Self-adaptive unmanned navigation control system for ship

Abstract

The invention discloses a self-adaptive unmanned navigation control system for a ship, which belongs to the technical field of ship navigation control and comprises a multivariate information sensing unit, a comprehensive feature recognition unit, an automatic route planning unit, a data management unit and a navigation information deduction unit. The multivariate information sensing unit collects environmental information in the ship navigation process and sends the environmental information to the data management unit, and the data management unit is in communication connection with all other units. The comprehensive feature recognition unit performs classification, motion feature recognition and contour feature recognition on the target in the environment information to obtain a motion feature and a contour feature of the target; the navigation information deduction unit deduces the motion track of the moving target based on the environment information and the motion features and the contour features of the target classified as the moving target; the automatic track planning unit plans and controls the track of the ship based on the movement track of the moving target; and the problem of potential safety hazards in the face of complex water areas or severe weather conditions is greatly avoided.

Description

Technical Field

[0001] The invention relates to an adaptive unmanned ship navigation control system, belonging to the technical field of ship navigation control. Background Art

[0002] When traditional ships are sailing, the crew can only rely on low-level and limited information about the working status of the ship's equipment and environmental perception information, and make decisions and operations on driving behavior completely manually to avoid ship targets or islands and reefs on the route, which poses a serious risk to navigation safety. In addition, with the continuous increase in application scenarios, ships are not only required to be able to navigate in conventional waters and sea areas, but also to be able to perform navigation tasks in unknown waters and waters with complex environments. The mission waters have a harsh environment and are no longer suitable for crew members to accompany them, so the demand for unmanned navigation control of ships is becoming more and more urgent.

[0003] Faced with the urgent need for unmanned navigation control technology for ships, relevant institutions have also carried out research on unmanned navigation control for ships. However, currently only a single data source is used, such as navigation radar information, optoelectronic equipment information, etc. as input data for unmanned navigation control. This situation is more effective for simple navigation environments. Once it involves complex waters or severe weather conditions, there will be great safety hazards. Summary of the invention

[0004] The purpose of the present invention is to provide an adaptive unmanned navigation control system for ships, so as to solve the potential safety hazard problem existing in the prior art when sailing in complex / harsh environments.

[0005] To achieve the above objectives, the present invention is implemented by adopting the following technical solutions: The present invention provides an adaptive unmanned navigation control system for ships, comprising: a multi-information perception unit, a comprehensive feature recognition unit, an automatic track planning unit, a data management unit and a navigation information deduction unit; The multi-information sensing unit collects environmental information of the ship's navigation process and sends it to the data management unit, and the data management unit is connected to all other units in communication; The comprehensive feature recognition unit classifies the target in the environmental information, recognizes the motion features and the contour features, and obtains the motion features and the contour features of the target; The navigation information deduction unit deduces the motion trajectory of the moving target based on the environmental information and the motion characteristics and contour characteristics of the target classified as the moving target; The automatic track planning unit plans and controls the track of the ship based on the motion track of the moving target.

[0006] Furthermore, the environmental information includes targets around the ship and their identity information, navigation information, and optoelectronic information, meteorological information, and hydrological information of the environment around the ship.

[0007] Furthermore, the identity information of the target is collected by the AIS device in the multi-information sensing unit, the navigation information of the target is collected by the navigation radar in the multi-information sensing unit, the photoelectric information is collected by the camera in the multi-information sensing unit, the meteorological information is collected by the meteorological instrument in the multi-information sensing unit, and the hydrological information of the water area is collected by the hydrological monitoring device in the multi-information sensing unit.

[0008] Further, the classifying, motion feature recognition and contour feature recognition of the target in the environmental information includes classifying the target in the environmental information, and the classifying of the target in the environmental information includes: Firstly, the targets are divided into moving targets and fixed targets according to their current motion state. Then, the moving targets are divided into ship targets, active moving targets and floating moving targets. The fixed targets are divided into fixed targets and reference targets. The classification criteria of the moving targets include: For targets with identity information, they are classified as ship targets; For targets without identity information, whether the target has autonomous navigation capability is determined based on the target's movement speed, the target's movement direction, the current water flow speed, and the current water flow direction obtained from the data management unit. For targets with autonomous navigation capability, they are classified as active motion targets, and for targets without autonomous navigation capability, they are classified as floating motion targets; The classification criteria for the fixed targets include: For targets whose closed outlines can be identified by navigation radar, they are classified as fixed targets; Targets whose partial outlines can only be identified by the navigation radar are classified as reference targets.

[0009] Further, the classifying, identifying motion features and contour features of the targets in the environmental information includes identifying motion features of the targets in the environmental information, and the identifying motion features of the targets in the environmental information includes: Real-time positioning of the target is achieved through target tracking based on deep learning, so as to obtain the motion characteristics of the target, and the real-time position of the target is stored in the data management unit.

[0010] Further, the classifying, identifying motion features and contour features of the targets in the environmental information includes contour feature recognition of the targets in the environmental information, and the contour feature recognition of the targets in the environmental information includes: For a ship target, extracting the contour information of the ship target according to the navigation radar echo data corresponding to the ship target in the navigation information obtained from the data management unit, and generating a rectangular envelope of the ship target accordingly; For an active moving target / floating moving target / fixed target, extracting contour information of the active moving target / floating moving target / fixed target according to navigation radar echo data corresponding to the active moving target / floating moving target / fixed target in the navigation information obtained from the data management unit, and generating a polygonal envelope of the active moving target / floating moving target / fixed target accordingly; For the reference target, the contour information of the reference target is extracted according to the navigation radar echo data corresponding to the reference target in the navigation information obtained from the data management unit, and the curve characteristics of the adjacent ship are generated accordingly.

[0011] Furthermore, the step of deducing the motion trajectory of the moving target from the motion features and contour features of the target classified as a moving target based on the environmental information includes: For ship targets and actively moving targets, curve fitting is performed on the longitude and latitude coordinate points of the target's motion process to obtain the target's motion mathematical expression, and the target's subsequent position coordinates are predicted based on the motion mathematical expression, thereby obtaining the target's motion trajectory; For floating moving targets, the target's moving direction and speed are directly fitted according to the real-time update information of the target's position coordinates to obtain the target's moving trajectory.

[0012] Furthermore, the planning and controlling of the ship's track based on the motion trajectory of the moving target includes: Taking the line segment from the current position of the ship to the terminal position as the center line, three obstacle points in the vicinity of the center line are identified by the navigation radar, wherein the three obstacle points include the point where the active moving target is located and the point where the floating moving target is located; The three obstacle points are connected into two line segments; The midpoints of the two line segments formed by the current position of the ship and the three obstacle points are connected to each other to obtain two track segments, and the shorter track segment is selected as the final track segment; The end point of the track segment is taken as the new current position of the ship, and all the above steps are repeated for polling until the current position of the ship coincides with the end point position. At this time, the planned track composed of all track segments is a broken line segment. The corners of the broken line segments in the planned track are transitioned into corner arcs to generate a smooth planned track, and the ship is controlled according to the smooth planned track.

[0013] Furthermore, in the process of controlling the ship according to the smooth planned track, the process further includes: Determine whether the planned trajectory and the moving trajectory of the moving target intersect; If there is an intersection, the intersection time interval of the intersection point is calculated according to the navigation information of the planned track and the navigation information of the moving target, and the minimum intersection distance of the intersection point is calculated according to the speed and heading information of the ship. If the minimum intersection distance is less than the safety distance, the ship's track is replanned until the minimum intersection distance is not less than the safety distance; If there is no intersection, the shortest distance between the planned track of the ship and the moving track of the moving target is calculated, and the minimum distance between the ship and the moving target during navigation is calculated based on the shortest distance. If the minimum distance is less than the safety distance, the track of the ship is replanned until the minimum distance is not less than the safety distance.

[0014] Furthermore, if the information of a certain target collected by the camera in the environmental information is not included in the information collected by the navigation radar, the automatic track planning unit controls the ship to perform an emergency avoidance action.

[0015] Compared with the prior art, the beneficial effects achieved by the present invention are: The present invention provides an adaptive unmanned navigation control system for ships. It collects target information around the ship through multi-information perception, deduces motion information of targets around the ship based on the collected multi-information, and automatically plans tracks and controls the navigation of the ship according to the deduction results, thereby greatly avoiding potential safety hazards in complex waters or severe weather conditions.

[0016] The multi-information sensing equipment is used to collect target information and environmental information around the ship, and the reference information obtained is more comprehensive, which plays a vital role in the subsequent feature analysis and extraction, track planning, navigation control, etc. Targets are divided into moving targets and fixed targets according to their motion characteristics. Moving targets are further divided into ship targets, active moving targets and floating moving targets. Fixed targets are further divided into fixed targets and reference targets. The detailed division of targets is more conducive to formulating corresponding collision avoidance strategies for different targets, which is more efficient and safer. Autonomously plan the track based on the motion characteristics of the targets around the ship, and continuously iterate until the planned navigation path meets the requirements for safe navigation; In addition to using navigation radar and AIS to monitor the ship's surrounding environment, optoelectronic equipment such as cameras are also installed to monitor the ship's surrounding environment to avoid collision accidents caused by navigation radar's monitoring blind spots and target information loss, thereby improving the safety of unmanned navigation of ships. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1This is a work flow chart of a ship adaptive unmanned navigation control system provided by an embodiment of the present invention; Figure 2 is a schematic diagram of a navigation control process provided by an embodiment of the present invention; Figure 3 It is a schematic diagram of track planning provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and are not intended to limit the protection scope of the present invention.

[0019] Example 1.

[0020] The present invention provides an adaptive unmanned navigation control system for ships, comprising: a multi-information perception unit, a comprehensive feature recognition unit, an automatic track planning unit, a data management unit and a navigation information deduction unit; The multi-information sensing unit collects environmental information of the ship's navigation process and sends it to the data management unit, and the data management unit is connected to all other units in communication; The comprehensive feature recognition unit classifies the target in the environmental information, recognizes the motion features and the contour features, and obtains the motion features and the contour features of the target; The navigation information deduction unit deduces the motion trajectory of the moving target based on the environmental information and the motion characteristics and contour characteristics of the target classified as the moving target; The automatic track planning unit plans and controls the track of the ship based on the motion track of the moving target.

[0021] The above-mentioned ship adaptive unmanned navigation control system is based on Figure 1 The process shown is to carry out adaptive unmanned navigation control of the ship. Information on targets around the ship is collected through multi-information perception. Based on the collected multi-information, the motion information of targets around the ship is deduced. Track planning and ship navigation are automatically carried out according to the deduction results, which greatly avoids safety hazards when facing complex waters or severe weather conditions.

[0022] Example 2.

[0023] The present invention provides an adaptive unmanned navigation control system for ships, comprising: a multi-information perception unit, a comprehensive feature recognition unit, an automatic track planning unit, a data management unit and a navigation information deduction unit.

[0024] The multi-information perception unit mainly realizes the perception and collection of environmental information during the navigation process, provides analysis data for subsequent comprehensive feature recognition, automatic track planning, etc., and summarizes the collected information into environmental information.

[0025] Environmental information includes identity information of surrounding targets, navigation information, optoelectronic information of the surrounding environment, meteorological information, hydrological information of the water area, etc., among which identity information is obtained through AIS (Automatic Identification System) (this information is only for targets equipped with AIS equipment), navigation information is obtained through navigation radar, optoelectronic information is obtained by camera, meteorological information is obtained by meteorological instrument, and hydrological information is obtained by hydrological monitoring device. The environmental information around the ship is comprehensively obtained and stored in the database (i.e. data management unit) to facilitate the later processing and information mining of the obtained data. Other units obtain the corresponding environmental information from the data management unit.

[0026] The navigation radar monitors the surrounding target position information and radar echo characteristics; the meteorological sensing equipment collects information such as wind speed, wind direction, humidity, temperature, etc. around the ship; the hydrological information collection device collects information such as water depth, flow velocity, flow direction, etc. at the current position of the ship; the optoelectronic equipment ships detect the environment around the ship.

[0027] Set the surrounding environment data collection range, collect features of targets within the collection range, and store the collected information in the database. The feature collection of the same target will continue for 3 minutes from the time of discovery. If the device loses the target, the information stored in the database will be cleared. If a new target is found, the new target will be stored in the database and collection will continue.

[0028] The comprehensive feature recognition unit classifies the target in the environmental information, recognizes the motion features and the contour features, and obtains the motion features and contour features of the target. The specific process is as follows: The location information of the surrounding ships is obtained through the AIS equipment, and the information of the targets around the ship is obtained through the navigation radar, which includes moving targets and fixed targets. The moving targets include ships, floating objects on the water surface, etc., and the fixed targets include small obstacles on the water surface, embankments, large water surface buildings, etc. The moving ship targets are screened by combining the AIS information and the navigation radar information, and the AIS targets are eliminated from the navigation radar target information. For other moving targets, it is necessary to combine the object movement speed and direction fed back by the navigation radar and the current water flow speed and direction to determine whether they have autonomous navigation capabilities. If the object movement speed and direction captured by the navigation radar are linearly related to the water flow speed and direction, it is judged that the object movement is caused by the water flow, and it does not have autonomous navigation capabilities. If the captured object has no movement, it is judged that it does not have autonomous navigation capabilities. Otherwise, it is judged that it has autonomous navigation capabilities. Finally, the moving targets are divided into ship targets, active moving targets and floating moving targets, and the fixed targets are divided into fixed targets and reference targets according to their volume size.

[0029] The classification criteria for sports targets include: For targets with identity information, they are classified as ship targets; For targets that do not have identity information, whether the target has autonomous navigation capability is judged based on the target's movement speed, target's movement direction, current water flow speed and current water flow direction obtained from the data management unit. For targets that have autonomous navigation capability, they are classified as active motion targets, and for targets that do not have autonomous navigation capability, they are classified as floating motion targets.

[0030] The classification criteria for fixed targets include: For targets whose closed outlines can be identified by navigation radar, they are classified as fixed targets; Targets whose partial outlines can only be identified by the navigation radar are classified as reference targets.

[0031] The motion features of the classified moving targets around the ship are identified, and the target position is located in real time based on deep learning target tracking. The real-time position information of the moving target is collected and stored in the data management unit.

[0032] The contour features of the acquired classified targets are identified: the contour information of the ship target is extracted according to the navigation radar echo data to generate a rectangular envelope; for active moving targets and floating moving targets, their contour features are extracted to generate a polygonal envelope; for fixed targets, their contour features are extracted to generate a polygonal envelope; for reference targets, their contour edge features are extracted to generate curve features of nearby ships.

[0033] For ship targets and active moving targets, the heading and speed information of the target are calculated according to the position update (data obtained in the motion feature recognition step), and its navigation track is predicted and deduced; the specific method is: according to the updated target longitude and latitude coordinates, the obtained longitude and latitude coordinates are curve fitted, and an approximate mathematical expression is obtained, and the position coordinates behind the target can be calculated according to the mathematical expression of the track deduced and fitted. In this embodiment, its navigation track within 5 minutes is deduced.

[0034] For floating moving targets, their current speed and direction of movement are calculated, and their movement trend and speed are predicted and deduced in combination with meteorological information and hydrological information. The specific method is as follows: floating moving targets move under the action of wind and water flow, and their movement trajectory is generally a straight line trajectory. Therefore, according to their position update situation, the movement direction and speed are directly fitted and calculated to deduce and predict their movement trend. In this embodiment, the navigation trajectory within 5 minutes is deduced.

[0035] After meeting the time requirement for information collection, the discovered targets are classified according to the collected information. When the radar discovered target coincides with the AIS discovered target, it is determined to be a ship target. The speed and heading information of the ship target are calculated based on the collected position update information and the radar wave echo characteristics of the target. The navigation track of the ship target is generated based on the collected data. The navigation track of the ship in the next 5 minutes is simulated and deduced based on the calculated heading and speed information. The navigation track will be updated in real time based on the real-time collected data.

[0036] The automatic track planning unit plans the ship's track as follows: Figure 3 As shown: Taking the line segment from the current position of the ship to the terminal position as the center line, three obstacle points in the vicinity of the center line are identified by the navigation radar, wherein the three obstacle points include the point where the active moving target is located and the point where the floating moving target is located; The three obstacle points are connected into two line segments; The midpoints of the two line segments formed by the current position of the ship and the three obstacle points are connected to each other to obtain two track segments, and the shorter track segment is selected as the final track segment; The end point of the track segment is taken as the new current position of the ship, and all the above steps are repeated for polling until the current position of the ship coincides with the end point position. At this time, the planned track composed of all track segments is a broken line segment. The corners of the broken line segments in the planned track are transitioned into corner arcs to generate a smooth planned track, and the ship is controlled according to the smooth planned track.

[0037] The process of controlling the vessel according to a smooth planned trajectory also includes: Figure 2 The track control process shown: Determine whether the planned trajectory and the moving trajectory of the moving target intersect; If there is an intersection, the intersection time interval of the intersection point is calculated according to the navigation information of the planned track and the navigation information of the moving target, and the minimum intersection distance of the intersection point is calculated according to the speed and heading information of the ship. If the minimum intersection distance is less than the safety distance, the ship's track is replanned until the minimum intersection distance is not less than the safety distance; If there is no intersection, the shortest distance between the planned track of the ship and the moving track of the moving target is calculated, and the minimum distance between the ship and the moving target during navigation is calculated based on the shortest distance. If the minimum distance is less than the safety distance, the track of the ship is replanned until the minimum distance is not less than the safety distance.

[0038] In the process of controlling the ship, a kinematic model of the ship (a mathematical model implemented through existing technology) is established based on the ship's own characteristics, and the track and navigation data that meet the requirements after automatic planning are sent to the ship end. Based on its kinematic model, the ship controls the ship to navigate accurately according to the track in combination with the current environmental information and hydrological information.

[0039] Cameras and other optoelectronic equipment are installed around the ship to monitor the environment around the ship in real time. The camera is equipped with a target detection function. When a target is found, it is compared with the target information of the navigation radar. Once it is confirmed that the target found by the camera is not within the target range identified by the navigation radar, an emergency avoidance action will be performed to avoid collision accidents caused by blind spots in the navigation radar's monitoring or target loss. Redundant environmental monitoring methods are used to improve the safety of ships during unmanned flight control.

[0040] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A ship adaptive unmanned navigation control system, characterized in that: include: Multi-information perception unit, comprehensive feature recognition unit, automatic track planning unit, data management unit and navigation information deduction unit; The multi-information sensing unit collects environmental information of the ship's navigation process and sends it to the data management unit, and the data management unit is connected to all other units in communication; The comprehensive feature recognition unit classifies the target in the environmental information, recognizes the motion features and the contour features, and obtains the motion features and the contour features of the target; The navigation information deduction unit deduces the motion trajectory of the moving target based on the environmental information and the motion characteristics and contour characteristics of the target classified as the moving target; The automatic track planning unit plans and controls the track of the ship based on the motion track of the moving target.

2. The ship adaptive unmanned navigation control system according to claim 1, characterized in that: The environmental information includes targets around the ship and their identity information, navigation information, and optoelectronic information, meteorological information, and hydrological information of the environment around the ship.

3. The ship adaptive unmanned navigation control system according to claim 2, characterized in that: The identity information of the target is collected by the AIS device in the multi-information sensing unit, the navigation information of the target is collected by the navigation radar in the multi-information sensing unit, the photoelectric information is collected by the camera in the multi-information sensing unit, the meteorological information is collected by the meteorological instrument in the multi-information sensing unit, and the hydrological information of the water area is collected by the hydrological monitoring device in the multi-information sensing unit.

4. The ship adaptive unmanned navigation control system according to claim 1, characterized in that: The classifying, motion feature recognition and contour feature recognition of the targets in the environmental information includes classifying the targets in the environmental information, and the classifying of the targets in the environmental information includes: Firstly, the targets are divided into moving targets and fixed targets according to their current motion state. Then, the moving targets are divided into ship targets, active moving targets and floating moving targets. The fixed targets are divided into fixed targets and reference targets. The classification criteria of the moving targets include: For targets with identity information, they are classified as ship targets; For targets without identity information, whether the target has autonomous navigation capability is determined based on the target's movement speed, the target's movement direction, the current water flow speed, and the current water flow direction obtained from the data management unit. For targets with autonomous navigation capability, they are classified as active motion targets, and for targets without autonomous navigation capability, they are classified as floating motion targets; The classification criteria for the fixed targets include: For targets whose closed outlines can be identified by navigation radar, they are classified as fixed targets; Targets whose partial outlines can only be identified by the navigation radar are classified as reference targets.

5. The ship adaptive unmanned navigation control system according to claim 1, characterized in that: The classifying, motion feature recognition and contour feature recognition of the target in the environmental information includes motion feature recognition of the target in the environmental information, and the motion feature recognition of the target in the environmental information includes: Real-time positioning of the target is achieved through target tracking based on deep learning, so as to obtain the motion characteristics of the target, and the real-time position of the target is stored in the data management unit.

6. The ship adaptive unmanned navigation control system according to claim 1, characterized in that: The classifying, identifying motion features and contour features of the targets in the environmental information includes performing contour feature recognition on the targets in the environmental information. The performing contour feature recognition on the targets in the environmental information includes: For a ship target, extracting the contour information of the ship target according to the navigation radar echo data corresponding to the ship target in the navigation information obtained from the data management unit, and generating a rectangular envelope of the ship target accordingly; For an active moving target / floating moving target / fixed target, extracting contour information of the active moving target / floating moving target / fixed target according to navigation radar echo data corresponding to the active moving target / floating moving target / fixed target in the navigation information obtained from the data management unit, and generating a polygonal envelope of the active moving target / floating moving target / fixed target accordingly; For the reference target, the contour information of the reference target is extracted according to the navigation radar echo data corresponding to the reference target in the navigation information obtained from the data management unit, and the curve characteristics of the adjacent ship are generated accordingly.

7. The ship adaptive unmanned navigation control system according to claim 1, characterized in that: The step of deducing the motion trajectory of the moving target based on the motion characteristics and contour characteristics of the target classified as the moving target according to the environmental information comprises: For ship targets and actively moving targets, curve fitting is performed on the longitude and latitude coordinate points of the target's motion process to obtain the target's motion mathematical expression, and the target's subsequent position coordinates are predicted based on the motion mathematical expression, thereby obtaining the target's motion trajectory; For floating moving targets, the target's moving direction and speed are directly fitted according to the real-time update information of the target's position coordinates to obtain the target's moving trajectory.

8. The ship adaptive unmanned navigation control system according to claim 1, characterized in that: The planning and control of the ship's track based on the motion trajectory of the moving target includes: Taking the line segment from the current position of the ship to the terminal position as the center line, three obstacle points in the vicinity of the center line are identified by the navigation radar, wherein the three obstacle points include the point where the active moving target is located and the point where the floating moving target is located; The three obstacle points are connected into two line segments; The midpoints of the two line segments formed by the current position of the ship and the three obstacle points are connected to each other to obtain two track segments, and the shorter track segment is selected as the final track segment; The end point of the track segment is taken as the new current position of the ship, and all the above steps are repeated for polling until the current position of the ship coincides with the end point position. At this time, the planned track composed of all track segments is a broken line segment. The corners of the broken line segments in the planned track are transitioned into corner arcs to generate a smooth planned track, and the ship is controlled according to the smooth planned track.

9. The ship adaptive unmanned navigation control system according to claim 8, characterized in that: The process of controlling the ship according to the smooth planned track also includes: Determine whether the planned trajectory and the moving trajectory of the moving target intersect; If there is an intersection, the intersection time interval of the intersection point is calculated according to the navigation information of the planned track and the navigation information of the moving target, and the minimum intersection distance of the intersection point is calculated according to the speed and heading information of the ship. If the minimum intersection distance is less than the safety distance, the ship's track is replanned until the minimum intersection distance is not less than the safety distance; If there is no intersection, the shortest distance between the planned track of the ship and the moving track of the moving target is calculated, and the minimum distance between the ship and the moving target during navigation is calculated based on the shortest distance. If the minimum distance is less than the safety distance, the track of the ship is replanned until the minimum distance is not less than the safety distance.

10. The ship adaptive unmanned navigation control system according to claim 3, characterized in that: If the information of a certain target collected by the camera in the environmental information is not included in the information collected by the navigation radar, the automatic track planning unit controls the ship to perform an emergency avoidance action.

Images