Unmanned ship path tracking control method based on lightweight self-adaptive sight guidance

Abstract

The invention provides an unmanned ship path tracking control method based on lightweight self-adaptive sight guidance. The method comprises a self-adaptive sight guidance algorithm, a foresight distance is adjusted in real time according to the current position and path deviation of an unmanned ship, ideal course information is obtained, and the path convergence speed and algorithm calculation efficiency are improved. And in the virtual simulation platform, marine environment information is combined, and unmanned ship dynamic model parameters are obtained through experiments. And constructing an objective function taking the path tracking error and the rudder angle change as evaluation indexes, and optimizing PID controller parameters by adopting an improved sparrow search algorithm. According to the algorithm, a dynamic additive updating strategy is introduced, so that the global search capability and the convergence stability are improved, and the optimal PID parameter is ensured to be obtained. The method has the advantages of being high in response speed, small in overshoot and high in path tracking precision in the complex marine environment, the real-time control performance and the environment adaptability of the unmanned ship can be remarkably improved, and the method is suitable for the unmanned ship path tracking control requirements in various task scenes.

Description

Technical Field

[0001] The present invention belongs to the technical field of unmanned boat control, and particularly relates to a path tracking control method for an unmanned boat based on lightweight adaptive line-of-sight guidance. Background Technique

[0002] With the continuous development of unmanned boat technology, its applications in the fields of ocean exploration, ocean resource exploration, ocean environment monitoring, ocean military, etc. have become increasingly widespread. In these application scenarios, the unmanned boat needs to execute tasks along a predetermined path for a long time and with high precision. However, due to the complexity of the ocean environment (such as factors like wind speed, ocean current, and surge), the path tracking control of the unmanned boat faces many challenges. How to ensure that the unmanned boat can achieve high-precision and high-stability path tracking in a complex environment has become an important research direction in the field of unmanned boat control.

[0003] Existing unmanned boat path tracking control methods mainly rely on traditional control algorithms, such as PID control, LQR control, etc. These methods can achieve certain effects in static or simple environments, but in a complex ocean environment, due to the influence of external disturbances (such as wind, waves, and currents), the path tracking accuracy is often reduced, and even large heading errors and overshoot phenomena occur. Especially when facing a dynamic and uncertain environment, the adaptability and robustness of traditional control methods are poor, and they cannot effectively solve the path tracking problem of unmanned boats in complex environments.

[0004] In the prior art, Chinese Patent CN111487966B discloses an adaptive path tracking control method for a surface unmanned boat based on waypoints, belonging to the technical field of control. It is mainly to solve the problem that when the turning angle of the unmanned boat is greater than 90°, the classical LOS guidance will generate a large overshoot, resulting in low turning tracking accuracy. The present invention calculates the basic line-of-sight angle based on the proposed adaptive LOS circle radius, compensates the basic line-of-sight angle according to the path deviation and heading deviation to obtain the final desired line-of-sight angle; then designs a turning strategy based on virtual points, and uses three small-angle turns to transition to a large-angle turn to overcome the serious overshoot problem generated when the turning angle is large. At the same time, the present invention also designs a speed calculator, an intelligent adaptive S-surface heading controller, and an intelligent adaptive integral S-surface speed controller, which can improve the tracking efficiency and anti-interference ability of the unmanned boat, and can also well handle the complexity and uncertainty of the unmanned boat model. It is mainly used for the adaptive path tracking control of surface unmanned boats.

[0005] However, the prior art still has the following deficiencies:

[0006] (1) Although the adaptive line-of-sight circle radius strategy is adopted, complex exponential function calculations are still required during the path tracking process, with a large computational overhead, which is not conducive to applications in scenarios with limited resources or high real-time requirements;

[0007] (2) In terms of parameter tuning, existing methods lack a real-time optimization mechanism for controller parameters. They often adopt off-line tuning or manual adjustment, making it difficult to adapt to dynamic environmental disturbances and system characteristics.

[0008] (3) The path tracking control system of the unmanned boat has not achieved real-time linkage with the dynamic model of the unmanned boat, and cannot effectively utilize dynamic feedback for parameter update and control precision improvement, reducing the intelligence and adaptive ability of the overall system.

[0009] Therefore, there is an urgent need to propose a path tracking control method for unmanned boats with higher control precision, lower computational complexity, and stronger adaptive ability to achieve efficient and stable operation in complex environments. Summary of the Invention

[0010] The purpose of the present invention is to provide a path tracking control method for unmanned boats based on lightweight adaptive line-of-sight guidance to overcome the defects of the existing technologies mentioned above.

[0011] The purpose of the present invention can be achieved through the following technical solutions:

[0012] On the one hand, the present invention provides a path tracking control method for unmanned boats based on lightweight adaptive line-of-sight guidance, including the following steps:

[0013] Step S1: Obtain the marine environment information around the unmanned boat, as well as the desired path information, the position information of the unmanned boat, and the heading information of the unmanned boat in a preset coordinate system.

[0014] Step S2: According to the desired path information and the position information of the unmanned boat in the preset coordinate system, obtain the ideal heading information of the unmanned boat in the current state through a lightweight adaptive line-of-sight guidance algorithm.

[0015] Step S3: Construct a dynamic model of the unmanned boat. According to the dynamic model of the unmanned boat, use the marine environment information to perform real-time simulation on a virtual simulation platform to obtain the parameters of the dynamic model of the unmanned boat under this environmental condition.

[0016] Step S4: Determine the dynamic model of the unmanned boat according to the parameters of the dynamic model of the unmanned boat. Through an improved sparrow search algorithm based on the determined dynamic model of the unmanned boat, obtain PID control parameters with the smallest overshoot.

[0017] Step S5: According to the heading information of the unmanned boat, combined with the ideal heading information of the unmanned boat in the current state, construct a real-time PID controller through the PID control parameters, obtain the ideal rudder angle increment of the unmanned boat, and perform real-time control on the rudder of the unmanned boat according to the ideal rudder angle increment of the unmanned boat.

[0018] Further, the dynamic model of the unmanned boat is the Nomoto first-order heading motion model, and its mathematical expression is:

[0019]

[0020] where Ψ is the heading angle of the unmanned boat, is the heading angular velocity of the unmanned boat, that is, the first derivative of the heading angle with respect to time, is the heading angular acceleration of the unmanned boat, that is, the second derivative of the heading angle with respect to time, δ is the rudder angle input, K1 is the gain coefficient, indicating the response degree of the rudder angle to the heading angular velocity, and T1 is the time constant, reflecting the response speed of the system to the input change.

[0021] Further, the marine environment information includes wind speed and direction data, sea current data, and surge data.

[0022] Further, the preset coordinate system is the XOY plane of the coordinate system determined based on the preset origin within the area where the desired path is located as the coordinate origin and on the principle that all desired paths fall in the first quadrant.

[0023] Further, the unmanned boat position information is two-dimensional coordinate information relative to the preset coordinate system;

[0024] The unmanned boat heading information is the heading angle Ψ of the unmanned boat relative to the preset coordinate system, where the heading angle is the angle between the longitudinal axis of the unmanned boat and the X axis of the preset coordinate system;

[0025] The desired path information includes a sequence of path points {P1, P2,..P k …, P n}, where each path point P k includes its two-dimensional coordinates (x k , y k ) in the preset coordinate system, which is used to describe the target navigation route that the unmanned boat should sequentially pass through along the path points.

[0026] Further, according to the desired path information and the unmanned boat position information in the preset coordinate system, the ideal heading information of the unmanned boat in the current state is obtained through a lightweight adaptive line-of-sight guidance algorithm, specifically including:

[0027] From the sequence of desired path points {P1, P2,..P k …, P n}, select the two consecutive path points P k-1 =(x k-1 , y k-1 ) and P k =(x k , y k), construct the path segment where the current unmanned boat is located;

[0028] Based on the current path segment P k-1 P k calculate the direction angle α of the target path k-1 , and the calculation formula is:

[0029] α k-1 = arctan 2(y k -y k-1 , x k -x k-1 )

[0030] Calculate the vertical deviation distance d from the current position (x, y) of the unmanned boat to the path segment P k-1 P k , and the formula is: e , the formula is:

[0031]

[0032] According to the current deviation distance d e , calculate the forward-looking distance R, and the formula is:

[0033]

[0034] where R is the forward-looking distance, representing the reference distance in front of the unmanned boat on the path segment, d s is a constant in the task requirements, controlling the convergence intensity of the unmanned boat, and R min is the preset minimum forward-looking distance, determined according to the task requirements;

[0035] According to the forward-looking distance R and the coordinates of the current path segment P k-1 P k calculate the coordinates of the forward-looking point P LOS =(x LOS , y LOS ), and the formula is:

[0036]

[0037] According to the coordinates of the forward-looking point P LOS =(x LOS , y LOS ) calculate the projection distance Δ along the path segment from the current position (x, y) of the unmanned boat to the path segment P k-1 P k , and the formula is:

[0038] Δ=(x LOS -x) 2 +(y LOS -y) 2 -d e2

[0039] Among them, Δ is the projection distance, representing the actual distance along the path segment from the current unmanned boat position to the front view point;

[0040] According to the vertical deviation distance d e , the projection distance Δ and the direction angle α k-1 , calculate the ideal course angle Ψ LOS :

[0041]

[0042] Among them, Ψ LOS represents the course angle that the unmanned boat needs to adjust according to the current state;

[0043] Obtain the ideal course information Ψ LOS .

[0044] Further, according to the unmanned boat dynamics model, use the marine environment information to perform real-time simulation on the virtual simulation platform to obtain the unmanned boat dynamics model parameters under this environmental condition, specifically including:

[0045] According to the dynamics model of the unmanned boat, use the obtained marine environment information to establish a virtual marine environment model in the virtual simulation platform, and perform a turning experiment and a Zigzag experiment on the unmanned boat through this dynamics model, where the turning experiment is used to test the turning performance of the unmanned boat, and the Zigzag experiment is used to test the course stability of the unmanned boat when affected by external disturbances;

[0046] During the experiment, by controlling the change of the rudder angle, make the unmanned boat move along the preset path, and at the same time record the rudder angle value and the position information of the unmanned boat at different times, including course and speed parameters;

[0047] Based on the recorded data of the turning experiment and the Zigzag experiment, use the control system theory and dynamics modeling technology, and through the least squares method or other data fitting methods, calculate the dynamics model parameters K and T of the unmanned boat, where K is the control gain coefficient and T is the time delay time constant, reflecting the relationship between the rudder angle input and the course change of the unmanned boat.

[0048] Further, determine the unmanned boat dynamics model according to the unmanned boat dynamics model parameters, and obtain the PID control parameters with the minimum overshoot through the improved sparrow search algorithm, specifically including:

[0049] According to the determined unmanned boat dynamics model, construct a PID control parameter optimization model with the weighted combination of overshoot, adjustment time and steady-state error as the objective function, and the PID control parameters include the proportional coefficient K p, integral coefficient K i and differential coefficient K d , using the improved sparrow search algorithm to iteratively optimize the PID control parameters, namely the proportional coefficient, integral coefficient, and differential coefficient. The algorithm initializes the population within the set search space and simulates the path tracking response, calculates the individual fitness value according to the control performance index of the unmanned boat, and dynamically adjusts the population distribution in combination with the early warning mechanism and position update strategy to obtain the optimal PID control parameters that minimize the system overshoot.

[0050] Furthermore, the objective function is as follows:

[0051]

[0052] where J c is the objective function of the PID controller performance index, e p (k) is the path tracking error of the unmanned boat at time k, r(k) is the rudder angle control amount at time k, w1 and w2 are proportional coefficients, and w1 >> w2.

[0053] Furthermore, according to the heading information of the unmanned boat, combining with the ideal heading information of the unmanned boat in the current state, a real-time PID controller is constructed through the PID control parameters to obtain the ideal rudder angle increment of the unmanned boat, which specifically includes:

[0054] Obtain the heading error e(k) at the current moment:

[0055] e(k) = Ψ k - Ψ LOS

[0056] where Ψ k is the heading information of the unmanned boat at the current moment, that is, the actual heading angle at time k, and Ψ LOS is the ideal heading information of the unmanned boat in the current state;

[0057] Based on the heading error e(k) and the current optimal PID control parameters K p 、K i 、K d , use the digital PID controller to calculate the ideal rudder angle increment u(k) of the unmanned boat. The formula is:

[0058]

[0059] where k is the sampling time index.

[0060] Compared with the prior art, the present invention has the following advantages:

[0061] (1) By designing an adaptive line-of-sight (LOS) guidance algorithm, the radius of the forward-looking distance circle is adjusted in real time. Combining the path deviation and heading deviation compensation strategies, a more accurate desired heading angle is obtained, realizing the adaptive convergence of the path deviation, effectively avoiding the problem of slow response of the traditional LOS guidance system in the case of large deviations, improving the convergence speed and heading accuracy of the unmanned surface vehicle (USV) path tracking, and having a lower computational complexity than the existing exponential LOS algorithm, making it more suitable for real-time control applications.

[0062] (2) By introducing an improved sparrow search algorithm to optimize the PID controller parameters online, the producer position update strategy in the traditional sparrow search algorithm is improved, enhancing the global search ability and convergence speed of the algorithm, achieving the adaptive optimal configuration of the controller parameters, significantly reducing the overshoot rate and adjustment time of the system response, and enhancing the stability and adaptability of the USV control system.

[0063] (3) By constructing a virtual simulation platform based on the feedback of the dynamic model and combining the real-time PID tuning mechanism, the dynamic closed-loop coupling of the controller parameters and the operating state of the USV is realized. The control parameters can be dynamically adjusted online to adapt to environmental disturbances and nonlinear model changes, achieving a path tracking control effect with higher precision and stronger robustness, and being more intelligent and generalized than the existing methods based on fixed models and preset parameters.

[0064] (4) The virtual turning point strategy and small-angle transition control mechanism adopted in the present invention can effectively address the path tracking challenges brought by a large turning angle of the track, avoiding the path overshoot and tracking oscillation problems caused by large-angle turning, and improving the smoothness and accuracy of the overall track tracking. Description of the Drawings

[0065] Figure 1 is a schematic flow chart of the high-precision USV path tracking control method according to the embodiment of the present invention;

[0066] Figure 2 is a block diagram of the high-precision USV path tracking control system according to the embodiment of the present invention;

[0067] Figure 3 is a schematic diagram of the LOS guidance algorithm according to the embodiment of the present invention.

[0068] Figure 4 is a schematic diagram of the variation curves of the forward-looking distance R with the deviation distance d in the proportional guidance algorithm, exponential guidance algorithm, and adaptive LOS guidance algorithm according to the embodiment of the present invention e ;

[0069] Figure 5 is a block diagram of the rudder angle PID control system of the USV in the simulation platform according to the embodiment of the present invention;

[0070] Figure 6 Path tracking effect diagrams under different guidance algorithms according to embodiments of the present invention;

[0071] Figure 7 Step response output curve diagrams under different algorithms according to embodiments of the present invention. Detailed implementation manners

[0072] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0073] Embodiment 1:

[0074] Figure 1 Exemplarily shown is a schematic flowchart of a path tracking control method for an unmanned boat based on lightweight adaptive line-of-sight guidance according to an embodiment of the present invention. The method includes:

[0075] Step S1: Obtain the marine environment information around the unmanned boat, as well as the desired path information, the unmanned boat position information, and the unmanned boat heading information in a preset coordinate system;

[0076] Step S2: According to the desired path information and the unmanned boat position information in the preset coordinate system, obtain the ideal heading information of the unmanned boat in the current state through a lightweight adaptive line-of-sight guidance algorithm;

[0077] Step S3: Construct a dynamic model of the unmanned boat. According to the dynamic model of the unmanned boat, use the marine environment information to perform real-time simulation on a virtual simulation platform to obtain the dynamic model parameters of the unmanned boat under this environmental condition;

[0078] Step S4: Determine the dynamic model of the unmanned boat according to the dynamic model parameters of the unmanned boat. According to the determined dynamic model of the unmanned boat, obtain the PID control parameters with the minimum overshoot through an improved sparrow search algorithm;

[0079] Step S5: According to the unmanned boat heading information, combined with the ideal heading information of the unmanned boat in the current state, construct a real-time PID controller through the PID control parameters to obtain the ideal rudder angle increment of the unmanned boat, and perform real-time control on the rudder of the unmanned boat according to the ideal rudder angle increment of the unmanned boat.

[0080] The dynamic model of the unmanned boat is the Nomoto first-order heading motion model, and its mathematical expression is:

[0081]

[0082] Where, Ψ is the heading angle of the unmanned boat, is the heading angular velocity of the unmanned boat, that is, the first-order derivative of the heading angle with respect to time, is the heading angular acceleration of the unmanned boat, that is, the second-order derivative of the heading angle with respect to time, δ is the rudder angle input, K1 is the gain coefficient, indicating the response degree of the rudder angle to the heading angular velocity, and T1 is the time constant, reflecting the response speed of the system to the input change.

[0083] The marine environment information includes wind speed and direction data, ocean current data, and surge data.

[0084] The preset coordinate system is the XOY plane of the coordinate system determined based on the preset origin within the area where the desired path is located as the coordinate origin and on the principle that all the desired paths fall in the first quadrant.

[0085] For example, use the sensors carried on the unmanned boat to obtain the marine environment information around the unmanned boat and the GPS module carried on the unmanned boat to obtain the position and heading information of the unmanned boat, and determine the desired path information in the preset coordinate system. Among them, the marine environment information includes wind speed and direction data, ocean current and surge data, and the preset coordinate system is the XOY plane of the coordinate system determined based on the preset origin within the area where the desired path is located as the coordinate origin and on the principle that all the desired paths fall in the first quadrant.

[0086] The unmanned boat position information is two-dimensional coordinate information relative to the preset coordinate system;

[0087] The unmanned boat heading information is the heading angle Ψ of the unmanned boat relative to the preset coordinate system, where the heading angle is the angle between the longitudinal axis of the unmanned boat and the X axis of the preset coordinate system;

[0088] The desired path information includes a sequence of path points {P1, P2,..P k …, P n}, where each path point P k includes its two-dimensional coordinates (x k , y k ) in the preset coordinate system, which is used to describe the target navigation route that the unmanned boat should pass through in sequence along the path points.

[0089] According to the desired path information and the unmanned boat position information in the preset coordinate system, through the lightweight adaptive line-of-sight guidance algorithm, the ideal heading information of the unmanned boat in the current state is obtained. The lightweight adaptive line-of-sight guidance algorithm of this embodiment is as Figure 3 shown, and specifically includes:

[0090] From the sequence of desired path points {P1, P2,..P k …, P n}, select the two consecutive path points P k-1 =(xk-1 , y k-1 ), and P k = (x k , y k ), construct the path segment where the current unmanned boat is located;

[0091] Based on the current path segment P k-1 P k calculate the direction angle α of the target path k-1 , and the calculation formula is:

[0092] α k-1 = arctan 2(y k - y k-1 , x k - x k-1 )

[0093] Calculate the vertical deviation distance d from the current position (x, y) of the unmanned boat to the path segment P k-1 P k , and the formula is: e The formula is:

[0094]

[0095] According to the current deviation distance d e , calculate the forward-looking distance R, and the formula is:

[0096]

[0097] Among them, R is the forward-looking distance, indicating the reference distance in front of the unmanned boat on the path segment, d s is a constant in the mission requirements, which controls the convergence intensity of the unmanned boat, R min is the preset minimum forward-looking distance, which is determined according to the mission requirements; d s is a boundary constant determined by the mission requirements, which determines the convergence intensity of the unmanned boat. R min is the minimum forward-looking distance. In the adaptive line-of-sight guidance algorithm, R min is a variable parameter determined by the environment. When the marine environmental conditions are good (i.e., the wind and current are slow and there are few obstacles), it is set to a larger value, which can make the path following smoother and the navigation efficiency higher. When the marine environmental conditions are poor (i.e., the wind and current are fast and there are many obstacles), it is set to a smaller value, so that the unmanned boat pays more attention to the convergence of the required path and improves the navigation safety. Figure 4 is a schematic diagram of the change curve of the forward-looking distance R with the deviation distance d in the proportional guidance algorithm, exponential guidance algorithm, and adaptive line-of-sight guidance algorithm. Among them, P e , P k-1 , P k , P k+1 are waypoints in the planned path, (xk , y k ) is the coordinate of the k-th waypoint, P LOS is the reference position of the forward-looking distance, d e is the deviation distance between the ship and the desired path, P t (x, y) is the position of the unmanned boat at time t. R is used to determine P LOS of the ship's forward-looking distance radius, Ψ LOS is the forward-looking distance angle, Δ is the forward-looking distance;

[0098] According to the forward-looking distance R and the current path segment P k-1 P k calculate the forward-looking point coordinate P LOS =(x LOS , y LOS ), and the formula is:

[0099]

[0100] According to the forward-looking point coordinate P LOS =(x LOS , y LOS ) calculate the projection distance Δ from the current unmanned boat position (x, y) to the path segment P k-1 P k along the path segment, and the formula is:

[0101] Δ=(x LOS -x) 2 +(y LOS -y) 2 -d e 2

[0102] where Δ is the projection distance, representing the actual distance along the path segment from the current unmanned boat position to the forward-looking point;

[0103] According to the vertical deviation distance d e , the projection distance Δ and the direction angle α k-1 , calculate the ideal course angle Ψ LOS :

[0104]

[0105] where Ψ LOS represents the course angle that the unmanned boat needs to adjust according to the current state;

[0106] Obtain the ideal course information Ψ LOS .

[0107] In this embodiment, based on the relationship between the current state of the unmanned boat and the desired path, a lightweight adaptive line-of-sight guidance algorithm is adopted to achieve high-precision path tracking control with low computational load, which is particularly suitable for the control task of unmanned boats significantly affected by disturbances in the marine environment. The setting of the algorithm steps has strict technical logic and significant overall control advantages.

[0108] First, by selecting the two path points P closest to the current position of the unmanned boat k-1 P k , a path segment is constructed for local path modeling. This selection method reduces the dependence on the global information of the entire path, improves the local responsiveness and real-time performance of the algorithm, enables the control system to dynamically reconstruct the path segment during the continuous navigation of the unmanned boat, and enhances the ability to adapt to the dynamically changing environment.

[0109] Next, by calculating the direction angle of the current path segment, the theoretical forward direction of the current unmanned boat in response to the path segment is obtained. This step provides a basic direction reference for subsequent deviation calculation and heading adjustment, effectively avoiding path judgment errors caused by the heading deviation of the unmanned boat itself.

[0110] After that, the vertical deviation distance between the current position of the unmanned boat and the path segment is used to quantify the degree of deviation of the unmanned boat from the path. This deviation information is not only used for error feedback, but also serves as the core variable of the adaptive adjustment parameter in the next step, which is used to dynamically adjust the forward viewing distance R, enabling the line-of-sight guidance to have the ability of directionality and intensity adjustment.

[0111] Particularly crucial is that the calculation formula of the forward viewing distance R proposed by the present invention has self-adaptive and piecewise response characteristics:

[0112] When the deviation is small, linear interpolation is used to gradually reduce the forward viewing distance, enhance the guidance sensitivity, and promote the rapid convergence of the heading;

[0113] When the deviation is large, the forward viewing distance is equal to the deviation itself, maintaining a stable line-of-sight guidance and avoiding system oscillations caused by sudden adjustments.

[0114] This method improves the system stability when the deviation is large and enhances the precision control when the deviation is small, achieving a dynamic balance of rigid and flexible guidance.

[0115] At the same time, the system sets R in combination with the actual environment min , that is, the lower limit of the forward viewing distance, so that the forward viewing distance can be automatically reduced in poor sea conditions (large wind, wave and current disturbances), prompting the unmanned boat to be closer to the path and reducing yaw; the forward viewing distance can be appropriately enlarged in good sea conditions to improve the navigation efficiency. This dynamic adjustment mechanism enhances the robustness of the path control strategy of the present invention to the marine disturbance environment, which is superior to the traditional fixed forward viewing distance method.

[0116] Next, the forward viewing point that satisfies the forward viewing distance constraint is solved by geometric methods, and the projection distance is further calculated. Then, combined with the direction angle, vertical deviation, and projection distance, through the line-of-sight angle calculation formula, the current optimal ideal heading is obtained. This formula integrates the geometric direction of the path and the degree of deviation, making the finally generated ideal heading have geometric rationality, error guidance, and adjustment smoothness.

[0117] Overall, this lightweight adaptive line-of-sight guidance algorithm of the present invention reduces global dependence and improves real-time performance through local path segment modeling; dynamically adjusts the forward viewing distance of the line of sight to improve guidance flexibility and robustness; geometrically projects and fuses errors to calculate the ideal heading to ensure the rationality of the navigation angle and the smoothness of control.

[0118] It not only realizes faster, smoother, and lower computational cost heading adjustment control than traditional proportional LOS or exponential LOS, but also significantly improves the path tracking stability and accuracy in complex dynamic environments. Especially in the face of large-angle turning, strong disturbances, or path mutation scenarios, the method of the present invention can quickly respond and gradually approach the target path, effectively reducing overshoot and control jitter.

[0119] Therefore, this step design enables the present invention to have good real-time response ability and environmental adaptability while ensuring control accuracy, provides accurate and dynamically traceable reference heading information for the subsequent PID controller, and constitutes an important part of the efficient cooperation mechanism of "adaptive guidance + intelligent optimization control" of the present invention.

[0120] According to the unmanned surface vehicle dynamics model, using the marine environment information to conduct real-time simulation on the virtual simulation platform to obtain the unmanned surface vehicle dynamics model parameters under this environmental condition, specifically including:

[0121] According to the dynamics model of the unmanned surface vehicle, using the obtained marine environment information, a virtual marine environment model is established in the virtual simulation platform, and the unmanned surface vehicle is subjected to a turning experiment and a Zigzag experiment through this dynamics model. The turning experiment is used to test the turning performance of the unmanned surface vehicle, and the Zigzag experiment is used to test the heading stability of the unmanned surface vehicle when subjected to external disturbances;

[0122] During the experiment, by controlling the change of the rudder angle, the unmanned surface vehicle is made to move along the preset path, and at the same time, the rudder angle value and the position information of the unmanned surface vehicle at different times are recorded, including heading and speed parameters;

[0123] Based on the recorded data of the turning experiment and the Zigzag experiment, using control system theory and dynamics modeling technology, through the least squares method or other data fitting methods, the dynamics model parameters K and T of the unmanned surface vehicle are calculated, where K is the control gain coefficient and T is the time delay time constant, reflecting the relationship between the rudder angle input and the heading change of the unmanned surface vehicle.

[0124] To improve the adaptability of the path-following control system to complex ocean environments, this paper constructs a dynamic ocean model that matches the current environment within a virtual simulation platform. Based on the dynamic characteristics of the unmanned vehicle, gyration and zigzag experiments are conducted. These experiments control the rudder angle to obtain the unmanned vehicle's motion response data under disturbance conditions, thereby dynamically identifying key dynamic parameters such as the control gain K and the time delay constant T.

[0125] Through this process, the present invention achieves modeling of the actual dynamic behavior of the unmanned vehicle under current sea conditions, making subsequent PID parameter tuning more targeted and precise, ensuring a faster control system response, smaller overshoot, and greater stability. This method effectively establishes a closed-loop chain from environmental perception, model identification, and controller adjustment, significantly enhancing the robustness and control accuracy of the entire path tracking system in dynamic environments.

[0126] The unmanned boat dynamics model is determined according to the unmanned boat dynamics model parameters. The PID control parameters with the minimum overshoot are obtained by using the improved sparrow search algorithm based on the determined unmanned boat dynamics model. Specifically, the parameters include:

[0127] According to the determined unmanned boat dynamics model, a PID control parameter optimization model is constructed with a weighted combination of overshoot, adjustment time and steady-state error as the objective function. The PID control parameters include the proportional coefficient K p , integral coefficient K i and differential coefficient K d , the improved sparrow search algorithm is used to iteratively optimize the proportional coefficient, integral coefficient and differential coefficient of the PID control parameters. The algorithm initializes the population in the set search space and simulates the path tracking response. The individual fitness value is calculated according to the control performance index of the unmanned boat. The population distribution is dynamically adjusted by combining the early warning mechanism and the position update strategy to obtain the optimal PID control parameters that minimize the system overshoot. The block diagram of the rudder angle PID control system of the unmanned boat simulation platform in this embodiment is shown in the figure. Figure 5 shown.

[0128] Considering the lack of stability of the sparrow search algorithm in solving the generalized Schwefel problem. Through careful analysis, the optimal solution of the generalized Schwefel problem is far from 0, while the optimal solutions of other test functions are close to 0. The reason for the instability in solving the generalized Schwefel problem is that the update strategy for moving in a small range in the standard sparrow search algorithm is in multiplicative form, while the update strategy for moving in a large range is in additive form. When the absolute value of the solution space is large, due to the existence of α random numbers, the product will cause a large number of sparrow position updates, which runs counter to the idea of gathering when there is no danger. For this reason, the present invention proposes a new additive expression update strategy. When the sparrow needs to search in a small range, ensure that the step size of position update is small enough. The position update strategy is as follows:

[0129]

[0130] where ρ is a constant used as the exponential decay coefficient, ω is the larger one of the absolute values of the upper and lower bounds of, represents the number of sparrows, and δ is the warning coefficient. The larger the value of δ, the higher the warning level, the insufficient convergence of the sparrow group, and it is more conducive to global exploration. The smaller the value of δ, the lower the warning level, the sparrow group has sufficient convergence, and it is more conducive to local search. When the optimal solution is close to 0, the above formula is close to the update strategy of the standard sparrow search algorithm.

[0131] The objective function is:

[0132]

[0133] where J c is the objective function of the PID controller performance index, e p (k) is the path tracking error of the unmanned boat at time k, r(k) is the rudder angle control amount at time k, and w1 and w2 are proportionality coefficients, and w1 >> w2.

[0134] In order to achieve high-precision and high-stability path tracking control of the unmanned boat in a complex marine environment, after obtaining the real-time dynamic model parameters, the present invention constructs a PID control parameter optimization model with the weighted combination of overshoot, adjustment time and steady-state error as the objective function based on this model. This model takes the actual tracking error and the change amount of the rudder angle of the controller as evaluation indexes, emphasizes the path accuracy (w1 >> w2) and does not overly rely on the change of the rudder angle, so as to achieve a balance between accuracy and control smoothness and effectively adapt to the control requirements of the unmanned boat for "high precision + high stability".

[0135] To improve the global search ability and convergence speed of controller parameter optimization, the present invention introduces an improved sparrow search algorithm. Aiming at the problem of insufficient stability of the standard algorithm when the solution space boundary or the optimal solution is far from the origin, an update strategy based on additive expression is proposed, enabling the sparrow group to dynamically adjust global and local search behaviors at different warning levels (δ), and enhancing the adaptability and robustness of the algorithm in high-dimensional and non-linear control problems. Finally, combined with the individual fitness values feedback by the simulation platform, the optimal PID parameters that can minimize the path tracking error and control fluctuation are quickly obtained.

[0136] Through the above method, the entire control system not only obtains more reasonable controller parameters, significantly reducing the overshoot phenomenon and response time of the unmanned boat in a disturbed environment, but also ensures the universality and stability of the algorithm optimization process in different complex paths and dynamic environments, thus greatly enhancing the control accuracy, robustness and intelligence level of the overall path tracking scheme of the present invention.

[0137] According to the heading information of the unmanned boat, combined with the ideal heading information of the unmanned boat in the current state, a real-time PID controller is constructed through PID control parameters to obtain the ideal rudder angle increment of the unmanned boat, specifically including:

[0138] Obtain the heading error e(k) at the current moment:

[0139] e(k) = Ψ k -Ψ LOs

[0140] where Ψ k is the heading information of the unmanned boat at the current moment, that is, the actual heading angle at the k-th moment, and Ψ LOs is the ideal heading information of the unmanned boat in the current state;

[0141] Based on the heading error e(k) and the current optimal PID control parameters K p 、K i 、K d , use the digital PID controller to calculate the ideal rudder angle increment u(k) of the unmanned boat, and the formula is:

[0142]

[0143] where k is the sampling time index.

[0144] According to this embodiment, the performance verification results of different line-of-sight guidance algorithms are as Figure 6 、 Figure 7As shown, the proportional guidance algorithm has the slowest convergence speed. When the deviation error of the unmanned boat is large, the forward-looking distance also increases, resulting in a small adjustment of the heading angle of the unmanned boat and a slow convergence speed. The convergence speed of the exponential guidance algorithm is similar to that of the adaptive guidance algorithm, but the computational complexity of the adaptive guidance algorithm is significantly less than that of the exponential guidance algorithm. The improved sparrow search algorithm improves the convergence performance of the algorithm by updating the position update strategy of the producers in the sparrow population. The PID parameters obtained by this algorithm are the most effective, and the response overshoot and adjustment time of the PID system are the smallest. The parameters of the real-time PID controller can better meet the control requirements of the unmanned boat, achieve more accurate path tracking, and have the smallest overshoot. The convergence time of this algorithm is the shortest, and there are significant improvements compared with other algorithms. This method can be applied to other scenarios of optimizing the parameters of real-time PID control systems.

[0145] Embodiment 2:

[0146] Figure 2 Exemplarily shown is a block diagram of a high-precision unmanned boat path tracking control system according to an embodiment of the present invention. The system includes:

[0147] A data preparation module that acquires information on the marine environment around the unmanned boat, as well as information on the desired path and the position and heading information of the unmanned boat in a preset coordinate system. Among them, the marine environment information includes wind speed and direction data, and sea wave and surge data. The preset coordinate system is determined with a preset origin within the area where the desired path is located as the coordinate origin, and the XOY plane of the coordinate system is determined based on the principle that all desired paths fall within the first quadrant;

[0148] A line-of-sight guidance module that, according to the desired path information and the relative position information of the unmanned boat in the preset coordinate system, obtains the ideal heading of the unmanned boat in the current state through a lightweight adaptive line-of-sight guidance algorithm;

[0149] An intelligent search module for optimizing the PID parameters according to the environmental information to obtain the PID parameters with the smallest overshoot;

[0150] A PID control module that, according to the heading information of the unmanned boat and in combination with the ideal heading information of the unmanned boat in the current state, constructs a real-time PID controller through the parameters of the unmanned boat dynamics model and the PID control parameters to obtain the ideal rudder angle of the unmanned boat;

[0151] An unmanned boat actuator that responds to specific operations according to the rudder angle information output by the PID control module.

[0152] If the above functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs.

[0153] As described above, the above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of various equivalent modifications or substitutions, and these modifications or substitutions should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A path tracking control method for an unmanned boat based on lightweight adaptive line-of-sight guidance, characterized in that The following steps are involved: Step S1: Obtaining the ocean environment information around the unmanned boat, as well as the expected path information, the position information of the unmanned boat, and the heading information of the unmanned boat in a preset coordinate system; Step S2: obtaining the ideal heading information of the unmanned boat in the current state through a lightweight adaptive line-of-sight guidance algorithm according to the expected path information and the position information of the unmanned boat in the preset coordinate system; Step S3: constructing a dynamic model of the unmanned boat, and using the marine environment information to simulate in real time on a virtual simulation platform according to the dynamic model of the unmanned boat, to obtain the dynamic model parameters of the unmanned boat under the environmental conditions; Step S4: determining the unmanned vehicle dynamics model according to the unmanned vehicle dynamics model parameters, and obtaining PID control parameters with minimum overshoot according to the determined unmanned vehicle dynamics model through an improved sparrow search algorithm; Step S5: Based on the unmanned boat heading information and the ideal heading information of the unmanned boat in the current state, a real-time PID controller is constructed through the PID control parameters to obtain the ideal rudder angle increment of the unmanned boat, and the unmanned boat steering gear is controlled in real time according to the ideal rudder angle increment of the unmanned boat.

2. The method for path tracking control of an unmanned boat based on lightweight adaptive line-of-sight guidance according to claim 1, wherein The dynamic model of the unmanned boat is the Nomoto first-order heading motion model, and its mathematical expression is: where Ψ is the heading angle of the unmanned boat, is the heading angular velocity of the unmanned boat, that is, the first derivative of the heading angle with respect to time, is the heading angular acceleration of the unmanned boat, that is, the second derivative of the heading angle with respect to time, δ is the rudder angle input, K1 is the gain coefficient, indicating the response degree of the rudder angle to the heading angular velocity, and T1 is the time constant, reflecting the response speed of the system to the input change.

3. A path tracking control method for an unmanned boat based on lightweight adaptive line-of-sight guidance according to claim 1, characterized in that The ocean environment information includes wind speed and direction data, ocean current data, and surge data.

4. A path tracking control method for an unmanned boat based on lightweight adaptive line-of-sight guidance according to claim 1, characterized in that The preset coordinate system is a coordinate system XOY plane determined based on a preset origin in the area where the expected path is located as the coordinate origin and the principle that all expected paths fall in the first quadrant.

5. A path tracking control method for an unmanned boat based on lightweight adaptive line-of-sight guidance according to claim 1, characterized in that The position information of the unmanned boat is two-dimensional coordinate information relative to a preset coordinate system; The unmanned boat heading information is the heading angle Ψ of the unmanned boat relative to the preset coordinate system, where the heading angle is the angle between the longitudinal axis of the unmanned boat and the X-axis of the preset coordinate system; The expected path information includes a sequence of path points {P1, P2,..P k …, P n}, where each path point P k includes its two-dimensional coordinates (x k , y k ) in a preset coordinate system, which is used to describe the target navigation route that the unmanned boat should pass through in sequence along the path points.

6. The path tracking control method of an unmanned boat based on lightweight adaptive line-of-sight guidance according to claim 1, characterized in that The method of obtaining the ideal heading information of the unmanned boat in the current state by using a lightweight adaptive line of sight guidance algorithm based on the expected path information and the position information of the unmanned boat in the preset coordinate system specifically includes: From the expected path point sequence {P1, P2,..P k …, P n}, select two consecutive path points P k-1 = (x k-1 , y k-1 ) and P k = (x k , y k ) that are closest to the current position of the unmanned boat, and construct the path line segment where the current unmanned boat is located; Calculate the direction angle α of the target path based on the current path segment P k-1 P k The calculation formula is as follows: k-1 , and the calculation formula is: α k-1 = arctan 2(y k - y k-1 , x k - x k-1 ) Calculate the vertical deviation distance d from the current position (x, y) of the unmanned boat to the path segment P k-1 P k , where the formula is: e The formula is: According to the current deviation distance d e , calculate the forward view distance R, and the formula is: wherein, R is the forward viewing distance, representing the reference distance in front of the unmanned boat on the path segment, d s is a constant in the mission requirements, controlling the convergence intensity of the unmanned boat, R min is the preset minimum forward viewing distance, determined according to the mission requirements; According to the forward viewing distance R and the current path segment P k-1 P k calculate the coordinates of the forward view point P LOS =(x LOS , y LOS ), and the formula is: According to the coordinates P of the forward view point LOS =(x LOS ,y LOS ), calculate the projection distance Δ along the path segment from the current position (x, y) of the unmanned boat to the path segment P k-1 P k . The formula is as follows: Δ=(x LOS -x) 2 +(y LOS -y) 2 -d e 2 Among them, Δ is the projected distance, which represents the actual distance along the path segment from the current UAV position to the foresight point; According to the vertical deviation distance d e , the projection distance Δ and the direction angle α k-1 , calculate the ideal course angle Ψ LOS : Among them, Ψ LOS represents the heading angle that the unmanned boat needs to adjust according to the current state; Obtain the ideal course information Ψ LOS .

7. A path tracking control method for an unmanned boat based on lightweight adaptive line-of-sight guidance according to claim 1, characterized in that, The method of using the marine environment information to simulate in real time on a virtual simulation platform based on the unmanned boat dynamics model to obtain the unmanned boat dynamics model parameters under the environmental conditions specifically includes: Based on the dynamic model of the unmanned boat and the acquired ocean environment information, a virtual ocean environment model was established in the virtual simulation platform. The unmanned boat was then subjected to a maneuvering experiment and a Zigzag experiment using this dynamic model. The maneuvering experiment was used to test the steering performance of the unmanned boat, and the Zigzag experiment was used to test the heading stability of the unmanned boat when subjected to external disturbances. During the experiment, the rudder angle was controlled to make the UAV move along the preset path, and the rudder angle value and the position information of the UAV at different times, including heading and speed parameters, were recorded. Based on the recorded data from the turning experiment and the Zigzag experiment, using control system theory and dynamic modeling techniques, the dynamic model parameters K and T of the unmanned boat are calculated by the least squares method or other data fitting methods, where K is the control gain coefficient and T is the time delay time constant, reflecting the relationship between the rudder angle input and the heading change of the unmanned boat.

8. A path tracking control method for an unmanned boat based on lightweight adaptive line-of-sight guidance according to claim 1, characterized in that, Determine the dynamic model of the unmanned boat according to the dynamic model parameters of the unmanned boat, and obtain the PID control parameters with the minimum overshoot through the improved sparrow search algorithm based on the determined dynamic model of the unmanned boat, specifically including: According to the determined unmanned boat dynamics model, by constructing a PID control parameter optimization model with the weighted combination of overshoot, regulation time and steady-state error as the objective function, the PID control parameters include the proportional coefficient K p , integral coefficient K i and differential coefficient K d . The improved sparrow search algorithm is used to iteratively optimize the PID control parameters of proportional coefficient, integral coefficient and differential coefficient. The algorithm initializes the population in the set search space and simulates the path tracking response, calculates the individual fitness value according to the control performance index of the unmanned boat, and dynamically adjusts the population distribution in combination with the early warning mechanism and position update strategy to obtain the optimal PID control parameters that minimize the system overshoot.

9. A path tracking control method for an unmanned boat based on lightweight adaptive line-of-sight guidance according to claim 8, characterized in that, The objective function is: Among them, J c is the objective function of the PID controller performance index, e p (k) is the path tracking error of the unmanned boat at time k, r(k) is the rudder angle control amount at time k, w1 and w2 are proportionality coefficients, and w1 >> w2.

10. A path tracking control method for an unmanned boat based on lightweight adaptive line-of-sight guidance according to claim 1, characterized in that According to the heading information of the unmanned boat, combined with the ideal heading information of the unmanned boat in the current state, a real-time PID controller is constructed through the PID control parameters to obtain the ideal rudder angle increment of the unmanned boat, specifically including: Obtain the heading error e(k) at the current moment: e(k) = Ψ k -Ψ LOs where, Ψ k is the heading information of the unmanned boat at the current moment, that is, the actual heading angle at time k, Ψ LOs is the ideal heading information of the unmanned boat under the current state; Based on the heading error e(k) and the current optimal PID control parameters K p , K i , K d , the digital PID controller is used to calculate the ideal rudder angle increment u(k) of the unmanned boat, and the formula is: where k is the sampling moment index.

Images