A multi-missile cooperative anti-windup control method based on autonomous obstacle avoidance and network connectivity maintenance

Abstract

This invention relates to the field of aircraft guidance and control, and discloses a multi-missile cooperative anti-saturation control method based on autonomous obstacle avoidance and network connectivity maintenance, comprising the following steps: Step S1, constructing a three-dimensional kinematic model of the missile swarm based on various distance constraints including communication distance, safety distance, and collision distance; Step S2, based on the input saturation model transformation of bounded functions, constructing a dynamic model with output constraints using the idea of ​​variable transformation, and introducing a new system state to expand the dimension of the original nonlinear model to obtain a new model with anti-saturation characteristics; Step S3, designing a multi-missile autonomous obstacle avoidance control law based on an artificial potential field function to satisfy the distance constraints of external obstacle avoidance, internal collision avoidance, and network connectivity maintenance of the missile swarm. The method of this invention significantly improves the autonomous control capability and flight safety of the missile swarm while alleviating system input saturation.

Description

Technical Field

[0001] This invention relates to the field of aircraft guidance and control, specifically to a multi-missile cooperative anti-saturation control method based on autonomous obstacle avoidance and network connectivity maintenance. Background Technology

[0002] Precision-guided weapons play a crucial role in modern warfare, carrying out long-range destruction and pinpoint elimination, reshaping the current political landscape and international relations. The future battlefield will inevitably present a new, complex battlefield form under system-on-system confrontation. Modern warfare has entered an era of collaborative confrontation centered on networked platforms. Missiles and other weapon systems, as important network nodes, play a vital role in the success or failure of war by improving their comprehensive utilization. Collaborative operations between multiple missiles and platforms are the direction of future battlefield development. Multiple aircraft cooperating to perform flight missions can be equivalent to a multi-agent system, which is a system composed of multiple agents with autonomous decision-making capabilities. Each agent has the ability to interact with surrounding agents and make control decisions based on the collected information. Compared to single-agent systems, the advantage of multi-agent systems lies in their ability to leverage collective strength, performing complex tasks that a single agent cannot accomplish through mutual cooperation. In 1987, Reynolds, based on observations of biological swarms, pioneered the introduction of swarms into the field of computer science and proposed the classic Flocking model. The Flocking model and a series of subsequent biomimetic swarm models have become classic control models for unmanned system swarms.

[0003] Modern battlefield environments are highly dynamic, with the constant threat of enemy radar detection and anti-missile units. Therefore, the design of cooperative control algorithms for aircraft must consider the impact of external threat zones. Missile swarm collision avoidance faces challenges from multiple levels, including high environmental dynamism and nonlinearity. Research on missile swarm collision avoidance methods involves multiple fields such as perception, communication, and cooperative control of unmanned systems. First, research on missile swarm collision avoidance helps deepen the understanding of swarm intelligence behavior. By studying the cooperative behavior and decision-making strategies of missile swarms in collision avoidance missions, we can better understand the mechanisms and laws of swarm intelligence behavior, which is of great significance for research and theoretical modeling in the field of swarm intelligence. Autonomous missile swarm collision avoidance requires the development of efficient cooperative control algorithms for multiple unmanned systems. By exploring problems such as communication, path planning, coordination, and conflict resolution among missile swarms, cooperative collision avoidance in complex environments can be achieved, improving the cooperative penetration and strike capabilities of multiple missiles. Most existing literature employs intelligent methods or artificial potential field methods to study autonomous collision avoidance problems among multiple missiles. Intelligent methods (such as pigeon swarm algorithms and multi-agent reinforcement learning algorithms) use model iteration and agent training to obtain active obstacle avoidance strategies, exhibiting strong model generalization capabilities; however, their control accuracy often fails to meet practical engineering requirements. Artificial potential field methods are commonly used in existing literature, enabling safe distance constraints between aircraft and obstacles. However, current literature typically only guarantees collision avoidance between aircraft and obstacles, failing to prevent internal collisions among missile swarm members. Furthermore, considering the limited communication distance between missile swarm members, exceeding communication distance constraints may trigger a chain reaction effect in the network, leading to the failure of the coordinated flight mission. Existing literature lacks research on control algorithms that simultaneously consider external obstacle avoidance, internal collision avoidance, and network connectivity maintenance within the missile swarm.

[0004] The coordinated motion control of missile swarms plays a crucial role in achieving long-range coordinated penetration and strike capabilities during the mid-course phase of missile swarm flight. Due to the physical limitations of the actuators, there is a trade-off between the response speed of the missile swarm to the planned trajectory and input saturation. To ensure that the actuators can respond appropriately to coordinated guidance commands, the system input needs to meet certain control constraints. These constraints primarily include engine thrust and control surface constraints. Anti-saturation design methods generally fall into two categories: the direct method and the compensation method. The basic idea of ​​the direct method is to design bounded control signals that satisfy the performance of the control system while considering control saturation. The compensation method, also known as the anti-saturation method, introduces the input-output difference of the actuators while ignoring control saturation, and then designs the controller to compensate for the saturation effect. The control saturation problem in coordinated missile swarm flight is also largely addressed using these two approaches. Some researchers have designed model predictive controllers for the stable tracking control problem of elastic hypersonic vehicles under control deflection and state constraints. However, model predictive control methods rely on real-time rolling optimization solutions. Applying these methods to the cooperative flight control of missile swarms still faces challenges related to real-time online optimization and determining the prediction time-domain step size. Some researchers have addressed aircraft input saturation and uncertain control by introducing hyperbolic tangent and Nussbaum functions to process the control inputs, reducing the required control quantity. They then design disturbance observers to compensate for external disturbances, achieving robust adaptive control of the aircraft while satisfying control constraints. Other researchers have designed auxiliary systems to mitigate input saturation; however, these methods are all passive anti-saturation methods that occur after saturation has occurred. Although some literature proposes methods based on saturated sliding surfaces to achieve active anti-saturation control by limiting the amplitude of the input signal, these methods typically only achieve asymptotic convergence of the system error. To alleviate the contradiction between input saturation and rapid convergence of system errors, further research into novel anti-saturation control methods is needed. Based on this, this invention proposes a multi-missile cooperative anti-saturation control method based on autonomous obstacle avoidance and network connectivity maintenance. Summary of the Invention

[0005] The purpose of this invention is to provide a multi-launch cooperative anti-saturation control method based on autonomous obstacle avoidance and network connectivity maintenance, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a multi-launch cooperative anti-saturation control method based on autonomous obstacle avoidance and network connectivity maintenance, comprising the following steps:

[0007] Step S1: Based on the three-dimensional aircraft dynamics model, establish a control-oriented aircraft model. In the construction of the aircraft model, the missile group has an undirected communication topology. Based on various distance constraints, including communication distance, safety distance, and collision distance, construct a three-dimensional kinematic model of the missile group.

[0008] Step S2: Based on the input saturation model transformation of bounded functions, the idea of ​​variable transformation is adopted. Based on the three-dimensional kinematic model of the projectile group in step S1, a dynamic model with output constraints is constructed, and a new system state is introduced to expand the dimension of the original nonlinear model, resulting in a new model with anti-saturation characteristics.

[0009] Step S3: Design a multi-missile autonomous obstacle avoidance control law based on artificial potential field functions. External obstacles are equivalent to spherical obstacles. The relative distances between the aircraft and obstacles, and between aircraft, are used as independent variables to construct various artificial potential field functions to satisfy the equal distance constraints of external obstacle avoidance, internal collision avoidance, and network connectivity maintenance of the missile swarm.

[0010] Preferably, step S1 specifically involves a group of... Multiple missile systems consisting of several aircraft; aircraft The dynamic model is

[0011] (1)

[0012] in , , aircraft The three-dimensional position, For quality, , , For velocity, trajectory inclination angle, and deflection angle, , , and These are the aerodynamic drag, lift, lateral force, and engine thrust, respectively. , , For the angle of attack, sideslip angle, and bank angle;

[0013] make , members respectively Position vector and velocity vector, If we consider the control force vector, then we have:

[0014] .

[0015] Preferably: step S1 is in formula Based on this, the following control-oriented dynamic model is obtained.

[0016]

[0017] In the formula

[0018] , ,

[0019]

[0020] in , and For external disturbances affecting the system, the threat region is uniformly modeled as having a radius of [missing information]. hemispherical obstacles, using a collection It means that among them Let the number of obstacles be denoted as , and let the sensing radius (i.e., communication distance) of the aircraft be denoted as . The minimum safe distance between members is , No. The location of the threat zone is , For members The position vector, then the member The set of neighboring nodes Threat zone node set Defined as

[0021]

[0022] The mathematical description of the multi-launch autonomous obstacle avoidance and cooperative motion control problem with network connectivity preservation is as follows:

[0023]

[0024] In the formula It is a time constant. and For members Relative to members Position and velocity biases.

[0025] Preferably, step S2 specifically involves, for bounded variables... Design the following input transformation function.

[0026]

[0027] in , Variables The function has the following properties: (finds) the maximum and minimum values ​​of the given information.

[0028]

[0029] From the formula The following input was obtained With variables Relationship

[0030]

[0031] Pair Differentiate, we have

[0032]

[0033] In the formula Define variables based on this. Ensure variables and Satisfying the following one-to-one correspondence

[0034]

[0035] in , .

[0036] Preferably: step S2 is based on formula Transform the input saturation problem into an input unconstrained problem, and define the following variables:

[0037]

[0038]

[0039] in , .

[0040] Considering the undirected communication topology, define the following members. Position consistency error and speed consistency error variable

[0041]

[0042] in , , , .

[0043] Based on model Japanese style The following consistency position error system is obtained.

[0044]

[0045] in , , , .

[0046] For variables ,definition:

[0047]

[0048]

[0049]

[0050] .

[0051] Preferably, step S3 specifically involves, based on members obstacles For autonomous obstacle avoidance, the following potential function is designed.

[0052]

[0053] in , It is an intelligent agent and threat zone The relative distance, The radius of the threat zone, To determine the distance at which the obstacle's potential field function begins to apply, construct the following potential function.

[0054]

[0055] in Member and distance, , ;

[0056] Based on members and members To satisfy the communication distance constraint, design the following potential function.

[0057]

[0058] in , ;

[0059] For members Design the following fixed-time convergent nonsingular sliding surface.

[0060]

[0061] in And there are

[0062]

[0063] in For odd numbers greater than 0, satisfying , , ,satisfy .

[0064] For sliding surfaces Differentiation has

[0065]

[0066] In the formula And there are

[0067] .

[0068] Preferably, step S3 introduces three types of potential function gradient information into the control law, and designs the following intermediate virtual control instructions.

[0069]

[0070]

[0071]

[0072] in express For variables The partial derivatives, For odd numbers greater than 0, satisfying , , ,parameter . For the model medium disturbance The estimated value is generated by the following disturbance observer.

[0073]

[0074] In the formula , , , ;

[0075] For virtual instructions Tracking, designing sliding surfaces And thus

[0076]

[0077] For the system The following cooperative obstacle avoidance control law is designed.

[0078]

[0079] in ,variable derivative The following TD differentiator is used for estimation:

[0080]

[0081] In the formula , For the constant to be designed, They are respectively and The estimated value.

[0082] Compared with the prior art, the beneficial effects of this invention are as follows:

[0083] 1) Design a novel input-constrained anti-saturation system. Compared with the traditional anti-saturation control method based on auxiliary systems, the method of this invention reduces the risk of system input saturation, effectively saves system input energy consumption, and improves the flight stability of the system.

[0084] 2) Compared with existing control methods that can only achieve autonomous obstacle avoidance, this invention designs a variety of potential field functions, enabling missile groups to meet various distance constraints in complex flight environments, such as external obstacle avoidance, internal collision avoidance, and network connectivity maintenance, thus expanding the applicability of the algorithm.

[0085] 3) By combining sliding mode nonlinear control theory and fixed convergence theorem, a multi-missile autonomous obstacle avoidance and cooperative control method is designed. This not only increases the robustness of the system, but also accelerates the convergence speed of system errors, effectively improving the response speed and autonomous control capability of the missile group in mid-course flight. Attached Figure Description

[0086] Figure 1 This is a flowchart of the method of the present invention;

[0087] Figure 2 This is a communication topology diagram of the missile swarm of the present invention.

[0088] Figure 3 It is a three-dimensional flight trajectory diagram of the missile swarm.

[0089] Figure 4 This is a diagram showing the distance between the aircraft and the obstacle.

[0090] Figure 5 This is a diagram showing the distances between aircraft.

[0091] Figure 6 It is the positional consistency error of the missile group. Curve graph

[0092] Figure 7 It is the positional consistency error of the missile group. Curve graph

[0093] Figure 8 It is the positional consistency error of the missile group. Curve graph

[0094] Figure 9 This is the Fx curve of the axial control input for the missile swarm.

[0095] Figure 10 The Fy curve is the normal control input of the missile group.

[0096] Figure 11 This is the curve of the control input Fz for the missile group normal direction. Detailed Implementation

[0097] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0098] Example

[0099] Please see Figure 1-11The illustrated method for multi-launch cooperative anti-saturation control based on autonomous obstacle avoidance and network connectivity maintenance includes the following steps:

[0100] Step S1: Based on the three-dimensional aircraft dynamics model, establish a control-oriented aircraft model. In the construction of the aircraft model, the missile group has an undirected communication topology. Based on various distance constraints, including communication distance, safety distance, and collision distance, construct a three-dimensional kinematic model of the missile group.

[0101] Step S2: Based on the input saturation model transformation of bounded functions, the idea of ​​variable transformation is adopted. Based on the three-dimensional kinematic model of the projectile group in step S1, a dynamic model with output constraints is constructed, and a new system state is introduced to expand the dimension of the original nonlinear model, resulting in a new model with anti-saturation characteristics.

[0102] Step S3: Design a multi-missile autonomous obstacle avoidance control law based on artificial potential field functions. External obstacles are equivalent to spherical obstacles. The relative distances between the aircraft and obstacles, and between aircraft, are used as independent variables to construct various artificial potential field functions to satisfy the equal distance constraints of external obstacle avoidance, internal collision avoidance, and network connectivity maintenance of the missile swarm.

[0103] In this embodiment, step S1 is to establish a three-dimensional kinematic model of the missile swarm: During the mid-flight phase of a missile swarm carrying out a coordinated penetration or strike mission, it often needs to pass through dangerous areas such as enemy radar detection and anti-missile units. In order to improve the flight safety of the missile swarm, it is necessary to carry out research on a cooperative control algorithm with autonomous obstacle avoidance function.

[0104] Consider a set of Multiple missile systems consisting of several aircraft; aircraft The dynamic model is

[0105]

[0106] in , , aircraft The three-dimensional position, For quality, , , For velocity, trajectory inclination angle, and deflection angle, , , and These are the aerodynamic drag, lift, lateral force, and engine thrust, respectively. , , These are the angle of attack, sideslip angle, and bank angle.

[0107] Furthermore, , members respectively Position vector and velocity vector, The control force vector is expressed as follows:

[0108]

[0109] The definition of variables in the above formula and the formula same.

[0110] In the formula Based on this, the following control-oriented dynamic model is obtained.

[0111]

[0112] In the formula, t is time, and

[0113] , ,

[0114]

[0115] in , and External disturbances experienced by the system For quality, , , For velocity, trajectory inclination angle, and deviation angle.

[0116] In this embodiment, without loss of generality, the threat region is uniformly modeled as having a radius of... hemispherical obstacles, using a collection It means that among them Let the number of obstacles be denoted as and the aircraft's sensing radius (communication distance) be denoted as . The minimum safe distance between members is , No. The location of the threat zone is , For members The position vector, then the member The set of neighboring nodes Threat zone node set Defined as

[0117]

[0118] Given constraints on spatial configuration, communication range, internal collision avoidance, external obstacle avoidance, and input saturation, the mathematical description of the multi-missile autonomous obstacle avoidance and cooperative motion control problem with network connectivity maintenance is as follows:

[0119]

[0120] In the formula It is a time constant. and For members Relative to members Position and velocity biases.

[0121] In step S2 of this embodiment, the input saturation model transformation based on bounded functions is performed: In order to solve the system input saturation problem, for bounded control inputs... Design the following input transformation function.

[0122]

[0123] in , Variables The function has the following properties: (finds) the maximum and minimum values ​​of the given information.

[0124]

[0125] From the formula The following input was obtained With variables Relationship

[0126]

[0127] Pair Differentiate, we have

[0128]

[0129] in for The derivative, for The derivative, for The derivative of, in the equation Define variables based on this. Ensure variables and Satisfying the following one-to-one correspondence

[0130]

[0131] in , .

[0132] Furthermore, utilizing the formula Transform the input saturation problem into an input unconstrained problem, and define the following variables:

[0133]

[0134]

[0135] in The base is the natural number; , .

[0136] Furthermore, considering the undirected communication topology, the following members are defined. Position consistency error and speed consistency error variable

[0137]

[0138] in , , , .

[0139] Based on model Japanese style The following consistency position error system can be obtained.

[0140]

[0141] in , , , .

[0142] In this embodiment, to facilitate the design of the missile swarm cooperative control law, the variables are... ,definition:

[0143]

[0144]

[0145]

[0146]

[0147] in

[0148]

[0149] Furthermore, Lemma 1: For nonlinear systems Suppose there exists a Lyapunov function. ,parameter Meet the conditions , , This makes the following equation true.

[0150]

[0151] System It is stable over a fixed time, and the convergent residuals satisfy...

[0152]

[0153] in Convergence time satisfy

[0154]

[0155] In this embodiment, step S3, design of multi-missile autonomous obstacle avoidance control law based on artificial potential field function: the potential field method is a control method that simulates the potential force formed by electric charge in space, which can exert an attractive or repulsive force on spatial particles. The artificial potential field method is one of the commonly used methods to realize the distance constraint control of aircraft. This invention uses this method to design the autonomous obstacle avoidance algorithm of missile swarm, so as to realize that the missile swarm meets a variety of distance constraint conditions.

[0156] In order to achieve members obstacles For autonomous obstacle avoidance, the following potential function is designed.

[0157]

[0158] in , It is an intelligent agent and threat zone The relative distance, The radius of the threat zone, This is the distance at which the obstacle's potential field function begins to apply.

[0159] To avoid members and members An internal collision occurs, and the following potential function is constructed.

[0160]

[0161] in Member and distance, , .

[0162] Furthermore, in order to achieve member and members To satisfy the communication distance constraint, design the following potential function.

[0163]

[0164] in , .

[0165] Furthermore, using methods such as potential function, sliding mode control, and disturbance observer, a multi-missile cooperative obstacle avoidance control law is designed for the system. To achieve rapid convergence of position and velocity consistency errors, for members Design the following fixed-time convergent nonsingular sliding surface.

[0166]

[0167] in And there are

[0168]

[0169] in For odd numbers greater than 0, satisfying , , ,satisfy , This represents the position consistency error of the i-th member in the j-th degree of freedom.

[0170] For sliding surfaces Differentiation has

[0171]

[0172] In the formula And there are

[0173]

[0174] To ensure that the missile swarm satisfies multiple range constraints, three types of potential function gradient information are introduced into the control law, and the following intermediate virtual control command is designed.

[0175]

[0176]

[0177]

[0178] in express For variables The partial derivatives, For odd numbers greater than 0, satisfying , , , The control gain matrix is ​​positive definite. For virtual instruction gain, For the nominal dynamic model, For the acceleration of neighbor j, parameters , For the model medium disturbance The estimated value is generated by the following disturbance observer.

[0179]

[0180] In the formula , , , .

[0181] In order to realize virtual instructions Tracking, designing sliding surfaces And thus

[0182]

[0183] in For dynamic decoupling matrix, These are unmodeled dynamic and disturbance terms, including uncompensated terms such as aerodynamic disturbances. This is for controlling the efficiency matrix.

[0184] For the system The following cooperative obstacle avoidance control law is designed.

[0185]

[0186] in ,variable derivative The following TD differentiator is used for estimation:

[0187]

[0188] In the formula , For the constant to be designed, They are respectively and The estimated value.

[0189] In this embodiment, the following theorem holds for the aforementioned cooperative control law:

[0190] Theorem 1: Consider a set of multiple missile systems with undirected communication topology. Given constraints on spatial configuration, input saturation, internal collision avoidance, external obstacle avoidance, and communication distance, design as follows: ,Mode The sliding surface shown is as follows: The first-order bounded system is as follows: The disturbance observer shown is as follows: The control law described above holds the following conclusion:

[0191] (i) No collisions occur between members or between members and obstacles, and the connectivity of the bullet swarm network is always maintained;

[0192] (ii) Sliding surface , It is bounded and converges in a fixed time.

[0193] (iii) Sliding surface After converging to the neighborhood of 0, the consistency error and It converges to the neighborhood of 0 within a fixed time.

[0194] Proof: Construct the following Lyapunov function.

[0195]

[0196] right Differentiate along the system trajectory and apply the equation Substitute

[0197]

[0198] in This is the adjacency matrix weight. Connecting weight to the leader, To control the allocation of weights, For the actual perturbation of member i, This is the estimated value from the perturbation observer. Let the gradient of the repulsive potential function of member i with respect to the obstacle be . Let the gradient of the anti-collision potential function of member i and its neighbors be denoted as . Maintain the gradient of the potential function for the grouping of member i with its neighbors.

[0199] Further simplification

[0200]

[0201] Note that by choosing appropriate parameters for the above equation, the disturbance estimation error can be made sufficiently small, i.e., a time constant exists. , ,satisfy At this point, the above equation can be simplified to

[0202]

[0203] because , ,but asymptotically converges to the origin, because If it is bounded, then the potential function All are bounded, indicating that members With members The distance between them did not exceed the maximum communication distance. ,member With members Between, members and threat zone Since no collision occurred, conclusion (i) is proven.

[0204] Since the potential function is continuous and bounded, its gradient is also bounded, therefore there exists a constant. ,satisfy

[0205]

[0206] Since the bounded error of the NDO perturbation estimation exists, there exists a constant. ,satisfy Let the maximum speed of the aircraft be... Constructing Lyapunov functions ,right Differentiating along the system trajectory and simplifying, we have

[0207]

[0208] because , , Then there is

[0209]

[0210] Substituting the above equation into the equation , can be obtained

[0211]

[0212] in , , ,

[0213] , .

[0214] By selecting appropriate controller parameters, , According to Lemma 1, At a fixed time The convergence time is when the neighborhood of 0 is converged. satisfy

[0215]

[0216] in That is, the sliding surface , It is bounded and converges in a real fixed time, so conclusion (ii) is proved.

[0217] When the sliding surface After converging to the neighborhood of 0, let's assume... According to the formula have

[0218]

[0219] in .

[0220] when When designing Lyapunov functions Differentiating it has

[0221]

[0222] According to Lemma 1, At a fixed time Converging to the region The convergence time satisfies

[0223]

[0224] when When designing Lyapunov functions Differentiating it has

[0225]

[0226] make Then there is

[0227]

[0228] Integrating both sides of the above equation simultaneously, The convergence time satisfies

[0229] (50)

[0230] final, At a fixed time Converging to 0, where

[0231]

[0232] Therefore, conclusion (iii) is true, and the proof is complete.

[0233] Furthermore, the above method was verified through algorithm simulation: To verify the effectiveness of the algorithm proposed in this invention, a multi-missile system consisting of four aircraft was considered, given the communication topology of the missile group. To verify the cooperative obstacle avoidance effect of the algorithm, hemispherical obstacles with a radius of 20km were set at positions [100,0,15] km and [170,0,-28] km, respectively. The aircraft's perception radius was set to 20km, and the minimum safe distance between members was set to 2km. Two obstacles were set in the simulation. , Representing members respectively Distance from obstacle 1 and obstacle 2 Representatives With members The distance.

[0234] The initial states of the missile swarm are shown in Table 1. The maximum axial and normal overloads of the aircraft are set to 5g and 10g, respectively. The reference trajectories of the four aircraft are generated by a virtual leader missile (denoted as M0). The leader missile cruises at an altitude of 10km and a speed of Mach 3. The positional offset of the follower missiles relative to the virtual leader missile is set to...

[0235]

[0236] Table 1 Initial State of the Missile Swarm Simulation

[0237] The designed algorithm parameters are set as follows: potential function parameters , 40km 2.5km 3km 10km 18km 100km, control law parameters set to , , , , , , , , The potential function gain coefficient is , , , , , , .

[0238] It should be noted that, in this invention, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus.

[0239] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A multi-launch cooperative anti-saturation control method based on autonomous obstacle avoidance and network connectivity maintenance, characterized in that, Includes the following steps: Step S1: Based on the three-dimensional aircraft dynamics model, establish a control-oriented aircraft model. In the construction of the aircraft model, the missile group has an undirected communication topology. Based on various distance constraints, including communication distance, safety distance, and collision distance, construct a three-dimensional kinematic model of the missile group. Step S2: Based on the input saturation model transformation of bounded functions, the idea of ​​variable transformation is adopted. Based on the three-dimensional kinematic model of the projectile group in step S1, a dynamic model with output constraints is constructed, and a new system state is introduced to expand the dimension of the original nonlinear model, resulting in a new model with anti-saturation characteristics. Step S3: Design of multi-missile autonomous obstacle avoidance control law based on artificial potential field function. External obstacles are equivalent to spherical obstacles. The relative distance between the aircraft and the obstacle, and between the aircraft and the aircraft are used as independent variables. Various artificial potential field functions are constructed to satisfy the equal distance constraints of external obstacle avoidance, internal collision avoidance and network connectivity maintenance of the missile group. The specific process of step S1 is as follows: a group of... Multiple missile systems consisting of several aircraft; aircraft The dynamic model is （1）； in , , aircraft The three-dimensional position, For quality, , , For velocity, trajectory inclination angle, and deflection angle, , , and These are the aerodynamic drag, lift, lateral force, and engine thrust, respectively. , , For the angle of attack, sideslip angle, and bank angle; make , members respectively Position vector and velocity vector, If we consider the control force vector, then we have: （2）； Based on equation (1), step S1 yields the following control-oriented dynamic model. （3）； In the formula , , ， in , and For external disturbances affecting the system, the threat region is uniformly modeled as having a radius of [missing information]. hemispherical obstacles, using a collection It means that, among them Let the number of obstacles be denoted as , and let the sensing radius (i.e., communication distance) of the aircraft be denoted as . The minimum safe distance between members is , No. The location of the threat zone is , For members The position vector, then the member The set of neighboring nodes Threat zone node set Defined as （4）； The mathematical description of the multi-launch autonomous obstacle avoidance and cooperative motion control problem with network connectivity preservation is as follows: （5）； In the formula It is a time constant. and For members Relative to members Position and velocity biases; Specifically, step S2 involves, for bounded variables... Design the following input transformation function. （6）； in , Variables The function has the following properties: (finds) the maximum and minimum values ​​of the given information. （7）； The following input is obtained from equation (6). With variables Relationship （8）； Differentiating equation (6), we have （9）； Based on equation (9), define variables. Ensure variables and Satisfying the following one-to-one correspondence （10）； in , ; Step S2 transforms the input saturation problem into an input unconstrained problem based on equation (6), defining the following variables. （11）； （12）； in , ; Considering the undirected communication topology, define the following members. Position consistency error and speed consistency error variable （13）； in , , , ; Based on model (3) and equation (10), the following consistency position error system is obtained. （14）； in , , , ; For variables ,definition: （15）； （16）； （17）； （18）； Specifically, step S3 involves, based on members obstacles For autonomous obstacle avoidance, the following potential function is designed. （22）； in , It is an intelligent agent and threat zone The relative distance, The radius of the threat zone, To determine the distance at which the obstacle's potential field function begins to apply, construct the following potential function. （23）； in Member and distance, , ; Based on members and members To satisfy the communication distance constraint, design the following potential function. （24）； in , ; For members Design the following fixed-time convergent nonsingular sliding surface. （25）； in And there are （26）； in For odd numbers greater than 0, satisfying , , ,satisfy ; For sliding surfaces Differentiation has （27）； In the formula And there are （28）。 2. The multi-launch cooperative anti-saturation control method based on autonomous obstacle avoidance and network connectivity maintenance according to claim 1, characterized in that: Step S3 introduces three types of potential function gradient information into the control law and designs the following intermediate virtual control instructions. ; （29）； （30）； （31）； in express For variables The partial derivatives, For odd numbers greater than 0, satisfying , , ,parameter ; For the perturbation in model (3) The estimated value is generated by the following disturbance observer. （32）； In the formula , , , ; For virtual instructions Tracking, designing sliding surfaces And thus （33）； The following cooperative obstacle avoidance control law is designed for system (14). , （34）； in ,variable derivative The following TD differentiator is used for estimation: （35）； In the formula , For the constant to be designed, They are respectively and The estimated value.

Images