Automatic control method for ship running

Abstract

The invention discloses an automatic control method for ship running. The method comprises a distance monitoring method and / or a fleet marshalling control method in an automatic berth state. When it is detected that the ship enters the berthing state, the offshore distance of the ship in the future time period is predicted through a distance monitoring method; and when it is detected that the ship is in the fleet marshalling navigation state, the driving state of the ship is controlled through a fleet marshalling control method. The method has the advantages that the berthing state of the ship can be monitored, the distance between the ship and the shoreside in a period of time in the future can be predicted, the course and the speed can be conveniently adjusted according to prediction, and therefore the requirement for the berthing state of the ship under automatic or manual control is met.

Description

Technical Field

[0001] The present invention relates to the field of ship travel control, and in particular to an automatic monitoring control system and method for ship travel. Background Art

[0002] As an important means of transportation, ships generally use drivers to manually control the course and speed to achieve ship driving. However, with the development of intelligent automation technology, autonomous driving has gradually been carried out in the field of ship driving, and has even entered the stage of loading and use. For example, the patent application number 201910994324.4 is an autonomous driving integrated system for unmanned ships, including: a perception module for sensing the ship's navigation environment and obtaining real-time waterway, hydrology, ship's own status, and traffic environment dynamic information; a communication system for transmitting data and instructions between the ship and the shore, and between system modules; a data processing module for processing the information obtained by the perception module; a decision module for identifying the current operating status and environment of the ship based on the data output by the data processing module, selecting the next action to be taken, and generating the corresponding operation instructions; an execution module for receiving the operation instructions issued by the decision module and using a PID controller to control the ship's propeller and rudder to change the ship's motion state. This autonomous driving integrated system for unmanned ships integrates functions such as environmental perception, behavioral decision-making, ship control, and remote control, which can greatly liberate manpower and ensure the safe navigation of ships.

[0003] This type of automatic ship control in the existing technology only involves the control of the ship's normal navigation phase. However, when the ship reaches its destination and enters the mooring state for parking control, as well as in some special circumstances, such as when the ship is traveling in formation, this type of automatic driving technology cannot meet the control requirements of the formation travel and the monitoring and judgment requirements of the mooring process when in the mooring state. Summary of the invention

[0004] The purpose of the present invention is to overcome the deficiencies of the prior art and provide a method for automatically controlling ship travel to meet the control requirements of convoy travel and the monitoring and judgment requirements of the berthing process when in a berthing state.

[0005] In order to achieve the above-mentioned purpose, the technical solution adopted by the present invention is: a method for automatic control of ship driving, which includes a distance monitoring method and / or a fleet formation control method in an automatic berthing state; when it is detected that the ship enters a berthing state, the distance from the shore of the ship in a future time period is predicted by the distance monitoring method; when it is detected that the ship is in a fleet formation sailing state, the driving state of the ship is controlled by the fleet formation control method.

[0006] After the ship starts to enter the mooring state, the ship's water flow speed, real-time offshore distance, heading, speed, and ship draft data are obtained in real time. Based on the ship's water flow speed, real-time offshore distance, heading, speed, and ship draft data, the ship's offshore distance in the future is predicted and displayed on the ship's large screen.

[0007] A neural network model is pre-built and trained, wherein the neural network model is a multiple linear regression model that takes water velocity, real-time offshore distance, heading, speed, and ship draft data as input and predicts offshore distance as output.

[0008] When mooring, the distance of the ship from the shore in the future is adjusted by adjusting the ship's heading and speed, so as to obtain a heading and speed that satisfies the distance less than the set value within the set time, so as to control the ship to complete the mooring within the set time.

[0009] In formation mode, identify and obtain the position information of the preceding ship that the own ship needs to follow, control the own ship to travel within the following ship's distance range, and then adjust the own ship's attitude so that the own ship and the preceding ship are consistent in driving direction and the axis from the bow to the stern of the two ships are consistent.

[0010] In formation mode, the ship uploads the navigation plan to the cloud platform. The cloud platform forms a formation for all the ships received in formation mode, and divides the ships whose routes overlap in the navigation plan and whose positions are within the set range into a group. The position and tonnage data of all ships in a group are obtained, and the sequence relationship of the front and rear ships in the group is divided accordingly and sent to each ship in the group. After each ship receives the position and heading information of the ship in front of it, it controls the ship to sail within the distance range of the ship in front, and then controls the ship to sail in the formation following manner.

[0011] The ships are sorted according to their positions and tonnage data, and the ships are preliminarily sorted based on their positions and heading directions. Then, the two adjacent ships in front and behind are sorted and adjusted according to their tonnage to obtain the fleet sorting relationship within the formation.

[0012] Each ship in the formation is equipped with a radar sensor to obtain the distance between the ship and the preceding ship. When the distance is less than the safe distance or the real-time distance is greater than the following distance threshold, an early warning is issued and the speed is adjusted to meet the distance requirement.

[0013] The ships in the formation are connected through a network. The acceleration, deceleration and turning status data of the front ship are sent to the rear ship in the formation through the network. After receiving the deceleration of the front ship, the rear ship in the formation performs corresponding acceleration, deceleration and turning according to the acceleration and deceleration parameters of the front ship.

[0014] An alignment module is provided at the bow or stern of the ship. During the driving process, the front and rear ships use the alignment module to detect in real time whether the following ships of the front and rear ships meet the requirements of keeping the driving direction consistent and the axis from the bow to the stern of the two ships are consistent; if not, the attitude of the ship is adjusted until it meets the requirements.

[0015] The advantages of the present invention are that the berthing status of the ship can be monitored and the distance to the shore in the future can be predicted, so that the course and speed can be adjusted according to the prediction, thereby meeting the requirements of the berthing status of the ship under automatic or manual control. At the same time, during long-distance navigation, an automatic team heading control method is provided, which can meet the special mode: team navigation of ships in the formation mode to reduce navigation resistance, improve navigation efficiency and save navigation energy. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The following is a brief description of the contents expressed in the drawings of the present invention and the symbols in the drawings:

[0017] Figure 1 Schematic diagram of two modes included in the ship automatic control method of the present invention;

[0018] Figure 2 This is a schematic diagram of the principle of predicting the offshore distance according to the present invention. DETAILED DESCRIPTION

[0019] The specific implementation of the present invention will be further explained in detail below by describing the optimal embodiment with reference to the accompanying drawings.

[0020] The present application provides a method for automatically controlling ship driving, including two modes: Mode 1: berthing mode; Mode 2: fleet sailing mode, each mode has a corresponding ship monitoring and processing strategy under the mode, specifically a distance monitoring method and a fleet formation control method under the automatic berthing state; when it is detected that the ship enters the berthing state, the distance from the shore of the ship in the future time period is predicted by the distance monitoring method; when it is detected that the ship is in the fleet formation sailing state, the driving state of the ship is controlled by the fleet formation control method.

[0021] Parking mode:

[0022] When the ship needs to enter the mooring mode, it enters the mooring mode through the user's manual trigger button or automatic identification. After the ship begins to enter the mooring state, the ship's water flow speed, real-time offshore distance, heading, speed, and ship draft data are obtained in real time. The ship's offshore distance in the future is predicted based on the ship's water flow speed, real-time offshore distance, heading, speed, and ship draft data, and displayed on the ship's large screen. The display can provide users with basic reference data when the ship is moored.

[0023] A neural network model is built and trained in advance. The neural network model is a multivariate linear regression model that takes water flow velocity, real-time offshore distance, heading, speed, and ship draft data as input and predicts the offshore distance as output. When mooring, the ship's offshore distance in the future is adjusted by adjusting the ship's heading and speed, so as to obtain a heading and speed that satisfies the distance less than the set value within the set time, so as to control the ship to complete the mooring within the set time. If the ship needs to complete the mooring within the set time, the predicted offshore distance after reaching the mooring time in the current state is still large, and the mooring demand cannot be met. At this time, the heading and sailing speed are adjusted manually or automatically until the predicted output offshore distance is less than the distance threshold within the set time (the mooring state requires the offshore distance to be close to 0).

[0024] When entering the mooring mode, the weight data and draft data of the current ship are obtained through the ship database, the water flow speed is obtained in real time through the water flow sensor, the offshore distance is obtained from the radar in real time, and the video information is obtained from the camera; the draft data, water flow data, offshore distance, heading, and speed data are input into the large model, which can predict the offshore distance in the future, so as to adjust the speed and heading accordingly and achieve automatic berthing.

[0025] The big model is a pre-built AI big model. The principle of the AI ​​big model is as follows: The AI ​​big model is based on deep learning, which uses a large amount of data and computing resources to train a neural network model. We collect data on the entire process of ship berthing, including the weight data of the ship, instantaneous data such as draft data, water flow data, distance from the shore, heading, and speed;

[0026] The first step is to complete the data cleaning and preprocessing, and delete missing values ​​and abnormal data.

[0027] The second step is data conversion: normalize the cleaned data to eliminate the dimension and range differences between different features.

[0028] The third step is modeling, using a linear regression model. The linear regression model is a basic prediction model that is suitable for establishing a linear relationship between independent variables and dependent variables. It can be used to predict continuous numerical results through data and linear equations. The data we collected has the following characteristics: ship weight data, draft data, water flow rate, offshore distance, heading, and speed. A multivariate linear regression model is constructed as follows: y = b + w1x1 + w2x2 + w3x3 + w4x4 + w5x5 + w6x6, where y represents the predicted offshore distance, b represents the deviation, wi represents the weight of the i-th feature, and xj represents the j-th feature value; the value range of i and j is 1-6. Specifically: w1 is the ship weight weight, x1 is the ship weight, w2 is the draft data weight, x2 is the draft data, w3 is the water flow weight, x3 is the water flow, w4 is the offshore distance weight, x4 is the offshore distance, w5 is the heading weight, x5 is the heading data, w6 is the speed weight, and x6 is the speed data.

[0029] The fourth step is to use the converted data to train the above model, and train the correct parameter information based on a large amount of existing data. The fifth step: Based on the trained model and real-time data, the offshore distance in the future can be predicted based on the input parameters.

[0030] The above solution can achieve the following: By combining radar, camera, flow sensor, ship model data and AI big model, the ship's distance from the shore in the future can be predicted. By combining radar, camera, flow sensor, ship model data and AI big model, the arrival time at the berth can be predicted. By combining radar, camera, flow sensor, ship model data and AI big model, the ship's navigation trajectory can be predicted.

[0031] In formation mode:

[0032] After the driver manually triggers the formation mode, in the formation mode, the position information of the preceding ship that the ship needs to follow is identified and obtained, and the ship is controlled to travel within the distance range of the preceding ship and the following ship, and then the attitude of the ship is adjusted so that the ship and the preceding ship are consistent in the driving direction and the axis from the bow to the stern of the two ships is consistent. An alignment module is set at the bow or stern of the ship. During the driving process, the leading and trailing ships use the alignment module to detect in real time whether the following ships of the leading and trailing ships meet the requirements of keeping the driving direction consistent and the axis from the bow to the stern of the two ships consistent; if not, the attitude of the ship is adjusted until it is satisfied. The alignment mode can be laser aligned. Each ship is equipped with an infrared laser alignment system. A laser transmitter is set at the bow position of each ship, and a laser receiver is set at the stern position. The position set for each ship is the central axis and the emitted laser overlaps with the central axis. The central axis of the ship refers to the axis that divides the ship into two parts, left and right, along the direction from the bow to the stern. Laser transmitters or laser receivers are set at the positions of the axis and the bow or stern, respectively. During the marshaling, if the front ship receives the laser signal from the rear ship or the rear ship receives the laser signal from the front ship, it means that the two ships are sailing in alignment, otherwise it means that the two ships are not sailing in alignment. If the laser transmitter is set at the bow, the receiver is set at the stern, otherwise the receiver is set at the bow and the transmitter is set at the stern. It is required that the laser transmitter and receiver of each ship in the marshaling are set in the same way. In this way, the ships can be aligned during the following ship sailing. Why do we need to align the central axis? Because the main purpose of marshaling sailing is to reduce resistance, such as wind resistance or sailing in the wake area of ​​the front ship to save energy. If it cannot be aligned, even if the ships are sailing forward and backward, the purpose of saving energy cannot be achieved.

[0033] In the formation mode, the ship uploads the ship's navigation plan to the cloud platform. The cloud platform forms a formation for all the ships received in the formation mode, and divides the ships whose routes overlap in the navigation plan and whose positions are within the set range into a group. The position and tonnage data of all ships in a group are obtained, and the order relationship between the front and rear ships in the group is divided accordingly and sent to each ship in the group. After each ship receives the position and heading information of the ship in front of it, it controls the ship to sail within the distance range of the ship in front, and then controls the ship to sail in the formation following the ship. The cloud platform is a data processing software platform that can perform interactive processing based on the acquired data. For multiple ships within a set distance range, if they have the same sailing route and direction, they will be grouped into one group. Then, according to the order of the grouping, the ships will communicate with each other and obtain the position information of the preceding ship corresponding to the present ship in the order, and then sail towards the position of the preceding ship until it reaches the following ship distance range. Then, the following ship's posture is adjusted so that the following ship posture meets the alignment requirements, and the alignment module detects that the preceding and following ships are in an aligned state. In this way, all ships are in a front-to-back aligned state, which can complete the navigation control of the group and meet the purpose of energy saving.

[0034] As for the sorting and division of the cloud platform, the ships are sorted according to their location and tonnage data. The ships are initially sorted based on their location and heading direction, and then the order of the two adjacent ships in front and behind is adjusted according to the tonnage to obtain the fleet sorting relationship within the formation. In the direction of travel, the ship in front is ranked first, so that the rear ship only needs to approach the front ship for formation.

[0035] Each ship in the formation is equipped with a radar sensor to obtain the distance between the ship and the preceding ship. When the distance is less than the safety distance or the real-time distance is greater than the threshold of the following distance, an early warning is issued and the speed is adjusted to meet the distance requirement, thereby avoiding the safety risks caused by too close a distance when following a ship and the defect of being unable to reduce resistance due to too large a distance. A threshold of the following distance is set, and real-time detection is performed and the acceleration and deceleration of the ship are dynamically controlled based on the detection results to keep the distance constant.

[0036] The ships in the formation are connected through a network. The acceleration, deceleration and turning status data of the front ship are sent to the rear ship in the formation through the network. After receiving the deceleration of the front ship, the rear ship in the formation performs corresponding acceleration, deceleration and turning according to the acceleration and deceleration parameters of the front ship, thereby achieving consistency in navigation between the front and rear ships and avoiding the risk of formation navigation caused by the front ship slowing down and the rear ship not responding in time.

[0037] The main component of the marshaling navigation in this scheme is the shipboard terminal, which includes a display screen, annunciator, processor, controller, and input device of the working computer system. Taking the two ships A and B as an example, the principle is introduced as follows:

[0038] 1. Ship A sets the destination on the shipboard terminal, and nearby ships can learn about the navigation plan of ship A through the signaler.

[0039] 2. Ship B learns that Ship A has the same destination as it during navigation, and operates the autonomous formation on the shipboard terminal.

[0040] The processor receives the command and initializes the formation structure, determining that ship A is the leading ship and ship B is the following ship. Ships A and B establish communication and synchronize navigation / positioning information. Ship B sends instructions to the controller to control the speed and heading of the ship and follow behind ship A. Ship B automatically follows ship A and the autonomous formation is completed.

[0041] 3. Other ships can form a fleet and follow A and B in turn.

[0042] 4. When ship A turns, accelerates, decelerates, etc., it notifies other following ships through the signaler, and the following ships take corresponding actions in turn to achieve autonomous formation and follow the action in sequence.

[0043] Marshalling navigation control can improve the fuel economy of ships. The following ship is in the wake area of ​​the front ship during sailing, which can greatly reduce the air resistance of the following ship and reduce the comprehensive fuel consumption by 7%-14%. The captain is more relaxed. The ship's autonomous formation system takes on most of the work instead of the captain, reducing the captain's work intensity and avoiding the possibility of fatigue navigation accidents to a large extent. It has higher safety. During navigation, the information between ships is interconnected, the operation status of the ship can be monitored at any time, and the corresponding actions can be completed in a very short time. The reaction time is much faster than that of humans, which greatly improves the safety of navigation.

[0044] Obviously, the specific implementation of the present invention is not limited to the above-mentioned methods. As long as various non-substantial improvements are made using the method concept and technical solution of the present invention, they are all within the protection scope of the present invention.

Claims

1. A method for automatic control of ship travel, characterized in that: The method includes a distance monitoring method and a fleet formation control method in an automatic berthing state; when it is detected that a ship enters a berthing state, the distance from the shore of the ship in a future time period is predicted by the distance monitoring method; when it is detected that the ship is in a fleet formation sailing state, the sailing state of the ship is controlled by the fleet formation control method.

2. A method for automatic control of ship travel according to claim 1, characterized in that: After the ship starts to enter the mooring state, the ship's water flow speed, real-time offshore distance, heading, speed, and ship draft data are obtained in real time. Based on the ship's water flow speed, real-time offshore distance, heading, speed, and ship draft data, the ship's offshore distance in the future is predicted and displayed on the ship's large screen.

3. A method for automatic control of ship travel as claimed in claim 2, characterized in that: A neural network model is pre-built and trained, wherein the neural network model is a multiple linear regression model that takes water velocity, real-time offshore distance, heading, speed, and ship draft data as input and predicts offshore distance as output.

4. A method for automatically controlling ship travel according to any one of claims 1 to 3, characterized in that: When mooring, the distance of the ship from the shore in the future is adjusted by adjusting the ship's heading and speed, so as to obtain a heading and speed that satisfies the distance less than the set value within the set time, so as to control the ship to complete the mooring within the set time.

5. A method for automatically controlling ship travel according to any one of claims 1 to 3, characterized in that: In formation mode, identify and obtain the position information of the preceding ship that the own ship needs to follow, control the own ship to travel within the following ship's distance range, and then adjust the own ship's attitude so that the own ship and the preceding ship are consistent in driving direction and the axis from the bow to the stern of the two ships are consistent.

6. A method for automatic control of ship travel as claimed in claim 5, characterized in that: In formation mode, the ship uploads the navigation plan to the cloud platform. The cloud platform forms a formation for all the ships received in formation mode, and divides the ships whose routes overlap in the navigation plan and whose positions are within the set range into a group. The position and tonnage data of all ships in a group are obtained, and the sequence relationship of the front and rear ships in the group is divided accordingly and sent to each ship in the group. After each ship receives the position and heading information of the ship in front of it, it controls the ship to sail within the distance range of the ship in front, and then controls the ship to sail in the formation following manner.

7. A method for automatic control of ship travel as claimed in claim 6, characterized in that: The ships are sorted according to their positions and tonnage data, and the ships are preliminarily sorted based on their positions and heading directions. Then, the two adjacent ships in front and behind are sorted and adjusted according to their tonnage to obtain the fleet sorting relationship within the formation.

8. A method for automatic control of ship travel according to any one of claims 5 to 7, characterized in that: Each ship in the formation is equipped with a radar sensor to obtain the distance between the ship and the preceding ship. When the distance is less than the safe distance or the real-time distance is greater than the following distance threshold, an early warning is issued and the speed is adjusted to meet the distance requirement.

9. A method for automatically controlling ship travel according to any one of claims 5 to 7, characterized in that: The ships in the formation are connected through a network. The acceleration, deceleration and turning status data of the front ship are sent to the rear ship in the formation through the network. After receiving the deceleration of the front ship, the rear ship in the formation performs corresponding acceleration, deceleration and turning according to the acceleration and deceleration parameters of the front ship.

10. A method for automatically controlling ship travel according to any one of claims 5 to 7, characterized in that: An alignment module is provided at the bow or stern of the ship. During the driving process, the front and rear ships use the alignment module to detect in real time whether the following ships of the front and rear ships meet the requirements of keeping the driving direction consistent and the axis from the bow to the stern of the two ships are consistent; if not, the attitude of the ship is adjusted until it meets the requirements.

Images