input. The physical input constraints of the drone need to satisfy ,in is the maximum allowed input for the drone, Expressed as The infinity norm of , that is, the maximum absolute value of each element in the vector.

[0035] Step 1.2, define UAV state variables , write the system state space expression:

[0036]

[0037] in and is the state vector, , , and Respectively represent input and output, the embodiment also records is the reference trajectory of the tracked UAV, The kinematic equations are abbreviated as: . is the state vector The time derivative, that is, the rate of change of state with time, is a nonlinear function that describes the state At the control input Dynamic behavior under action.

[0038] Step 2: Construct the formation error based on the obstacle avoidance shift function from tracking formation to obstacle avoidance, the adjacency matrix based on the fault agent, the status of different UAVs, and the relative distance between UAVs; Step 2.1: Construct an obstacle avoidance shift function based on obstacle distance, collision cone detection range, and rotation angle. Construct an obstacle avoidance shift function from tracking formation to obstacle avoidance based on the collision cone mechanism. :

[0039] in For the safety range, is the collision cone detection range, It is the rotation angle of the drone when it needs to avoid obstacles. and represent the obstacle and the UAV’s mission vector respectively. for exist The projection on is the vector pointing from the drone to the obstacle, is the current velocity vector of the UAV. is the obstacle avoidance trajectory of the UAV, , its detailed expression is as follows:

[0040] in, is the transfer matrix used for coordinate transformation, and its specific form is as follows:

[0041] angle The vector from the current position of the drone to the target position The vector from the current position of the drone to the desired position for obstacle avoidance The specific form of the angle is as follows:

[0042] Achieve a safe distance for the drone The obstacle avoidance safety trajectory , its detailed expression is as follows:

[0043] in are the coordinates of the obstacle, For a safe distance, The current time is expressed as follows:

[0044] is the x-coordinate position of the obstacle, , Vertical obstacle Ordinal coordinate position, is the x-direction coordinate of the collision cone point, is the current time, is the linear velocity of the UAV in three-dimensional space.

[0045] The dynamic adjustment logic of the above obstacle avoidance shift function includes: When the distance between the obstacle and the drone is within the collision warning range and meets the obstacle avoidance direction conditions, the obstacle avoidance trajectory is output; When the distance is equal to the preset safety distance, the obstacle avoidance safety trajectory is output; When the distance exceeds the obstacle avoidance detection range or the obstacle avoidance conditions are not met, the original reference trajectory is maintained.

[0046] The obstacle avoidance trajectory is generated by calculating the angle between the vector from the current position to the target position and the vector to the desired obstacle avoidance position, transforming the reference trajectory through the rotation matrix, and generating a circumventing path through the obstacle coordinates, safety radius, and time-related angle function.

[0047] Step 2.2: Construct an adjacency matrix based on faulty agents ,When a faulty UAV is detected, the adjacency matrix is ​​reconstructed to isolate the faulty node and maintain the formation communication topology;:

[0048] in, For intelligent agents exist The fault status at the moment, For intelligent agents exist Fault status at the moment. =1 when the agent No trouble, normal operation, =0 when the agent Completely loses effect, output is fixed at zero. Indicated by The original adjacency matrix composed of is an element of the adjacency matrix. =1 when the agent No trouble, normal operation, =0 when the agent Completely loses effect, output is fixed at zero. The zero value can eliminate , thereby eliminating defective agents For intelligent agents Therefore, the agent’s loss is taken into account in the cost function, which enables the generation of correct trajectories to achieve the task.

[0049] In practice, faulty drones are handled by assigning a fault status identifier to each drone. This identifier assumes a first value when operating normally and a second value when completely faulty. This identifier is used to dynamically adjust the adjacency matrix elements to eliminate the faulty drone's impact on the formation. When the fault indicator in the adjacency matrix reaches zero, the faulty drone node is automatically isolated and its data is removed from the formation error calculation.

[0050] Step 2.3, set relative distance , construct formation error :

[0051] in, is the adjacency matrix element, For the UAV system output. , and They are and The relative distance to the tracked drone. and is the obstacle avoidance shift function, is a diagonal matrix element, when When the UAV is tracked, it can receive information from the tracked UAV.

[0052] The calculation of formation tracking error includes fusing the relative positions, obstacle avoidance displacement values ​​and reference trajectory deviations between UAVs, and introducing the weighted communication adjacency relationship of fault state identifiers.

[0053] Step 3: Construct a collision avoidance function between drones, and construct a cost function for drones based on the drone formation error and the optimal input of the system, and finally give the constraint conditions; Step 3.1: Build a collision avoidance function between drones :

[0054] in Representative drones, Represents the first UAV coordinates ( , , ) and UAV coordinates ( , , ) is the Euclidean distance between them. is the minimum safe distance, is the collision weight parameter, In the cost function, the collision avoidance function generates an exponential collision penalty term based on the Euclidean distance between drones and the safety distance threshold. The collision avoidance weight increases as the distance decreases, i.e., the closer the distance is to the safety threshold, the higher the penalty intensity.

[0055] Step 3.2, set the cost function including the collision avoidance function, and construct the first Cost function for flying a drone:

[0056] in, is the differential element of the integral variable, used to represent the time interval The objective function is accumulated internally. For the Cost function for flying a drone; is the prediction time domain, is the number of time steps, is a time point within the forecast time range, , Divided into time steps, each step is , that is, the sampling period, For control input The constraints that belong to For the The predicted value of the control input signal of the UAV, Indicates that in the optimization process, we are looking for the objective function Minimized control signal mathematical operations. is the predicted formation error, Indicates time When, The predicted formation error vector of the UAVs In the weight matrix The squared norm under . and yes Moment drone, the The predicted state of the drone, Indicates time When, The predictive control input vector of the UAV In the weight matrix The squared norm under . is the symmetric weighted matrix of formation error, is the output of the controller optimized by PSO.

[0057] Step 3.3, give the corresponding model predictive controller constraints:

[0058]

[0059]

[0060]

[0061] formula represents the kinematic constraints of the drone for predicting future trajectories, for Moment The derivative of the predicted state of the drone with respect to time, Represents the initial state of the predicted state, that is hour. Indicates time When, The predictive control input vector of the UAV The infinite norm of , that is, the maximum absolute value of each element in the vector, Physical constraints for the drone. is the constructed Lyapunov function, yes The time derivative, ,in , yes The time derivative, is based on The constructed state vector is used to assist in the derivation of the controller. and are all positive control constants. For the drones Always in prediction state The system predicts input when .

[0062] This time is a nonlinear auxiliary controller derived from the Lyapunov stability theorem, , which is specifically expressed as:

[0063] Among them, the degree matrix , is the adjacency matrix element, Indicates drone Is it possible to obtain information from the tracked drone? If so, ,on the contrary , yes The reciprocal of is the response characteristic matrix of the UAV to the control input, yes The second derivative of 、 They are drone, the Velocity vector of a drone. and They are and The second derivative of . For the The translation matrix from the drone body fixed frame to the earth fixed frame, is the translation matrix The derivative with respect to time.

[0064] According to the above model predictive controller constraints, the cost function satisfies the stability constraints based on the Lyapunov function, the infinity norm of the control input does not exceed the maximum allowed value, and the initial consistency constraints of the predicted state and the actual state.

[0065] 2. PSO Optimization Controller Design PSO optimization controller and output the optimal solution Provide a collision avoidance attitude controller for a quadcopter drone. The specific implementation steps in the embodiment are as follows: Step 4: Construct the fitness function of the PSO algorithm by defining the formation error performance index and energy consumption index; Step 4.1, define the formation error performance index, that is, the tracking accuracy index :

[0066] Indicates time When, The predictive control input vector of the UAV In the weight matrix The squared norm under .

[0067] Step 4.2, define energy consumption indicators :

[0068] Indicates time When, Optimal predictive control input vector for a UAV In the optimal weight matrix The squared norm under . For the formula The optimal solution obtained.

[0069] Step 4.3, construct fitness function :

[0070] Step 5: According to the classic speed and position update formula of the PSO algorithm, the optimal value of the fitness function is solved and the output PSO optimization controller is constructed; Step 5.1, the particle velocity and position update formula is as follows:

[0071]

[0072] in, and Represents the particle at time speed and position, and Represents the particle at time +1 for speed and position, is the inertia weight, and is the learning factor, and is a random number, is the optimal position found by the particle itself, is the optimal position found in the particle’s group, and is the next time step of the particle Position and velocity status.

[0073] Step 5.2, the PSO optimization controller is designed as:

[0074] in, express The inverse function of .

[0075] like Figure 1 As shown, by obtaining the reference trajectory of the UAV and current location , construct the fitness function , including formation error indicators and energy consumption indicators Then, the nonlinear MPC solution method is used to solve the minimum value of the fitness function as the fitness of the particle, and finally the PSO algorithm is used to solve the position of the optimal particle. And use it as the system input weight of the model predictive controller.

[0076] In specific implementation, control can be performed through a collaborative control framework, and the optimal control input can be applied to the UAV actuator to achieve obstacle avoidance collaboration and formation tracking.

[0077] The collision avoidance attitude controller uses a model predictive control algorithm to optimize the formation tracking error, the collision avoidance cost between drones, and the control energy consumption, and solves the optimal control input under the conditions of satisfying the dynamic constraints and input constraints. A PSO optimization controller is used to construct a fitness function including a formation tracking performance index and an energy consumption index, dynamically calculate the optimal weight parameters through a particle swarm optimization algorithm, and output the optimal weight parameters to the collision avoidance posture controller; The operating logic of the collaborative control framework is as follows: the PSO optimization controller updates the weight parameters based on real-time formation tracking performance and energy consumption data, and the collision avoidance attitude controller uses the updated weight parameters to solve the optimal control input. The dual controllers form a closed-loop optimization mechanism.

[0078] In order to facilitate understanding of the technical effects of the present invention, Figure 2 The effectiveness of the formation controller proposed in this invention is verified by using the PX4 quadrotor model on the Gazebo physical simulation platform shown in the figure. The formation communication topology is shown in the figure. Figure 3 As shown, there are multiple quad-rotor drones including UAV0 and UAV1, and UAV0 is the leader.

[0079] In order to evaluate the effectiveness and innovation of the proposed control strategy, some comparative tests were carried out, and attempts were made to compare the obstacle avoidance efficiency and formation maintenance ability of the collision cone-based model predictive controller, the formation maintenance ability after adding a fault handling mechanism, and the formation tracking performance under the PSO algorithm optimization with the performance of the existing quadrotor formation model predictive controller.

[0080] A. Performance comparison test of quadrotor UAV collision avoidance attitude controller The controller adopts the collision cone-based obstacle avoidance method. The UAV formation trajectory and tracking error in the XYZ direction generated by this method are as follows: Figure 4 and 6 (a), where Figure 4 Part (a) is the formation trajectory curve of the UAVs in the XY plane, and part (b) is the formation trajectory curve of the UAVs in the XYZ three-dimensional space. Figure 6 Part (a) is the error of the collision cone-based obstacle avoidance method in the XYZ direction. The performance of the original controller is as follows Figure 5 and 6 As shown in (b), Figure 5 Part (a) is the formation trajectory curve of the original controller drone in the XY plane, and part (b) is the formation trajectory curve of the drone in the XYZ three-dimensional space. Figure 6 Part (b) shows the error of the potential function-based obstacle avoidance method in the XYZ direction. By comparison, we can conclude that the collision cone-based obstacle avoidance method has better timeliness and formation tracking capabilities. The trajectory comparison of the drone in the XY plane during obstacle avoidance is shown in the figure below. Figure 7 As shown, Figure 7Part (a) is the obstacle avoidance method based on collision cones, and part (b) is the obstacle avoidance method based on potential functions. By comparing the changes in the trajectories of the two over time, the timeliness of the obstacle avoidance method based on collision cones is further verified. Figure 8 As shown, Figure 8 Part (a) is the solution with a fault, and part (b) is the solution without a fault. It can be seen that the quadrotor UAV collision avoidance attitude controller has good obstacle avoidance performance and formation maintenance capabilities. B. Comparative test of PSO optimized controller The performance of selecting the output of the PSO optimization controller of the present invention as the weight parameter is as follows Figure 9 and 10 As shown, Figure 9 and Figure 10 The error curves of the drone in the XY direction are shown when the optimization parameters are 1.33 and 0.51 respectively. Figure 9 and Figure 10 ,It can be concluded that the error in the tracking direction of using the optimized parameters of the optimizer is smaller than that of using custom parameters, which can further optimize the tracking performance of the formation while ensuring lower energy consumption.

[0081] In specific implementation, the method proposed in the technical solution of the present invention can be automatically run by those skilled in the art using computer software technology. System devices that implement the method, such as computer-readable storage media that store the corresponding computer program of the technical solution of the present invention and computer equipment that runs the corresponding computer program, should also be within the scope of protection of the present invention.

[0082] The following describes the electronic device for coordinated collision avoidance attitude control of a four-rotor UAV formation provided by the present invention. The electronic device for coordinated collision avoidance attitude control of a four-rotor UAV formation described below and the coordinated collision avoidance attitude control method for a four-rotor UAV formation described above can be referenced to each other.

[0083] The electronic device may include: a processor, a communications interface, a memory, and a communications bus, wherein the processor, the communications interface, and the memory communicate with each other via the communications bus. The processor may call logic instructions in the memory to execute a coordinated collision avoidance attitude control method for a formation of four-rotor drones, which primarily includes the software processing portion of the aforementioned steps.

[0084] Furthermore, the logical instructions in the aforementioned memory can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage media include various media capable of storing program code, such as USB flash drives, mobile hard drives, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0085] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the software processing part of the coordinated collision avoidance attitude control method for a four-rotor drone formation provided by the above-mentioned methods.

[0086] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the software processing part of the coordinated collision avoidance attitude control method for a formation of four-rotor drones provided by the above-mentioned methods.

[0087] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0088] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0089] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A coordinated collision avoidance attitude control method for a quadrotor UAV formation, characterized in that: Build a distributed control framework, including a collaborative quadrotor UAV collision avoidance attitude controller and a particle swarm optimization controller; The collision avoidance gesture controller performs the following operations: Establish the kinematic state model of the quadrotor drone; Design an obstacle avoidance displacement function based on the collision cone model and dynamically adjust the reference trajectory according to the distance to the obstacle; When a faulty UAV is detected, the adjacency matrix is ​​reconstructed to isolate the faulty node and maintain the formation communication topology; Combining the formation tracking error, obstacle avoidance function, and inter-UAV collision avoidance function, a cost function of model predictive control is constructed to solve the optimal control instructions that satisfy kinematic constraints, input constraints, and stability constraints. The particle swarm optimization controller performs the following operations: Define formation tracking accuracy and energy consumption indicators, and construct a comprehensive evaluation fitness function; The particle swarm optimization algorithm is used to optimize the control weight parameters in real time and output them to the collision avoidance attitude controller to balance tracking accuracy and energy consumption.

2. The coordinated collision avoidance attitude control method for a quadrotor UAV formation according to claim 1, characterized in that: The dynamic adjustment logic of the obstacle avoidance shift function includes: When the distance between the obstacle and the drone is within the collision warning range and meets the obstacle avoidance direction conditions, the obstacle avoidance trajectory is output; When the distance is equal to the preset safety distance, the obstacle avoidance safety trajectory is output; When the distance exceeds the obstacle avoidance detection range or the obstacle avoidance conditions are not met, the original reference trajectory is maintained.

3. The coordinated collision avoidance attitude control method for a quadrotor UAV formation according to claim 1, characterized in that: According to the fault identification identifier in the adjacency matrix, the faulty drone node is automatically isolated and the node data is eliminated in the formation error calculation.

4. The method for coordinated collision avoidance attitude control of a quadrotor UAV formation according to claim 1, characterized in that: The calculation of the formation tracking error includes fusing the relative positions, obstacle avoidance displacement values ​​and reference trajectory deviations between UAVs, and introducing the weighted communication adjacency relationship of fault state identifiers.

5. The coordinated collision avoidance attitude control method for a quadrotor UAV formation according to claim 1, characterized in that: The design of the collision avoidance function includes generating an exponential collision penalty term based on the Euclidean distance between drones and a safety distance threshold, where the closer the distance is to the safety threshold, the higher the penalty intensity.

6. The coordinated collision avoidance attitude control method for a quadrotor UAV formation according to claim 1, characterized in that: The cost function satisfies the stability constraints based on the Lyapunov function, the infinity norm of the control input does not exceed the maximum allowed value, and the initial consistency constraints of the predicted state and the actual state.

7. The coordinated collision avoidance attitude control method for a quadrotor UAV formation according to claim 1, characterized in that: The execution of the particle swarm optimization algorithm includes taking the sum of squares of formation errors as a tracking accuracy indicator and the sum of squares of control inputs as an energy consumption indicator, weighting the two to form a fitness function; and solving the optimal weight coefficient by iteratively updating the particle position and velocity.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, a coordinated collision avoidance attitude control method for a four-rotor drone formation is implemented as described in any one of claims 1 to 7.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, a coordinated collision avoidance attitude control method for a four-rotor drone formation is implemented as described in any one of claims 1 to 7.

10. A computer program product comprising a computer program, characterized in that: When the computer program is executed by a processor, a coordinated collision avoidance attitude control method for a four-rotor drone formation is implemented as described in any one of claims 1 to 7.

Images