Unmanned ship path planning method and system based on fusion model predictive control

Abstract

The invention discloses an unmanned ship path planning method and system based on fusion model predictive control, and belongs to the field of unmanned ship path planning and obstacle avoidance. Comprising the following steps: S1, carrying out uncertainty modeling and processing on external interference of an unmanned ship marine environment; s2, constructing a dual-mode control strategy according to the real-time state of the unmanned ship and the change of the marine environment, and introducing a collaborative successive linearization method; s3, reasonably optimizing the weight of each element in the unmanned ship path planning parameter matrix; designing a cost function of unmanned ship path planning; and S4, constructing a simulation environment conforming to the ocean task scene in the actual task execution of the unmanned ship, recording the path trajectory, the navigation time and the key information of the control input sequence in the whole process in detail, analyzing the simulation result, and verifying the validity. According to the technical scheme provided by the invention, the problem that unmanned ship path planning and obstacle avoidance cannot be accurately and efficiently carried out in the prior art is effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned ship navigation and obstacle avoidance, and in particular to an unmanned ship path planning method and system based on fusion model predictive control. Background Art

[0002] Unmanned ships are playing an increasingly important role in today's marine applications. However, their operation in the marine environment faces many complex challenges. The marine environment is full of uncertainties, such as unpredictable wind and wave interference, unpredictable sea conditions, and potential obstacles. In such a complex marine environment, it is very important to achieve accurate path planning for unmanned ships. Traditional path generation methods, including A* and its improved algorithms and artificial potential field APF and its improved algorithms, cannot meet actual needs. Summary of the invention

[0003] In view of the defects of related technologies, the present invention provides an unmanned ship path planning method and system based on fusion model predictive control, aiming to solve the problem in the prior art that the unmanned ship cannot be predicted and controlled based on water environment interference to accurately achieve path planning for collision avoidance.

[0004] The technical solution is as follows:

[0005] A method for unmanned ship path planning based on fusion model predictive control includes the following steps:

[0006] Step S1: Model and process the uncertainty of external interference to the unmanned ship's marine environment, define interference boundary parameters, consider the maximum possible value range of external interference under a certain probability level, optimize the calculation of uncertain disturbance sets, determine the impact range of external interference on the unmanned ship state, and make the unmanned ship path planning fully consider uncertainty;

[0007] Step S2: A dual-mode control strategy is constructed according to the real-time status of the unmanned ship and the changes in the ocean environment. A simple control strategy is used in the safe area to ensure stable navigation. Beyond the safe area, the optimization problem is solved to obtain the optimal control input to deal with complex situations. At the same time, a collaborative successive linearization method is introduced to reduce the computational complexity of the dynamic model, and further optimization solutions are performed based on the linear time-varying model under different control modes.

[0008] Step S3: According to the requirements of the unmanned ship to perform tasks, the conditions of the marine environment and its own resource conditions, dynamically balance the system performance and robustness, and reasonably optimize the weights of each element in the unmanned ship path planning parameter matrix according to the complexity of the marine environment and the type of task; design the cost function of the unmanned ship path planning, and adjust the parameters in the cost function in different marine environments and task scenarios by reasonably determining the value range of the penalty coefficient parameter to ensure that the unmanned ship path planning is balanced between performance and robustness;

[0009] Step S4: Construct a simulation environment that matches the marine mission scenario in the actual execution of the unmanned ship, and obtain detailed results by executing path planning; in the scene setting link, comprehensively consider the task-related factors, and in the process of executing path planning, use the fusion and improved method to evaluate the impact of the current wind, wave and current interference on the state of the unmanned ship at each sampling moment based on the uncertainty model constructed in the early stage, and then determine the control input with the help of the optimized control strategy, update the state through the unmanned ship dynamics model, and record the path trajectory, navigation time, and key information of the control input sequence in the whole process in detail, analyze the simulation results, and verify the effectiveness.

[0010] An unmanned ship path planning system based on fusion model predictive control, the system comprising:

[0011] The external environment construction module is used to model and process the uncertainty of external interference to the unmanned ship's marine environment, define interference boundary parameters, consider the maximum possible value range of external interference under a certain probability level, optimize the calculation of uncertain disturbance sets, determine the scope of influence of external interference on the state of the unmanned ship, and make the path planning of the unmanned ship fully consider uncertainty;

[0012] The fusion model predictive control module is used to build a dual-mode control strategy based on the real-time status of the unmanned ship and the changes in the marine environment. A simple control strategy is used in the safe area to ensure stable navigation. Beyond the safe area, the optimization problem is solved to obtain the optimal control input to deal with complex situations. At the same time, the collaborative successive linearization method is introduced to reduce the calculation complexity of the dynamic model, and further optimization solutions are performed based on the linear time-varying model under different control modes;

[0013] The path planning module is used to dynamically balance system performance and robustness according to the requirements of the unmanned ship to perform tasks, the conditions of the marine environment, and its own resource conditions. According to the complexity of the marine environment and the type of task, the weights of each element in the unmanned ship path planning parameter matrix are reasonably optimized; the cost function of the unmanned ship path planning is designed, and by reasonably determining the value range of the penalty coefficient parameter, the various parameters in the cost function are adjusted in different marine environments and mission scenarios to ensure that the unmanned ship path planning is balanced between performance and robustness;

[0014] The simulation verification module is used to build a simulation environment that matches the marine mission scenario in the actual execution of the unmanned ship, and obtain detailed results by executing path planning; in the scene setting link, the task-related factors are fully considered, and in the process of executing path planning, the fusion and improved method is used to evaluate the impact of the current wind, wave and current interference on the state of the unmanned ship at each sampling moment based on the uncertainty model constructed in the early stage, and then determine the control input with the help of the optimized control strategy, update the state through the unmanned ship dynamics model, and record the path trajectory, navigation time, and control input sequence key information of the whole process in detail, analyze the simulation results, and verify the effectiveness.

[0015] The present invention has the following beneficial effects:

[0016] The present invention provides an unmanned ship path planning method and system based on fusion model predictive control, which accurately models and processes the uncertainty of the marine environment faced by the unmanned ship, defines similar interference boundary parameters, optimizes the calculation of uncertain disturbance sets, and provides a more accurate uncertainty information basis for the unmanned ship path planning; at the same time, in terms of control strategy, a dual-mode control strategy and a collaborative successive linearization method are introduced to flexibly select appropriate control strategies according to the real-time status of the unmanned ship and changes in the marine environment, thereby improving the path planning capability and accuracy of the unmanned ship in complex situations; in addition, through a comprehensive trade-off between system performance and robustness, parameters are dynamically adjusted according to the task, environment and resource conditions, and the path planning performance and robustness are balanced to ensure the optimal performance of the unmanned ship in different tasks and environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a flow chart of an unmanned ship path planning method based on fusion model predictive control provided by an embodiment of the present invention;

[0018] Figure 2 It is a flowchart of uncertainty modeling and processing of external interference of the unmanned ship marine environment provided by an embodiment of the present invention;

[0019] Figure 3 It is a flow chart of the dual-mode control strategy of an unmanned ship provided by an embodiment of the present invention;

[0020] Figure 4 is a flow chart of unmanned ship path planning provided by an embodiment of the present invention;

[0021] Figure 5 This is a simulation verification flow chart of the unmanned ship actually performing tasks provided by an embodiment of the present invention;

[0022] Figure 6 It is a structural diagram of an unmanned ship path planning system based on fusion model predictive control provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0024] It will be understood by those skilled in the art that, unless expressly stated, the singular forms "one", "said", and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be intermediate elements. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling. The term "and / or" used herein includes all or any unit and all combinations of one or more associated listed items.

[0025] See also Figure 1 The present invention provides an unmanned ship path planning method based on fusion model predictive control, comprising the following steps:

[0026] Step S1: Model and process the uncertainty of external interference to the unmanned ship's marine environment, define interference boundary parameters, consider the maximum possible value range of external interference under a certain probability level, optimize the calculation of uncertain disturbance sets, determine the impact range of external interference on the unmanned ship state, and make the unmanned ship path planning fully consider uncertainty;

[0027] Step S2: A dual-mode control strategy is constructed according to the real-time status of the unmanned ship and the changes in the ocean environment. A simple control strategy is used in the safe area to ensure stable navigation. Beyond the safe area, the optimization problem is solved to obtain the optimal control input to deal with complex situations. At the same time, a collaborative successive linearization method is introduced to reduce the computational complexity of the dynamic model, and further optimization solutions are performed based on the linear time-varying model under different control modes.

[0028] Step S3: According to the requirements of the unmanned ship to perform tasks, the conditions of the marine environment and its own resource conditions, dynamically balance the system performance and robustness, and reasonably optimize the weights of each element in the unmanned ship path planning parameter matrix according to the complexity of the marine environment and the type of task; design the cost function of the unmanned ship path planning, and adjust the parameters in the cost function in different marine environments and task scenarios by reasonably determining the value range of the penalty coefficient parameter to ensure that the unmanned ship path planning is balanced between performance and robustness;

[0029] Step S4: Construct a simulation environment that matches the marine mission scenario in the actual execution of the unmanned ship, and obtain detailed results by executing path planning; in the scene setting link, comprehensively consider the task-related factors, and in the process of executing path planning, use the fusion and improved method to evaluate the impact of the current wind, wave and current interference on the state of the unmanned ship at each sampling moment based on the uncertainty model constructed in the early stage, and then determine the control input with the help of the optimized control strategy, update the state through the unmanned ship dynamics model, and record the path trajectory, navigation time, and key information of the control input sequence in the whole process in detail, analyze the simulation results, and verify the effectiveness.

[0030] See also Figure 2 ,Step S1: Model and process the uncertainty of external interference to the marine environment of the unmanned ship, define the interference boundary parameters, consider the maximum possible value range of external interference under a certain probability level, optimize the calculation of the uncertain disturbance set, determine the impact range of external interference on the state of the unmanned ship, and make the path planning of the unmanned ship fully consider the uncertainty, specifically:

[0031] Step S11: define the interference boundary parameters of the external interference of the unmanned ship's marine environment, where the interference of the unmanned ship to the marine environment is ω i (t), define σ max is the maximum value of interference at the 95% confidence level, and the interference limit ρ i To represent the interference ω i (t) intensity range, when ρ i ≤ρ -i When the marine environment is considered suitable for unmanned ships to travel, otherwise it is considered unsuitable for unmanned ships to travel. -i is defined as:

[0032]

[0033] Among them, the positive definite matrix Q i It is the weight matrix related to the performance of the unmanned ship system, reflecting the importance of the unmanned ship position and tracking error performance indicators; the positive definite matrix P i is the weighted norm matrix of the state parameters of the unmanned ship system, reflecting whether the unmanned ship meets the stability requirements; α i and εi is the design parameter of the unmanned ship system terminal, which is determined according to the system design and stability requirements; T is the prediction time domain parameter, which determines the time range of future states considered in path planning; L i is a constant related to the unmanned ship system model and is related to the nonlinear characteristics of the unmanned ship dynamics model; the matrix eigenvalues ​​λ and are the minimum and maximum eigenvalues ​​of the matrix;

[0034] In the actual application process, firstly according to the system design analysis of the unmanned ship, through theoretical calculation, experimental test or empirical data, the positive definite matrix Q is preliminarily determined. i , P i , design parameter α i and ε i , predict the time domain T and parameter L i Then calculate the matrix Where R i is the matrix that limits the control input energy consumption, K i is the gain matrix related to the system feedback control, and solves the minimum eigenvalue of the corresponding matrix and the maximum eigenvalue of the corresponding matrix

[0035] Then according to σ max and ρ -i The relationship between σ max The value of, for example, σ max The value of can be slightly smaller than ρ -i , σ max The value of is used as an important basis for judging the interference intensity and adjusting the control input when calculating the control input during path planning. i The norm of (t) is close to σ max When the unmanned ship is in a state of disturbance, the path planning algorithm will appropriately increase the adjustment range of the control input according to the pre-designed rules, such as increasing the adjustment amount of the rudder angle or the propulsion force, so that the unmanned ship can better resist interference and keep sailing near the predetermined path, ensuring that the unmanned ship maintains a certain performance under interference;

[0036] Step S12: Optimize the calculation of the uncertain disturbance set to determine the impact range of external disturbance on the state of the unmanned ship. The offline dynamic model of the unmanned ship is x k+1 =f(x k ,u k ,ω k ), where x k is the state vector of the unmanned ship system at time k, including the position, speed, and heading information of the unmanned ship, u k is the control input vector, including the rudder angle command and the thrust force, ω kis the interference of the marine environment; the offline dynamic model of the unmanned ship is linearized and approximated as A k , B k and E k is the matrix related to the system state and control input, obtained by linearizing the dynamic model at the current state; Assume the uncertain perturbation dynamic equation is Δω k+1 =F k Δω k +G k ξ k , F k and G k is the matrix related to system dynamics and disturbance characteristics, ξ k For k The probability distribution of the random noise vector is related to the uncertainty perturbation set X. Δ (k) = {Δω k :||Δω k ||≤σ max},calculate Where W k is a compact set associated with random noise, whose size is determined by the interference ω k The probability distribution of X is determined by real-time monitoring and calculation Δ (k) and X Δ (k+1), adjust the control strategy according to its changing trend, and increase the control input adjustment amplitude when the disturbance set increases to keep the unmanned ship sailing stably;

[0037] See also Figure 3 , Step S2: Construct a dual-mode control strategy based on the real-time status of the unmanned ship and the changes in the ocean environment. A simple control strategy is used in the safe area to ensure stable navigation. Beyond the safe area, the optimization problem is solved to obtain the optimal control input to deal with complex situations; at the same time, the collaborative successive linearization method is introduced to reduce the calculation complexity of the dynamic model, and the optimization solution is further performed based on the linear time-varying model under different control modes, specifically:

[0038] Step S21: Introduce the dual-mode control strategy. When the state of the unmanned ship meets When the unmanned ship is in a safe area, a control strategy based on linear feedback is adopted, which is determined through system stability analysis and performance index design; sensors are used to monitor the state of the unmanned ship in real time and calculate and with ε i Compare and decide whether to switch the control mode;

[0039] When the unmanned ship approaches an obstacle or deviates from the planned route, that is, the state exceeds the safe area, it switches to the optimized control mode. In the optimized control mode, the interference ω is set k The probability distribution ofk ), add constraint P(ω k ∈W k )≥p0, p0 is the preset probability threshold, which is set according to the reliability requirements of the task; the control input u is retained min ≤u k ≤u max and state x min ≤x k ≤x max Constraints, where u min and u max are the lower and upper limits of the control input, depending on the physical limitations of the UAV actuator, x min and x max Reasonable range of the state of the unmanned ship; determine the optimization problem:

[0040]

[0041] in, tk is the time value, and the definitions of matrices T, Q, R, and P are the same as in step S11.

[0042] Step S22: The real-time state and environment of the unmanned ship are coordinated using a successive linearization processing model to calculate the nonlinear dynamic model of the unmanned ship at the sampling time k. Linearize and obtain the initial control input trajectory u at the current moment by translating the optimal control input sequence at the previous moment 0 (i|k), apply this initial control input trajectory to obtain the initial state trajectory x 0 (i|k), the interference input uses the predicted mean Then the nonlinear dynamic model is linearized to obtain x(i+1|k)=x 0 (i+1|k)+A(i|k)Δx(i|k)+B(i|k)Δu(i|k)+C(i|k)Δω(i|k), where A(i|k), B(i|k), and C(i|k) are Jacobian matrices;

[0043] Step S23: Under the dual-mode control strategy, the model after successive linearization is used in coordination for path planning according to the real-time status and environment of the unmanned ship; in the safe area, the control input is calculated using the simple model after successive linearization to reduce the consumption of computing resources. In the optimization control mode, the optimization problem is solved based on the successive linearization model. When the path needs to be adjusted, a more appropriate control input is obtained based on the linearization model in the optimization control mode to ensure that the unmanned ship can cope with complex environments and mission requirements.

[0044] See also Figure 4, Step S3: According to the requirements of the unmanned ship to perform tasks, the conditions of the marine environment and its own resource conditions, dynamically balance the system performance and robustness, and reasonably optimize the weights of each element in the unmanned ship path planning parameter matrix according to the complexity of the marine environment and the type of task; design the cost function of the unmanned ship path planning, and adjust the parameters in the cost function in different marine environments and task scenarios by reasonably determining the value range of the penalty coefficient parameter to ensure that the unmanned ship path planning is balanced between performance and robustness; specifically:

[0045] Step S31: Adjust the weights according to the intensity of wind, wave and current in the ocean environment, the distribution of obstacles and the transport task or monitoring task of the unmanned ship. In the transport task, increase the weights of adjacent path points in the area with strong winds and waves to keep the path of the unmanned ship stable and avoid damage to the cargo; in the monitoring task, adjust the weights of the path points according to the importance of the monitoring area to ensure comprehensive monitoring while improving robustness;

[0046] Step S32: Design the cost function of the unmanned ship path planning. By reasonably determining the value range of the penalty coefficient parameter, adjust the parameters in the cost function under different marine environments and mission scenarios. The cost function of the unmanned ship path planning is:

[0047]

[0048] Among them, x ref,k is the state vector corresponding to the reference path point, Q k The matrix is ​​a weight matrix that is dynamically adjusted according to the surrounding environment of the waypoint, which is related to the complexity of the environment and the task requirements. The R matrix is ​​also a weight matrix that is used to limit the energy consumption of the control input. i is the penalty coefficient, and its value range is reasonably determined by the relationship between stability and interference bound; Φ(x N ) is a function of the state of the unmanned ship at the terminal moment N, which is used to consider the constraints or targets of the terminal state. It is a deviation function between the position of the unmanned ship at the terminal moment and the target position, or a constraint function on the terminal speed and heading state of the unmanned ship; ω i (s;t k ) represents the unmanned ship at time s, the range of s is from t k to k +T, the i-th interference vector received.

[0049] See also Figure 5,Step S4: Construct a simulation environment that matches the marine mission scenario in the actual mission of the unmanned ship, and obtain detailed results by executing path planning; in the scene setting link, comprehensively consider the task-related factors, and in the process of executing path planning, use the fusion and improved method to evaluate the impact of the current wind, wave and current interference on the state of the unmanned ship at each sampling moment based on the uncertainty model constructed in the early stage, and then determine the control input with the help of the optimized control strategy, update the state through the unmanned ship dynamics model, and record the path trajectory, navigation time, and key information of the control input sequence in the whole process in detail, analyze the simulation results, and verify the effectiveness, specifically:

[0050] Step S41: Setting simulation scene parameters, ω k It follows the normal distribution N(0,0.1), and sets parameters based on the actual performance of the unmanned ship to ensure that the simulation is within the actual operation range. The maximum speed of the unmanned ship v max =5m / s, minimum turning radius r min =10m, the control input constraint is the thrust of the thruster δ T ∈[-1000,1000]N, rudder angle δ R ∈[-30 0 ,30 0 ]; Multiple obstacles are randomly distributed in the driving area, and their positions and shapes are determined according to the actual environment;

[0051] Step S42: At each sampling time k, according to the uncertainty model improved in step S1, consider the current wind wave current interference ω k For the impact on the state of the unmanned ship, the control input is determined using the control strategy optimized in step S2; if the unmanned ship is in open waters and the interference is small, the control input is calculated according to the simple control mode in the dual-mode control strategy combined with the successive linearization model; if it is close to a port or a complex area and faces greater interference, it is switched to the optimization control mode, and the optimization problem is solved based on the successive linearization model to obtain the optimal control input; then, the state of the unmanned ship is updated through the unmanned ship dynamics model, and information such as the path trajectory, navigation time, and control input sequence are recorded; this is repeated until the unmanned ship reaches the target position or meets the termination condition, and a complete path planning result including detailed data such as the actual path, the time to reach the target, and the control input at each moment is obtained to verify the path planning effect.

[0052] See also Figure 6 The present invention provides an unmanned ship path planning system 100 based on fusion model predictive control, the system comprising:

[0053] The external environment construction module 101 is used to model and process the uncertainty of the external interference of the unmanned ship's marine environment, define the interference boundary parameters, consider the maximum possible value range of the external interference under a certain probability level, optimize the calculation of the uncertain disturbance set, determine the impact range of the external interference on the unmanned ship state, and make the unmanned ship path planning fully consider the uncertainty;

[0054] The fusion model prediction control module 102 is used to build a dual-mode control strategy according to the real-time status of the unmanned ship and the changes in the ocean environment. A simple control strategy is used in the safe area to ensure stable navigation. Beyond the safe area, the optimization problem is solved to obtain the optimal control input to deal with complex situations. At the same time, a collaborative successive linearization method is introduced to reduce the calculation complexity of the dynamic model, and further optimization solutions are performed based on the linear time-varying model under different control modes;

[0055] The path planning module 103 is used to dynamically balance the system performance and robustness according to the requirements of the unmanned ship to perform tasks, the conditions of the marine environment in which it is located, and its own resource conditions, and reasonably optimize the weights of each element in the unmanned ship path planning parameter matrix according to the complexity of the marine environment and the type of task; design the cost function of the unmanned ship path planning, and adjust the parameters in the cost function in different marine environments and task scenarios by reasonably determining the value range of the penalty coefficient parameter to ensure that the unmanned ship path planning is balanced between performance and robustness;

[0056] The simulation verification module 104 is used to construct a simulation environment that matches the marine mission scenario in the actual execution of the unmanned ship, and obtain detailed results by executing path planning; in the scene setting link, the task-related factors are fully considered, and in the process of executing path planning, the fusion and improved method is used to evaluate the impact of the current wind, wave and current interference on the state of the unmanned ship at each sampling moment based on the uncertainty model constructed in the early stage, and then determine the control input with the help of the optimized control strategy, update the state through the unmanned ship dynamics model, and record the path trajectory, navigation time, and control input sequence key information in detail during the whole process, analyze the simulation results, and verify the effectiveness.

Claims

1. A path planning method for an unmanned ship based on fusion model predictive control, characterized in that: The following steps are involved: Step S1: Model and process the uncertainty of external interference to the unmanned ship's marine environment, define interference boundary parameters, consider the maximum possible value range of external interference under a certain probability level, optimize the calculation of uncertain disturbance sets, determine the impact range of external interference on the unmanned ship state, and make the unmanned ship path planning fully consider uncertainty; Step S2: A dual-mode control strategy is constructed according to the real-time status of the unmanned ship and the changes in the ocean environment. A simple control strategy is used in the safe area to ensure stable navigation. Beyond the safe area, the optimization problem is solved to obtain the optimal control input to deal with complex situations. At the same time, a collaborative successive linearization method is introduced to reduce the computational complexity of the dynamic model, and further optimization solutions are performed based on the linear time-varying model under different control modes. Step S3: According to the requirements of the unmanned ship to perform tasks, the conditions of the marine environment and its own resource conditions, the system performance and robustness are dynamically balanced, and the weights of each element in the unmanned ship path planning parameter matrix are reasonably optimized according to the complexity of the marine environment and the type of mission; Design the cost function of the unmanned ship path planning, reasonably determine the value range of the penalty coefficient parameter, adjust the parameters in the cost function in different marine environments and mission scenarios, and ensure that the unmanned ship path planning is balanced between performance and robustness; Step S4: Construct a simulation environment that matches the marine mission scenario in the actual execution of the unmanned ship, and obtain detailed results by executing path planning; in the scene setting link, comprehensively consider the task-related factors, and in the process of executing path planning, use the fusion and improved method to evaluate the impact of the current wind, wave and current interference on the state of the unmanned ship at each sampling moment based on the uncertainty model constructed in the early stage, and then determine the control input with the help of the optimized control strategy, update the state through the unmanned ship dynamics model, and record the path trajectory, navigation time, and key information of the control input sequence in the whole process in detail, analyze the simulation results, and verify the effectiveness.

2. The unmanned ship path planning method based on fusion model predictive control as claimed in claim 1, characterized in that: The step S1 specifically includes: Step S11: define the interference boundary parameters of the external interference of the unmanned ship's marine environment, where the interference of the unmanned ship to the marine environment is ω i (t), define σ max is the maximum value of interference at the 95% confidence level, and the interference limit ρ i To represent the interference ω i (t) intensity range, when ρ i ≤ρ -i When the marine environment is considered suitable for unmanned ships to travel, otherwise it is considered unsuitable for unmanned ships to travel. -i is defined as: Among them, the positive definite matrix Q i It is the weight matrix related to the performance of the unmanned ship system, reflecting the importance of the unmanned ship position and tracking error performance indicators; the positive definite matrix P i is the weighted norm matrix of the state parameters of the unmanned ship system, reflecting whether the unmanned ship meets the stability requirements; α i and ε i is the design parameter of the unmanned ship system terminal, which is determined according to the system design and stability requirements; T is the prediction time domain parameter, which determines the time range of future states considered in path planning; L i It is a constant related to the unmanned ship system model and is related to the nonlinear characteristics of the unmanned ship dynamics model; the matrix eigenvalue λ and are the minimum and maximum eigenvalues ​​of the matrix; Step S12: Optimize the calculation of the uncertain disturbance set to determine the impact range of external disturbance on the state of the unmanned ship. The offline dynamic model of the unmanned ship is x k+1 =f(x k ,u k ,ω k ), where x k is the state vector of the unmanned ship system at time k, including the position, speed, and heading information of the unmanned ship, u k is the control input vector, including the rudder angle command and the thrust force, ω k is the interference of the marine environment; the offline dynamic model of the unmanned ship is linearized and approximated as A k , B k and E k is the matrix related to the system state and control input, obtained by linearizing the dynamic model at the current state; Assume the uncertain perturbation dynamic equation is Δω k+1 =F k Δω k +G k ξ k , F k and G k is the matrix related to system dynamics and disturbance characteristics, ξ k For k The probability distribution of the random noise vector is related to the uncertainty perturbation set X. Δ (k) = {Δω k :||Δω k ||≤σ max },calculate Where W k is a compact set associated with random noise, whose size is determined by the interference ω k The probability distribution of X is determined by real-time monitoring and calculation Δ (k) and X Δ (k+1), adjust the control strategy according to its changing trend, and increase the control input adjustment amplitude when the disturbance set increases to keep the unmanned ship sailing stably.

3. The unmanned ship path planning method based on fusion model predictive control as claimed in claim 1, characterized in that: The step S2 specifically includes: Step S21: Introduce the dual-mode control strategy. When the state of the unmanned ship meets When the unmanned ship is in a safe area, a control strategy based on linear feedback is adopted, which is determined through system stability analysis and performance index design; sensors are used to monitor the state of the unmanned ship in real time and calculate and with ε i Compare and decide whether to switch the control mode; When the unmanned ship approaches an obstacle or deviates from the planned route, that is, the state exceeds the safe area, it switches to the optimized control mode. In the optimized control mode, the interference ω is set k The probability distribution of k ), add constraint P(ω k ∈W k )≥p0, p0 is the preset probability threshold, which is set according to the reliability requirements of the task; the control input u is retained min ≤u k ≤u max and state x min ≤x k ≤x max Constraints, where u min and u max are the lower and upper limits of the control input, depending on the physical limitations of the UAV actuator, x min and x max Reasonable range of the state of the unmanned ship; determine the optimization problem: in, tk is the time value, and the definitions of matrices T, Q, R, and P are the same as in step S11. Step S22: The real-time state and environment of the unmanned ship are coordinated using a successive linearization processing model to calculate the nonlinear dynamic model of the unmanned ship at the sampling time k. Linearize and obtain the initial control input trajectory u at the current moment by translating the optimal control input sequence at the previous moment 0 (i|k), apply this initial control input trajectory to obtain the initial state trajectory x 0 (i|k), the interference input uses the predicted mean Then the nonlinear dynamic model is linearized to obtain x(i+1|k)=x 0 (i+1|k)+A(i|k)Δx(i|k)+B(i|k)Δu(i|k)+C(i|k)Δω(i|k), where A(i|k), B(i|k), and C(i|k) are Jacobian matrices; Step S23: Under the dual-mode control strategy, the model after successive linearization is used in coordination for path planning according to the real-time status and environment of the unmanned ship; in the safe area, the control input is calculated using the simple model after successive linearization to reduce the consumption of computing resources. In the optimization control mode, the optimization problem is solved based on the successive linearization model. When the path needs to be adjusted, a more appropriate control input is obtained based on the linearization model in the optimization control mode to ensure that the unmanned ship can cope with complex environments and mission requirements.

4. The unmanned ship path planning method based on fusion model predictive control as claimed in claim 1, characterized in that: The step S3 specifically includes: Step S31: Adjust the weights according to the intensity of wind, wave and current in the ocean environment, the distribution of obstacles and the transport task or monitoring task of the unmanned ship. In the transport task, increase the weights of adjacent path points in the area with strong winds and waves to keep the path of the unmanned ship stable and avoid damage to the cargo; in the monitoring task, adjust the weights of the path points according to the importance of the monitoring area to ensure comprehensive monitoring while improving robustness; Step S32: Design the cost function of the unmanned ship path planning. By reasonably determining the value range of the penalty coefficient parameter, adjust the parameters in the cost function under different marine environments and mission scenarios. The cost function of the unmanned ship path planning is: Among them, x ref,k is the state vector corresponding to the reference path point, Q k The matrix is ​​a weight matrix that is dynamically adjusted according to the surrounding environment of the waypoint, which is related to the complexity of the environment and the task requirements. The R matrix is ​​also a weight matrix that is used to limit the energy consumption of the control input. i is the penalty coefficient, and its value range is reasonably determined by the relationship between stability and interference bound; Φ(x N ) is a function of the state of the unmanned ship at the terminal moment N, which is used to consider the constraints or targets of the terminal state. It is a deviation function between the position of the unmanned ship at the terminal moment and the target position, or a constraint function on the terminal speed and heading state of the unmanned ship; ω i (s;t k ) represents the unmanned ship at time s, the range of s is from t k to k +T, the i-th interference vector received.

5. The unmanned ship path planning method based on fusion model predictive control as claimed in claim 1, characterized in that: The step S4 specifically includes: Step S41: Setting simulation scene parameters, ω k It follows the normal distribution N(0,0.1), and sets parameters based on the actual performance of the unmanned ship to ensure that the simulation is within the actual operation range. The maximum speed of the unmanned ship v max =5m / s, minimum turning radius r min = 10m, the control input constraint is the thrust of the propeller δ T ∈[-1000,1000]N, rudder angle δ R ∈[-30°,30°]; Multiple obstacles are randomly distributed in the driving area, and their positions and shapes are determined according to the actual environment; Step S42: At each sampling time k, according to the uncertainty model improved in step S1, consider the current wind wave current interference ω k For the impact on the state of the unmanned ship, the control input is determined using the control strategy optimized in step S2; if the unmanned ship is in open waters and the interference is small, the control input is calculated according to the simple control mode in the dual-mode control strategy combined with the successive linearization model; if it is close to a port or a complex area and faces greater interference, it is switched to the optimization control mode, and the optimization problem is solved based on the successive linearization model to obtain the optimal control input; then, the state of the unmanned ship is updated through the unmanned ship dynamics model, and information such as the path trajectory, navigation time, and control input sequence are recorded; this is repeated until the unmanned ship reaches the target position or meets the termination condition, and a complete path planning result including detailed data such as the actual path, the time to reach the target, and the control input at each moment is obtained to verify the path planning effect.

6. An unmanned ship path planning system 100 based on fusion model predictive control, characterized in that: The system includes: The external environment construction module 101 is used to model and process the uncertainty of the external interference of the unmanned ship's marine environment, define the interference boundary parameters, consider the maximum possible value range of the external interference under a certain probability level, optimize the calculation of the uncertain disturbance set, determine the impact range of the external interference on the unmanned ship state, and make the unmanned ship path planning fully consider the uncertainty; The fusion model prediction control module 102 is used to build a dual-mode control strategy according to the real-time status of the unmanned ship and the changes in the ocean environment. A simple control strategy is used in the safe area to ensure stable navigation. Beyond the safe area, the optimization problem is solved to obtain the optimal control input to deal with complex situations. At the same time, a collaborative successive linearization method is introduced to reduce the calculation complexity of the dynamic model, and further optimization solutions are performed based on the linear time-varying model under different control modes; The path planning module 103 is used to dynamically balance the system performance and robustness according to the requirements of the unmanned ship to perform tasks, the conditions of the marine environment in which it is located, and its own resource conditions, and reasonably optimize the weights of each element in the unmanned ship path planning parameter matrix according to the complexity of the marine environment and the type of task; design the cost function of the unmanned ship path planning, and adjust the parameters in the cost function in different marine environments and task scenarios by reasonably determining the value range of the penalty coefficient parameter to ensure that the unmanned ship path planning is balanced between performance and robustness; The simulation verification module 104 is used to construct a simulation environment that matches the marine mission scenario in the actual execution of the unmanned ship, and obtain detailed results by executing path planning; in the scene setting link, the task-related factors are fully considered, and in the process of executing path planning, the fusion and improved method is used to evaluate the impact of the current wind, wave and current interference on the state of the unmanned ship at each sampling moment based on the uncertainty model constructed in the early stage, and then determine the control input with the help of the optimized control strategy, update the state through the unmanned ship dynamics model, and record the path trajectory, navigation time, and control input sequence key information in detail during the whole process, analyze the simulation results, and verify the effectiveness.

Images