Intelligent navigation situation multi-dimensional sensing terminal

Abstract

The invention belongs to the technical field of intelligent navigation situation sensing, and particularly relates to an intelligent navigation situation multi-dimensional sensing terminal, which generates a forward sensing domain covering a water surface and an underwater area in a preset space-time range in front of a route based on a target ship route trajectory, identifies candidate obstacles in the domain through multi-sensor fusion, and obtains a target navigation situation. Constructing an obstacle original data set containing position, size and medium type, performing confidence cross validation with collaborative perception data of ships in an overlapped perception domain, removing low-confidence candidate obstacles, generating a verified obstacle list, and combining real-time draught, attitude and tide data of the ships to obtain an obstacle data set; a three-dimensional electronic fence is constructed for each obstacle in the list, whether the collision risk exists in the navigation relative to the target ship or not is judged, finally, the electronic fences and the risk state are superposed to an electronic chart for visual display, high-reliability environment sensing data are provided for the autonomous navigation of the intelligent ship, and the obstacle avoidance efficiency in the navigation environment is remarkably improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of intelligent navigation situation awareness, and specifically relates to an intelligent navigation situation multi-dimensional perception terminal. BACKGROUND

[0002] The navigation situation multi-dimensional perception terminal is a key equipment of modern intelligent ships, integrates radar, sonar, vision and other sensor technologies, undertakes the core task of real-time collection, processing and analysis of navigation environment information, and the performance thereof is directly related to the improvement of sea navigation safety and shipping efficiency.

[0003] In a complex sea navigation environment, accurate perception of the navigation obstacle situation is a core link to ensure navigation safety. Obstacles such as reefs, icebergs and other ships often have the characteristics of strong concealment, rapid dynamic change and wide distribution range. If the position, speed and motion trend of the obstacles cannot be accurately perceived, the probability of collision accidents will be greatly increased, and the safety of people, ships and cargo will be seriously threatened. Therefore, breaking through the technical bottleneck of accurate perception of the navigation obstacle situation in complex sea conditions is a key challenge to realize high-level intelligent navigation of ships.

[0004] However, the navigation obstacle situation perception scheme in the prior art still has significant limitations in dealing with the above challenges, mainly in the following aspects: 1. The prior art often defines a static region with a fixed distance threshold or a fixed angle as a situation perception domain, which cannot be dynamically adjusted according to the speed, turning speed, braking performance and other differences of the ship, and cannot meet the actual navigation requirements of the ship. More importantly, the prior art does not construct a unified three-dimensional dynamic perception domain under water and on water, which cannot capture the navigation situation in all directions, and is prone to information loss and misjudgment.

[0005] 2. The prior art has a single obstacle identification method and insufficient confidence, lacks effective cross-media correlation fusion and cross-ship cooperative verification mechanism, and is prone to missed detection and false detection, thereby reducing the obstacle identification accuracy.

[0006] 3. The prior art often uses a simple two-dimensional warning circle or a fixed spacing, lacks a three-dimensional electronic fence that accurately represents the complex three-dimensional shapes of obstacles and ships and their dynamic relationship, and is based on a static ship size model, a fixed safety distance threshold and static chart depth information to evaluate the collision risk of navigation obstacles. The influence of the actual draft, attitude of the ship and the change of the tide on the underwater safety space is not considered in real time, which leads to a disconnection between the risk evaluation result and the actual navigation risk, and threatens the navigation safety. SUMMARY

[0007] In order to overcome the shortcomings in the background art, the embodiments of the present application provide an intelligent navigation situation multi-dimensional perception terminal, which can effectively solve the problems involved in the above background art.

[0008] The object of the present application can be achieved by the following technical solution: an intelligent navigation situation multi-dimensional perception terminal, comprising: a forward perception domain generation module, a multi-source obstacle identification module, a cooperative verification processing module, an electronic fence construction module and a navigation risk visualization module.

[0009] The forward perception domain generation module is connected with the multi-source obstacle identification module, the multi-source obstacle identification module is connected with the cooperative verification processing module, the cooperative verification processing module is connected with the electronic fence construction module, and the electronic fence construction module is connected with the navigation risk visualization module.

[0010] The forward perception domain generation module generates a forward perception domain covering the water surface and underwater area within a preset space-time range in front of the route according to the route trajectory of the target ship.

[0011] The multi-source obstacle identification module identifies candidate obstacles in the forward perception domain through multi-sensor fusion to generate an obstacle raw data set containing spatial position information, geometric size information and medium type information of each candidate obstacle.

[0012] The cooperative verification processing module performs confidence cross verification on the obstacle raw data set and the cooperative perception data of the overlapping perception domain ship, eliminates candidate obstacles with a confidence lower than a preset threshold, and generates a verified obstacle list.

[0013] The electronic fence construction module constructs a three-dimensional electronic fence for each obstacle in the verified obstacle list based on the real-time draft depth, attitude data and tide data of the target ship, and judges whether there is a collision risk with the target ship.

[0014] The navigation risk visualization module superimposes the three-dimensional electronic fence on the electronic chart interface and displays it in real time on the target ship display screen, and highlights the electronic fence with a collision risk.

[0015] Compared with the prior art, the embodiments of the present application have at least the following advantages or beneficial effects:

[0016] (1) The present application generates a forward perception domain covering the water surface and underwater area within a preset space-time range in front of the route according to the route trajectory of the target ship, dynamically considers the ship speed, turning state and detection depth limit, breaks through the limitations of the existing solid-state detection mode, and dynamically adapts to strong and always accurate coverage of the key area in front of the route.

[0017] (2) The fusion of radar and sonar sensors is used to realize the detection of candidate obstacles on the water surface and underwater, effectively lock the same obstacle entity through cross-media association, and cross-verify the identification confidence of the candidate obstacles through the retrieval of overlapping perception domain ships to provide high-reliability environment perception data for intelligent ship autonomous navigation, and promote the technical upgrade from single-sensor independent judgment to multi-source fusion intelligent decision.

[0018] (3) The real-time draft depth, attitude data and tide data of the target ship are deeply integrated, a three-dimensional electronic fence is constructed for each obstacle in the verified obstacle list, and the relative target ship navigation collision risk of the obstacle is judged based on the three-dimensional electronic fence, so that the three-dimensional precise quantitative analysis of the spatial relationship between the ship and the obstacle is realized, and the obstacle avoidance efficiency in the navigation environment is significantly improved. BRIEF DESCRIPTION OF DRAWINGS

[0019] The application will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of the following drawings.

[0020] Figure 1 It is a schematic diagram of the module connection of the application.

[0021] Figure 2 It is a schematic diagram of the confidence cross-verification logic in the cooperative verification processing module of the application.

[0022] Figure 3 It is a schematic diagram of the three-dimensional electronic fence construction logic in the electronic fence construction module of the application. DETAILED DESCRIPTION

[0023] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only a part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the application.

[0024] Referring to Figure 1 The application provides an intelligent navigation situation multi-dimensional perception terminal, which comprises a forward perception domain generation module, a multi-source obstacle identification module, a cooperative verification processing module, an electronic fence construction module and a navigation risk visualization module.

[0025] The forward perception domain generation module is connected with the multi-source obstacle identification module, the multi-source obstacle identification module is connected with the cooperative verification processing module, the cooperative verification processing module is connected with the electronic fence construction module, and the electronic fence construction module is connected with the navigation risk visualization module.

[0026] The forward perception domain generation module generates a forward perception domain covering the water surface and underwater area in a preset space-time range in front of the route of the target ship according to the route trajectory of the target ship.

[0027] In a preferred embodiment of the present application, the forward perception domain generation process comprises: loading a preset safety rule library based on the type of the target ship, and extracting an emergency braking time length matching the current speed from the preset safety rule library; superimposing a preset response time delay on the emergency braking time length, and performing product operation in combination with the current speed of the target ship to obtain a time-varying safety envelope length.

[0028] Taking the current position of the target ship as a spatial vertex and the current heading angle as the direction of the main shaft, a basic cone angle is mapped according to the geometric relationship between the preset ship body width of the target ship and the time-varying safety envelope length, wherein when the turning angle speed of the target ship does not reach a preset turning state judgment threshold, the basic cone angle is used to construct a symmetrical perception cone, and when the turning angle speed of the target ship reaches or exceeds the judgment threshold, an asymmetric cone correction is performed, and the asymmetric cone correction operation comprises expanding the cone angle outside the sailing direction and contracting the cone angle inside the sailing direction.

[0029] It should be noted that the specific process of mapping the basic cone angle according to the geometric relationship between the preset ship body width of the target ship and the time-varying safety envelope length is: inputting the ratio of the preset ship body width of the target ship to the time-varying safety envelope length into an arctangent function to output the basic cone angle in degrees.

[0030] The expansion of the cone angle outside the sailing direction in the asymmetric cone correction operation is based on the positive correlation between the basic cone angle in degrees and the turning angle speed of the target ship, and the cone angle outside the sailing direction is calculated according to a linear gain rule, which can exemplarily be: performing cumulative operation on a preset adjustable proportionality coefficient, the turning angle speed of the target ship and the basic cone angle in degrees, and superimposing the cumulative operation result on the basic cone angle in degrees to obtain the cone angle outside the sailing direction.

[0031] The contraction of the cone angle inside the sailing direction is based on the inverse proportional relationship between the basic cone angle in degrees and the turning angle speed of the target ship, and the cone angle inside the sailing direction is calculated according to a dynamic convergence rule, which can exemplarily be: taking the sum of the turning angle speed of the target ship and 1 as the denominator, and taking the basic cone angle in degrees as the numerator to expand the ratio operation, and the ratio operation result is the cone angle inside the sailing direction.

[0032] The constructed cone is extended to the terminal maximum detectable depth in the water depth direction to generate a three-dimensional truncated cone space, which is used as the forward perception domain.

[0033] The embodiment of the present application generates a forward perception domain covering the water surface and underwater area in a preset space-time range in front of the route according to the route trajectory of the target ship, dynamically considers the ship speed, turning state and detection depth limit, breaks through the limitation of the existing solid-state detection mode, and dynamically adapts to the key area in front of the route and always accurately covers the key area.

[0034] The multi-source obstacle identification module identifies the candidate obstacles in the forward perception domain through multi-sensor fusion, and generates an obstacle original data set containing the spatial position information, geometric size information and medium type information of each candidate obstacle.

[0035] In a preferred embodiment of the present application, the candidate obstacle identification process in the forward perception domain includes: obtaining effective echo points on the water surface and underwater in the forward perception domain through radar and sonar sensors respectively.

[0036] It should be noted that the above-mentioned effective echo points on the water surface and underwater in the forward perception domain are obtained by performing sea wave clutter suppression processing on the echo detected by the radar sensor and water noise suppression processing on the echo detected by the sonar sensor respectively.

[0037] The working principle of the sea wave clutter suppression processing is to dynamically adjust the echo intensity threshold based on real-time sea state parameters, filter out low-energy scattering signals, separate low-speed sea wave clutter using Doppler velocity difference and identify the widely distributed clutter area through spatial continuity analysis, and identify depolarized target echo through dual polarization channels.

[0038] The working principle of the water noise suppression processing is to establish a water background noise spectrum baseline, perform adaptive spectral subtraction and dynamic gain compensation based on the baseline, filter out incoherent noise based on array signal coherence, and suppress parasitic signals in the narrowband interference frequency band.

[0039] The effective echo points on the water surface and underwater are respectively clustered to form a plurality of independently connected water surface echo sub-regions and underwater echo sub-regions, and each sub-region is an initial candidate obstacle.

[0040] The coordinate systems of the radar sensor and the sonar sensor are dynamically aligned, the water surface echo sub-regions and the underwater echo sub-regions are mapped into a unified spatial coordinate system, and the spatial projection errors of each water surface echo sub-region and each underwater echo sub-region are quantified.

[0041] It should be noted that the specific acquisition process of the spatial projection error of each water surface echo sub-region and each underwater echo sub-region is to project the key point coordinates of each water surface echo sub-region and each underwater echo sub-region to the horizontal plane to obtain corresponding projection coordinates, and the spatial projection error of any one water surface echo sub-region and any one underwater echo sub-region is defined as the normalized Euclidean distance of the projection coordinates of the key points of the two sub-regions, wherein the key points can exemplarily be the center of mass points.

[0042] When the spatial projection error between the water surface echo sub-region and the underwater echo sub-region is less than a preset correlation threshold, the two are correlated as the same candidate obstacle entity, and the medium type thereof is marked as a water surface penetration type, and the remaining uncorrelated sub-regions are marked as water surface floating type or underwater immersion type according to the source.

[0043] In a preferred embodiment of the present application, the candidate obstacle identification process in the forward perception domain further includes: constructing a minimum circumscribed cube based on the point cloud data of each water surface echo sub-region and underwater echo sub-region.

[0044] Taking the horizontal projection size of the minimum circumscribed cube as the length and width scale of the corresponding candidate obstacle, and taking the vertical axis size as the height or draft depth of the corresponding candidate obstacle.

[0045] The spatial position of the corresponding candidate obstacle is located by the centroid coordinates of the minimum circumscribed cube, so as to obtain the spatial position information and geometric size information of each candidate obstacle in the obstacle original data set.

[0046] The cooperative verification processing module performs confidence cross verification on the obstacle original data set and the cooperative perception data of the overlapping perception domain ship, eliminates candidate obstacles with a confidence lower than a preset threshold, and generates a verified obstacle list.

[0047] Referring to Figure 2 As shown in the figure, in a preferred embodiment of the present application, the confidence cross verification process includes: dynamically switching the terminal communication channel to retrieve the overlapping perception domain ship, and the overlapping perception domain ship is a ship whose navigation track corresponds to a forward perception domain that spatially overlaps with the current forward perception domain of the target ship within a preset time window.

[0048] Obtain the obstacle perception results of each overlapping perception domain ship in the spatially overlapping region through a ship communication network to form a cooperative verification data set.

[0049] Match and verify the cooperative verification data set and the obstacle original data set of the target ship, including spatial position coincidence verification, geometric size similarity verification, and medium attribute consistency verification, linearly weight and fuse the matching verification results, and output the confidence of each candidate obstacle in the current forward perception domain of the target ship.

[0050] It should be noted that the above matching verification process specifically includes: randomly selecting the spatial position information, geometric size information and medium type information of a candidate obstacle in the target ship obstacle original data set, recording the candidate obstacle as a target obstacle, retrieving the obstacle perceived by the overlapping perception domain ship with the highest comprehensive matching degree of spatial position information, geometric size information and medium type information of the target obstacle from the cooperative verification data set as the paired obstacle of the target obstacle, and taking the matching degree of the paired obstacle in the spatial position information, geometric size information and medium type information as the corresponding verification result.

[0051] The acquisition process of the spatial position information matching degree is: calculating the distance between the centroid coordinates of the paired obstacle and the centroid coordinates of the target obstacle, taking the ratio of the distance to the preset verification distance tolerance threshold as the spatial position deviation degree, and taking the reciprocal of the spatial position deviation degree as the spatial position information matching degree, wherein, in order to avoid the case that the denominator is 0, a preset constant is added to the spatial position deviation degree before the reciprocal operation.

[0052] The acquisition process of the geometric size information matching degree is: calculating the absolute deviation ratio of the paired obstacle and the target obstacle with respect to the length-width scale, height or draft depth, respectively, and then taking the reciprocal of the accumulated value to obtain the geometric size information matching degree.

[0053] The acquisition process of the medium type information matching degree is: if the medium types of the paired obstacle and the target obstacle are consistent, setting the medium type information matching degree as 1, otherwise setting it as 0.

[0054] The comprehensive matching degree is the cumulative result of the spatial position information matching degree, the geometric size information matching degree and the medium type information matching degree.

[0055] It should be further noted that the linear weight distribution of the above matching verification result can be set according to the experience of the industry, or can be obtained through a limited number of test data, such as collecting the verification results of the candidate obstacles in the historical data, including spatial position coincidence, geometric size similarity and medium attribute consistency, and counting the appearance frequency of each verification item when successfully identifying the obstacle. Taking the obstacle identification confidence as the dependent variable and the three verification results as the independent variables, using regression analysis or logistic regression analysis to determine the contribution degree of each verification item, and normalizing the contribution degree to a weight with a total sum of 1.

[0056] In a preferred embodiment of the present application, the dynamic switching of the terminal communication frequency channel to retrieve the overlapping perception domain ship includes: dynamically selecting the ship communication protocol and the working frequency channel based on the spatio-temporal attributes of the current forward perception domain of the target ship.

[0057] It should be noted that the above ship communication protocol and the dynamic selection of working channel, the essence is to analyze the spatial range and time characteristics of the forward sensing domain, match the coverage ability, transmission rate and anti-interference characteristics of different communication technologies. For example, the space is close distance, and the high-bandwidth protocol is selected, the space is far distance, and the satellite protocol is selected, the time efficiency requirement is driven to high frequency or low delay protocol, and the optimal communication efficiency is maintained in the diversified navigation scene of the ship.

[0058] Broadcasting a sensing domain cooperation request containing a space-time identifier in the selected channel, the space-time identifier containing a target ship position, a heading and a forward sensing domain parameter.

[0059] Receiving a metadata set in response to the feedback of the responding ship, the metadata at least including a responding ship pose, a forward sensing domain feature and a data access interface.

[0060] According to the preset judgment condition, the overlapping sensing domain ship is screened from the metadata, and a directional data channel is established to obtain the obstacle sensing result thereof.

[0061] It should be noted that the above preset judgment condition includes the following contents: a) the sensing domain spatial overlap area ratio is greater than a preset area ratio threshold.

[0062] b) the data timestamp and the current time difference value are less than the time window.

[0063] c) the historical obstacle verification accuracy is greater than a preset experience threshold, and the verification accuracy is the correlation coefficient of the past reported obstacles of the ship and the actual navigation dangerous events.

[0064] The embodiment of the application fuses radar and sonar sensors to realize candidate obstacle detection on the water surface and underwater, and effectively locks the same obstacle entity through cross-media association, and on this basis, the overlapping sensing domain ship is searched to perform candidate obstacle identification confidence cross verification, to provide high-reliability environment sensing data for intelligent ship autonomous navigation, and promote the technical upgrade from single-sensor independent judgment to multi-source fusion intelligent decision.

[0065] The electronic fence construction module constructs a three-dimensional electronic fence for each obstacle in the verified obstacle list based on the real-time draft depth, attitude data and tide data of the target ship, and judges whether there is a collision risk relative to the target ship navigation.

[0066] Referring to Figure 3 As shown in the preferred embodiment of the application, the three-dimensional electronic fence is constructed for each obstacle in the verified obstacle list, including: generating a dynamic ship envelope representing the underwater safety space requirement according to the target ship type, loading the static ship body model and the designed waterline stored in the preset safety rule library, and combining the real-time draft depth and the attitude data.

[0067] Extract the minimum circumscribed cube of each obstacle in the verified obstacle list to construct a basic three-dimensional model, combine the deviation amount of the sea-based surface determined by the real-time tidal data, and update the obstacle three-dimensional model by size superposition.

[0068] According to the preset navigation safety rules, calculate the required safety margin of the dynamic ship envelope, perform a three-dimensional space inflation operation on the real-time updated obstacle three-dimensional model to generate an associated three-dimensional safety buffer zone, and use the boundary of the three-dimensional safety buffer zone as a three-dimensional electronic fence.

[0069] It should be noted that the above-mentioned preset navigation safety rules mainly refer to a series of quantitative or logical safety parameters formulated by the navigation safety distance rules in the pre-set safety rule library. The content of the present application mainly includes static safety margin, dynamic safety margin, and obstacle medium type factor. According to the preset navigation safety rules, the required safety margin of the dynamic ship envelope includes: matching the static safety margin in the pre-set safety rule library based on the target ship type, and combining the real-time navigation state and environmental state of the ship to load the corresponding dynamic safety margin in the pre-set safety rule library.

[0070] Retrieve each obstacle medium type factor stored in the pre-set safety rule library, and perform coupling operation on the static safety margin, the dynamic safety margin, and the obstacle medium type factor to output the required safety margin value of the dynamic ship envelope in the three-dimensional space axis direction for different obstacles.

[0071] In a preferred embodiment of the present application, the dynamic ship envelope representing the underwater safety space requirement of the target ship includes: extracting the trim angle and the roll angle in the real-time attitude data of the target ship, and determining the real-time attitude transformation matrix of the target ship through a three-dimensional rotation transformation formula.

[0072] Vertically translating the designed waterline of the target ship static hull model according to the real-time draft depth, applying the real-time attitude transformation matrix to the vertically translated static hull model, and obtaining the real-time immersed hull model of the target ship.

[0073] The dynamic ship envelope representing the underwater safety space requirement of the target ship is generated by performing minimum bounding box calculation on the real-time immersed hull model of the target ship.

[0074] In a preferred embodiment of the present application, the sea-based surface deviation amount based on real-time tidal data includes: normalizing the vertical reference datum of the real-time tidal data and the original vertical datum of the obstacle list.

[0075] According to the offset amount of the normalized real-time tidal value relative to the preset reference surface, the real-time deviation amount of the sea-based surface in the form of axial offset parameters in the three-dimensional coordinate system is calculated and converted.

[0076] In a preferred embodiment of the present application, the condition for judging that the three-dimensional electronic fence relative to the target ship sailing has a collision risk is that the shortest spatial distance between the dynamic ship envelope on the sailing track of the target ship and the three-dimensional electronic fence of the obstacle is less than or equal to a preset distance warning threshold.

[0077] It should be noted that some reference indicators or recommended values about the spatial distance between the dynamic ship envelope and the three-dimensional electronic fence of the obstacle reaching the warning standard will be given in the relevant maritime navigation industry standards, specifications or technical guidelines, i.e., the above-mentioned preset distance warning threshold, for example, the International Regulations for Preventing Collisions at Sea document explicitly has mandatory requirements for the safe meeting distance of ships, or the minimum safe field spacing parameters for different ship types are formulated by classification societies such as DNV-GL, CCS and ABS.

[0078] The embodiment of the present application deeply integrates the real-time draft depth, attitude data and tide data of the target ship, constructs a three-dimensional electronic fence for each obstacle in the verified obstacle list, and judges the collision risk of the target ship sailing relative to the three-dimensional electronic fence on this basis, thereby realizing the three-dimensional precise quantitative analysis of the spatial relationship between the ship and the obstacle, and significantly improving the obstacle avoidance efficiency in the sailing environment.

[0079] The navigation risk visualization module superimposes the three-dimensional electronic fence on the electronic chart interface, displays it in real time on the target ship display screen, and highlights the electronic fence with a collision risk.

[0080] The above content is only an example and description of the structure of the present application, and those skilled in the art can make various modifications or supplements or use similar ways to replace the described specific embodiments, as long as they do not deviate from the structure of the present application or exceed the scope defined by the present application, which should belong to the protection scope of the present application.

Claims

1. An intelligent navigation situation multi-dimensional perception terminal, characterized by: include: The forward perception domain generation module generates a forward perception domain covering the surface and underwater areas within a preset time and space range ahead of the target ship according to the target ship's route trajectory; A multi-source obstacle recognition module identifies candidate obstacles in the forward perception domain through multi-sensor fusion and generates an obstacle raw data set containing spatial position information, geometric size information, and medium type information of each candidate obstacle; A collaborative verification processing module performs confidence cross-validation on the original obstacle dataset and the collaborative perception data of ships in the overlapping perception domain, eliminates candidate obstacles with confidence levels lower than a preset threshold, and generates a verified obstacle list; An electronic fence construction module, which constructs a three-dimensional electronic fence for each obstacle in the verified obstacle list based on the real-time draft, attitude data and tidal data of the target ship, and determines whether there is a collision risk relative to the target ship; The navigation risk visualization module superimposes the three-dimensional electronic fence on the electronic chart interface, displays it on the target ship display in real time, and highlights the electronic fences with collision risks.

2. The intelligent navigation situation multi-dimensional perception terminal according to claim 1, characterized in that: The forward perception domain generation process includes: loading a preset safety rule library based on the target ship type, extracting the emergency braking duration that matches the current speed, adding a preset response delay to the emergency braking duration, and performing a product operation with the current speed of the target ship to obtain a time-varying safety envelope length; With the current position of the target ship as the spatial vertex and the current heading angle as the principal axis direction, a basic cone angle is mapped according to the geometric relationship between the preset hull width of the target ship and the time-varying safety envelope length. When the turning angular velocity of the target ship does not reach a preset turning state determination threshold, a symmetrical perception cone is constructed using the basic cone angle. When the ship's turning angular velocity reaches or exceeds the determination threshold, a cone asymmetry correction is performed. The cone asymmetry correction operation includes expanding the cone angle outside the sailing direction and contracting the cone angle inside the sailing direction. The constructed cone is extended along the water depth direction to the maximum detectable depth of the terminal to generate a three-dimensional truncated cone space as the forward perception domain.

3. The intelligent navigation situation multi-dimensional perception terminal according to claim 1, characterized in that: The candidate obstacle identification process in the forward perception domain includes: obtaining effective echo points on the water surface and underwater in the forward perception domain by radar and sonar sensors respectively; Clustering the effective water surface and underwater echo points respectively to form a number of independently connected water surface echo sub-regions and underwater echo sub-regions, each sub-region being used as an initial candidate obstacle; Dynamically aligning the coordinate systems of the radar sensor and the sonar sensor, mapping the surface echo sub-region and the underwater echo sub-region into a unified spatial coordinate system, and quantifying the spatial projection error between each surface echo sub-region and each underwater echo sub-region; When the spatial projection error between the surface echo sub-area and the underwater echo sub-area is less than the preset association threshold, the two are associated as the same candidate obstacle entity, and their medium type is marked as water surface penetrating type. The remaining unassociated sub-areas are marked as surface floating type or underwater immersed type according to their source.

4. The intelligent navigation situation multi-dimensional perception terminal according to claim 3, characterized in that: The candidate obstacle identification process in the forward perception domain further includes: constructing a minimum circumscribed cube based on the point cloud data of each water surface echo sub-region and each underwater echo sub-region; The horizontal projection size of the minimum circumscribed cube is taken as the length and width of the corresponding candidate obstacle, and the vertical axial size is taken as the height or draft of the corresponding candidate obstacle; The spatial position of the corresponding candidate obstacle is located using the centroid coordinates of the minimum circumscribed cube, thereby obtaining the spatial position information and geometric size information of each candidate obstacle in the original obstacle data set.

5. The intelligent navigation situation multi-dimensional perception terminal according to claim 1, characterized in that: The confidence cross-validation process includes: dynamically switching the terminal communication channel to retrieve overlapping perception domain ships, wherein the overlapping perception domain ships are ships whose forward perception domains corresponding to their navigation trajectories within a preset time window spatially overlap with the current forward perception domain of the target ship; Obstacle perception results of ships in each overlapping perception domain in the spatial overlapping area are obtained through the ship communication network to form a collaborative verification data set; The collaborative verification dataset is matched and verified with the original dataset of the target ship obstacle, including spatial position consistency verification, geometric size similarity verification and medium property consistency verification. The matching verification results are linearly weighted fused to output the confidence of each candidate obstacle in the current forward perception domain of the target ship.

6. The intelligent navigation situation multi-dimensional perception terminal according to claim 5, characterized in that: The method of dynamically switching the terminal communication channel to retrieve the overlapping perception domain ships includes: Dynamically select the ship communication protocol and working channel based on the spatiotemporal attributes of the target ship's current forward perception domain; Broadcasting a perception domain cooperation request including a time and space identifier on a selected channel, wherein the time and space identifier includes the target ship's position, heading, and forward perception domain parameters; Receive a metadata set fed back by a responding ship, wherein the metadata at least includes a position of the responding ship, forward perception domain features, and a data access interface; Based on the metadata, the ships in the overlapping perception domain are screened according to the preset judgment conditions, and a directional data channel is established to obtain their obstacle perception results.

7. The intelligent navigation situation multi-dimensional perception terminal according to claim 4, characterized in that: Constructing a three-dimensional electronic fence for each obstacle in the verified obstacle list, including: loading a static hull model and designed waterline stored in a preset safety rule library according to the target ship type, combining real-time draft and attitude data, and generating a dynamic ship envelope representing underwater safety space requirements; Extracting the minimum circumscribed cube of each obstacle in the verified obstacle list to construct a basic three-dimensional model, and updating the three-dimensional obstacle model by dimension superposition in combination with the sea base deviation determined by real-time tidal data; The safety margin required for the dynamic ship envelope is calculated according to preset navigation safety rules, and a three-dimensional space expansion operation is performed on the real-time updated three-dimensional model of the obstacle to generate an associated three-dimensional safety buffer zone, with the boundary of the three-dimensional safety buffer zone serving as a three-dimensional electronic fence.

8. The intelligent navigation situation multi-dimensional perception terminal according to claim 7, characterized in that: The method of constructing a dynamic ship envelope representing the underwater safety space requirement of the target ship includes: extracting the pitch angle and the roll angle from the real-time attitude data of the target ship, and determining the real-time attitude transformation matrix of the target ship through a three-dimensional rotation transformation formula; The design waterline of the static hull model of the target ship is vertically translated according to the real-time draft depth, and a real-time attitude transformation matrix is ​​applied to the static hull model after vertical translation adjustment to obtain the real-time submerged hull model of the target ship; The minimum bounding box of the real-time submerged hull model of the target ship is calculated to generate a dynamic ship envelope that represents the underwater safety space requirement of the target ship.

9. The intelligent navigation situation multi-dimensional perception terminal according to claim 7, characterized in that: The determining of the sea base deviation based on the real-time tidal data includes: normalizing the vertical reference datum of the real-time tidal data with the original vertical datum of the obstacle list; According to the offset of the normalized real-time tide value relative to the preset reference surface, the real-time deviation of the sea base surface in the form of axial offset parameters of the three-dimensional space coordinate system is calculated and converted.

10. The intelligent navigation situation multi-dimensional perception terminal according to claim 7, characterized in that: The condition for judging whether there is a collision risk between the three-dimensional electronic fence and the target ship is that the shortest spatial distance between the dynamic ship envelope of the target ship on its navigation trajectory and the obstacle three-dimensional electronic fence is less than or equal to a preset distance warning threshold.

Images