Autonomous navigation decision-making method, equipment and medium based on digital twin navigation environment

Abstract

The present invention relates to an autonomous navigation decision-making method, device and medium based on a digital twin navigation environment, and belongs to the technical field of intelligent navigation of ships. The autonomous navigation decision-making method based on the digital twin navigation environment comprises: determining a feasible maneuvering range of a ship based on a constructed digital twin navigation environment model, a ship motion model considering environmental interference, an adaptive PID control method and a speed barrier algorithm; constructing a ship dynamic collision risk model based on the feasible maneuvering range of the ship, and then constructing a multi-ship encounter collision avoidance model in combination with an improved speed barrier algorithm; constructing an autonomous navigation decision-making scheme based on the ship dynamic collision risk model, a constructed ship track tracking model and a ship state estimation model, and a multi-ship encounter collision avoidance model; constructing an adaptive PID control model based on an adaptive PID control method, and controlling the ship to execute the autonomous navigation decision-making scheme based on the adaptive PID control model, so as to realize adaptive autonomous navigation decision-making of ships in nearshore waters.

Description

Technical Field

[0001] The present invention relates to the field of ship intelligent navigation technology, and in particular to an autonomous navigation decision-making method, device and medium based on a digital twin navigation environment. Background Art

[0002] The development of smart ships and unmanned ships is an important area of ​​maritime research. The great progress of autonomous navigation has improved the convenience of maritime operations. With the rapid development of technology, autonomous navigation research in the maritime field is becoming increasingly rich and complete.

[0003] The field of ship autonomy has been of great interest due to its potential to achieve autonomous navigation. Although autonomous navigation in open waters has been extensively studied, less attention has been paid to nearshore waters due to the complexity of the environment and traffic flows.

[0004] However, affected by factors such as the navigation environment and ship characteristics, the research on autonomous navigation decision-making in offshore waters needs further study, especially the complex navigation environment in offshore waters and obvious restrictions on ship maneuvering, which put forward higher requirements for autonomous navigation decision-making. Summary of the invention

[0005] In view of this, it is necessary to provide an autonomous navigation decision-making method, equipment and medium based on a digital twin navigation environment to solve the technical problem of low accuracy of autonomous navigation decision-making due to complex offshore environment and limited ship maneuverability.

[0006] In order to solve the above problems, the present invention provides an autonomous navigation decision-making method based on a digital twin navigation environment, comprising:

[0007] Determine the feasible maneuvering range of the ship based on the constructed digital twin navigation environment model, the ship motion model considering environmental interference, the adaptive PID control method and the speed obstacle algorithm;

[0008] After constructing a ship dynamic collision risk model based on the feasible maneuvering range of the ship, a multi-ship encounter collision avoidance model is constructed in combination with an improved speed barrier algorithm;

[0009] An autonomous navigation decision-making scheme is constructed based on the ship dynamic collision risk model, the constructed ship track tracking model and the ship state estimation model, and the multi-ship encounter and collision avoidance model;

[0010] An adaptive PID control model is constructed based on the adaptive PID control method, and the ship is controlled to execute an autonomous navigation decision-making plan based on the adaptive PID control model.

[0011] In a possible implementation, the method of determining the feasible maneuvering range of a ship based on the constructed digital twin navigation environment model, the ship motion model considering environmental interference, the adaptive PID control method, and the speed obstacle algorithm includes:

[0012] Build digital twin navigation environment models for static and dynamic environments based on electronic nautical charts;

[0013] Considering the influence of wind disturbance, water flow disturbance and shallow water effect on ship motion, a ship motion model considering environmental disturbance is constructed;

[0014] Determine the relative speed between the own ship and the target ship based on the digital twin navigation environment model and the ship motion model considering environmental interference;

[0015] Determine the absolute collision interval based on the ship motion model considering environmental interference, the adaptive PID control method, the speed obstacle algorithm and the relative speed;

[0016] Determine a feasible speed range based on the absolute collision range in combination with the collision avoidance rule;

[0017] The ship's clock is obtained, and a feasible maneuvering range of the ship is determined based on the feasible speed range and the clock.

[0018] In a possible implementation, the calculation formula of the ship motion model considering environmental interference is:

[0019] ,

[0020] in, For the quality of the ship, , are the additional masses of the ship in the x-axis and y-axis directions, , , , are the speed and acceleration of the ship in the x-axis and y-axis directions, , are the angular velocity and angular acceleration of the ship on the x-axis and y-axis respectively, , are the ship's moment of inertia and additional moment of inertia, , , are the forces on the bare hull, propeller and rudder in the x-axis direction, , , are the forces on the bare hull, propeller and rudder in the y-axis direction, is the force acting on the ship, , , The bare hull, oars and rudder are respectively. , , External disturbance of wind.

[0021] In a possible implementation, the calculation formula for the feasible maneuvering range of the ship is:

[0022] ,

[0023] ,

[0024] ,

[0025] in, is the absolute collision interval, is the relative collision interval, Target ship The speed of time, is the Minkowski vector and operator, For this ship The speed of time, is the feasible speed range for own ship to avoid collision, For speed ranges not allowed by collision avoidance rules, is the union operator, is the feasible maneuvering range of the ship, A ship's bell.

[0026] In a possible implementation, the constructing of a ship dynamic collision risk model based on the ship feasible maneuvering range and the geometric model includes:

[0027] Determining a dangerous maneuver set and a feasible maneuver set of the ship based on the feasible maneuvering range of the ship;

[0028] Based on the dangerous maneuver set and the feasible maneuver set, a ship dynamic collision risk model is constructed according to a geometric model.

[0029] In a possible implementation, the calculation formula of the ship dynamic collision risk model is:

[0030] ,

[0031] ,

[0032] in, is the dynamic collision risk of ships, is the number of dangerous maneuvers set for the ship, is the number of all feasible manipulation sets, For ship maneuvering decisions, is the weight corresponding to the ship maneuvering decision, The range of values ​​for the heading of a ship’s dangerous maneuvers.

[0033] In a possible implementation, the calculation formula of the multi-ship collision avoidance model is:

[0034] ,

[0035] in, is the feasible maneuvering range for multiple ships, for The feasible maneuvering range of the ship at the point for The risk of ship collision at the point In order to consider the influence of shallow water, is the Minkowski vector and operator.

[0036] In a possible implementation, the controlling the ship to execute an autonomous navigation decision-making scheme based on the adaptive PID control model includes:

[0037] Acquire multi-source heterogeneous information, integrate the multi-source heterogeneous information into the digital twin navigation environment model, and obtain the motion state of the target ship based on the digital twin navigation environment model;

[0038] Obtaining the motion state of the ship based on the digital twin navigation environment model, the ship motion model considering environmental interference, the adaptive PID control method and the ship state estimation model;

[0039] Based on the motion state of the own ship and the motion state of the target ship, combined with the ship dynamic collision risk model, determine whether there is a collision risk. When there is a collision risk, perform an autonomous collision avoidance decision based on the autonomous navigation decision-making plan, and use the time rolling model to dynamically correct the error or deviation in the autonomous collision avoidance decision to obtain a corrected navigation path, and control the own ship to navigate based on the corrected navigation path. When there is no collision risk, control the own ship to navigate based on the navigation path determined by the ship track tracking model.

[0040] In another aspect, the present invention further provides an electronic device, comprising: a processor and a memory;

[0041] The memory stores a computer-readable program executable by the processor;

[0042] When the processor executes the computer-readable program, the steps in the autonomous navigation decision-making method based on the digital twin navigation environment as described above are implemented.

[0043] On the other hand, the present invention also provides a computer-readable storage medium, which stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps in the autonomous navigation decision-making method based on the digital twin navigation environment as described above.

[0044] The beneficial effects of the present invention are as follows: a digital twin navigation environment model is constructed to include all potential environmental factors that offshore ships may encounter. By abstracting elements such as shape and speed into computer-recognizable formulas and integrating data from different sources, these models can simulate the navigation environment in real time, construct a ship dynamic collision risk model to determine the time for collision avoidance, and quantify the urgency of collision risk between ships through a ship collision risk index, providing an accurate measurement method that reflects the different degrees of collision urgency in different maritime scenarios, and also takes into account the maneuverability of the target ship and actual maritime navigation experience. The ship dynamic collision risk model is combined with an improved speed barrier algorithm to construct a multi-ship encounter collision avoidance model, so that the ship can achieve autonomous navigation in a multi-ship encounter scenario in complex waters. The ship is controlled to execute an autonomous navigation decision-making plan based on an adaptive PID control model, taking into account the complex offshore environment and the navigation of multiple ships in offshore waters, resulting in limited ship maneuverability, the maneuverability of multiple target ships and the inaccuracy of ship motion prediction, and the prediction error in continuous time intervals is iteratively corrected, thereby significantly improving the accuracy of autonomous navigation decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 A flow chart of an embodiment of an autonomous navigation decision-making method based on a digital twin navigation environment provided by the present invention;

[0046] Figure 2 A schematic diagram of a digital twin navigation environment model of an autonomous navigation decision-making method based on a digital twin navigation environment provided by the present invention;

[0047] Figure 3 A schematic diagram of an adaptive PID control method for an autonomous navigation decision-making method based on a digital twin navigation environment provided by the present invention;

[0048] Figure 4 A schematic diagram of a speed obstacle algorithm for an autonomous navigation decision-making method based on a digital twin navigation environment provided by the present invention;

[0049] Figure 5 A schematic diagram of collision avoidance in a multi-ship encounter of an autonomous navigation decision-making method based on a digital twin navigation environment provided by the present invention;

[0050] Figure 6 A schematic diagram of ship position deviation of the autonomous navigation decision-making method based on the digital twin navigation environment provided by the present invention;

[0051] Figure 7 A structural schematic diagram of an embodiment of an autonomous navigation decision-making device based on a digital twin navigation environment provided by the present invention;

[0052] Figure 8 A schematic structural diagram of an embodiment of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0053] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not used to limit the scope of the present invention.

[0054] The present invention discloses an autonomous navigation decision-making method, device and medium based on a digital twin navigation environment, which can be used in a computer. The method, device or computer-readable storage medium involved in the present invention can be integrated with the above-mentioned device or can be relatively independent.

[0055] A specific embodiment of the present invention discloses an autonomous navigation decision-making method based on a digital twin navigation environment, which can be executed by a computer, specifically by one or more processors of the computer. Figure 1 is a flowchart of an autonomous navigation decision-making method based on a digital twin navigation environment provided by an embodiment of the present invention. Figure 1 , the autonomous navigation decision-making methods based on the digital twin navigation environment include:

[0056] S101. Determine the feasible maneuvering range of the ship based on the constructed digital twin navigation environment model, the ship motion model considering environmental interference, the adaptive PID control method, and the speed obstacle algorithm;

[0057] S102, after constructing a ship dynamic collision risk model based on the feasible maneuvering range of the ship, a multi-ship encounter collision avoidance model is constructed in combination with an improved speed barrier algorithm;

[0058] S103, constructing an autonomous navigation decision-making plan based on the constructed ship track tracking model and ship state estimation model, multi-ship encounter and collision avoidance model, and ship dynamic collision risk model;

[0059] S104, constructing an adaptive PID control model based on the adaptive PID control method, and controlling the ship to execute an autonomous navigation decision-making plan based on the adaptive PID control model.

[0060] Among them, the elements related to the navigation environment, static obstacles and the maneuverability of the target ship are taken into consideration, and a digital twin navigation environment model of the static environment and the dynamic environment is constructed to realize the digitization of the traffic environment in order to fully understand the navigation environment. The influence of wind interference, water flow interference and shallow water effect on the operation of the ship is considered, and a high-precision ship motion model of environmental interference is constructed to accurately simulate the movement of the ship. A ship state estimation model including along the waterway and not along the waterway is constructed to predict the future motion state of the ship. A ship track tracking model is constructed through the line of sight algorithm to determine the navigation trajectory of the ship.

[0061] Compared with the prior art, the autonomous navigation decision-making method based on the digital twin navigation environment provided in this embodiment determines the feasible maneuvering range of the ship based on the constructed digital twin navigation environment model, the ship motion model considering environmental interference, the adaptive PID control method and the speed barrier algorithm, takes into account the influence of environmental interference on the ship motion and the collision avoidance rules, constructs a ship dynamic collision risk model based on the feasible maneuvering range and the geometric model of the ship, and constructs a multi-ship encounter collision avoidance model in combination with the improved speed barrier algorithm. Through the improved ship dynamic collision risk model, the maneuverability of the target ship and the actual maritime navigation experience are taken into account, the urgency of the collision risk between ships is quantified, and an accurate measurement method is provided to reflect the The different degrees of collision urgency in different maritime scenarios are taken into account through the multi-ship encounter and collision avoidance model, which takes into account the situation of multiple ships sailing in near-shore waters, so that ships can realize autonomous navigation in multi-ship encounter scenarios in complex waters. An autonomous navigation decision-making scheme is constructed based on the ship dynamic collision risk model, the constructed ship track tracking model and the ship state estimation model as well as the multi-ship encounter and collision avoidance model. The ship is controlled to execute the autonomous collision avoidance decision-making scheme based on the adaptive PID control model. The complex offshore environment, the maneuverability of multiple ships and the inaccuracy of ship motion prediction are taken into account. The time rolling algorithm is used to continuously update and improve the ship's autonomous navigation decision. The prediction error in continuous time intervals is iteratively corrected to significantly improve the accuracy of autonomous navigation decisions.

[0062] In some embodiments, in step S101, the feasible maneuvering range of the ship is determined based on the constructed digital twin navigation environment model, the ship motion model considering environmental interference, the adaptive PID control method and the speed obstacle algorithm. First, the digital twin navigation environment models of static and dynamic environments are constructed based on the electronic nautical chart, and all potential environmental factors that the ship may encounter, including static environmental factors and dynamic environmental factors, are considered to construct a static environment model and a dynamic environment model. The digital twin navigation environment model is constructed based on the static environment model and the dynamic environment model, and all potential environmental factors (including static and dynamic factors) that the ship may encounter are included. By abstracting elements such as shape and speed into computer-recognizable formulas and integrating data from different sources, these models can simulate the navigation environment in real time;

[0063] The static environment refers to stationary objects with zero speed relative to the ground. These elements contain multiple elements and can be divided into two categories according to their characteristics. The first category includes elements related to the navigation environment, such as channel boundaries and separated navigation areas. The second category includes static obstacles, such as shipwrecks, docks, reefs, drilling platforms, islands and navigation equipment. Considering that the speed of static environment elements is zero, modeling only needs to consider their shape and size; in order to better represent the navigable waterway, the navigable waterway is divided into multiple segments. Each line segment can be regarded as a closed quadrilateral defined by four points. The description of the channel is:

[0064] ,

[0065] in, For waterway In part, For the region, , , , For static obstacles The four vertices of are the coordinates of the four vertices. However, the shapes of warning areas, isolation areas and coastal traffic areas vary greatly and are often circular or polygonal. In addition to waterways, other static obstacles in coastal waters also need to be modeled. Obstacles with shapes close to circles, such as navigation equipment, shipwrecks, coral reefs, etc., are selected as mathematical models. The circular mathematical model is:

[0066] ,

[0067] in, is the center of the circular area, is the radius of the circle; for docks and islands, a polygonal mathematical model is used, and its polygonal mathematical model is:

[0068] ,

[0069] in, are coordinate points, and the area is composed of these coordinate points;

[0070] The dynamic environment generally refers to various types of ships in progress, except the ship itself, whose speed on the ground is greater than zero. In addition to ordinary ships, there are various types of powered ships, such as tugboats, fishing boats and other types of ships. Therefore, a variety of mathematical models, such as ellipses, circles, rectangles, etc., are used to represent them. The ellipse represents the mathematical representation of the dynamic environment of a single ship, while ships engaged in fishing activities, such as double trawlers and purse seine boats, are usually gathered together and represented by the mathematical model of a circle. The mathematical model of ship formation navigation and towing navigation can be represented by a rectangle, and the calculation formula for the side length of the rectangle is:

[0071] ,

[0072] in, For the field width of ships sailing in formation and towing, The length of the field for ships sailing in formation and towing, For ships Length, is the number of ships, For the The length of the ship, is the position coordinates of the ship;

[0073] The digital twin navigation environment model is constructed based on the static environment model and the dynamic environment model, which can be used as the input of computer calculation. Therefore, the environment of the electronic nautical chart can be converted into the digital twin navigation environment model. For the schematic diagram of the digital twin navigation environment model, please refer to Figure 2 ,like Figure 2 As shown, dynamic obstacles include 4 target ships determined based on AIS data, and other factors are static obstacles. Symbols 1, 2, 3, 4, and 5 represent islands. Symbols 4 and 5 represent islands with irregular shapes, which ships should avoid entering or approaching. Symbol A represents an anchorage area, and symbols B, C, and D represent channel boundaries.

[0074] Secondly, considering the influence of wind interference, water flow interference and shallow water effect on ship motion, a ship motion model considering environmental interference is constructed. In the process of ship navigation and collision avoidance, it is generally not necessary to consider the ship's roll, pitch and heave motion; during ship navigation, the ship is often exposed to winds of different directions and intensities. Considering the influence of wind on ship motion, a wind interference model is established. The calculation formula of the wind interference model is:

[0075] ,

[0076] in, is the lateral wind pressure acting on the hull, is the longitudinal wind pressure acting on the hull, is the moment acting on the hull, is the length of the ship, is the orthographic projection area of ​​the ship above the waterline, is the side projection area of ​​the ship above the waterline, is the wind pressure coefficient, is the ratio of the wind pressure action point to the length of the ship, is the relative wind speed, is the wind speed angle, is the wind pressure resultant angle, is the air density at the location of the ship; under the influence of the westerly wind, the ship moves eastward, and under the influence of the north wind, the center of the ship's turn moves southward. These observations are consistent with maritime practice, proving the influence of wind on ship movement.

[0077] Uniform water flow will not generate hydrodynamic force on the ship. The water flow only affects the ship's speed according to the kinematic principle, while the heading of the bow remains unchanged. The calculation formula of the water flow interference model is:

[0078] ,

[0079] in, is the longitudinal speed of the ship in the water, is the transverse speed of the ship in the water, is the velocity of the ocean current in a fixed coordinate system, is the direction of the ocean current in a fixed coordinate system, It is the heading angle. The direction of the ocean current for ships sailing at sea is different from the wind direction. It has been verified that under the influence of the easterly current, the ship moves eastward. Similarly, under the influence of the southerly current, the ship moves southward. This is consistent with maritime practice and proves the influence of water current on ship movement.

[0080] Shallow water effect refers to the increase in the mass of a ship when it sails in waters with a depth less than three times the draft of the ship, which has a serious impact on the movement of the ship. A shallow water model is constructed, and the calculation formula of the shallow water model is:

[0081] ,

[0082] in, and are the additional masses in the x and y directions in shallow water, , are the additional masses in the x and y directions in deep water, is the additional moment of inertia in shallow water, is the additional moment of inertia in deep water, For water depth, For the ship's draft, is the ship width; turning radius verification at water depths of 30 meters and 20 meters was carried out to verify the shallow water model. The ship's draft is 14.5m. The results show that the shallower the water depth, the larger the turning radius, highlighting the significant impact of shallow water on the maneuverability of the ship, indicating that the ship's turning ability is seriously affected by the water depth.

[0083] Furthermore, in order to accurately simulate the motion of the ship, a ship motion model considering environmental interference is constructed based on the wind interference model, the water flow interference model and the shallow water model. The ship motion model considering environmental interference is a three-degree-of-freedom MMG model. The calculation formula of the ship motion model considering environmental interference is:

[0084] ,

[0085] in, For the quality of the ship, , are the additional masses of the ship in the x-axis and y-axis directions, , , , are the speed and acceleration of the ship in the x-axis and y-axis directions, , are the angular velocity and angular acceleration of the ship on the x-axis and y-axis respectively, , are the ship's moment of inertia and additional moment of inertia, , , are the forces on the bare hull, propeller and rudder in the x-axis direction, , , are the forces on the bare hull, propeller and rudder in the y-axis direction, is the force acting on the ship, , , The bare hull, oars and rudder are respectively. , , External disturbance of wind.

[0086] The relative speed between the own ship and the target ship is determined based on the digital twin navigation environment model and the ship motion model considering environmental interference. The absolute collision interval ACC is determined based on the ship motion model considering environmental interference, the adaptive PID control method, the speed obstacle algorithm and the relative speed. The calculation formula of the absolute collision interval is:

[0087] ,

[0088] in, is the absolute collision interval, is the relative collision interval, Target ship The speed of time, is the Minkowski vector and operator;

[0089] Based on the absolute collision range and the collision avoidance rules, the feasible speed range FVO is determined, that is, the feasible speed range for the ship to avoid collision. The calculation formula of the feasible speed range is:

[0090] ,

[0091] in, For this ship The speed of time, is the feasible speed range for own ship to avoid collision, To avoid collisions and maintain good seamanship, is the union operator;

[0092] During the ship maneuvering process, the ship's clock is usually used to control the speed, so that the feasible maneuvering range including all decision-making processes of ship collision avoidance can be determined. The ship's clock is obtained, and the ship's feasible maneuvering range is determined based on the feasible speed range and the clock. The calculation formula of the ship's feasible maneuvering range is:

[0093] ,

[0094] in, is the feasible maneuvering range of the ship, For the ship's bell, , which respectively represent full speed forward, half speed forward, slow speed forward, and slow speed forward.

[0095] For the schematic diagram of its adaptive PID control method, please refer to Figure 3 ,like Figure 3 As shown, the input of the adaptive PID control method includes the heading deviation and the angular velocity error , that is, the heading deviation and the heading deviation change rate in the system. The output is the PID parameter , , According to the PID control principle, the parameters , , The impact on the output characteristics of the control system is mainly manifested in: parameters The impact on the system response speed is mainly determined by the deviation between the target heading and the actual heading; It mainly affects the steady-state characteristics of the system and is determined by the cumulative deviation between the target heading and the actual heading; It mainly affects the dynamic characteristics of the system, characterizes the speed of the ship approaching the target route, adaptively adjusts the PID parameters through the optimal parameter control algorithm, and realizes the adaptive optimal control of the heading according to the required PID parameters.

[0096] In some embodiments, in step S102, a ship dynamic collision risk model is constructed based on the feasible maneuvering range of the ship. First, a dangerous maneuvering set and a feasible maneuvering set of the ship are determined based on the feasible maneuvering range of the ship. From the perspective of the captain and the driver, the larger the feasible operation set for avoiding collision, the lower the urgency. Therefore, the ship collision risk can be characterized by the ratio of the current dangerous maneuvering set that may cause a collision to the feasible maneuvering set. Secondly, based on the dangerous maneuvering set and the feasible maneuvering set, a ship dynamic collision risk model is constructed according to a geometric model, and the geometric model calculation formula is:

[0097] ,

[0098] in, is the dynamic collision risk of ships, A set of dangerous and feasible maneuvers for ships, is the set of all feasible maneuvers for the ship, that is, the number of all available executable sets. Considering the maneuverability of the ship, the greater the required turning or speed change, the more challenging the maneuver. Therefore, the ship collision risk is corrected to obtain the ship dynamic collision risk model. The calculation formula of the ship dynamic collision risk model is:

[0099] ,

[0100] in, is the dynamic collision risk of ships, For ship maneuvering decisions, is the weight corresponding to the ship maneuvering decision, The range of values ​​for the heading of a ship’s dangerous maneuvers.

[0101] A multi-ship collision avoidance model is constructed based on the ship dynamic collision risk model combined with the improved speed barrier algorithm. For a schematic diagram of the speed barrier algorithm, please refer to Figure 4 ,The traditional speed barrier algorithm VO, such as Figure 4 As shown in (a), it is generally applicable to ships sailing in open waters. However, in coastal waters, its applicability may be affected due to the maneuvering behavior of multiple target ships. The improved speed barrier algorithm IVO, such as Figure 4 As shown in (b), the operating system does not predict the action of the target ship TS at time t because there is no immediate risk of collision. However, in reality, when the target ship TS changes its course, a collision risk will arise. If the two ships are very close at that time, a collision accident is likely to occur, such as Figure 4As shown in (b), the reason why the ship OS misjudged the situation at time t is that the target ship TS changed its course. The ship introduces a virtual target ship to represent the position of the target ship at time t. The calculation formula is:

[0102] ,

[0103] in, is the position of the fictitious target ship TS, for The ship's heading at the time, for The initial position of the ship at time for Click the ship's position.

[0104] The decision-making process of multi-ship encounter collision avoidance is based on the dynamic collision risk index (CRI) of ships and the feasible maneuvering range of each target ship. When the CRI of the target ship is greater than 0.4, collision avoidance actions such as changing course and speed should be taken. When the CRI of multiple ships exceeds 0.4, the feasible maneuvering range is determined by the multi-ship encounter collision avoidance model as the intersection of the feasible maneuvering ranges of each ship. The calculation formula of the multi-ship encounter collision avoidance model is:

[0105] ,

[0106] in, is the feasible maneuvering range for multiple ships, for The feasible maneuvering range of the ship at the point for The risk of ship collision at the point In order to consider the influence of shallow water, is the Minkowski vector and operator; for the schematic diagram of collision avoidance in multi-ship encounter, please refer to Figure 5 ,like Figure 5 As shown in the figure, at time t0, there is a collision risk between the own ship and the target ship, and the collision risk exceeds 0.4. In response, a single-ship collision avoidance action is taken. At time t1, the collision risk between the own ship and the second target ship and the third target ship remains above 0.4. The feasible maneuvering interval is determined by the multi-ship encounter and collision avoidance model, which represents the intersection of the feasible intervals of each target ship. This ensures that the own ship takes into account the risks brought by multiple target ships and takes appropriate autonomous navigation decision-making plans.

[0107] In some embodiments, in step S103, a ship track tracking model and a ship state estimation model are constructed, and an autonomous navigation decision scheme considering time series rolling is constructed based on the ship dynamic collision risk model, the ship track tracking model, the ship state estimation model and the multi-ship encounter collision avoidance model. Through the autonomous navigation decision scheme, the safe guidance of the ship from the starting point to the destination is achieved. However, errors from various sources will affect the navigation process, including imperfect ship mathematical models, environmental interference and imperfect collision avoidance decisions. In order to reduce these errors, a timed rolling algorithm is introduced to continuously compensate for errors by updating the ship state at fixed time intervals. By incorporating real-time information and making corresponding adjustments, the navigation performance of the ship can be improved and more reliable autonomous navigation can be achieved;

[0108] Construct a ship track tracking model. During the voyage, the ship may deviate from the scheduled route due to various factors such as wind, waves, currents, and collision avoidance. When it is determined that the ship has deviated from the scheduled route, it needs to return to the scheduled route. For a schematic diagram of the ship position deviation, please refer to Figure 6 ,like Figure 6 As shown, for example, the ship position is given by Determined, the deviation of the ship's position is determined by the turning point and Determine that the vertical distance from the ship to the two turning points at time t is calculated as:

[0109] ,

[0110] in, for The vertical distance from own ship to two turning points at this moment, os the ship to the next turning point The distance For own ship to the last turning point The distance It is the vertical distance from own ship to the two turning points.

[0111] The line of sight (LOS) algorithm is used to guide the ship to maintain a straight track segment. The line of sight (LOS) algorithm is a tracking method. When the ship has a navigation deviation, the line of sight (LOS) algorithm is used to correct the ship's navigation. The calculation formula for the target heading is:

[0112] ,

[0113] in, For the target heading, Turning point and The angle between is the intersection angle between the target point and the planned route, is the ship's heading, and the calculation formula for the intersection angle between the target point and the planned route is:

[0114] ,

[0115] in, is the intersection angle between the target point and the planned route, is the radius of the circle for the ship to resume navigation, for The vertical distance from own ship to the two turning points at this moment.

[0116] Construct a ship state estimation model. In coastal waters, there are two types of ships according to their positions relative to the waterway: ships along the waterway and ships not along the waterway. Different ship motion prediction models are applied according to the classification of ship motion. To ensure effective ship avoidance, a 20-minute time is set to predict feasible maneuvering paths based on the navigation characteristics of nearshore waters. When a ship is sailing on a waterway, its trajectory is closely related to the general direction of the waterway. Therefore, the position at the next moment can be estimated based on the current speed of the ship and the general direction of the traffic lane. To realize the state of a ship sailing along the waterway, first, calculate the time it takes for the ship to pass through each section. , and its calculation formula is:

[0117] ,

[0118] in, To reach the next turning point. For the The position of the ship at the turning point, is the ship position at the initial moment, For this ship The speed of time.

[0119] Secondly, by comparing the time it takes for the ship to pass through each leg, it is determined that the ship is at the waypoint and The channel between , assuming that the total heading of this section of the channel is ,Sure The ship position at the time is calculated as:

[0120] ,

[0121] in, for The position of the ship at the moment, For this ship speed at the moment; when the ship is not sailing along the channel, The calculation formula for the ship's position at the moment is:

[0122] ,

[0123] in, for The position of the ship at the moment, The current heading of the ship.

[0124] In some embodiments, in step S104, an adaptive PID control model is constructed based on an adaptive PID control method, and the ship is controlled to execute an autonomous navigation decision plan based on the adaptive PID control model; multi-source heterogeneous information is obtained, and the multi-source heterogeneous information is integrated into the digital twin navigation environment model, and the motion state of the target ship is obtained based on the digital twin navigation environment model; in the initial stage, the multi-source heterogeneous information is integrated into the digital twin environment model to realize the digitization of the traffic environment, so as to fully understand the navigation environment. The initial information is obtained from key equipment such as GPS, AIS and other related sources. This information can be used as a basis for establishing an environmental model and providing data about multiple target ships to obtain the motion state of the target ship; based on the digital twin navigation environment model, the multi-source heterogeneous information is integrated into the digital twin environment model to realize the digitization of the traffic environment, so as to fully understand the navigation environment. The twin navigation environment model, the ship motion model considering environmental interference, the adaptive PID control method, and the ship state estimation model are used to obtain the motion state of the own ship. Based on the motion state of the own ship and the motion state of the target ship, combined with the ship dynamic collision risk model, it is determined whether there is a collision risk. When there is a collision risk, the feasible maneuvering range of the own ship is determined by the multi-ship encounter collision avoidance model, and the autonomous collision avoidance decision is executed based on the autonomous navigation decision-making scheme. The error or deviation in the autonomous collision avoidance decision is dynamically corrected using the time rolling model to obtain the corrected navigation path, and the own ship is controlled to navigate based on the corrected navigation path. When there is no collision risk, the own ship is controlled to navigate based on the navigation path determined by the ship track tracking model.

[0125] In order to realize dynamic adaptive ship autonomous navigation decision, firstly, in the initial stage, multi-source heterogeneous information is integrated into the digital twin navigation environment model to realize the digitization of the traffic environment; it can fully understand the navigation environment; secondly, in the decision-making, the IVO algorithm and trajectory tracking model are combined to obtain autonomous navigation decisions, allowing the ship to determine the best path and make wise decisions to avoid collisions and navigate safely. Once the decision is made, the ship enters the execution step, executes the navigation decision according to the determined trajectory, and controls the movement and action of the ship accordingly. Finally, after dynamically correcting the error or deviation in the ship's decision using the time rolling model, the ship is controlled to navigate according to the corrected navigation path, ensuring that the ship can adapt and adjust its navigation strategy in real time. Taking into account the changes in the offshore environment and potential interference, the time rolling model continuously updates and refines the ship's decision to realize dynamic adaptive ship autonomous navigation decision. The whole process is executed in the order of information integration, decision-making, execution, and dynamic correction in the ship's autonomous navigation.

[0126] When a collision risk is detected, the multi-ship collision avoidance model is executed to generate an avoidance strategy. When there is no collision risk, the ship track tracking model continues to guide the ship along its planned trajectory. During the whole process, the time rolling model is used to continuously update and improve the ship's autonomous collision avoidance decision. In the process of updating the information according to any changes in the environment and the status of the ship, it should be noted that The smaller the value, the more frequent the update, in order to reduce errors in the decision-making process, choose 1 second The value matches the GPS update frequency, ensuring timely control of the ship.

[0127] In order to verify the effectiveness of the improved VO algorithm in nearshore waters, the AIS historical data was applied to the autonomous navigation platform and set as the target ship to better reflect the real environment. The results show that in coastal waters, the autonomous navigation decision-making scheme considering time series rolling can enable the ship to sail safely to the destination, while effectively avoiding collisions with other target ships and staying on the predetermined route. The autonomous navigation decision-making scheme has good adaptability to complex scenarios involving multiple ships and is easy to implement.

[0128] In order to better implement the autonomous navigation decision-making method based on the digital twin navigation environment in the embodiment of the present invention, on the basis of the autonomous navigation decision-making method based on the digital twin navigation environment, correspondingly, Figure 7 As shown, the embodiment of the present invention further provides an autonomous navigation decision-making device based on a digital twin navigation environment. The autonomous navigation decision-making device 700 based on a digital twin navigation environment includes:

[0129] The ship feasible maneuvering range module 701 is used to determine the ship feasible maneuvering range based on the constructed digital twin navigation environment model, the ship motion model considering environmental interference, the adaptive PID control method and the speed obstacle algorithm;

[0130] The multi-ship collision avoidance module 702 is used to construct a multi-ship collision avoidance model based on the feasible maneuvering interval of the ship, and then construct a dynamic collision risk model of the ship in combination with the improved speed barrier algorithm;

[0131] The autonomous navigation decision module 703 is used to construct an autonomous navigation decision plan based on the ship dynamic collision risk model, the constructed ship track tracking model and the ship state estimation model, and the multi-ship encounter collision avoidance model;

[0132] The autonomous navigation execution control module 704 is used to construct an adaptive PID control model based on the adaptive PID control method, and control the ship to execute the autonomous navigation decision plan based on the adaptive PID control model.

[0133] like Figure 8As shown, the autonomous navigation decision method based on the digital twin navigation environment, the present invention also provides an electronic device 800 accordingly, the electronic device 800 can be a computing device such as a mobile terminal, a desktop computer, a notebook, a palm computer and a server. The electronic device 800 includes a processor 801, a memory 802 and a display 803. Figure 8 Only some components of the electronic device 800 are shown, but it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0134] In some embodiments, the memory 802 may be an internal storage unit of the electronic device 800, such as a hard disk or memory of the electronic device 800. In other embodiments, the memory 802 may also be an external storage device of the electronic device 800, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device 800. Further, the memory 802 may also include both an internal storage unit of the electronic device 800 and an external storage device. The memory 802 is used to store application software and various types of data installed in the electronic device 800, such as program codes installed in the electronic device 800. The memory 802 may also be used to temporarily store data that has been output or is to be output. In one embodiment, an autonomous navigation decision program based on a digital twin navigation environment is stored on the memory 802, and the autonomous navigation decision program based on a digital twin navigation environment can be executed by the processor 801, thereby realizing the autonomous navigation decision method based on a digital twin navigation environment of each embodiment of the present invention.

[0135] In some embodiments, the processor 801 can be a central processing unit (CPU), a microprocessor or other data processing chip, used to run the program code or process data stored in the memory 802, such as an autonomous navigation decision-making method based on a digital twin navigation environment.

[0136] In some embodiments, the display 803 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, etc. The display 803 is used to display identification information of the autonomous navigation decision-making program based on the digital twin navigation environment and to display a visualized user interface. The components 801-803 of the electronic device 800 communicate with each other through a system bus.

[0137] In some embodiments, when the processor 801 executes the autonomous navigation decision-making program based on the digital twin navigation environment in the memory 802, the various steps of the autonomous navigation decision-making method based on the digital twin navigation environment as described in the above embodiments are implemented. Since the autonomous navigation decision-making method based on the digital twin navigation environment has been described in detail above, it will not be repeated here.

[0138] In summary, the autonomous navigation decision-making method, device and medium based on the digital twin navigation environment provided by the present invention determine the feasible maneuvering range of the ship based on the constructed digital twin navigation environment model, the ship motion model considering environmental interference, the adaptive PID control method and the speed barrier algorithm; after constructing the ship dynamic collision risk model based on the ship feasible maneuvering range, a multi-ship encounter and collision avoidance model is constructed in combination with the improved speed barrier algorithm; an autonomous navigation decision-making scheme is constructed based on the ship dynamic collision risk model, the constructed ship track tracking model and the ship state estimation model and the multi-ship encounter and collision avoidance model; an adaptive PID control model is constructed based on the adaptive PID control method, and the ship is controlled to execute the autonomous navigation decision-making scheme based on the adaptive PID control model, so as to realize adaptive autonomous navigation decision-making of ships in nearshore waters and improve the accuracy of autonomous navigation decision-making.

[0139] Those skilled in the art will appreciate that all or part of the processes of the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, wherein the computer-readable storage medium is a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0140] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. An autonomous navigation decision-making method based on digital twin navigation environment, It is characterized in that include: The feasible maneuvering range of the ship is determined based on the constructed digital twin navigation environment model, the ship motion model considering environmental interference, the adaptive PID control method and the speed obstacle algorithm, wherein the feasible maneuvering range of the ship is determined based on the constructed digital twin navigation environment model, the ship motion model considering environmental interference, the adaptive PID control method and the speed obstacle algorithm, including: Build digital twin navigation environment models for static and dynamic environments based on electronic nautical charts; Considering the influence of wind interference, water flow interference and shallow water effect on ship motion, a ship motion model considering environmental interference is constructed, wherein the calculation formula of the ship motion model considering environmental interference is: Where, m is the mass of the ship, m x , m y are the additional masses of the ship in the x-axis and y-axis directions, u, v, are the speed and acceleration of the ship in the x-axis and y-axis directions, r, are the angular velocity and angular acceleration of the ship on the x-axis and y-axis respectively, I zz , J zz are the ship’s moment of inertia and additional moment of inertia, X H , X P , X R are the forces and moments of the hull, propeller and rudder in the transverse direction, Y H , Y P , Y R are the forces and moments of the hull, propeller and rudder in the longitudinal direction, N is the force on the ship, H, P, R are the forces of the hull, propeller and rudder, respectively, X w , Y w is the external disturbance of wind, N w is the wind force and moment in the yaw direction, N H is the force and moment of the hull in the bow rolling direction, N R is the force and moment of the rudder in the heading direction, N P are the propeller force and moment in the bow pitch direction; Determine the relative speed between the own ship and the target ship based on the digital twin navigation environment model and the ship motion model considering environmental interference; Determine the absolute collision interval based on the ship motion model considering environmental interference, the adaptive PID control method, the speed obstacle algorithm and the relative speed; Determine a feasible speed range based on the absolute collision range in combination with the collision avoidance rule; The ship's clock is obtained, and a feasible maneuvering range of the ship is determined based on the feasible speed range and the clock, wherein the calculation formula of the feasible maneuvering range of the ship is: Among them, ACC is the absolute collision interval, RCC is the relative collision interval, is the speed of the target ship at time t, is the Minkowski vector and operator, is the speed of the ship at time t, FVO is the feasible speed range of the ship to avoid collision, RL is the speed range not allowed by the collision avoidance rules, ∪ is the union operator, MVO is the feasible maneuvering range of the ship, T i is the i-th ship's clock, i is the number of clocks; After constructing the ship dynamic collision risk model based on the ship feasible maneuvering interval, a multi-ship encounter collision avoidance model is constructed in combination with an improved speed barrier algorithm, wherein the ship dynamic collision risk model is constructed based on the ship feasible maneuvering interval, including: Determining a dangerous maneuver set and a feasible maneuver set of the ship based on the feasible maneuvering range of the ship; Based on the dangerous maneuver set and the feasible maneuver set, a ship dynamic collision risk model is constructed according to a geometric model, wherein the calculation formula of the ship dynamic collision risk model is: Among them, CRI is the dynamic collision risk of the ship, n(DT) is the number of dangerous maneuver sets of the ship, and n(AT) is the number of all feasible maneuver sets. For ship maneuvering decisions, is the weight corresponding to the ship maneuvering decision, k is the range of the ship's dangerous maneuvering heading, i is the number of clocks, and j is the steering degree; The calculation formula of the multi-ship encounter collision avoidance model is: Among them, MVO is the feasible maneuvering range of multiple ships, MVO k is the feasible maneuvering range of the ship at point k, CRI k is the ship collision risk at point k, ξ is the factor considering shallow water, is the Minkowski vector and operator, U i＝1,2,...,n is the union of n feasible maneuvering intervals of multiple ships, n is the number of feasible maneuvering intervals; An autonomous navigation decision-making scheme is constructed based on the ship dynamic collision risk model, the constructed ship track tracking model and the ship state estimation model, and the multi-ship encounter and collision avoidance model; An adaptive PID control model is constructed based on the adaptive PID control method, and the ship is controlled to execute an autonomous navigation decision-making scheme based on the adaptive PID control model, wherein the ship is controlled to execute an autonomous navigation decision-making scheme based on the adaptive PID control model, including: Acquire multi-source heterogeneous information, integrate the multi-source heterogeneous information into the digital twin navigation environment model, and obtain the motion state of the target ship based on the digital twin navigation environment model; Obtaining the motion state of the ship based on the digital twin navigation environment model, the ship motion model considering environmental interference, the adaptive PID control method and the ship state estimation model; Based on the motion state of the own ship and the motion state of the target ship, combined with the ship dynamic collision risk model, determine whether there is a collision risk. When there is a collision risk, perform an autonomous collision avoidance decision based on the autonomous navigation decision-making plan, and use the time rolling model to dynamically correct the error or deviation in the autonomous collision avoidance decision to obtain a corrected navigation path, and control the own ship to navigate based on the corrected navigation path. When there is no collision risk, control the own ship to navigate based on the navigation path determined by the ship track tracking model.

2. An electronic device, It is characterized in that including memory and processor; The memory stores a computer-readable program executable by the processor; When the processor executes the computer-readable program, it implements the steps in the autonomous navigation decision-making method based on the digital twin navigation environment as described in claim 1.

3. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps in the autonomous navigation decision-making method based on the digital twin navigation environment as described in claim 1.

Images