CN-115903840-B - Ship dynamic autonomous navigation method and device based on domain gradient field

CN115903840BCN 115903840 BCN115903840 BCN 115903840BCN-115903840-B

Abstract

The invention discloses a dynamic autonomous navigation method of a ship based on domain gradient fields, which comprises the following steps of establishing a domain gradient field decreasing model of a ship meeting scene, establishing a ship navigation motion constraint model, establishing domain gradient field models of scenes corresponding to different ship meeting scenes after establishing the ship navigation motion constraint model, converting different ship meeting scene navigation path design problems into a nonlinear optimization problem with constraint and solving, and realizing local navigation path design of the ship and reaching target points. In the running process of the ship, the intelligent autonomous device is used for collecting information of the ship, the other ship and the obstacle, and effective movement decision is provided by judging the positions and the states of all the parties in the running process, so that the safety of the ship is ensured.

Inventors

LIU CHENGUANG
ZHANG KANG
HE ZHIBO
CHU XIUMIN
WANG CHENGUANG
LI TAILONG
LAI LONGHUA

Assignees

武汉理工大学

Dates

Publication Date: 20260505
Application Date: 20221220

Claims (6)

1. The dynamic autonomous navigation method of the ship based on the domain gradient field is characterized by comprising the following steps of: establishing a domain gradient field decreasing model of a ship meeting scene; Establishing a ship navigation motion constraint model; After establishing a ship navigation motion constraint model, establishing domain gradient field models of scenes corresponding to different ship meeting scenes; Converting the problem of the design of the ship capable of meeting the scene into a nonlinear optimization problem with constraint and solving the nonlinear optimization problem, so as to realize the design of the local ship path and reach the target point; the specific steps of establishing a domain gradient field decreasing model of a ship meeting scene are as follows: When a ship encounters a stationary object, the ship establishes a detection area by taking the position of the ship as the center, and establishes a static domain gradient field of a circular field intensity influence range by taking the size of the detection area as the radius to calculate the field intensity; when a ship encounters a moving object, a dynamic domain gradient field is established by considering the domain field existing in the moving object; The method for calculating the field intensity in the static domain gradient field of the domain gradient field decreasing model formed by combining the static domain gradient field and the dynamic domain gradient field comprises the following steps: In the middle of Is the field strength of the static obstacle, Is the field strength of the target point, Is the distance between the vessel and the stationary object, Adjustable parameters representing the risk of different static objects, A constant representing the detection area of the static object, Is the radius of the static unit, pmin is the minimum intensity value, And The distance from the ship and the starting point to the end point respectively; The Coldwell ship field and the quaternary ship field are combined to obtain a dynamic domain gradient field, which is expressed as: wherein QSD is the field of quaternary ships, x is the x-axis direction of the coordinate axis established by taking the center of the ship body as the origin, y is the y-axis direction of the coordinate axis established by taking the center of the ship body as the origin, Represents a boundary of the object and, Is a captain; the solved nonlinear optimization problem with constraints at least comprises the following contents: and (3) ship motion constraint: Ship rudder angle constraint: ship bow angle constraint: constraint of rudder angle to bow angular speed of ship: objective function: description of the vessel in the first Time of day, prediction step size The internal expectation accumulation is minimum; Wherein, the Is the first The heading of the ship at the moment, Is the first The position of the ship at the moment in time, Is the wind and flow interference to the navigation of the ship, Is the unit time interval of the ship motion simulation, For the discretization step size, Representing the number of discrete steps, Is the first The advancing speed of the bow at the moment, Is a preset minimum value of rudder angle constraint of the ship, Is the maximum value of the constraint of the preset rudder angle of the ship, In order to be the rudder angle, Is the first The rudder angle at the moment of time, Is the first The angular velocity of the bow at the moment, For the angular velocity of the bow of the ship, Is a nonlinear coefficient, F is the predicted minimum field strength, The field strength at each sampling instant i.
2. The method for dynamically autonomous navigation of a vessel based on domain gradient fields according to claim 1, wherein the specific method for establishing the vessel navigation motion constraint model is established according to a first-order linear Nomoto model, which is expressed as: In the formula, Are operational coefficients of the ship, respectively represent a gyratory index and a following index of the ship, For the angular velocity of the bow swing, Is the rudder angle.
3. The method for dynamically autonomous sailing of a ship based on a domain gradient field according to claim 1, wherein the ship meeting scene at least comprises a ship meeting scene, a ship crossing meeting scene and a ship overtaking meeting scene, wherein the ship meeting scene further comprises a port ship meeting scene, a front ship meeting scene and a starboard ship meeting scene, the ship overtaking meeting scene further comprises ① that the ship is overtaken, the ship keeps the original course and speed to continue running, and the ship is turned by the opposite side to avoid the scene encountered by the two ships and the ② ship overtaken opposite side, and the ship needs to turn to a yielding scene.
4. The domain gradient field-based ship dynamic autonomous navigation method of claim 1, wherein model prediction strategies are introduced when different ships meet scene route design problems to be converted into nonlinear optimization problems with constraints and solved.
5. The method for dynamically autonomous navigation of a vessel based on domain gradient fields according to claim 4, wherein the algorithm principle of converting the different vessel meeting scene route design problem into the nonlinear optimization problem with constraint and solving is as follows At the moment, according to the state information of the ship and the predicted environment information, the step length is predicted Obtaining the minimum cost motion sequence of the ship, and taking the first step in the predicted minimum cost motion sequence as the first step of the ship ) Outputting motion instructions at the moment.
6. An apparatus for dynamic autonomous navigation of a ship using a domain-based gradient field as set forth in claim 1, comprising an information flow transmission unit, a data processing unit, a design output unit, and an effect evaluation unit, wherein, The information flow transmission unit is used for transmitting the position, the size, the displacement and the time information of other units captured by the ship sensing system to the data processing unit, and the data storage unit is used for analyzing and processing the information; The data processing unit receives various information and data streams transmitted by the information stream transmission unit, adopts a grid method to carry out map modeling, establishes a field intensity influence range for each unit, and carries out processing on various information data of ships and obstacles on the basis; The design output unit is used for issuing an optimal route solved by an algorithm program in the storage medium to a ship control end and guiding the ship to run under the optimal route; and the effect evaluation unit is used for evaluating the design effect of the optimal route by comparing and analyzing the actual advancing route of the ship with the optimal route obtained by solving, and providing basic data for device improvement and optimization.

Description

Ship dynamic autonomous navigation method and device based on domain gradient field Technical Field The invention belongs to the field of waterway traffic, and particularly relates to a dynamic autonomous navigation method and device for a ship based on a domain gradient field. Background With the development of shipping industry, the density of ships is continuously increased, the channels are more crowded, and the risks of accidents such as collision of the ships are increased. And ship accidents can cause great loss to life, property and environment. In recent years, with the rapid development of technologies such as artificial intelligence, big data, advanced sensors, beidou satellite navigation and the like, autonomous water surface vessels have become a necessary trend for the intelligent development of shipping industry and ship industry. Because ships have nonlinear, high-hysteresis, large-inertia and other dynamic characteristics, marine traffic rules need to be complied with, so how to realize autonomous safe running of the ship in a complex meeting scene of the ship has challenges. Most current ship movement route design methods consider that the design is relatively simple and do not consider time attributes, and when multiple ships meet, a safe driving route is difficult to generate. Disclosure of Invention The invention aims to provide a dynamic autonomous navigation method and device for a ship based on domain gradient fields, which are characterized in that in the running process of the ship, an intelligent autonomous device is utilized to collect information of the ship, other ships and obstacles, and effective movement decisions are provided by judging the positions and states of all parties in the running process, so that the safety of the ship is ensured. In order to solve the technical problems, the technical scheme of the invention is that the dynamic autonomous navigation method of the ship based on the domain gradient field is applied to a dynamic autonomous navigation device of the ship based on the domain gradient field, the device comprises an information flow transmission unit, a data processing unit, a design output unit and an effect evaluation unit, wherein, The information flow transmission unit is used for transmitting the position, the size, the displacement and the time information of other units captured by the ship sensing system to the data processing unit, and the data storage unit is used for analyzing and processing the information; The data processing unit receives various information and data streams transmitted by the information stream transmission unit, adopts a grid method to carry out map modeling, establishes a field intensity influence range for each unit, and carries out processing on various information data of ships and obstacles on the basis; The design output unit is used for issuing an optimal route solved by an algorithm program in the storage medium to a ship control end and guiding the ship to run under the optimal route; The effect evaluation unit is used for evaluating the design effect of the optimal route by comparing and analyzing the actual advancing route of the ship with the optimal route obtained by solving, and providing basic data for device improvement and optimization; The method comprises the following steps: establishing a domain gradient field decreasing model of a ship meeting scene; Establishing a ship navigation motion constraint model; After establishing a ship navigation motion constraint model, establishing domain gradient field models of scenes corresponding to different ship meeting scenes; and converting the problem of the design of the ship capable of meeting the scene into a nonlinear optimization problem with constraint and solving the nonlinear optimization problem, so as to realize the design of the local ship path and reach the target point. The specific steps of establishing a domain gradient field decreasing model of a ship meeting scene are as follows: When a ship encounters a stationary object, the ship establishes a detection area by taking the position of the ship as the center, and establishes a static domain gradient field of a circular field intensity influence range by taking the size of the detection area as the radius to calculate the field intensity; when a ship encounters a moving object, a dynamic domain gradient field is established by considering the domain field existing in the moving object; and combining the static domain gradient field and the dynamic domain gradient field to form a domain gradient field decreasing model. The method for calculating the field intensity in the static domain gradient field comprises the following steps: In the formula, Is the distance between the vessel and the stationary object,Adjustable parameters representing the risk of different static objects,A constant representing the detection area of the static object,Is the radius of the static unit, pmin is the minimum in