A ship dynamic autonomous navigation method and device based on domain type gradient field

Abstract

This invention discloses a method for dynamic autonomous navigation of ships based on domain-type gradient fields, comprising the following steps: establishing a domain-type gradient field decreasing model for ship encounter scenarios; establishing a ship navigation motion constraint model; after establishing the ship navigation motion constraint model, establishing corresponding domain-type gradient field models for different ship encounter scenarios; transforming the route design problem for different ship encounter scenarios into a constrained nonlinear optimization problem and solving it to achieve local route design and reach the target point. During ship operation, this invention utilizes an intelligent autonomous device on board to collect information about the ship itself, other ships, and obstacles. By judging the positions and states of all parties during navigation, it provides effective motion decisions to ensure ship safety.

Description

Technical Field

[0001] This invention belongs to the field of waterway transportation, specifically relating to a method and device for dynamic autonomous navigation of ships based on a domain gradient field. Background Technology

[0002] With the development of the shipping industry, ship density is constantly increasing, waterways are becoming increasingly congested, and the risk of ship collisions and other accidents is rising sharply. Ship accidents can cause enormous losses to life, property, and the environment. In recent years, with the rapid development of technologies such as artificial intelligence, big data, advanced sensors, and BeiDou satellite navigation, autonomous surface vessels have become an inevitable trend in the intelligent development of the shipping and shipbuilding industries. Due to the nonlinear, high-hysteresis, and large-inertia dynamic characteristics of ships, as well as the need to comply with maritime traffic rules, achieving autonomous and safe navigation of ships in complex encounter scenarios is challenging. Currently, most ship motion path design methods consider relatively simple encounters and do not take into account time attributes, making it difficult to generate safe navigation routes when multiple ships encounter each other. Summary of the Invention

[0003] The purpose of this invention is to provide a method and device for dynamic autonomous navigation of ships based on domain gradient fields. During the operation of a ship, an intelligent autonomous device is used on board to collect information about the ship itself, other ships, and obstacles. By judging the position and status of each party during the navigation process, effective motion decisions are made to ensure the safety of the ship.

[0004] To solve the above-mentioned technical problems, the technical solution of the present invention is: a method for autonomous ship dynamic navigation based on a domain-type gradient field, applied to a device for autonomous ship dynamic navigation based on a domain-type gradient field, the device comprising an information flow transmission unit, a data processing unit, a design output unit, and an effect evaluation unit; wherein,

[0005] The information flow transmission unit transmits the position, size, displacement, and time information of other units captured by the ship's sensing system to the data processing unit, where the data storage unit analyzes and processes the information.

[0006] The data processing unit receives various information and data streams from the information flow transmission unit, and uses the grid method to perform map modeling, establishes the field strength influence range of each unit, and processes various information data of ships and obstacles on this basis; the data processing unit also includes a storage medium, which contains an algorithm program for solving the optimal route.

[0007] The design output unit sends the optimal route, solved by the algorithm program in the storage medium, to the ship control terminal to guide the ship to travel along the optimal route;

[0008] The effect evaluation unit compares and analyzes the actual forward path of the ship with the optimal path obtained by the solution, evaluates the design effect of the optimal path, and provides basic data for the improvement and optimization of the device.

[0009] This method includes the following steps:

[0010] Establish a domain gradient field decreasing model for ship encounter scenarios;

[0011] Establish a constraint model for ship navigation motion;

[0012] After establishing the ship navigation motion constraint model, domain gradient field models are established for different ship encounter scenarios.

[0013] The route design problem for different ship encounter scenarios is transformed into a constrained nonlinear optimization problem and solved to achieve local route design for ships and reach the target point.

[0014] The specific steps for establishing a domain gradient field decreasing model for a ship encounter scenario are as follows:

[0015] When a ship encounters a stationary object, the ship establishes a detection area centered on its own position, and calculates the field strength by establishing a static domain gradient field with the size of the detection area as the radius of the field strength influence range.

[0016] When a ship encounters a moving object, the field of the moving object itself is considered, and a dynamic domain gradient field is established.

[0017] A decreasing gradient field model is formed by combining static and dynamic domain gradient fields.

[0018] The method for calculating the field strength in a static domain gradient field is as follows:

[0019]

[0020]

[0021]

[0022]

[0023] In the formula, It is the distance between the ship and a stationary object. Adjustable parameters representing different static risks. A constant representing the detection region of a static object. Pmin is the radius of a stationary unit, and Pmin is the minimum intensity value. and These refer to the ship and the distance from the starting point to the destination, respectively.

[0024] Combining the Coldwell ship domain and the quaternary ship domain, we obtain a dynamic domain gradient field, expressed as:

[0025]

[0026] In the formula, QSD represents the quaternary ship domain, x is the x-axis direction of the coordinate system established with the ship's center as the origin, and y is the y-axis direction. Represents the boundary. Captain.

[0027] The specific method for establishing the ship's navigation motion constraint model is as follows: it is established based on the first-order linear Nomoto model, expressed as:

[0028]

[0029] In the formula, These are all ship maneuverability coefficients, representing the ship's turning index and following index, respectively. The bow roll rate is angular velocity. For rudder angle.

[0030] Ship encounter scenarios include at least the following: ship-to-ship encounter, ship-crossing encounter, and ship-overtaking encounter. Among them, ship-to-ship encounter scenarios include port-side encounter, ship-to-direction encounter, and ship-to-starboard encounter. Ship-overtaking encounter scenarios include ① the scenario where the ship is overtaken, the ship maintains its original course and speed, and the other ship turns to avoid a collision, and ② the scenario where the ship overtakes the other ship and the ship needs to turn to give way.

[0031] The route design problem for different ship encounter scenarios is transformed into a constrained nonlinear optimization problem, and a model prediction strategy is introduced when solving it.

[0032] The algorithm principle for transforming the route design problem for different ship encounter scenarios into a constrained nonlinear optimization problem and solving it is as follows: In the first... At any given moment, based on the ship's own status information and predicted environmental information, the prediction step size is determined. The minimum cost motion sequence of the ship is obtained internally, and the first step in the predicted minimum cost motion sequence is taken as the ( ) The output motion command at that moment.

[0033] The constrained nonlinear optimization problem to be solved must include at least the following:

[0034] Ship motion constraints:

[0035] Ship rudder angle constraints:

[0036] Ship bow angle constraints:

[0037] Constraints on the ship's rudder angle to bow angular velocity:

[0038] Objective function: Describing the ship in the Time, Predicting Step Size Minimize the cumulative inner expectation;

[0039] in, For the first The bow direction at that moment, For the first The ship's position at any given time. For wind and current interference to ship navigation, It is a unit time interval for simulating ship motion. To discretize the step size, Represents the number of discretization steps. For the first The speed of the bow forward at any given moment. This is the preset minimum value for the ship's rudder angle constraint. This is the preset maximum value of the ship's rudder angle constraint. As the rudder angle, For the first The rudder of time, For the first The bow angular velocity at time [time]. The bow angular velocity, Here, is the nonlinear coefficient, and F is the predicted minimum field strength. The magnitude of the field strength at each sampling time i.

[0040] A ship dynamic autonomous navigation device based on a domain gradient field is also provided, comprising an information flow transmission unit, a data processing unit, a design output unit, and an effect evaluation unit; wherein,

[0041] The information flow transmission unit transmits the position, size, displacement, and time information of other units captured by the ship's sensing system to the data processing unit, where the data storage unit analyzes and processes the information.

[0042] The data processing unit receives various information and data streams from the information flow transmission unit, and uses the grid method to perform map modeling, establishes the field strength influence range of each unit, and processes various information data of ships and obstacles on this basis; the data processing unit also includes a storage medium, which contains an algorithm program for solving the optimal route.

[0043] The design output unit sends the optimal route, solved by the algorithm program in the storage medium, to the ship control terminal to guide the ship to travel along the optimal route;

[0044] The effect evaluation unit compares and analyzes the actual forward path of the ship with the optimal path obtained by the solution, evaluates the design effect of the optimal path, and provides basic data for the improvement and optimization of the device.

[0045] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention establishes the relationship between the field of a stationary unit and other ships during navigation, as well as the gravitational field of a point, and establishes a ship motion model considering specified constraints. It proposes a gradient field descent method that integrates model prediction strategies to solve the optimal motion command sequence within the prediction step in real time during ship operation, and uses the first step of the obtained sequence as the motion output at the next moment. This ensures that the ship moves along the point of minimum field strength while avoiding obstacles, thus solving the problem of ship encounter route design and ensuring the ship safely arrives at its destination. Attached Figure Description

[0046] Figure 1 This is a flowchart illustrating the method of an embodiment of the present invention;

[0047] Figure 2 This is a schematic diagram of the structure of the device according to an embodiment of the present invention;

[0048] Figure 3 This is a diagram showing the positional relationship of all parties during a "starboard approach encounter" in an embodiment of the present invention.

[0049] Figure 4 This is a diagram showing the positional relationship of the parties during a "cross-encounter" in an embodiment of the present invention;

[0050] Figure 5 This is a diagram showing the positional relationship of each party during the "overtaking" process in an embodiment of the present invention. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0052] The technical problem this invention aims to solve is to implement a ship dynamic autonomous maneuvering device and method based on a decreasing field. During ship operation, an intelligent autonomous device on board collects information about the ship itself, other ships, and obstacles. By judging the positions and states of all parties involved in the navigation process, it provides effective motion decisions to ensure ship safety. To achieve the above objective, the technical solution adopted by this invention is: a ship dynamic autonomous navigation device and method based on a decreasing field strength in a specific region.

[0053] include:

[0054] A ship dynamic autonomous navigation device based on domain-type gradient fields is proposed, specifically as follows:

[0055] To address the stationary elements such as bridge piers and buoys, as well as moving objects like other vessels, that may exist in the navigation environment, and considering the safety of the vessel itself, a device for autonomous navigation route planning is proposed, integrating an information flow transmission unit, a data processing unit, a design output unit, and an effect evaluation unit. The device includes:

[0056] The information flow transmission unit transmits the position, size, displacement, and time information of other units captured by the ship's sensing system to the data processing unit, where the data storage unit analyzes and processes the information.

[0057] The data processing unit receives various information and data streams from the information flow transmission unit, and uses a grid method to perform map modeling, establishing the field strength influence range for each unit. Based on this, it processes various information data about ships and obstacles. The data processing unit also includes a storage medium containing the program for solving the optimal route.

[0058] The design output unit sends the optimal route, solved by the algorithm program in the storage medium, to the ship control terminal to guide the ship to navigate in the safest and most energy-efficient manner.

[0059] The effect evaluation unit evaluates the effectiveness of the device's route design by comparing and analyzing the actual route with the pre-designed route, providing basic data for device improvement and optimization.

[0060] A method for autonomous ship navigation based on domain-type gradient fields is proposed, specifically as follows:

[0061] Step 1: Establish a domain field reduction model for ship encounter scenarios

[0062] 1.1 During the ship's forward movement, it will encounter various stationary objects. A circular field strength influence range can be established with the object's own position as the center and the size of the detection area as the radius. The field strength calculation method is as follows:

[0063]

[0064] It is the distance between the ship and a stationary object. Adjustable parameters representing different static risks. A constant representing the detection region of a static object. It is the radius of a stationary unit.

[0065] The electric field strength is greatest at the center and least at the edges. Ships can be considered safe when outside the potential field's influence. However, the static obstacles encountered by ships during actual navigation vary, and their collision risk impacts differ. Therefore, the following method is used for calculation:

[0066] Pmin is the minimum strength value.

[0067] The endpoint itself also has a field influence range, which is calculated as follows:

[0068]

[0069]

[0070] in and These are the distances from the ship itself and from the starting point to the destination, respectively.

[0071] 1.2 During the operation of a ship, it may encounter moving obstacles. These objects exist in a field that other ships cannot enter, and this field can be further expanded into a "domain-type" decreasing field.

[0072] The four-dimensional ship domain model offers a more concise description. The Coldwell ship domain can be approximated as a variant of the four-dimensional ship domain. Combining the Coldwell and four-dimensional ship domains yields a new ship domain model that reduces the impact of ship-to-ship effects and meets ship safety requirements. The new ship domain is as follows:

[0073] ,in Captain.

[0074] In summary, a field model was established where the encounter time between ships decreases with distance.

[0075] Step 2: Establish a ship motion constraint model

[0076] After establishing the field model for ship encounters, the constraints during ship navigation must be considered to ensure that the pre-set route is applicable to actual conditions. Internationally unified standards must be followed when the ship is moving forward. Considering the ship's three degrees of freedom motion (bow roll, drift, and forward movement), a constraint model was established using a first-order linear Nomoto model.

[0077]

[0078] in, This is the ship's maneuverability coefficient, representing the ship's turning index and following index. The bow roll rate is angular velocity. For rudder angle.

[0079] Step 3: After establishing the constraint model for ship encounter scenarios, different constraint methods are proposed for different ship encounter modes.

[0080] 3.1 For encounter scenarios, a route design method based on hypothetical ships is proposed.

[0081] Ship encounters are generally classified into three types: encounters with a vessel approaching from the port side, encounters with a vessel directly ahead, and encounters with a vessel approaching from the starboard side. For these encounter situations, the rule stipulates that both vessels should turn to starboard and pass each other on the port side. The purpose of these ship movement constraints is to prevent vessels from turning to port during encounters, thereby reducing the danger of such encounters.

[0082] When other ships appear to the port side and directly ahead of the ship, the optimal trajectory route design can be achieved by finding the point with the minimum field value and heading there, based on the improved ship field and the established distance reduction field.

[0083] However, if a vessel appears on the starboard side of the ship, continuing with the previous decreasing field model strategy would actually increase the danger encountered. Therefore, a method for designing the ship's navigation route based on a hypothetical vessel is proposed when another vessel appears on the starboard side of the ship.

[0084] The positional relationship between this vessel and the encountering vessels and hypothetical vessels is as follows: Figure 1 As shown, the specific coordinate relationships are as follows:

[0085]

[0086] in The coordinates of the other ship's position in the geodetic coordinate system. Let these be the coordinates of the hypothetical ship's position in the Earth's coordinate system. These are the coordinates of the ship's position in the geodetic coordinate system.

[0087] When the ship is turning to right , Imagine the ship's position coordinates are infinitely close to those of the real ship. By safely navigating the simulated ship, we can ensure that there is no possibility of an encounter between the ship and the real ship, thus guaranteeing safety.

[0088] 3.2 For scenarios involving intersections and encounters, a driving route design method based on virtual locations is proposed.

[0089] Regarding cross-ship encounters, the following provisions apply: When a vessel approaches from the port side, the vessel is not required to give way except in cases of emergency; when a vessel approaches from the starboard side, the other vessel has priority, and the vessel must steer around it and cannot pass in front of it. Based on the "chasing method" proposed by some ship operators for cross-ship encounters on the starboard side, a method is proposed that involves chasing the fictitious position behind the other vessel and then proceeding to the end. The positional relationships of the aforementioned parties are as follows: Figure 3 As shown. The specific positional relationships are as follows:

[0090]

[0091] in Let the coordinates of the other ship be the coordinates of its position in the geodetic coordinate system. These are the position coordinates of the hypothetical region in the geodetic coordinate system. To indicate the course of other ships, In the quaternary domain of a ship, the radius of the stern section is defined as follows:

[0092] The specific procedure of this method is as follows: when a vessel approaches from the starboard side, an imaginary, non-existent area is created at the boundary of the vessel's territory directly behind the other vessel. The vessel then turns its course in this area. Once the other vessel has passed by the bow and is heading towards the port side, the vessel turns its course again in the opposite direction to return to its original planned course. This method not only complies with maritime traffic regulations but also generates a safer route of travel.

[0093] 3.3 For overtaking scenarios, a detour method based on fictitious location points is proposed.

[0094] According to regulations, in good visibility conditions, any vessel overtaking another vessel must give way. This can be further divided into two situations:

[0095] 1. If this vessel is overtaken, it shall maintain its original course and speed and continue sailing while the other vessel makes a turn to avoid an encounter.

[0096] 2. If this vessel overtakes the other vessel, this vessel must turn to give way.

[0097] If this vessel is at a relative rudder angle of 112.5 to another vessel. 。 Up to 247.5 。 If the vehicle is within the range and its speed is greater, a left or right turn is required to avoid it.

[0098] When this vessel is overtaking, the other vessel does not need to perform any operations and only needs to maintain its original course and speed. However, this vessel needs to turn to bypass and overtake. Under the premise that the two vessels are in the same territory during the "overtaking", this vessel can adopt a detour strategy of turning to the left or right.

[0099] This paper proposes to generate an imaginary point at a safe distance to the right rear of the overtaken vessel (sufficient safe distance means that the vessel can safely overtake even after turning around after reaching the imaginary point). This point exerts an attractive force on the vessel. The resultant force of this attractive force, the repulsive force of the other vessel on the vessel, and the attractive force of the actual terminal point points to the upper right. This resultant force pulls the vessel gradually toward the imaginary point. After the vessel reaches the point, it turns around and moves forward until the overtaking is completed, and then proceeds toward the destination.

[0100] The relative positions of all parties are shown in the attached figure. Figure 4 As shown, the specific coordinate positions are as follows:

[0101]

[0102] in Let the coordinates of the other ship be the coordinates of its position in the geodetic coordinate system. The coordinates of the fictional point in the geodetic coordinate system. d4 is the angle between the reverse extension of the other ship's course and the imaginary point, and d4 is the distance from that point to the other ship.

[0103] Step 4: Transform the different encounter problems of ships into constrained nonlinear optimization problems and solve them.

[0104] Based on steps one through three, a decreasing field model and a ship motion constraint model for ship encounters have been established, and the basic modeling is complete. For different ship encounter scenarios, different field and force influence range models are proposed. The traditional method uses a decreasing field strength approach to solve such encounter problems. However, this traditional method can lead to delays in avoidance when considering ship kinematic constraints, while model prediction strategies can effectively address this issue.

[0105] The principle of this algorithm is based on the first... At any given moment, based on the ship's own status information and predicted environmental information, the prediction step size is determined. The minimum cost motion sequence of the ship is obtained internally. The first step of the predicted minimum cost motion sequence is then used as the... The next step is to transform the problem of pre-generating the ship's navigation path under different conditions into a constrained nonlinear optimization problem.

[0106] Ship motion constraints:

[0107] Ship rudder angle constraints:

[0108] Ship bow angle constraints:

[0109] Constraints on the ship's rudder angle to bow angular velocity:

[0110] Objective function: Describing the ship in the Time, Predicting Step Size The cumulative internal expectation is minimized.

[0111] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for autonomous ship navigation based on domain-type gradient fields, characterized in that, Includes the following steps: Establish a domain gradient field decreasing model for ship encounter scenarios; Establish a constraint model for ship navigation motion; After establishing the ship navigation motion constraint model, domain gradient field models are established for different ship encounter scenarios. The route design problem for different ship encounter scenarios is transformed into a constrained nonlinear optimization problem and solved to achieve local route design for ships and reach the target point; The specific steps for establishing a domain gradient field decreasing model for ship encounter scenarios are as follows: When a ship encounters a stationary object, the ship establishes a detection area centered on its own position, and calculates the field strength by establishing a static domain gradient field with the size of the detection area as the radius of the field strength influence range. When a ship encounters a moving object, the field of the moving object itself is considered, and a dynamic domain gradient field is established. Among them, the method for calculating the field strength of the static domain gradient field, which combines the static domain gradient field and the dynamic domain gradient field to form a domain gradient field decreasing model, is as follows: In the formula It is the electric field strength of a static obstacle. It is the field strength at the target point. It is the distance between the ship and a stationary object. Adjustable parameters representing different static risks. A constant representing the detection region of a static object. Pmin is the radius per unit at rest, and Pmin is the minimum intensity value. and These are the ship's location and the distance from the starting point to the destination, respectively. Combining the Coldwell ship domain and the quaternary ship domain, we obtain a dynamic domain gradient field, expressed as: In the formula, QSD represents the four-dimensional ship domain, x is the x-axis direction of the coordinate system established with the ship's center as the origin, and y is the y-axis direction of the coordinate system established with the ship's center as the origin. Represents the boundary. Captain; The constrained nonlinear optimization problem to be solved must include at least the following: Ship motion constraints: Ship rudder angle constraints: Ship bow angle constraints: Constraints on the ratio of rudder angle to bow angular velocity: Objective function: Describing the ship in the Time, Predicting Step Size Minimize the cumulative inner expectation; in, For the first The bow direction at that moment, For the first The ship's position at any given time. For wind and current interference to ship navigation, It is a unit time interval for simulating ship motion. To discretize the step size, Represents the number of discretization steps. For the first The speed of the bow forward at any given moment. This is the preset minimum value for the ship's rudder angle constraint. This is the preset maximum value of the ship's rudder angle constraint. As the rudder angle, For the first The rudder of time, For the first The bow angular velocity at time [time]. The bow angular velocity, Here, is the nonlinear coefficient, and F is the predicted minimum field strength. The magnitude of the field strength at each sampling time i.

2. The method for autonomous ship navigation based on a domain gradient field according to claim 1, characterized in that, The specific method for establishing the ship's navigation motion constraint model is as follows: it is established based on the first-order linear Nomoto model, expressed as: In the formula, These are all ship maneuverability coefficients, representing the ship's turning index and following index, respectively. The bow roll angular velocity, For rudder angle.

3. The method for autonomous ship navigation based on a domain-type gradient field according to claim 1, characterized in that, Ship encounter scenarios include at least the following: ship-to-ship encounter, ship-crossing encounter, and ship-overtaking encounter. Among them, ship-to-ship encounter scenarios include port-side encounter, ship-to-direction encounter, and ship-to-starboard encounter. Ship-overtaking encounter scenarios include ① the scenario where the ship is overtaken, the ship maintains its original course and speed, and the other ship turns to avoid a collision, and ② the scenario where the ship overtakes the other ship and the ship needs to turn to give way.

4. The method for autonomous ship navigation based on a domain-type gradient field according to claim 1, characterized in that, The route design problem for different ship encounter scenarios is transformed into a constrained nonlinear optimization problem, and a model prediction strategy is introduced when solving it.

5. The method for autonomous ship navigation based on a domain gradient field according to claim 4, characterized in that, The algorithm principle for transforming the route design problem for different ship encounter scenarios into a constrained nonlinear optimization problem and solving it is as follows: In the first... At any given moment, based on the ship's own status information and predicted environmental information, the prediction step size is determined. The minimum cost motion sequence of the ship is obtained internally, and the first step in the predicted minimum cost motion sequence is taken as the ( ) The output motion command at that moment.

6. An apparatus for using a method for autonomous ship navigation based on a domain gradient field as described in claim 1, characterized in that, It includes an information flow transmission unit, a data processing unit, a design output unit, and an effect evaluation unit; among which, The information flow transmission unit transmits the position, size, displacement, and time information of other units captured by the ship's sensing system to the data processing unit, where the data storage unit analyzes and processes the information. The data processing unit receives various information and data streams from the information flow transmission unit, and uses the grid method to perform map modeling, establishes the field strength influence range of each unit, and processes various information data of ships and obstacles on this basis; the data processing unit also includes a storage medium, which contains an algorithm program for solving the optimal route. The design output unit sends the optimal route, solved by the algorithm program in the storage medium, to the ship control terminal to guide the ship to travel along the optimal route; The effect evaluation unit compares and analyzes the actual forward path of the ship with the optimal path obtained by the solution, evaluates the design effect of the optimal path, and provides basic data for the improvement and optimization of the device.

Images