A path planning cooperative control method for shipboard heterogeneous strong coupling unmanned equipment

Abstract

The application provides a path planning cooperative control method for ship-borne heterogeneous strong coupling unmanned equipment, relates to the technical field of intelligent control and path planning, and comprises the following steps: considering the longitudinal, transverse and yaw motion of an unmanned tractor and an unmanned carrier aircraft on the deck plane of an aircraft carrier, a three-degree-of-freedom coupling motion mathematical model of the heterogeneous unmanned equipment is established; a virtual force field is constructed according to an improved artificial potential field method, a field strength adjusting operator and a heading disturbance injection term are introduced, and discrete path points are generated; the discrete path points are converted into continuous expected trajectories by using a Gaussian smoothing function; precise trajectory tracking of the unmanned equipment is realized through an adaptive neural network double-layer control architecture, the upper layer controls the unmanned carrier aircraft, the lower layer controls the unmanned tractor, and cooperative control is realized by using the coupling relationship. The application introduces a first-order filter (dynamic surface control technology), effectively suppresses the differential explosion phenomenon, reduces the algorithm complexity, and improves the engineering applicability of the controller.

Description

Technical Field

[0001] This invention relates to the technical field of intelligent control and path planning, and more particularly to a path planning and collaborative control method for shipborne heterogeneous and strongly coupled unmanned equipment. Background Technology

[0002] With the development of artificial intelligence technology, future aircraft carrier operations will exhibit characteristics of deep intelligence, distributed collaboration, and cross-domain integration. Carrier-based aircraft and towing vehicles, as widely used deck equipment, require the development of intelligent unmanned carrier-based aircraft and towing vehicles to provide key technological support for efficient and safe aircraft carrier deck operations. Furthermore, the movement of carrier-based aircraft on the deck requires towing vehicles as auxiliary equipment to provide necessary operational support. For example, when unmanned towing vehicles and unmanned carrier-based vehicles are transferred together, they form heterogeneous unmanned equipment, and the path planning and control of such equipment are significantly more complex than that of individual pieces of equipment.

[0003] However, the limited deck operating area and the precise planning and control of highly dynamic, strongly coupled, and nonlinear heterogeneous unmanned equipment on the ship deck are among the core problems restricting future unmanned aircraft carrier deck operations. Several key points need to be addressed to achieve path planning and control of unmanned equipment on the ship deck:

[0004] (1) There are significant differences between unmanned heterogeneous equipment and single equipment in terms of physical properties (mass, moment of inertia), dynamic model structure and mission objectives (unmanned tractor active traction and precise trajectory tracking, unmanned carrier-based aircraft passively follow but need to maintain stability);

[0005] (2) Heterogeneous unmanned equipment consisting of unmanned tractor vehicles and shipborne mechanisms has characteristics such as strong coupling and nonlinearity. Compared with the control of single equipment, the dynamics of heterogeneous unmanned equipment are more complex and difficult to model accurately. In addition, the motion of such equipment on the ship deck is inevitably affected by unknown bounded external disturbances such as deck wind and ship motion caused by waves.

[0006] (3) The strong coupling relationship of unmanned heterogeneous equipment requires that path planning must support multi-objective and multi-task parallel processing to ensure the overall efficiency of deck operations. At the same time, path planning for heterogeneous unmanned equipment is significantly more complex than that for single equipment due to its large space occupation and the need for real-time obstacle avoidance.

[0007] In response to the aforementioned problems, experts and scholars both domestically and internationally have proposed many effective methods. Regarding path planning, A... Algorithms, Dijkstra's algorithm, and artificial potential fields were subsequently proposed. However, A... Both the algorithm and Dijkstra's algorithm rely excessively on static environment assumptions, making it difficult to respond promptly to dynamic obstacle changes, resulting in slightly insufficient environmental adaptability. Furthermore, both typically focus on the globally optimal path for a single piece of equipment, failing to meet the real-time requirements of collaborative heterogeneous unmanned equipment. In addition, A... Algorithms like the Dijkstra algorithm tend to generate suboptimal paths in dynamic environments, leading to reduced efficiency in the limited and highly dynamic conditions of deck operations. In contrast, the artificial potential field method, a path planning method based on physical principles, is widely used in the field of dynamic obstacle avoidance for robots due to its computational efficiency and simplicity. It guides equipment movement by simulating target point induction forces and obstacle avoidance forces. However, the artificial potential field method is prone to getting stuck in local optima in the complex environment of an aircraft carrier deck, exhibiting difficulties in escaping local minima. Furthermore, it lacks sufficient support for dynamic obstacle avoidance and task collaboration among heterogeneous equipment. Therefore, improving the artificial potential field method to meet the robustness requirements of highly dynamic environments has become a key direction for improving the collaborative efficiency of unmanned equipment. The core challenge currently facing path planning technology lies in the fact that traditional graph-based methods (such as A) are insufficient for dynamic obstacle avoidance and task collaboration among heterogeneous equipment. While Dijkstra's algorithm offers deterministic results, its static environment assumptions make it ill-suited for handling dynamic obstacles and multi-equipment collaborative scenarios. There is an urgent need to develop more advanced dynamic optimization path planning technologies to meet the complex demands of future aircraft carrier deck operations.

[0008] In terms of collaborative control technology, traditional centralized or decoupled control methods are insufficient to effectively handle the nonlinear coupling, model parameter uncertainty, and disturbance problems in such heterogeneous and strongly coupled unmanned equipment. For example, the unmanned tractor-towing vehicle and the unmanned carrier-based aircraft (heterogeneous unmanned equipment) on an aircraft carrier deck exhibit significant heterogeneity and strong coupling. The unmanned tractor-towing vehicle and the unmanned carrier-based aircraft differ significantly in physical properties (mass, moment of inertia), dynamic model structure, and mission objectives (active tractor-towing and precise trajectory tracking, unmanned carrier-based aircraft passively following but needing to maintain stability). Simultaneously, the two are rigidly connected through a hinge point, resulting in complex nonlinear geometric constraints on their motion states (position, attitude, velocity) and force states (traction / tension, torque). The motion of the unmanned carrier-based aircraft is directly affected by the traction force applied by the unmanned tractor-towing vehicle, while the motion of the unmanned tractor-towing vehicle is also significantly constrained by the reaction force applied by the unmanned carrier-based aircraft through the hinge point, forming a two-way, nonlinear, strong coupling of force and motion. In addition, the system also faces the following challenges: (1) The system dynamics are complex and have strong nonlinear coupling, making it difficult to model accurately; (2) It is inevitably affected by unknown bounded external disturbances such as deck wind and sea waves causing ship motion; (3) The parameters of the system inertial matrix and Coriolis force matrix are unknown or time-varying; (4) The deck working space is limited, requiring the unmanned carrier-based aircraft to move along a given trajectory with high precision.

[0009] In summary, existing path planning and cooperative control technologies have many shortcomings in dealing with the complex dynamic environment of heterogeneous unmanned equipment on aircraft carrier decks, and cannot meet the requirements for efficient cooperative operation of heterogeneous unmanned equipment. Therefore, there is an urgent need for a path planning and cooperative control method for shipborne heterogeneous and strongly coupled unmanned equipment to achieve efficient coordination of path planning, collision avoidance, and trajectory tracking, thereby improving the operational efficiency and robustness of unmanned equipment in complex dynamic environments. Summary of the Invention

[0010] Based on the technical problems mentioned in the background section above, a path planning and cooperative control method for shipborne heterogeneous and strongly coupled unmanned equipment is provided. The path planning layer mainly performs path planning and collision avoidance for the heterogeneous unmanned equipment, the trajectory smoothing layer mainly transforms the discrete points generated by path planning into the desired trajectory information that the heterogeneous unmanned equipment can execute, and the cooperative control layer mainly performs trajectory tracking for the heterogeneous unmanned equipment.

[0011] The technical means employed in this invention are as follows:

[0012] A path planning and cooperative control method for shipborne heterogeneous and strongly coupled unmanned equipment includes the following steps:

[0013] Step 1: Consider the longitudinal, lateral, and yaw motions of the unmanned tractor and unmanned carrier-based aircraft on the aircraft carrier deck plane, and establish a three-degree-of-freedom coupled motion mathematical model of the heterogeneous unmanned equipment; initialize environmental information, and determine the starting point coordinates of the heterogeneous unmanned equipment, the coordinates of obstacles, the coordinates of the target point, the maximum influence radius of obstacles, and the step size of the heterogeneous unmanned equipment.

[0014] Step 2: Construct a virtual force field based on the improved artificial potential field method, introduce the field stress modulation operator and the heading disturbance injection term, and generate discrete path points;

[0015] Step 3: Use the Gaussian smoothing function to transform discrete path points into continuous desired trajectories;

[0016] Step 4: Achieve precise trajectory tracking of unmanned equipment through an adaptive neural network two-layer control architecture. The upper layer controls the unmanned carrier-based aircraft, and the lower layer controls the unmanned tractor, using the coupling relationship to achieve collaborative control.

[0017] Furthermore, the environmental information initialization in step 1 includes: the starting point coordinates of the unmanned carrier-based aircraft. The distance from the center of gravity to the nose of the unmanned carrier-based aircraft Target point coordinates Coordinates of the obstacle radius of the obstacle Maximum range of influence and the number of obstacles And the step size of heterogeneous unmanned equipment.

[0018] Furthermore, step 2 includes the following steps:

[0019] Step 21: Set up the potential field model, including defining the obstacle avoidance potential field function, obstacle avoidance function, target-induced potential field function, and target-induced function, to lay the foundation for path planning;

[0020] Step 22: Add the expansion coefficient of the heterogeneous unmanned equipment and calculate the safe envelope boundary points of the target point. To the boundary of the safe envelope shaft and The furthest distance from the axis;

[0021] Step 23: Introduce the field stress adjustment operator to adjust the effects of avoidance force and induction force in real time.

[0022] Furthermore, step 3 includes the following steps:

[0023] Step 31: Extract the position sequence using the adaptive dynamic potential field method, and obtain the time sequence for each time step through the path planning iterative process. corresponding coordinate sequence ;

[0024] Step 32: Calculate the unmanned equipment in Cumulative distance traveled in the direction ;

[0025] Step 33: Construct a continuous smooth function using Gaussian smoothing integral. In the formula It is the original path function defined at discrete time points, that is, in The value at time is Step function:

[0026] ;

[0027] in, Denotes the smoothing parameter, satisfying , This indicates an adjustable proportional parameter, typically selected between 1 and 5; This represents the time integration variable, which can be iterated over during the integration process. All moments;

[0028] Step 34, respectively and Path points in the direction are extracted and smoothed to obtain the unmanned equipment in... direction and The expected signals in the directions are as follows: , The expected trajectory of the intelligent unmanned equipment is then:

[0029] .

[0030] Furthermore, in step 4, the adaptive neural network dual-layer control architecture includes: upper-layer unmanned carrier-based aircraft desired traction force and coupling angle design and articulation point decoupling and lower-layer unmanned tractor controller.

[0031] Furthermore, the design of the desired traction force and coupling angle of the upper-level unmanned carrier-based aircraft, as well as the decoupling of the articulation point, includes the following steps:

[0032] Desired traction input for designing unmanned carrier-based aircraft To enable the unmanned carrier-based aircraft to input the desired traction force Tracking the desired trajectory ;

[0033] The tracking error and velocity error are defined as follows:

[0034] ;

[0035] ;

[0036] Define the virtual control law as ,in This represents the positive definite design matrix to be designed. Represents the rotation matrix;

[0037] Introducing dynamic surface control: ;

[0038] in, , This represents the time constant for dynamic surface control;

[0039] Approximating unknown nonlinear terms using radial basis function neural networks: ;

[0040] in, , Represents the weight vector. Represents the radial basis function vector. Indicates the approximation error;

[0041] An adaptive robust term is introduced to correct for external environmental factors and approximation errors. ,in , Represents positive integers;

[0042] get ;

[0043] in, , Represents a constant;

[0044] ;

[0045] ;

[0046] Design the desired traction control law for unmanned carrier-based aircraft:

[0047] ;

[0048] in, This represents the positive definite design matrix to be designed. The estimated value, This represents the upper bound estimation vector of the perturbation. , Represents a constant. Represents the diagonal matrix of the hyperbolic tangent function;

[0049] Design an adaptive law to update the neural network weights and estimate the upper bound of the perturbation:

[0050] ;

[0051] ;

[0052] in, and Both represent positive definite adaptive gain matrices. and All of these represent design parameters that are greater than zero;

[0053] By constructing the following Lyapunov function

[0054] ;

[0055] ;

[0056] Prove that all signals in the upper-level controller are bounded;

[0057] The expected traction force of the unmanned carrier-based aircraft was obtained through calculation. and coupling angle The resistance term of the unmanned tractor is obtained by conversion, and the reference trajectory of the unmanned tractor is calculated using the coupling angle. .

[0058] Furthermore, the lower-level unmanned tractor controller utilizes the upper-level output Calculating the resistance of the unmanned tractor using coupling formulas Coupling angle and reference trajectory Design of control input for unmanned tractor Make it track ;

[0059] Define tracking error: ;

[0060] Designing virtual control laws: ,in It is a positive definite design matrix.

[0061] Approximating unknown nonlinear terms using RBFNN: Let The input vector is , Represents the weight vector. Represents the radial basis function vector. This represents the approximation error; to avoid oscillations caused by differentiation, a first-order filter is used to output the vector. to replace , , , This represents the time constant of the filter;

[0062] An adaptive robust term is introduced to correct for external environmental factors and approximation errors. ,in, , Represent positive integers; obtain

[0063] ;

[0064] in, , ; , ;

[0065] The control law for the unmanned tractor is designed as follows: ;

[0066] in, Represents a positive definite design matrix. The estimated value, This represents the estimated vector of the combined perturbation upper bound. , ;

[0067] Design Adaptive Law:

[0068] ;

[0069] ;

[0070] in, and It is the positive definite adaptive gain matrix to be designed. and All are design parameters greater than 0;

[0071] By constructing the following Lyapunov function

[0072] ;

[0073] ;

[0074] This proves that all signals in the upper-level controller are bounded.

[0075] Furthermore, step 21 includes the following steps:

[0076] First, the initial environmental information is read, a virtual force field is constructed, and a potential field model is set; the potential field model includes an obstacle avoidance potential field function. Obstacle avoidance function Target induced potential field function and target induced function ;

[0077] The obstacle avoidance potential field function is set as follows:

[0078] ;

[0079] in, Let represent the potential function for avoiding the potential field. Indicates the distance between heterogeneous unmanned equipment and obstacles. Indicates the first Safe distance from each obstacle This represents the strength parameter of the control avoidance potential field. Indicates the number of obstacles;

[0080] The target induced potential field function is set as follows:

[0081] ;

[0082] in, Let represent the potential field function of the induced field. This indicates the distance from the heterogeneous unmanned equipment to the target point. To control the intensity parameters of the induced field.

[0083] Furthermore, in step 22, the expansion coefficient of the heterogeneous unmanned equipment is taken into account. First, the safety envelope boundary equation for heterogeneous unmanned equipment is established as follows:

[0084] ;

[0085] in, This represents the straight-line distance from the unmanned carrier-based aircraft to the edge of its nose.

[0086] Secondly, calculate the distance from the obstacle to the safety envelope boundary separately. shaft and The furthest distance of the axis is:

[0087] ;

[0088] ;

[0089] Among them, the x-coordinate of the intersection point of the farthest distance from the obstacle to the boundary of the safety envelope. for:

[0090] ;

[0091] ;

[0092] Furthermore, due to the expansion coefficient being considered in the path planning of heterogeneous unmanned equipment... Therefore, the target point cannot be considered as just a single point. Based on the equation of the safety envelope boundary of heterogeneous unmanned equipment, the equation of the target safety envelope region is mapped as follows:

[0093] ;

[0094] Calculate the target point's safe envelope boundary points To the boundary of the safe envelope shaft and The furthest distance of the axis is:

[0095] ;

[0096] ;

[0097] in, The coordinates of the intersection point from the target point to the boundary of the safety envelope of the heterogeneous unmanned equipment are:

[0098] ;

[0099] ;

[0100] ;

[0101] .

[0102] Furthermore, in step 23, firstly, a new target induced potential field function is constructed for the induced force source:

[0103] ;

[0104] ;

[0105] The obstacle avoidance potential field function is improved as follows:

[0106] ;

[0107] ;

[0108] in, Indicates the radius of the obstacle. Indicates the first The distance from each obstacle to the target point Indicates the range of influence of the obstacle. and This indicates the field emphasis operator.

[0109] Compared with the prior art, the present invention has the following advantages:

[0110] (1) Existing technologies usually use a single dynamic model or ignore strong coupling characteristics, making it difficult to accurately describe the interaction between heterogeneous unmanned equipment. However, this invention proposes a simplified three-degree-of-freedom nonlinear coupling model for heterogeneous unmanned equipment, which fully considers the differences in mass, dynamic model and mission objectives, and describes the force and motion constraint relationship between the unmanned tractor and the unmanned carrier-based aircraft.

[0111] (2) In existing technologies, the traditional artificial potential field method is prone to getting trapped in local minima and has poor adaptability to dynamic environments. However, this invention significantly improves the robustness of path planning by introducing a field intensity modulating operator and a heading disturbance injection term. The field intensity modulating operator dynamically adjusts the effects of the induction force and the avoidance force in real time to ensure efficient obstacle avoidance in complex environments. The heading disturbance injection term effectively solves the local minima escape problem by dynamically optimizing the direction and intensity of the induction force, ensuring path continuity and planning efficiency.

[0112] (3) Existing technologies directly use discrete path points, resulting in non-differentiable trajectories and sudden acceleration changes. In contrast, this invention transforms discrete path points into continuously differentiable desired trajectories through a Gaussian smoothing function, which greatly improves path smoothness, reduces sudden acceleration changes during motion, and significantly enhances the trajectory tracking capability of unmanned equipment.

[0113] (4) Existing technologies for the coordinated control of heterogeneous equipment mostly adopt a centralized strategy, which makes it difficult to take into account both independent targets and coupled constraints. However, this invention uses a two-layer control architecture to handle the position tracking targets of the unmanned carrier-based aircraft and the tractor separately, and realizes information interaction through the coupling relationship of the articulated point (force transmission, kinematic constraints). Combining the RBF neural network to approximate the unknown nonlinear dynamics (model uncertainty and external disturbances) online, and introducing an adaptive robust term to suppress the approximation error and disturbance effect, the coordinated control under strong coupling constraints is finally realized.

[0114] (5) Existing backstepping methods suffer from the problem of differential explosion of virtual control laws, resulting in high computational complexity and difficulty in engineering implementation. However, this invention introduces a first-order filter (dynamic surface control technology) to effectively suppress the differential explosion phenomenon, reduce algorithm complexity, and improve the engineering applicability of the controller. Attached Figure Description

[0115] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0116] Figure 1 This is an architecture diagram of a path planning and cooperative control method for shipborne heterogeneous and strongly coupled unmanned equipment according to the present invention.

[0117] Figure 2 Schematic diagram of heterogeneous unmanned equipment.

[0118] Figure 3 Force analysis diagram of the hinge point.

[0119] Figure 4 Schematic diagram of the safety envelope boundary of heterogeneous unmanned equipment.

[0120] Figure 5 Schematic diagram of collision avoidance for heterogeneous unmanned equipment.

[0121] Figure 6 This is a flowchart of the heterogeneous unmanned equipment path planning method based on the improved artificial potential field method of the present invention.

[0122] Figure 7 This is a flowchart of the algorithm in the path planning method of the present invention.

[0123] Figure 8 This is a schematic diagram of the potential field interaction region in the method of the present invention.

[0124] Figure 9 This is a schematic diagram of the force analysis of heterogeneous unmanned equipment under the improved artificial potential field method in the present invention.

[0125] Figure 10 This is a schematic diagram illustrating the transformation of discrete points into continuous smooth curves after path planning in the method of this invention.

[0126] Figure 11 The block diagram of the two-layer adaptive neural network tracking control structure proposed in this invention. Detailed Implementation

[0127] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. 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 should fall within the scope of protection of the present invention.

[0128] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0129] like Figure 1-11 As shown, the present invention provides as follows Figure 1 The proposed path planning and cooperative control method for shipborne heterogeneous and strongly coupled unmanned equipment includes the following steps:

[0130] Step 1: Consider the longitudinal, lateral, and yaw motions of the unmanned tractor and unmanned carrier-based aircraft on the aircraft carrier deck plane, and establish a three-degree-of-freedom coupled motion mathematical model of the heterogeneous unmanned equipment; initialize environmental information, and determine the starting point coordinates of the heterogeneous unmanned equipment, the coordinates of obstacles, the coordinates of the target point, the maximum influence radius of obstacles, and the step size of the heterogeneous unmanned equipment.

[0131] To describe the motion of heterogeneous unmanned equipment on an aircraft carrier deck, the longitudinal, lateral, and yaw motions of unmanned tractors and unmanned carrier-based aircraft on the carrier deck plane are considered. A system is established as follows: Figure 2 The mathematical model of three-degree-of-freedom coupled motion of the heterogeneous unmanned equipment (unmanned tractor-unmanned carrier-based aircraft) is shown. For inertial coordinate system, and These are the coordinate systems of the unmanned shipborne aircraft and the unmanned tractor, respectively. The origin of the coordinate system is fixed at the center of mass of both the unmanned carrier-based aircraft and the unmanned towing vehicle. , Pointing to the right in the direction of travel of the unmanned carrier-based aircraft and the unmanned tractor, respectively, and perpendicular to the direction of travel. Figure 2The image shows an unmanned towing vehicle (without a lever) lifting the front wheel of an unmanned carrier-based aircraft to complete the towing operation. The articulation point is located at the center point of the rear axle of the unmanned towing vehicle.

[0132] Figure 2 middle Position of the unmanned carrier-based aircraft in inertial frame of reference. The position of the unmanned tractor under inertial frame. For the yaw angle of unmanned carrier-based aircraft, For the quality of unmanned carrier-based aircraft, For the quality of unmanned tractor vehicles, For the rotational inertia of the unmanned carrier-based aircraft, For the moment of inertia of the unmanned tractor, The distance from the center of gravity of the unmanned carrier-based aircraft to its front wheel. The distance from the center of gravity of the unmanned tractor to the front and rear axles. The difference between the two yaw angles , For unmanned carrier-based aircraft to be tractioned by unmanned towing vehicles along Axial component force, For unmanned carrier-based aircraft to be tractioned by unmanned towing vehicles along Axial component force, The unmanned tractor is pulled along by the unmanned carrier-based aircraft. Axial component force, The unmanned tractor is pulled along by the unmanned carrier-based aircraft. Axial component force.

[0133] The unmanned tractor unit (UTU) connects to the unmanned carrier-based aircraft via its own clamping device (located at the rear axle of the UTU). The two exhibit strong nonlinear coupling during traction motion, and the speed of the traction system on the deck is severely limited. Therefore, considering the influence of external environmental disturbances and the pulling force of the UTU during the UMU's movement, a simplified three-degree-of-freedom dynamic model of the UMU is established:

[0134]

[0135] in, For the inertial matrix of unmanned carrier-based aircraft, The Coriolis force matrix, This represents the external environmental disturbance vector. The velocity vector of the unmanned carrier-based aircraft (longitudinal, lateral, and yaw angular velocities). The traction force input vector provided for the unmanned towing vehicle.

[0136] The manned towing vehicle is independently driven on four wheels, simultaneously providing force and torque in the unmanned towing vehicle's body coordinate system. The tension force on the unmanned carrier-based aircraft is transmitted to the unmanned towing vehicle through the articulation point, becoming the drag force of the unmanned towing vehicle. Therefore, considering the thrust of the thrusters, the drag provided by the unmanned carrier-based aircraft, and the influence of external environmental disturbances during the movement of the unmanned towing vehicle, a simplified three-degree-of-freedom mathematical model is established as follows:

[0137]

[0138] in, The inertial matrix of the unmanned tractor (including the equivalent mass of the unmanned carrier-based aircraft). For the Coriolis force matrix of the unmanned tractor, This represents the external environmental disturbance vector. Let V be the velocity vector of the unmanned tractor. The drag / torque vector exerted by the unmanned carrier-based aircraft on the unmanned tractor. This is the control input vector for the unmanned tractor's thruster.

[0139] Consider as Figure 2 The force analysis of the articulation point between the unmanned carrier-based aircraft and the unmanned tractor is shown in the figure. Point A is the articulation point, located at the center of the rear axle of the unmanned tractor, and point B is the center of mass of the unmanned tractor. The following nonlinear relationship can be obtained.

[0140] ;

[0141] in, For unmanned carrier-based aircraft to be tractioned by unmanned towing vehicles along Axial component force, For unmanned carrier-based aircraft to be tractioned by unmanned towing vehicles along Axial component force, The unmanned tractor is pulled along by the unmanned carrier-based aircraft. Axial component force, The unmanned tractor is pulled along by the unmanned carrier-based aircraft. Axial component force, The difference in yaw angle .

[0142]

[0143] in, This represents the position and attitude of the unmanned carrier-based aircraft and the unmanned towing vehicle in the inertial coordinate system, respectively. This is a rotation matrix. The relative positions of the unmanned tractor and the unmanned carrier-based aircraft are as follows:

[0144] ;

[0145] Given the desired trajectory of an unmanned carrier-based aircraft At that time, the expected trajectory of the unmanned tractor for:

[0146] ;

[0147] like Figure 6 The diagram illustrates an improved artificial potential field method for heterogeneous unmanned equipment surface path planning, which includes the following steps:

[0148] Initialize environmental information, determining the starting point coordinates of the heterogeneous unmanned equipment, the coordinates of obstacles, the coordinates of the target point, the maximum influence radius of obstacles, and the step size of the heterogeneous unmanned equipment; further, initializing environmental information specifically includes: the system needs to initialize environmental information, including determining the starting point coordinates of the unmanned carrier-based aircraft. The distance from the center of gravity to the nose of the unmanned carrier-based aircraft Target point coordinates Coordinates of the obstacle radius of the obstacle Maximum range of influence and the number of obstacles The step size of the heterogeneous unmanned equipment is set to 1.

[0149] Step 21: Set up the potential field model, which specifically includes defining the obstacle avoidance potential field function, the obstacle avoidance function, the target-induced potential field function, and the target-induced function, laying the foundation for path planning;

[0150] First, the initial environmental information is read, a virtual force field is constructed, and a potential field model is set. The potential field model includes an obstacle avoidance potential field function. Obstacle avoidance function Target induced potential field function and target induced function This lays the foundation for subsequent route planning.

[0151] The obstacle avoidance potential field function is set as follows:

[0152] ;

[0153] in, To avoid the potential field function of the potential field, The distance between heterogeneous unmanned equipment and obstacles. For the first Safe distance from each obstacle To control the intensity parameters of the avoidance potential field, This represents the number of obstacles.

[0154] The target induced potential field function is set as follows:

[0155] ;

[0156] in, Let be the potential field function of the induced field. The distance from the heterogeneous unmanned equipment to the target point. To control the intensity parameters of the induced field.

[0157] Step 22: Add the expansion coefficient of heterogeneous unmanned equipment

[0158] The path planning of heterogeneous unmanned equipment differs significantly from that of individual equipment. In particular, for heterogeneous unmanned equipment combining unmanned tractor vehicles and unmanned carrier-based aircraft, an expansion coefficient needs to be considered in the path planning of the unmanned carrier-based aircraft. Combining Figure 4 As shown, Let be the distance from the nose of the carrier-based aircraft to the nose of the tractor unit in the heterogeneous unmanned equipment. Since the articulation point of the heterogeneous unmanned equipment can rotate, this distance is allowed if and only if the angle between the tractor unit and the carrier-based aircraft is 0. The maximum value is obtained, and it will be named as follows. .

[0159] Considering the expansion coefficient in heterogeneous unmanned equipment ,according to Figure 5 The diagram illustrating collision avoidance of heterogeneous unmanned equipment first establishes the safety envelope boundary equation for the heterogeneous unmanned equipment as follows:

[0160] ;

[0161] in, This is the straight-line distance from the unmanned carrier-based aircraft to the edge of its nose.

[0162] Secondly, calculate the distance from the obstacle to the safety envelope boundary separately. shaft and The furthest distance of the axis is:

[0163] ;

[0164] ;

[0165] Among them, the x-coordinate of the intersection point of the farthest distance from the obstacle to the boundary of the safety envelope. for:

[0166] ;

[0167] ;

[0168] Furthermore, due to the expansion coefficient being considered in the path planning of heterogeneous unmanned equipment... Therefore, the target point cannot be considered as just a single point. Based on the equation of the safety envelope boundary of heterogeneous unmanned equipment, the equation of the target safety envelope region is mapped as follows:

[0169] ;

[0170] Calculate the target point's safe envelope boundary points To the boundary of the safe envelope shaft and The furthest distance of the axis is:

[0171] ;

[0172] ;

[0173] in, The coordinates of the intersection point from the target point to the boundary of the safety envelope of the heterogeneous unmanned equipment.

[0174] ;

[0175] ;

[0176] ;

[0177] ;

[0178] Step 23: Introduce field moderation operator: Based on the traditional induced force and obstacle avoidance function, introduce field moderation operator to adjust the effect of avoidance force and induced force in real time, so as to ensure that heterogeneous unmanned equipment avoids collisions in path planning and achieves cooperative obstacle avoidance effect;

[0179] Because traditional artificial potential field methods have many drawbacks, a new potential field function for induced force and obstacle avoidance is proposed based on the influence range of the induced force and the magnitude of gravity and obstacle avoidance force. Figure 7 As shown, by adding a field intensity modulation operator to the traditional artificial potential field function, heterogeneous unmanned equipment can avoid the above situation without affecting the overall performance, and reach the target point more smoothly.

[0180] First, a new target induced potential field function is constructed for the induced force source:

[0181] ;

[0182] ;

[0183] The obstacle avoidance potential field function is improved as follows:

[0184] ;

[0185] ;

[0186] in, Indicates the radius of the obstacle. For the first The distance from each obstacle to the target point The range of influence of the obstacle. and It is a field-emphasis operator, whose size is equal to... Relevant. From the perspective of the improved induced potential field, when hour, The value is 1, according to the conventional target induced function: when Heterogeneous unmanned equipment is in the transition zone of induced force, due to The change in velocity direction and magnitude of the heterogeneous unmanned equipment reduces the inductive force effect more quickly, making obstacle avoidance easier. For the new obstacle avoidance function, when... When the force is large, heterogeneous unmanned equipment is in the zone where it cannot avoid forces; when At that time, the heterogeneous unmanned equipment is in the transition zone of evasive forces. It can be seen that its influence range is wider than that of traditional evasive forces, and the velocity direction of the heterogeneous unmanned equipment changes more significantly. Furthermore, due to time... The presence of [something], and the linear increase in avoidance force, will also make the obstacle avoidance path smoother; when [something] exists, the linear increase in avoidance force .... When the traditional avoidance force is applied, the specific potential field region is as follows: Figure 8 As shown.

[0187] Improve the target induction function to solve the problem of unreachable targets: Further optimize the target induction function so that unmanned equipment can overcome the target unreachability problem encountered in the traditional artificial potential field method and improve the coverage of path planning;

[0188] The specific improvements to the target induced potential field function are as follows:

[0189] ;

[0190] in, For the positive parameters to be designed, they must satisfy... . This represents the distance that heterogeneous unmanned equipment can reach in the target area. When the heterogeneous unmanned equipment reaches the target point, the gravity is 0.

[0191] When heterogeneous unmanned equipment moves away from the target

[0192] ;

[0193] Compared with the traditional target-induced potential field function, the improved new induced force potential field is stronger; when the heterogeneous unmanned equipment encounters the problem of not being able to reach the target, due to the addition of part of the gravitational field, the resultant force of the heterogeneous unmanned equipment will make it move towards the target and eventually reach the target.

[0194] Introducing a heading disturbance injection term: An injection term is added to the target induction function to dynamically adjust the direction and intensity of the induction force, break the local minimum trap, ensure the continuity of the path planning process, and enable the unmanned equipment to continuously move toward the target and avoid stagnation;

[0195] By adding an injection term to the target induction function, the direction and intensity of the induction force are dynamically adjusted, breaking the local minimum trap, ensuring the continuity of the path planning process, and enabling unmanned equipment to continuously move towards the target and avoid stagnation. For example... Figure 9 As shown, the guiding force of the target point on the heterogeneous unmanned equipment and the obstacle avoidance force reach a balance of 180° with each other. With the addition of the heading disturbance injection term, a new guiding force breaks the balance relationship, causing the heterogeneous unmanned equipment to move towards the target.

[0196] like Figure 8 As shown, an inertial coordinate system is established. ,when At that time, induced force and Angle between axes Let the new inductive force be related to... The angle of the axis is The new inductive force is:

[0197] ;

[0198] Injecting terms for heading disturbances, new induced forces and Angle between axes yes:

[0199] ;

[0200] So when Then, the new expression for the induced force is:

[0201] ;

[0202] Heading disturbance injection term Size depends only on angle It is relevant, but it needs to meet the following conditions:

[0203] ;

[0204] An improved artificial potential field method was applied to find paths for heterogeneous unmanned equipment, and the optimal path was obtained through multiple iterations of optimization.

[0205] Calculate the resultant force: At each moment, calculate the resultant force from all the induction fields and avoidance fields based on the current position of the unmanned equipment (refer to (2)).

[0206] By finding the direction of the resultant force using the gradient descent method, the direction of the next movement can be determined.

[0207] Determine the direction of movement and update the position: The gradient descent method is used to accurately calculate the direction of the resultant force on the unmanned equipment and its corresponding velocity vector. Then, based on the displacement principle, the path is iteratively updated to ensure that each step is taken along the direction of the fastest decrease in potential energy until the target point is safely reached.

[0208] The target point exerts an inductive force on the unmanned equipment, the intensity of which increases as the distance decreases. The gradient of the inductive force exerted by the target point on the unmanned equipment can be expressed as:

[0209]

[0210] in, It is the proportionality coefficient of the induction force. It is the location of unmanned equipment. These are the coordinates of the target point for the unmanned equipment.

[0211] Each obstacle generates a avoidance force for the unmanned equipment (UAV) to prevent it from getting too close. The avoidance force typically employs an exponential decay model, meaning it weakens rapidly with increasing distance. The gradient of the avoidance force for each obstacle can be expressed as:

[0212]

[0213] in, It is the avoidance force proportionality coefficient. These are the coordinates of the obstacle's position. It is the distance from the current position of the heterogeneous unmanned equipment to the obstacle. It is the radius of influence of the obstacle.

[0214] Adding the gradients of all attraction and avoidance forces together yields the total gradient. :

[0215]

[0216] To ensure that the unmanned equipment moves along the direction of the fastest decrease in potential energy, the gradient descent method is used to find the direction of the resultant force. That is, along... The direction in which the potential field decreases the fastest is the direction of movement, i.e., its negative direction of movement, and its velocity vector is... It can be calculated using the following formula:

[0217]

[0218] in, It is the step size, which controls the amount of movement in each iteration.

[0219] Finally, based on the velocity vector Update the position of the unmanned equipment. This step involves the actual movement and is the final output of path planning.

[0220]

[0221] in, For unmanned equipment Location at any given moment For unmanned equipment Location at any given moment This is the time interval. This update process will repeat at each time interval until the unmanned equipment reaches the target point and stops.

[0222] An improved artificial potential field method is applied to find paths for heterogeneous unmanned equipment, enabling it to safely reach its target point.

[0223] The trajectory smoothing method employs, as follows: Figure 10 The discrete points shown are converted into a continuous smooth curve. This is mainly achieved by calculating the discrete path using discrete point data generated by path planning, constructing the original function, and finally integrating using a Gaussian smoothing function to obtain a continuous smooth curve suitable for trajectory tracking of unmanned traction equipment. The specific operation steps are as follows:

[0224] Discrete path point extraction is performed by extracting the position sequence using the adaptive dynamic potential field method described in step 2, and obtaining the time sequence at each time step through the path planning iteration process. corresponding coordinate sequence ;

[0225] Cumulative movement distance calculation, calculating the distance traveled by unmanned equipment. Cumulative distance traveled in the direction ;

[0226] Gaussian smoothing function integral continuity: Constructing a continuous smooth function using Gaussian smoothing integrals. In the formula It is the original path function defined at discrete time points, that is, in The value at time is Step function:

[0227]

[0228] in, For smoothing parameters, satisfying , This is an adjustable proportional parameter, typically selected between 1 and 5. It is a time-integral variable that can be iterated over during the integration process. Every moment.

[0229] Desired trajectory generation, after the above operations, respectively, the trajectory generated in step 2 and By extracting and smoothing path points along the direction, the unmanned equipment can obtain the path points in the direction of the unmanned equipment. direction and The expected signals in the directions are respectively as follows: , The expected trajectory of the intelligent unmanned equipment is then... .

[0230] Adaptive neural network two-layer collaborative control, the control method employs as follows: Figure 11 The illustrated two-layer architecture mainly consists of the upper layer's design for the desired traction force and coupling angle of the unmanned carrier-based aircraft, the articulated point decoupling, and the lower layer's unmanned tractor controller. The specific operating steps are as follows:

[0231] Upper-level controller (for unmanned carrier-based aircraft): Designing the desired traction input for the unmanned carrier-based aircraft To enable it to accurately track the desired trajectory .

[0232] Within the backstepping design framework, the tracking error and velocity error are defined as follows:

[0233] ;

[0234] ;

[0235] Design the virtual control law as follows: ,in Let be the positive definite design matrix to be designed. It is a rotation matrix.

[0236] Introducing dynamic surface control (first-order filter): To avoid differential explosion, among which , is the filter time constant.

[0237] Approximating unknown nonlinear terms using radial basis function neural networks (RBFNN): ,in , For the weight vector, Let them be radial basis function vectors. This is the approximation error.

[0238] An adaptive robust term is introduced to correct for external environmental factors and approximation errors. ,in , It is a positive constant. Therefore, we can obtain... ,in , It is a constant. , .

[0239] Design the desired traction control law for unmanned carrier-based aircraft:

[0240]

[0241] in, Let be the positive definite design matrix to be designed. for The estimated value, This is the vector for estimating the upper bound of the perturbation. , It is a constant. Let be the diagonal matrix of the hyperbolic tangent function.

[0242] Design an adaptive law to update the neural network weights and estimate the upper bound of the perturbation:

[0243]

[0244]

[0245] in, and All are positive definite adaptive gain matrices. and All are design parameters that are greater than zero.

[0246] By constructing the following Lyapunov function , This proves that all signals in the upper-level controller are bounded.

[0247] (2) Decoupling at the hinge point

[0248] The desired traction force of the unmanned carrier-based aircraft can be calculated through step (1). and coupling angle ,according to Figure 3 Force analysis at the articulation point can be used to obtain the resistance term of the unmanned tractor, and the reference trajectory of the unmanned tractor can be calculated using the coupling angle. .

[0249] (3) Lower-level controller (for unmanned tractor): utilizing the upper-level output Calculating the resistance of the unmanned tractor using coupling formulas Coupling angle and reference trajectory Design of control input for unmanned tractor Make it track ;

[0250] Define tracking error: , .

[0251] Designing virtual control laws: ,in It is a positive definite design matrix.

[0252] Approximating unknown nonlinear terms using RBFNN: Let The input vector is , Represents the weight vector. Represents the radial basis function vector. This represents the approximation error; to avoid oscillations caused by differentiation, a first-order filter is used to output the vector. to replace , , , This represents the time constant of the filter.

[0253] An adaptive robust term is introduced to correct for external environmental factors and approximation errors. ,in , It is a positive constant. Therefore, we can obtain... ;

[0254] In the formula, , . , .

[0255] The control law for the unmanned tractor is designed as follows: ,in, For positive definite design matrices, for The estimated value, For the comprehensive disturbance (including the drag term) Upper bound estimate vector, , .

[0256] Design Adaptive Law:

[0257]

[0258]

[0259] in, and It is the positive definite adaptive gain matrix to be designed. and All of them are design parameters greater than 0.

[0260] By constructing the following Lyapunov function:

[0261] ;

[0262] This proves that all signals in the upper-level controller are bounded.

[0263] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments. In the above embodiments of the present invention, the descriptions of each embodiment have their own emphasis; parts not described in detail in a certain embodiment can be referred to in the relevant descriptions of other embodiments. It should be understood that the disclosed technical content in the several embodiments provided in this application can be implemented in other ways.

[0264] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A path planning and cooperative control method for shipborne heterogeneous and strongly coupled unmanned equipment, characterized in that, Includes the following steps: Step 1: Consider the longitudinal, lateral, and yaw motions of the unmanned tractor and unmanned carrier-based aircraft on the aircraft carrier deck plane, and establish a three-degree-of-freedom coupled motion mathematical model of the heterogeneous unmanned equipment; initialize environmental information, and determine the starting point coordinates of the heterogeneous unmanned equipment, the coordinates of obstacles, the coordinates of the target point, the maximum influence radius of obstacles, and the step size of the heterogeneous unmanned equipment. Step 2: Construct a virtual force field based on the improved artificial potential field method, introduce the field stress modulation operator and the heading disturbance injection term, and generate discrete path points; Step 3: Use a Gaussian smoothing function to transform discrete path points into a continuous desired trajectory; Step 3 includes the following steps: Step 31: Extract the position sequence using the adaptive dynamic potential field method, and obtain the time sequence for each time step through the path planning iterative process. corresponding coordinate sequence ; Step 32: Calculate the unmanned equipment in Cumulative distance traveled in the direction ; Step 33: Construct a continuous smooth function using Gaussian smoothing integral. In the formula It is the original path function defined at discrete time points, that is, in The value at time is The step function, where, : ； in, Denotes the smoothing parameter, satisfying , This indicates an adjustable proportional parameter, typically selected between 1 and 5; This represents the time integration variable, which can be iterated over during the integration process. All moments; Step 34, respectively and Path points in the direction are extracted and smoothed to obtain the unmanned equipment in... direction and The expected signals in the directions are as follows: , The expected trajectory of the intelligent unmanned equipment is then: ； Step 4: Achieve precise trajectory tracking of unmanned equipment through an adaptive neural network two-layer control architecture. The upper layer controls the unmanned carrier-based aircraft, and the lower layer controls the unmanned tractor, using the coupling relationship to achieve collaborative control.

2. The path planning and cooperative control method for shipborne heterogeneous and strongly coupled unmanned equipment according to claim 1, characterized in that, The environmental information initialized in step 1 includes: the starting point coordinates of the unmanned carrier-based aircraft. The distance from the center of gravity to the nose of the unmanned carrier-based aircraft Target point coordinates Coordinates of the obstacle radius of the obstacle Maximum range of influence and the number of obstacles And the step size of heterogeneous unmanned equipment.

3. The path planning and cooperative control method for shipborne heterogeneous and strongly coupled unmanned equipment according to claim 1, characterized in that, Step 2 includes the following steps: Step 21: Set up the potential field model, including defining the obstacle avoidance potential field function, obstacle avoidance function, target-induced potential field function, and target-induced function, to lay the foundation for path planning; Step 22: Add the expansion coefficient of the heterogeneous unmanned equipment and calculate the safe envelope boundary points of the target point. To the boundary of the safe envelope shaft and The furthest distance from the axis; Step 23: Introduce the field stress adjustment operator to adjust the effects of avoidance force and induction force in real time.

4. The path planning and cooperative control method for shipborne heterogeneous and strongly coupled unmanned equipment according to claim 1, characterized in that, In step 4, the adaptive neural network dual-layer control architecture includes: upper-layer unmanned carrier-based aircraft desired traction force and coupling angle design and articulation point decoupling and lower-layer unmanned tractor controller.

Images