A real-time decision-making method for safe movement of high-speed unmanned vessels

Abstract

This invention discloses a real-time decision-making method for the safe movement of a high-speed unmanned vessel, which relates to the technical field of unmanned vessel motion planning. The method includes designing a preset performance function based on the Froude number that characterizes the unmanned vessel's navigation state, which is adaptively adjusted as the Froude number changes. A high-order obstacle Lyapunov function is designed based on the unmanned vessel's motion deviation value, and a nominal guidance law for the unmanned vessel's high-speed motion is designed using backstepping. A high-order control obstacle function condition is constructed based on the position and speed of obstacles detected in real time by the unmanned vessel to ensure obstacle avoidance safety. Using the high-order control obstacle function condition as a safety constraint and integrating the unmanned vessel system's drive saturation constraint, the method establishes and solves a quadratic programming problem in real time, with the optimization objective of minimizing the deviation between the optimized guidance law and the nominal guidance law, and outputs real-time guidance instructions for the unmanned vessel's control system. This invention effectively ensures the navigation safety of high-speed unmanned vessels in complex dynamic environments.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned vessel motion planning, and in particular to a real-time decision-making method for the safe motion of a high-speed unmanned vessel. Background Art

[0002] With the rapid growth in demand for marine resource development and intelligent exploration, high-speed unmanned vessels, with their significant advantages such as small size, strong maneuverability, and high risk tolerance, are playing an increasingly important role in complex dynamic scenarios such as ocean monitoring and waterway inspections. However, dynamic obstacle avoidance and real-time decision-making at high speeds remain technical bottlenecks restricting their application.

[0003] Traditional control obstacle functions (CBFs) exhibit two critical flaws when operating in high-speed motion systems. First, low-order CBFs cannot constrain higher-order dynamic states, resulting in overly aggressive obstacle avoidance maneuvers. Second, the decoupling of safety constraints from mission objectives often sacrifices trajectory tracking accuracy for collision avoidance. These technical shortcomings severely restrict the ability of high-speed unmanned vessels to navigate complex and dynamic waters. There is an urgent need for an integrated decision-making framework that can simultaneously implement high-order safety constraints, precise trajectory tracking, strong anti-interference capabilities, and rapid response. Summary of the Invention

[0004] In order to overcome the above problems existing in the prior art, the present invention proposes a real-time decision-making method for safe movement of a high-speed unmanned vessel.

[0005] The technical solution adopted by the present invention to solve the technical problem is: a real-time decision-making method for safe movement of a high-speed unmanned ship, comprising the following steps:

[0006] Step 1: Based on the Froude number that characterizes the navigation state of the unmanned ship, design a preset performance function that is adaptively adjusted as the Froude number changes. ;

[0007] Step 2: preset the performance function in step 1. Based on the above, a high-order obstacle Lyapunov function is designed according to the motion deviation value of the unmanned ship, and the nominal guidance law of the unmanned ship high-speed motion is designed using the backstepping method;

[0008] Step 3: Based on the position and speed of obstacles detected by the unmanned vessel in real time, a high-order control obstacle function condition is constructed to ensure obstacle avoidance safety.

[0009] In step 4, the high-order control obstacle function condition described in step 3 is used as a safety constraint, and the unmanned ship system drive saturation constraint is integrated. The deviation between the optimized guidance law and the nominal guidance law in step 2 is minimized as the optimization goal. A quadratic programming problem is established and solved in real time, and real-time guidance instructions acting on the unmanned ship control system are output as the safe motion decision of the high-speed unmanned ship.

[0010] In the above-mentioned real-time decision-making method for safe movement of a high-speed unmanned vessel, step 1 specifically includes:

[0011] Step 1.1, according to the acceleration of gravity 、Unmanned boat waterline length And the current speed of the unmanned ship obtained in real time , calculate the Froude number :

[0012] ;

[0013] Step 1.2: Divide the navigation state according to the Froude number threshold:

[0014] when It is determined to be in displacement navigation state;

[0015] when It is judged as transition state or semi-gliding state;

[0016] when When it is determined to be in the sliding state, is the critical Froude number for high-speed navigation;

[0017] Step 1.3: Design the preset performance function of the unmanned ship's motion :

[0018] ;

[0019] Where, and They are the preset initial and final performance values ​​respectively; , is the logistic function, Represents the adjustment coefficient.

[0020] In the above-mentioned real-time decision-making method for safe movement of a high-speed unmanned vessel, step 2 specifically includes:

[0021] Step 2.1: The motion error of the unmanned ship Limited to the preset performance function ;

[0022] Step 2.2, construct the fourth-order tangent barrier Lyapunov function:

[0023] ;

[0024] Step 2.3, The error dynamics equation is obtained by differentiation:

[0025] ;

[0026] Step 2.4, design control parameters make sure , the inverse solution is used to obtain the nominal guidance law of the unmanned ship motion .

[0027] In the above-mentioned real-time decision-making method for safe movement of a high-speed unmanned vessel, step 3 specifically includes:

[0028] Step 3.1: Select the obstacle control function based on the real-time motion information of the unmanned ship and the detected obstacle information. And define the security collection :

[0029] ;

[0030] in, is the dynamic safety distance positively correlated with the ship speed, and are the positions of the unmanned ship and the obstacle respectively;

[0031] Step 3.2: For the unmanned vessel's vertical speed , construct the first-order control barrier function condition:

[0032] ;

[0033] in, for Class functions, , for The first-order differential of The corresponding security set is expressed as ;

[0034] Step 3.3: Based on the first-order control obstacle function condition obtained in step 3.2, the bow angular velocity of the unmanned ship is Construct the second-order control barrier function condition:

[0035] ;

[0036] in, for Class functions, is the first-order control barrier function condition The first-order differential of The corresponding security set is expressed as ;

[0037] Step 3.4: According to the control obstacle function conditions of steps 3.2 and 3.3, the guidance law is calculated. Constraint matrix inequality:

[0038] ;

[0039] in, , is the constraint coefficient matrix, is the constraint vector.

[0040] In the above-mentioned real-time decision-making method for safe movement of a high-speed unmanned vessel, step 4 specifically includes:

[0041] Step 4.1: Establish the quadratic programming objective function to minimize the optimized guidance law With nominal guidance law The deviation is the optimization goal :

[0042] ;

[0043] Step 4.2, construct the quadratic programming problem:

[0044] ;

[0045] in, To optimize the goal, is the bow angular velocity of the unmanned ship, is the unmanned boat's vertical swing speed, is the maximum longitudinal speed of the unmanned ship, is the maximum bow angular velocity of the unmanned ship, is the constraint coefficient matrix, is the constraint vector;

[0046] Solve the above problems in real time and calculate the optimal guidance law , which is the optimal decision instruction for the safe movement of high-speed unmanned boats.

[0047] The beneficial effect of the present invention is that, through the collaborative innovative design of the high-order obstacle Lyapunov function (HOBLF) and the high-order control barrier function (HOCBF), combined with the hydrodynamic characteristics of high-speed unmanned ships and real-time environmental perception, an effective trade-off between motion efficiency and safe navigation is achieved, which greatly improves the reliability of unmanned ships in highly dynamic tasks such as patrol surveillance and emergency rescue, and provides an integrated decision-making solution with strict safety guarantees and real-time response for complex ocean scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a schematic diagram of the process of the present invention;

[0049] Figure 2 Schematic diagram of simulation results in an embodiment of the present invention. DETAILED DESCRIPTION

[0050] In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0051] like Figure 1 As shown, this embodiment discloses a real-time decision-making method for safe movement of a high-speed unmanned vessel, which specifically includes the following steps:

[0052] Step 1: Based on the Froude number that characterizes the navigation state of the unmanned ship, a preset performance function that is adaptively adjusted as the Froude number changes is designed.

[0053] Specifically include:

[0054] Step 1.1, according to the acceleration of gravity 、Unmanned boat waterline length And the current speed of the unmanned ship obtained in real time , calculate the Froude number:

[0055] ;

[0056] Step 1.2: Divide the navigation state according to the Froude number threshold:

[0057] when It is determined to be in displacement navigation state;

[0058] when It is determined to be a transitional (or semi-planing) navigation state;

[0059] when It is judged as the sliding state;

[0060] in, is the critical Froude number for high-speed navigation. ;

[0061] Step 1.3, design the preset performance function of the unmanned ship movement:

[0062] ;

[0063] Where, and They are the preset initial and final performance values ​​respectively. , is the logistic function, Represents the adjustment coefficient.

[0064] Step 2: Based on the preset performance function described in step 1, the HOBLF is designed according to the motion deviation value of the unmanned ship, and the nominal guidance law for high-speed motion of the unmanned ship is designed using the backstepping method.

[0065] Step 2 specifically includes:

[0066] Step 2.1: The motion error of the unmanned ship Limited to the preset performance function obtained in step 1.3 ;

[0067] Step 2.2, construct the fourth-order tangent barrier Lyapunov function (HOBLF):

[0068] ;

[0069] Step 2.3, The error dynamics equation is obtained by differentiation:

[0070] ;

[0071] Step 2.4, design control parameters make sure , the inverse solution is used to obtain the nominal guidance law of the unmanned ship motion .

[0072] Step 3: Based on the position and speed of the obstacle detected by the unmanned boat in real time, the HOCBF condition to ensure obstacle avoidance safety is constructed.

[0073] Step 3 specifically includes:

[0074] Step 3.1: Select the obstacle control function based on the real-time motion information of the unmanned ship and the detected obstacle information. And define the security collection :

[0075] ;

[0076] in, is the dynamic safety distance positively correlated with the ship speed, and are the positions of the unmanned ship and the obstacle respectively;

[0077] Step 3.2, relative order is the core concept in control theory that characterizes the dynamic response relationship between system input and output. Its technical meaning is: the output function of a nonlinear system Relative order of control input Defined as the smallest integer that satisfies the following conditions:

[0078] ;

[0079] Where, For function Along the vector field The k-th order Lie derivative of , For function Along the vector field The r-1 Lie derivative of Along the vector field The Lie derivative of is the system state vector. (relative order is 1), construct the first-order control barrier function condition:

[0080] ;

[0081] in, for Class functions, , for The first-order differential of The corresponding security set is expressed as ;

[0082] Step 3.3: Based on the first-order control obstacle function condition obtained in step 3.2, the bow angular velocity of the unmanned ship is (Relative order is 2) Construct the second-order control barrier function condition:

[0083] ;

[0084] in, for Class functions, is the first-order control barrier function condition The first-order differential of The corresponding security set is expressed as .

[0085] Step 3.4: Based on the control obstacle function conditions in steps 3.2 and 3.3, calculate the guidance law constraint matrix inequality:

[0086] ;

[0087] in, , is the constraint coefficient matrix, is the constraint vector.

[0088] In step 4, the HOCBF condition described in step 3 is used as a safety constraint, and the unmanned vessel system drive saturation constraint is integrated. With minimizing the deviation between the optimized guidance law and the nominal guidance law in step 2 as the optimization goal, a quadratic programming problem is established and solved in real time, and the optimal real-time guidance command acting on the unmanned vessel control system is output as the safe motion decision of the high-speed unmanned vessel.

[0089] Step 4 specifically includes:

[0090] Step 4.1: Establish the quadratic programming objective function to minimize the difference between the optimized guidance law and the nominal guidance law. Deviation:

[0091] ;

[0092] Step 4.2: Based on the HOCBF obstacle avoidance constraints and the UAV control input constraints in step 3.4, construct a quadratic programming problem:

[0093] .

[0094] Solve the above problems in real time and calculate the optimal guidance law , which is the optimal decision instruction for the safe movement of high-speed unmanned ship. In the control obstacle function Under the effect, for The optimized guidance law at any moment can ensure that the unmanned ship system state is always in the safe subset That is to say, it can ensure that the unmanned ship avoids collision with environmental obstacles during the entire process of high-speed movement.

[0095] The experimental object was a high-speed unmanned vessel with a waterline length of 6.34 meters. The target motion information for the unmanned vessel was derived from a sea trial, during which the vessel approached the target at a speed of 8 meters per second. To verify the effectiveness of the real-time decision-making method for safe motion disclosed in this embodiment, a simulation experiment was conducted on the MATLAB platform using a portable computer equipped with an i5-13500HX CPU. Based on the above unmanned vessel sea trial data, a virtual obstacle was placed on the motion trajectory. The center of the obstacle was set at (61m, 0), and a circle with a radius of 5 meters, centered around the obstacle, was designated as a prohibited area for the unmanned vessel. The unmanned vessel was set to detect the obstacle when it was 20 meters from the obstacle center, at which point it should initiate collision avoidance maneuvers.

[0096] The simulation results are as follows Figure 2 The black thick solid line is the nominal guidance law The trajectory of the unmanned boat under control. The red solid line is the unmanned boat under the optimal guidance law. The trajectory of motion under control. It can be seen that in the optimal guidance law Under this effect, the unmanned ship can make the required collision avoidance decisions based on obstacle information while sailing at high speed, and the collision avoidance trajectory is relatively smooth.

[0097] The above embodiments are merely exemplary embodiments of the present invention and are not intended to limit the present invention. Those skilled in the art may make various modifications or equivalent substitutions to the present invention within the spirit and scope of protection of the present invention, and such modifications or equivalent substitutions shall also be deemed to fall within the scope of protection of the present invention.

Claims

1. A real-time decision-making method for safe movement of high-speed unmanned vessels, characterized by: The steps include: Step 1: Based on the Froude number that characterizes the navigation state of the unmanned ship, design a preset performance function that is adaptively adjusted as the Froude number changes. ; Step 2: preset the performance function in step 1. Based on the above, a high-order obstacle Lyapunov function is designed according to the motion deviation value of the unmanned ship, and the nominal guidance law of the unmanned ship high-speed motion is designed using the backstepping method; Step 3: Based on the position and speed of obstacles detected by the unmanned vessel in real time, a high-order control obstacle function condition is constructed to ensure obstacle avoidance safety. Step 4: Using the high-order control obstacle function condition described in step 3 as a safety constraint and integrating the unmanned vessel system drive saturation constraint, with minimizing the deviation between the optimized guidance law and the nominal guidance law in step 2 as the optimization goal, a quadratic programming problem is established and solved in real time, and real-time guidance instructions acting on the unmanned vessel control system are output as the safe motion decision of the high-speed unmanned vessel; The step 2 specifically includes: Step 2.1: The motion error of the unmanned ship Limited to the preset performance function ; Step 2.2, construct the fourth-order tangent barrier Lyapunov function: ； Step 2.3, The error dynamics equation is obtained by differentiation: ； Step 2.4, design control parameters make sure , the inverse solution is used to obtain the nominal guidance law of the unmanned ship ; The step 3 specifically includes: Step 3.1: Select the obstacle control function based on the real-time motion information of the unmanned ship and the detected obstacle information. And define the security collection : ； in, is the dynamic safety distance positively correlated with the ship speed, and are the positions of the unmanned ship and obstacles respectively; Step 3.2: For the unmanned vessel's vertical speed , construct the first-order control barrier function condition: ； in, for Class functions, , for The first-order differential of The corresponding security set is expressed as ; Step 3.3: Based on the first-order control obstacle function condition obtained in step 3.2, the bow angular velocity of the unmanned ship is Construct the second-order control barrier function condition: ； in, for Class functions, is the first-order control barrier function condition The first-order differential of The corresponding security set is expressed as ; Step 3.4: According to the control obstacle function conditions of steps 3.2 and 3.3, the guidance law is calculated. Constraint matrix inequality: ； in, , is the constraint coefficient matrix, is the constraint vector.

2. A real-time decision-making method for safe movement of a high-speed unmanned vessel according to claim 1, characterized in that: The step 1 specifically includes: Step 1.1, according to the acceleration of gravity 、Unmanned boat waterline length And the current speed of the unmanned ship obtained in real time , calculate the Froude number : ； Step 1.2: Divide the navigation state according to the Froude number threshold: when It is determined to be in displacement navigation state; when It is judged as transition state or semi-gliding state; when When it is determined to be in the sliding state, is the critical Froude number for high-speed navigation; Step 1.3: Design the preset performance function of the unmanned ship's motion : ； Where, and They are the preset initial and final performance values ​​respectively; , is the logistic function, Represents the adjustment coefficient.

3. The real-time decision-making method for safe movement of a high-speed unmanned vessel according to claim 1 is characterized in that: The step 4 specifically includes: Step 4.1: Establish the quadratic programming objective function to minimize the optimized guidance law With nominal guidance law The deviation is the optimization goal : ； Step 4.2, construct the quadratic programming problem: ； in, To optimize the goal, is the bow angular velocity of the unmanned ship, is the unmanned boat's vertical swing speed, is the maximum longitudinal speed of the unmanned ship, is the maximum bow angular velocity of the unmanned ship, is the constraint coefficient matrix, is the constraint vector; Solve the above problems in real time and calculate the optimal guidance law , which is the optimal decision instruction for the safe movement of high-speed unmanned boats.

Images