Radar positioning processing and related aspects

Abstract

A computer implementation method for determining the geographical location of a vessel using radar (300, 1500) includes: acquiring at least two time-series radar scan images (302); processing the radar images by compensating for the movement of the vessel between at least the radar scan images (306) and suppressing radar echoes from moving objects (308) (304); deriving a composite radar scan image from the processed radar scan images (310); inputting the vessel's position; matching the composite radar scan image to the coastline position of one or more nautical charts closest to each estimated vessel's position for a plurality of estimated vessel's positions within a search area including at least one estimated vessel's position (312); and determining the geographical location of the vessel based on the estimated vessel's position that yields the best match between the composite radar scan image and the coastline position of one or more nautical charts (314).

Description

Technical Field

[0001] The present disclosure relates to identifying the position of a ship using radar, for example, using automatic radar positioning to identify the position of a ship, and related aspects.

[0002] In particular, but not exclusively, the present disclosure relates to a method for automatically radar positioning a ship using a radar system that can evaluate its own accuracy, and a method for automatically evaluating the accuracy of a ship position identified using a method for automatically radar positioning a ship. The ship may be, in some embodiments, an autonomous or remotely controlled ship, or another type of ship having an unmanned bridge. The method may be performed by the ship, and references to the ship herein include references to the own ship that autonomously executes the disclosed technology, unless the context explicitly excludes this.

[0003] Background Art By way of example and not limitation, large ships such as aircraft carriers and destroyers, as well as large merchant ships such as oil tankers and gas tankers, grain carriers, container carriers for transportation, large cruise ships, and large fishing trawlers, which need to locate themselves using all available means, may be subject to regulations, but small ships also require positioning for navigation purposes.

[0004] Positioning of such ships may need to be performed with various accuracies depending on the ship's environment. For example, when maneuvering within a port or dock, positioning to a distance of less than 1 meter may be required, but for example, in international waters, at 15 nautical miles from any nearby object or landmass, positioning may be within or around a few hundred meters. Some areas may also impose regulations for a particular level of positioning to be met, for example, how accurately the absolute position or map-based position of the ship is specified within a certain range of latitude / longitude and / or with a specified form of accuracy, such as within + / - a few meters.

[0005] Several regulations impose requirements on all available means used to locate a vessel. These include, for example, navigation radar required by the International Maritime Organization (IMO) for all large commercial and civilian vessels, and similar regulations may apply to naval vessels. Global Navigation Satellite Systems (GNSS), such as the Baidu Navigation System (BDS), Galileo, the Global Positioning System (GPS), the Global Positioning Satellite System (GLONASS), and the Indian Regional Navigation Satellite System (IRNSS) / Navigation Indian Constellation (NavIC), are added to these requirements because they are not sufficiently reliable for safe navigation in all waters and under all conditions. In addition, naval vessels must be resistant to GNSS interference and spoofing, and manual procedures for safe navigation are widely practiced.

[0006] The development of automated, autonomously navigating and / or operating vessels, including, for example, remotely controlled vessels, may require automated positioning with a higher degree of precision and reliability than may be required for manned and human-operated vessels.

[0007] Existing navigation systems use radar positioning and may rely on human interpretation to determine a vessel's position. Figure 1 shows an example of a radar image superimposed on an electronic chart in an electronic chart display and information system ECDIS. The ECDIS chart complies with IMO Regulations V / 19 & V / 27 of the amended SOLAS Convention by displaying selected information from the System Electronic Navigation Chart (SENC). ECDIS equipment compliant with SOLAS requirements can be used as an alternative to paper charts. Radar images may be presented alongside electronic charts or even paper charts, but this may not be practical or safe in all situations. For example, a ship's navigator or radar operator or other crew member must then evaluate the similarity of the radar features to the features of the comparison chart to determine whether the vessel appears to be correctly positioned at a given GNSS location. This visual evaluation requires constant attention from a focused human. The accuracy achieved varies greatly depending on the selected range scale of the radar image. Other drawbacks include the inherent risk of collisions occurring because comparing radar to charts distracts from the collision avoidance task. This risk is amplified when charts are superimposed on or below the radar, as the radar image becomes cluttered with chart information, potentially requiring greater concentration from the crew performing the visual assessment.

[0008] A more precise alternative involves manually tracking smaller objects in waters with known geographical locations, such as navigation aids (ATONs), or tracking them by automated means. Specifically, an ATON is any device located outside a ship or aircraft that is intended to assist a navigator in determining its position or a safe route, or to warn the navigator of hazards or obstacles to navigation.

[0009] Some radar systems can track the position of their own vessel by designating a tracked ATON or other type of tracked object as the vessel's reference target. The vessel's own position can then be determined by tracking the radar echoes of multiple tracked known objects using the onboard radar system, for example by assigning a known position to each tracked object from a location shown in chart data, and the position of the vessel hosting the radar system can then be determined based on the distance and bearing to the known position of each object.

[0010] While this other method of tracking a reference target is more accurate, it generates a target loss alert indicating that radar positioning has been lost, requiring manual intervention whenever the reference target moves out of line of sight.

[0011] Therefore, such known methods are labor-intensive, as they require the navigator to constantly prepare new reference targets along with their position from the ENC chart, as well as to check for warnings or manually remove objects that disappear as soon as the vessel moves along its course. Because it involves manual labor, this second method, although more accurate than radar positioning of the vessel, is rarely used by ship navigators, resulting in an increased risk of grounding.

[0012] Summary of the Invention The disclosed technology relates to a system and method for an vessel to automatically determine its position using radar positioning. In some embodiments, the estimated quality of the radar positioning position is also provided along with the vessel's position.

[0013] The disclosed technology is limited to environments in which the vessel itself is within range of radar reflectors. This may limit the use of the disclosed technology to, for example, coastal areas and / or, for example, when the vessel itself is located within 100 nautical miles (or less, depending on the maximum range of the vessel's navigation radar system) of the coastline. Conventional navigation radar systems may be used without external position input once the positioning system is initialized at an approximate known position. The initial approximate position may be provided manually or using GNSS. Once initialized, the proposed radar positioning technique does not rely on an external reference position system such as GNSS and uses radar information and heading sensor information that can be obtained, for example, from a compass.

[0014] According to a first aspect of the disclosed technology, a computer implementation method for determining the geographical location of a vessel using radar includes: acquiring at least two time-series radar scan images; processing the radar images by compensating for the movement of the vessel between at least the radar scan images and suppressing radar echoes from moving objects; deriving a composite radar scan image from the processed radar scan images; inputting the vessel's position for the composite radar scan; matching the composite radar scan image with the coastline position of one or more nautical charts closest to each of the multiple estimated vessel positions within a search area including the input vessel position; and determining the geographical location of the vessel based on the estimated vessel position that yields the best match between the composite radar scan image and the coastline position of one or more nautical charts.

[0015] The disclosed embodiments of the first embodiment of the method do not require any preprocessing of radar data, including detection or thresholding of signals, noise, or clutter. Instead, the first embodiment of the method and its described embodiments seek the most likely match between the chart and the radar data. In this way, as long as there are known objects in land or water visible to the ship's radar system, it is highly likely that the ship's position can be accurately determined using the ship's radar system. For example, even if there are weak radar echoes hidden within clutter caused by ocean waves and rain clouds, the ship positioning method may still be reliably used, along with inter-scan signal augmentation.

[0016] In some embodiments, the coastline position closest to each estimated vessel position is obtained by acquiring coastline chart data based on the first possible vessel position and identifying the position of the coastline within the line of sight for each possible vessel position within the search area from the acquired coastline chart data.

[0017] In some embodiments, in each iteration of Method 300, matching includes projecting the nearest coastline position data from an electronic chart onto a composite radar scan to determine the best fit for all possible ship positions.

[0018] In some embodiments, the best fit is determined by finding the possible ship position within a search area that has the maximum sum of radial gradients in amplitude radar echo image data at the predicted position of coastline data closest to the possible ship position.

[0019] In some embodiments, processing of the radar image (304) results in a blurred representation of one or more or all of moving objects, clutter, noise, and interference in the composite radar scan, which has a lower radial gradient of echo signal intensity along the line of sight from the vessel.

[0020] In some embodiments, the processing further includes one or both of the following: enhancing the signal-to-noise ratio of the radar echo signal in each motion-compensated acquired scan, or filtering out noise from each motion-compensated acquired scan.

[0021] In some embodiments, the nearest coastline data also includes other stationary objects shown in electronic chart data from which radar scan echoes can be generated.

[0022] In some embodiments, coastlines may include coastlines within tidal zones, and time and date may be taken into consideration when acquiring line-of-sight coastline data closest to a possible vessel position. In some embodiments, coastlines with shallow beaches where the coastline data changes above a threshold over a period of time are excluded from the coastline dataset used to match the coastlines with features in radar images.

[0023] In some embodiments, the method further includes generating information on the speed and route of a vessel, and using the generated information on the speed and route of a vessel to input an estimated starting position of the vessel in a subsequent iteration of the method according to any one of the preceding claims.

[0024] In some embodiments, the method further includes generating information on the speed and route of a vessel, and using the generated information on the speed and route of a vessel to perform motion compensation in subsequent iterations of the method according to any one of the preceding claims.

[0025] In some embodiments, the method further includes generating an estimate of the accuracy of the estimated ship position, the method comprising obtaining chart coverage area data for the estimated ship position, identifying coastline data from the coverage area, identifying the coastline data closest to the estimated ship position, obtaining at least one of the radar range accuracy and the azimuth accuracy with respect to the position of the closest coastline, combining the estimated radar range accuracy and azimuth accuracy for each of the closest coastlines over all azimuths, and outputting a ship position accuracy indicator.

[0026] Another second aspect of the disclosed technology includes a method for estimating the accuracy of a ship position determined using the method according to any of the first aspect and its embodiments disclosed herein, the method comprising obtaining chart coverage area data for the estimated ship position, identifying coastline data from the coverage area, identifying the coastline data closest to the estimated ship position, obtaining at least one of the radar range accuracy and the azimuth accuracy with respect to the position of the closest coastline, combining the estimated radar range accuracy and azimuth accuracy for each of the closest coastlines over all azimuths, and outputting a ship position accuracy indicator.

[0027] In some embodiments, the ship position accuracy indicator includes one or more of an audible or visual value metric of accuracy and a visual shape contour superimposed on an image of a synthetic radar scan.

[0028] Another third aspect of the disclosed technology includes an apparatus comprising at least a memory, one or more processors or processing circuits, and computer code, the computer code being stored in the memory and, when loaded and executed by the one or more processors or processing circuits, causing the apparatus to perform the method according to any one or both of the first or second aspects disclosed herein or any one of their embodiments.

[0029] Another third aspect of the disclosed technology includes a computer-readable medium containing computer code, which is configured to cause a device to implement a method according to one or both of the first or second aspects disclosed herein or any one of their embodiments when loaded from the memory and executed by one or more processors or processing circuits.

[0030] Another third aspect of the disclosed technology includes a ship having an unmanned bridge, including a device configured to implement a method according to one or both of the first or second aspects disclosed herein or any one of their embodiments.

[0031] In some embodiments, the ship includes an autonomous ship or a remotely controlled ship.

[0032] Therefore, embodiments of the disclosed technology provide an automatic radar positioning system that can enhance navigation safety because it reduces the dependence of the electronic navigation system on the GNSS system. In the case of a ship equipped with a radar scanner, the scan is available 24 hours a day and 7 days a week, and using the disclosed method for determining the radar-positioned ship position enables the ship's position to be updated every 1 - 2 seconds. The radar-positioned ship position can be used for comparison with the GNSS ship position generated by a satellite contemporaneously. If the position obtained by either system deviates beyond a threshold amount, the system can be used to trigger a navigation warning.

[0033] Therefore, the disclosed technology may be used to reduce the vigilance of a navigator who may not always be fully concentrated throughout a shift due to distractions such as other tasks to be performed while on the bridge, and thus can improve the safety of the ship.

[0034] Furthermore, in the event of a complete GNSS failure (e.g., jamming, spoofing, or another type of system failure), automated radar positioning for ship location may be used as a backup service for determining the ship's position and also as a ground speed sensor.

[0035] In some embodiments, which may include preferred embodiments, the dependent claims…

[0036] Advantageously, the disclosed embodiment of the technology automates the radar positioning process in a more reliable and accurate manner. The proposed method is robust and works well in both harsh weather conditions and calm waters.

[0037] In some embodiments, the preprocessing of the radar data does not involve the detection or thresholding of any signals, noise, or clutter, but instead relies on seeking the most likely match between the chart and the radar data.

[0038] Therefore, as long as known objects on land or in water are within radar range, the automated radar positioning system using the disclosed technology will output an accurate position. In addition to strong radar echoes, weak radar echoes hidden within clutter caused by ocean waves and rain clouds may also be used for radar positioning for inter-scan signal augmentation in motion-compensated schemes, which are implemented in some embodiments.

[0039] The method is self-aware in the sense that it generates an estimate of the positioning accuracy for the radar position based on the surrounding scenery, in other words, radar-visible objects around the vessel, and therefore it can also generate an alert if the method provides the position with an accuracy below a desired threshold, which may be a regulated threshold.

[0040] Hereinafter, some embodiments of the disclosed technology are described with reference to the accompanying drawings, which are merely illustrative examples. [Brief explanation of the drawing]

[0041] [Figure 1] This figure shows an example radar image with chart information. [Figure 2A] This figure shows an exemplary set of time-series radar scans that may be used in some exemplary embodiments of the disclosed technology. [Figure 2B] This figure shows an example of how motion compensation can be performed in some exemplary embodiments of the disclosed technology. [Figure 2C] This figure shows exemplary synthetic echo intensity images according to several exemplary embodiments of the disclosed technology. [Figure 3] This diagram schematically illustrates an example of a method for automatically performing radar positioning of a ship, according to several embodiments of the disclosed technology. [Figure 4A] Figure 2 shows an example of a coastline from a chart that may be used in some exemplary embodiments of the method shown. [Figure 4B] Figure 2 shows an example of how the coastline in Figure 4A may move relative to the radar image in some exemplary embodiments of the method shown. [Figure 5A] Figure 4B shows how the radar scans shown can be mapped to positions on a heatmap. [Figure 5B] Figure 4B shows how the radar scans shown can be mapped to positions on a heatmap. [Figure 6A] This figure shows an example of a synthesized radar echo image. [Figure 6B] This figure shows an example of the nearest coastline data. [Figure 7] This figure shows an example of a low fit of radar scan images to the nearest coastline data. [Figure 8] This figure shows the offset between the ship's position where a poor fit was found and the updated ship's position estimate, which obtained the best fit for radar scan images of the nearest coastline data. [Figure 9]This diagram illustrates how ship positions can be monitored using radar data across shipping lanes. [Figure 10] This figure shows an example of feature points in a time-series radar scan. [Figure 11] This figure shows an example of feature points in a time-series radar scan. [Figure 12] This figure shows an illustrative flowchart of a method for determining the speed and route of a vessel using the disclosed technology. [Figure 13] This figure schematically illustrates an example of a method for estimating the accuracy of a ship's position obtained using an embodiment of the method shown in Figure 3 or Figure 15. [Figure 14] This figure schematically shows an example of an uncertainty search region around the estimated ship position obtained using the method in Figure 3 or Figure 15. [Figure 15] This figure schematically shows an exemplary embodiment of the method described in Figure 3. [Figure 16] This diagram schematically illustrates a system for implementing several embodiments of the disclosed technology. [Figure 17] This figure schematically shows an exemplary apparatus for implementing the system shown in Figure 16.

[0042] Modes for carrying out the invention Figure 1 shows an exemplary composite image 100 containing electronic chart information overlaid with radar image data information, as is known in the art. In Figure 1, a vessel 102 is navigating a body of water bounded by various coastlines in the chart, two of which are labeled 104a and 104b in Figure 1. Coastline 104a is the coastline closest to the vessel 102, within the line of sight (LoS). Coastline 104b is a coastline that is part of the landmass on one side of the body of water in which the vessel 102 is navigating. Figure 1 also shows several examples of radar features 106 overlaid on the electronic chart data.

[0043] The disclosed method for locating a vessel using radar involves using radar positioning to locate the vessel and enabling automated monitoring of the vessel's position using radar.

[0044] The method begins by acquiring a series of radar scans, for example, radar scans 200a–200g shown in Figure 2A, which are roughly stacked relative to each other and offset from each other to provide some indication of the ship's movement when the scans were acquired.

[0045] The acquired radar scans are then preprocessed. In some embodiments, preprocessing includes motion compensation and moving object echo suppression. For example, motion compensation may be performed by adjusting each scan according to the ship's movement since the last scan. In some embodiments, the ship's movement may be automatically entered, if determined using a ship model such as the updated ship model output 1620. However, the initial ship's position and movement may also be entered manually.

[0046] Time-series radar scans are not limited to temporally adjacent radar scans. When temporally adjacent or nearly temporally adjacent radar scan images are acquired for use in the manner of the disclosed technique, the ship's position can be adjusted by translating the scan image in the direction of the ship's course by an amount based on the ship's speed between each radar scan used, based on the 1-2 seconds between each scan acquired, which typically does not result in any significant change to the radar scan.

[0047] In some embodiments, acquired radar scans that are not adjacent in time are used, but using only every other scan, for example, results in a loss of statistical reliability. Since the radar scan image processing algorithm can be run in real time using all acquired radar scans 200, this leads to greater reliability in the accuracy of the ship's position as it was identified using the radar scans. In some embodiments, it is also possible to use filters within the ship model to weight each measurement of the ship's position acquired using the method of locating the ship using radar 300 according to the disclosed technology. In this way, if there are sufficiently frequent position measurements, each measurement can be low-pass filtered by the ship model to exclude measurements that are noisier than the expected dynamics of the actual ship.

[0048] Figure 2B schematically shows an example of how motion compensation can be performed for radar scan images 200a to 200g in Figure 2A by vertically aligning the motion offsets of radar scans 200a to 200b shown in the offset stack formation in Figure 2A.

[0049] The acquired radar scan image data may also be preprocessed to remove any other noise sources using appropriate techniques, for example, by decluttering the radar scan image by removing or suppressing data representing non-stationary echoes, for example, by averaging the echo intensity from a number of sampled radar scans 200a to 200g at each individual location.

[0050] Figure 2C shows an example of a composite radar scan image 202 obtained by compositing motion-compensated radar scan image data from each acquired scan 200a onto multiple acquired radar scans 200a-200g. The acquired radar scans may also be processed in some embodiments to remove noise from stationary objects in the scan and / or enhance the SNR before and after the scan is motion-compensated. Non-stationary echo clutter may be reduced by compensating for motion and compositing, for example, by averaging the radar scan images. The composite radar scan may also be appropriately denoised to enhance echo signals in the radar scan image that would otherwise be from stationary objects.

[0051] In this way, after processing to compensate for the movement of the vessel, a clean radar scan image 202, as shown in Figure 2C, is obtained by synthesizing, for example, averaging, the echo intensities from multiple acquired radar scans 200a-200g shown in Figure 2A. However, it will be apparent to those skilled in the art that there are many other filtering and signal-to-noise enhancement or suppression techniques that can be performed to obtain a radar image 202 with less radar clutter than that shown in the individual radar scans 200a-200g. Some embodiments of the disclosed technique can use a weighted averaging technique to weight features that are very similar in scans 200a-g with higher similarity than any feature with lower similarity.

[0052] The technique disclosed herein for synthesizing a set of acquired radar scans 200a-200g acquired over a fixed time interval, such as every 15 seconds, every 30 seconds, or every 60 seconds, has the advantage of blurring the echo signals from moving objects, since their positions change with each scan, even after the movement of the vessel 102 has been compensated for. This blurring results in a radial gradient in the radar echo scan signal intensity (with the radial direction centered on the vessel 102) being smaller than the radial gradient of the intensity or amplitude of the radar echo signal generated by stationary objects in the radar echo scan.

[0053] Even without preprocessing the radar images to remove radar echoes from all non-stationary objects, the radial gradient of the radar echo signal amplitude for such non-stationary objects will be smaller than the radial gradient of the radar echo signal amplitude for stationary objects, so that any remaining non-stationary objects are unlikely to excessively affect the matching of the composite or filtered radar image scan with features of electronic charts such as Land and Navigation Support (AToN).

[0054] The amount of time over which the individually acquired radar scan echo signal images 200 are sampled to generate the composite radar scan image 202 may be based on a predetermined time interval or the number of scans 200. In some embodiments, the amount of time may be configured by the user or set to a preset value.

[0055] In other embodiments of the disclosed technology, other techniques may be used to suppress echoes from non-stationary objects. For example, in some embodiments, one or more averaging and filtering techniques, which would be obvious to those skilled in the art, may be used instead.

[0056] Figure 3 shows an example of a computer implementation method 300 for determining the geographical location of a vessel using radar, according to the disclosed technology. Method 300 includes acquiring at least two time-series radar scan images S302, processing the time-series radar scan images S304 by, for example, compensating for the movement of the vessel between the radar scan images S306, and generating a composite radar scan in S310 from the motion-compensated scans in which radar echoes from non-stationary objects are suppressed. In some embodiments, other processing steps S308 may be performed, such as denoising the images to remove unwanted artifacts.

[0057] The composite radar scan 202 generated in S310 may contain a clearer radar scan image than any of the individual radar scans 200 acquired in S302, because radar signals from moving objects associated with blurry elements of the motion-compensated radar scan are suppressed in the composite radar scan 202.

[0058] The composite radar scan 202 may be obtained in S310 by, for example, compositing the processed radar scan images from S304 using any suitable technique. The method then performs matching of the composite radar scan image with the coastline position of one or more nautical charts closest to each possible ship position in S312 for several possible ship positions, and the method further includes determining the geographical position of the ship based on the estimated ship position that yields the best match of the composite radar scan image with the coastline position of one or more nautical charts in S314.

[0059] In some embodiments, the coastline position closest to each estimated vessel position is obtained by acquiring coastline chart data based on the initial vessel position at the start of the acquired radar scan and identifying the position of the coastline within the line of sight of the initial vessel position from the coastline chart data. In some embodiments, the coastline is represented by polygons or points having their geographical locations (e.g., latitude, longitude). This allows the scale of the coastline data to be irrelevant here.

[0060] If the application implementing Method 300 is hosted on a device having a suitable user interface and an ENC chart application interface, and the ENC chart application is configured to accept queries from such a device, the nearest coastline location can be found by scanning a rasterized chart or by directly querying the coastline. For example, the coastline location can be determined by querying the ENC software using a “stranding check” function query in a form appropriate to the area, specifying a safe depth of 0 as input to the ENC software application. This type of query returns all coastlines in the queried area. It is also possible to create queries to specifically obtain coastlines using software development kits (SDKs) for some ENCs. Coastlines around a ship's position can be obtained by scanning all bearings and using ENC software chart information to find the first hit of a coastline (or other known object). This function may be provided by ray tracing in a chart, for example, provided by a chart SDK, or in some embodiments, it may be implemented as a post-processing step after querying all coastlines. These are the points that define the nearest coastline. For example, land areas behind islands or peninsulas are not included because whether or not radar echoes exist from such areas depends on the height of the radar antenna mounted on the ship relative to the height of the inland landmass on such island and / or peninsula.

[0061] Coastlines are represented by vectors containing polygons or points with their precise geographical locations (latitude, longitude). By projecting coastlines onto radar scan images using a vector-based system, relative scales and coordinate systems between the coastlines and radar scan images are not required.

[0062] The coverage area for generating chart coastlines can be predetermined based on a known maximum radar range, or it can be dynamically changed based on the vessel's position and distance to the shore. In other words, to obtain coastline data, the initial or previously determined vessel position can be entered into the ENC software application (or automatically updated based on the vessel model), and the nearest coastline data is then automatically searched for within a given coverage area, in other words, around the area where land areas are being explored in the electronic chart of the area around the vessel's position.

[0063] Method 300 by the disclosed technology matches the line-of-sight nearest coastline position for several possible ship positions with the strongest radar signal amplitude in a composite radar scan in order to find the best fit in S316. Based on the best fit obtained, the geographical position of the ship can be determined in S318.

[0064] Figure 3 also shows how ENC data can be selectively used to obtain an estimate of the positioning accuracy of the geographical location of the vessel identified in S318.

[0065] Figure 3 also optionally shows that a composite radar scan may be used to determine the speed and route by performing Method 1200, which is described in more detail below. In a subsequent iteration of Method 300, for example, by updating the ship model input 1602 to an algorithm having the ship's speed and route using 1200 for the next set of acquired radar images used to generate the composite radar scan, the motion compensation in S306 can compensate for the motion between that set of acquired radar images using the updated ship's speed and route.

[0066] Figures 4A and 4B schematically illustrate an example in which the nearest coastline data is matched with a radar scan image. In Figure 4A, based on the estimated initial position X0 of the vessel 102 on the electronic navigation chart 400, the coastline 402 closest to the initially estimated geographical location of the vessel is determined by identifying one or more sections of coastline that lie within the line of sight of the vessel 102 based on the initial estimate of the vessel's position X0.

[0067] The initial estimated position X0 may first be provided by the user entering an initial ship estimate and / or by using another navigation ship position system such as GNSS. If Method 300 is used to monitor ship position using radar, the previous ship position identified using a previous iteration of Method may be used to provide the estimated ship position in a subsequent iteration of Method 300.

[0068] In Figure 4A, the solid coastline 402 represents the coastline facing the position X0 of the vessel 102. The dotted coastline is not used by the radar positioning method 200 due to the radar shadow that naturally falls behind the solid line-of-sight coastline 402.

[0069] Figure 4B shows two composite radar scan images 202 generated when the vessel is at position XX. The lower image shows the image of the nearest coastline data (shown as a solid white line for contrast) superimposed on the composite radar chart image 202, based on the estimated vessel position X0. In this case, there is no good fit because the vessel 102 is at position XX and the radar chart is fitted to coastline data generated based on a vessel that is instead at position X0. However, if the nearest coastline data is obtained for a different estimated initial position XX, there is a much better fit, as shown in the upper image.

[0070] Embodiments of this method correspond to sliding and / or translating the closest coastline data on the composite radar image 202 for multiple possible ship positions within the maximum search area for the estimated ship position. To find the best fit, each coastline image should have the same scale as the radar scan image, and the coastline chart-based image and the radar scan image may also need to be rotated to align the compass bearing. This can be achieved by providing heading information as an initial input to Method 300 along with the acquired radar scan (see also Figure 16 below for more details of an example of a system input for carrying out at least one embodiment of the disclosed technology).

[0071] The maximum search area may be user-defined in some embodiments, but may also have default and / or preset values. In some embodiments, a brute-force grid search may be used, but in some embodiments, an optimization method such as SGD (Great Descent) may be used instead to search for which estimated ship position within the search area yields the best match between the composite radar scan image and the nearest coastline image. In some embodiments, the SGD method determines, for each estimated ship position, the sum of all radial gradients of the composite radar scan signal before the nearest coastline geometry is acquired, where the radial gradient is the change in radar scan signal amplitude seen radially or azimuth-directly from the ship. A sudden increase in radar echo signal amplitude along the radial dimension indicates that the radar pulse has hit an object other than water.

[0072] Figure 5A reproduces the features of Figure 4B and schematically shows how the two estimated ship positions in Figure 4B can be represented in a heatmap, such as the heatmap in Figure 5B, when Method 300 uses the radial gradient descent method to search for the estimated position of ship 102. In Figure 5B, the heatmap 500 shows the sum of the radial gradients of the amplitude of the radar echo signal at different offsets of the coastline image estimated ship position closest to the electronic chart data, based on the actual ship position along the north-south and east-west directions across all bearings 0 to 360 around the ship.

[0073] The peak amplitude in the heatmap in Figure 5B is generated when the closest coastline data is used for the estimated ship position, which is where the ship is actually located. This results in the radar signal changing amplitude most strongly when it reaches land. In other words, the echo signal has the steepest gradient as it approaches land due to the flat water surface and rises due to the shape of the rising coastline. Cliffs produce a larger gradient accordingly than shallow beaches. The leading edge of the radar gradient is used, which is the gradient that progresses from the low echo intensity at the water surface compared to the high echo intensity when it hits land. By checking all radial directions from ship 102, in other words bearings, and summing the radial gradients across all bearings from 0 to 360, a heatmap can be generated for each possible ship position. The actual ship position is located at the peak of the heatmap, which is the ship position where the sum of radial gradients, in other words, the change in radar echo signal amplitude, changes the most across the entire bearing from 0 to 360 around the ship. The intensity of the heatmap peak is at least partially influenced by the amount of coastline used to generate the estimated ship position and the radar signal sensitivity. Brighter heatmap peaks may indicate a greater presence of coastlines and / or a more rapid increase in radar signal intensity upon reaching land.

[0074] Therefore, in some embodiments of Method 300, signal suppression S308, which includes radar echoes from moving objects, results in a blurred representation of moving objects in a composite radar scan with a lower radial gradient along the line of sight to the vessel.

[0075] This is useful because it blurs the signal gradient for non-stationary (e.g., moving) objects, automatically reducing the radial gradient radar echo signal. As a result, when searching for the geographical location of a ship, radar scan echo signals returned from non-stationary objects are given less weight than radar echo signals with any higher gradients generated by radar scan signals returned from stationary objects such as land and navigation aids (AToN).

[0076] To demonstrate in more detail how the geographical location of a vessel can be obtained using an exemplary embodiment of method 300 in which a grid search is performed to determine the estimated location of vessel 102, Figure 6A shows an exemplary composite radar scan image, including an indication of which direction is north. The direction of north can be determined using compass readings or by any other suitable technique known in the art. Figure 6B shows an exemplary image of coastline data obtained for the estimated location "X" of vessel 102. It is not necessary to generate an actual image of coastline data for use in the disclosed embodiments. Instead, vectors representing the locations of features (e.g., with respect to their latitude and longitude) in the coastline data obtained from electronic chart information are projected onto the features of the radar scan image to find the best match.

[0077] Figure 7 schematically illustrates an exemplary 10x10 search grid containing 100 possible ship locations, where the electronic chart data of the nearest coastline (e.g., in Figure 6B), indicated by the dashed lines, does not closely match the radar feature 106 shown in the composite radar scan image (e.g., the radar scan image in Figure 6A) at the possible ship locations indicated by the shaded circles in Figure 7. In Figure 7, the actual ship locations within the search grid are also indicated by the solid black circles. The concentric circles indicate the radar scan signals emanating from the possible shaded ship locations within the search grid, although in reality, the radar scan images emanate from the ships at the possible black circle locations.

[0078] Figure 8 schematically illustrates another possible ship position within the search grid of Figure 7, in which case the best estimated ship position, indicated by the black dots in the search grid, can be found by sliding the coastline data image over the radar scan image to obtain a much better fit between the radar features and the nearest coastline data. In Figure 8, the nearest coastline data is generated for the possible ship position indicated by the black dots in a 10x10 search grid, in which case the possible ship position provides the best fit for all possible ship positions in the search grid and is obtained as the actual ship position.

[0079] Figure 8 also schematically illustrates an example in which, after performing Method 300, the possible ship position indicated by the shaded circle in the search grid may be acquired as the initial ship position and updated with an offset of 900 relative to the actual ship position. The updated estimated ship position, indicated by the black dot in the search grid where the best fit was obtained when matching the nearest coastline chart data with the radar echo scan image, may then be acquired as the starting or initial ship position for subsequent iterations of Method 300 for the next composite radar scan 202.

[0080] Figure 9 shows how the position of vessel 102, estimated using the radar positioning of Method 300 (shown as black dots in each search grid in Figure 9), can be continuously monitored independently of other positioning techniques by repeating Method 300 for multiple composite radar scans 202N, 202N+1, 202N+2, and 202N+3. A similar monitoring scheme may be used in conjunction with the radial descent gradient technique schematically shown in Figure 5B to locate the position of vessel 102 in other embodiments of Method 300.

[0081] As shown in Figure 9, multiple time-series composite radar scans 202N, 202N+1, 202N+2, and 202N+3 (which are not necessarily temporally adjacent) are acquired, each representing a motion-compensated, non-stationary object echo-suppressed radar scan 202 generated from multiple acquired radar scans, such as scans 200a to 200g shown in Figure 2A.

[0082] Figure 9 shows, as an example, multiple 10x10 search grids, each containing 100 possible ship positions, which are searched to determine the actual ship position every N…N+3 composite radar scans. By matching the nearest coastline data image and / or radar scan image obtained for each of the 100 possible ship positions in each search grid with the composite radar scan image, by sliding around the nearest coastline data image and / or radar scan image until the best match is found for the superimposed image, it is possible to determine which of the 100 possible ship positions in the 10x10 search grids provides the best fit, and then obtain that position as the actual geographical position of the ship at various points in time along the ship's voyage.

[0083] As shown in Figure 9, for scan N, the best fit for the coastline data indicated by the dashed line is obtained at the indicated location from which both the arrows indicating the ship's route by the ship model (dashed arrows) and the ship's route by the chart matching method (solid arrows) originate. As just one example, each composite radar scan image 200 may be generated in some embodiments by processing multiple temporally adjacent acquired radar scan images 200, for example, 10 acquired scans 200 may be composited at once.

[0084] As a mere example, in some embodiments, a composite radar scan image 202 may be generated by processing acquired radar scan images that are not temporally adjacent, e.g., radar scan images 1, 3, 5, 7, and 9, instead of each temporally adjacent radar scan image, e.g., scans 1-10. In some embodiments, the number of acquired radar scans 200, or the duration over which the acquired radar scans 200 are synthesized to form the composite radar scan image 202, is determined dynamically based on the ship's environment. In this way, if one or more or all of the environment around the ship is changing rapidly, and the ship is moving rapidly, and the time intervals between radar scans are longer, the number of acquired radar scan images 200 used to form the composite radar scan image 202 may be smaller.

[0085] Each of the 202N…202N+3 composite radar echo signal scans shown in Figure 9 is generated by processing multiple acquired radar scans 200 and can compensate for at least the movement of the vessel between acquired radar scans, and can also suppress or remove noise and / or echoes from moving objects within the acquired radar scans 202N…N+3.

[0086] As the vessel moves from left to right in Figure 10, the vessel position determined for scan 202N can be updated for the next composite radar scan image 202N+1 using a vessel model that includes at least the vessel's route and speed, over the time interval between composite radar scan 200N and composite radar scan 200N+1.

[0087] When a new composite radar scan 202N+1 is generated from multiple acquired radar scans 200, a new search grid is generated and the ship positions identified from the ship model are updated with the estimated ship positions obtained using method 300. The new positions in the search grid to be used by method 300 for the composite scan 200N+1 are indicated by a diagonal circle in that search grid and linked by a solid arrow to the ship positions in the search grid of scan 202N.

[0088] Next, chart coastline data for the area around the vessel, also referred to herein as the coastline chart coverage area, is determined based on a rough position obtained, for example, using the planned vessel route and speed from the previous vessel position estimated by Method 300.

[0089] Next, the nearest coastline within this chart coverage area is determined for each possible location of vessel 102 in the search grid for the composite radar scan image 202N_1. Then, an estimate of the vessel's position in the search grid 202N+1 can be obtained by checking which of the 100 possible vessel positions in the 10x10 search grid shown in Figure 10 yields the best fit to the image of coastline chart position data that is closest to the features in the composite radar scan image 202N+1. Alternatively, the estimate of the vessel's position can be obtained using the strongest radial gradient descent technique shown in Figure 5A by updating the heatmap for the new search grid. The updated estimate resulting from the vessel's position in the search grid for the composite radar scan image 202N+1 can then be repeated for scan 202N+2, then scan 202N+3, and so on, allowing the vessel's position to be estimated using radar against the nearest coastline chart data. This technique is more reliable when more coastline data is available for the radar scan image to be fitted, and therefore, although limited, it is also possible to use Method 300 to determine the course and speed of a vessel as long as there is some coastline data or other stationary objects shown in the electronic chart data that generate a sufficiently strong radar echo signal. Therefore, Method 300 can be used independently of other techniques to monitor the position of a vessel.

[0090] To monitor a vessel over time, some embodiments of Method 300 further include generating estimates of the vessel's route and speed (indicated by solid arrows in Figure 9). These estimates can also be used to perform motion compensation S306 in processing S304 of Method 300. By compensating for the vessel's motion, in S308, it becomes possible to suppress echoes from any non-stationary objects detected by the radar scan. Thus, to reduce reliance on estimates of the vessel's route and speed generated using GNSS or another technique, some embodiments of Method 300 employ feature extraction followed by feature point matching to identify features in temporally adjacent or nearly adjacent acquired radar scans. Other methods that may be used in addition to, or instead of, other embodiments of the disclosed Method 300 include image correlation and optical flow, but any suitable technique that would be obvious to those skilled in the art may be used.

[0091] Figures 10 and 11 show an example in which a pair of composite radar scans 202a, 202b is used to determine the course and speed of a vessel. Figure 10 shows how feature point A in the acquired radar scan 200a has the corresponding feature point AA shown in the subsequently acquired radar scan 200b shown in Figure 11. Figures 10 and 11 show multiple pairs of corresponding feature points linked by horizontal or horizontal stripes, including a horizontal stream 1100 between A and AA. The pair of scans 202a, 202b are motion-compensated images, and determining the speed and course of the vessel in the vessel model (see 1620 in Figure 16) may involve a self-adjusting feedback loop in some embodiments of the disclosed technology. Referring briefly to Figure 16, it is also possible to provide a ship model including an initialized ship position (which may be entered manually) as an initial input 1602 to a system 1600 implementing the method disclosed herein, and in some embodiments, the ship model input 1602 to the system may also be initialized with an optional known speed and route in order to perform motion compensation as shown in 1612 (see also S306 in Figure 3).

[0092] Figure 12 shows in more detail one embodiment of method 1200 for determining the speed and course of a vessel 102, which in some embodiments of method 300 can be used to compensate for movement between adjacent radar scans 200a, 200b, such as those shown in Figures 11A and 11B. Method 1200 shown in Figure 12 acquires a pair of time-series radar scans 200a, b in S1202, which may be a pair of time-series radar scans that are temporally adjacent or otherwise reasonably close, separated by an interval of time T. For example, radar scans 200a and 200b shown in Figures 10 and 11 can form a pair of radar scans.

[0093] In method 1200 shown in Figure 12, in S1204, features are first extracted from each radar scan in the pair of radar scans being processed. In S1206, the next corresponding feature points in each of the radar scans are identified, and in S1208, for multiple feature points in each radar image of the pair, the positions of the corresponding feature points are identified. The direction of the offset and the direction of the distance offset can then be identified. The stripes linking the scan images 200a and b in Figures 10 and 11 between the feature points reflect the offset distance and direction to some extent, but the distance and offset between scan image positions are not considered in Figures 10 and 11.

[0094] The composite offset distance and direction are determined by method 1200 in S1210. If the bearing and radar bearing of the vessel 102 are known, the direction of the offset can be determined as the vessel's route, and the vessel's speed can be determined as long as the scale of the radar image and the length of time between the generated radar images are known, which allows the actual distance between matching feature points to be determined. The results can be output in S1214 and may be used in some embodiments of method 300 to perform motion compensation S306 (see motion compensation S306 in Figure 3).

[0095] Some embodiments of Method 300 also estimate the accuracy of the geographical location of a vessel identified using Method 300.

[0096] Figure 13 shows an exemplary method 1300 for estimating the accuracy of a ship's radar positioning location obtained using method 300.

[0097] Method 1300 includes obtaining chart coverage area data around the estimated vessel position in S1302, and then obtaining coastline data from the chart coverage area in S1304. Next, the closest coastline data for the estimated vessel position is obtained in S1306, for example, the closest coastline may be found across all bearings from 0 to 360 around the possible vessel position.

[0098] Next, in S1308, the method obtains radar range accuracy and azimuth accuracy for each of the nearest coastlines. For example, if the radar range resolution is + / 3m at 100 miles from the coast and + / -6m at 200 miles from the coast, the vehicle position based on radar matching with the accurate coastline at 100 miles from the coast is limited to at least + / -3m by radar resolution alone, while if the nearest coastline is at 200 miles from the coast, the vessel position identified using the radar system cannot be better than + / -6m due to the limitations of the radar scanner resolution. These resolution limitations are exacerbated by any limitations in azimuth accuracy.

[0099] The estimated radar range accuracy and heading accuracy are combined in S1310 for all possible headings from 0 to 360 around the ship's position, and the results are output in S1312 as a ship's position accuracy indicator. The output may be an audible alarm or message, a displayed alarm or message, or may take the form of a value, percentage, or shape contour superimposed on a composite radar scan image.

[0100] Figure 14 schematically shows the estimated ship position P_estimate obtained by Method 300, which has the best fit with the nearest coastline data around its radar scan image, and also shows how a suitable search area for ship positioning can be determined based on the estimation accuracy obtained by Method 1300, which is obtained from the analysis of coastlines within the “coverage area”.

[0101] In Figure 14, the search area 1400 is determined from the coastline within the coverage area, which provides an estimate of how accurate the radar scan is likely to be at a distance from the vessel where the nearest coastline is located. In Figure 14, the estimated vessel position is P_estimate, and another estimated vessel position is offset from P_estimate by P_delta, shown in Figure 14 as the circle defining the position of the search grid 1400.

[0102] Figure 14 also shows an example of a coverage area 1402 used for coastline search for the nearest coastline, and an example of a maximum radar scan range 1404 from a vessel 102. The nearest coastlines are shown as 1406a, b, and c. The nearest coastline 1406b is on an island, and land areas within the radar shadow from P_estimate have a bright dot pattern fill to distinguish them from land areas not within the radar shadow.

[0103] In some embodiments, the uncertainty search grid may have a smaller grid size than the grid size used in method 300, in which case it is possible to obtain a more accurate estimated ship position using the precision method.

[0104] The radar positioning accuracy of Method 300 depends on the amount of land surrounding the coastline. For example, accuracy is higher when land is present in all directions within a short range. In contrast, a large, distant landmass in only one direction reduces the accuracy of the automatically determined radar positioning. The probability that Method 300 will provide a position with a certain accuracy can also be determined by using the chart coverage area to combine the range and azimuth resolutions identified by the radar scan with features on the coastline.

[0105] Figure 15 schematically shows an example of Method 1500, which includes one embodiment of Method 300, in which the steepest radial gradient technique is used to determine the ship's position by matching radar scan echo image data with image data of the nearest coastline within the search area. As shown in Figure 15, Method 1500 obtains the ship's position in S1502, which may be an initial position entered by the user or another navigation system, or may be obtained by performing a previous iteration of Method 1500 or 300. Next, in S1504, the nearest coastline data for the ENC coverage area is obtained for that ship's position. Next, in S1506, for all bearings 0 to 360, the expected radar range to the nearest coastline for the position determined in S1502 is determined in S1508, and the radial gradient from the radar echo in the expected range of the nearest coastline data is determined in S1510. Then, in S1512, the sum of the radial gradients for all bearings where a coastline or other known chart object exists is determined and stored as fitting information. If fit information is stored for all possible ship positions within the search area (S1514), the best fit is determined by finding the ship position associated with the fit information showing the highest gradient, and the ship's position is updated in S1516. Otherwise, the method returns to step S1502 to find another possible ship position within the search area, and steps S1504 to S1514 are repeated for that possible ship position.

[0106] Figure 16 schematically shows a system 1600 comprising an apparatus such as the apparatus 1700 of Figure 17 described below, configured to carry out one embodiment of the technology disclosed using, for example, appropriate methods 300, 1200, 1300, and 1500.

[0107] Figure 16 shows a system input 1602 which includes the ship's speed and route 1604, multiple acquired radar scans 200, and the ship's heading 1606, used to generate multiple motion-compensated radar images 1612 and then processed to generate a composite radar scan image 202. From the composite radar scan image 202, along with chart object data shown as the nearest coastline 1628 obtained from ENC chart coverage data 1610, the system generates, for example, a ship speed estimate 1616 and a ship position estimate 1618. The results may be filtered over time, for example, by applying a Kalman filter or another filtering technique used to improve the accuracy of the resulting position estimate 1622 in some embodiments that form the system output 1630. The system output 1630 may also include an optional accuracy estimation indicator 1624 and / or a position quality or reliability indicator 1626, similarly generated in some embodiments of the system 1600 using nearest shoreline data 1628 acquired from the ENC coverage area input 1610. In some embodiments of the system 1600, the system output 1630 includes the speed and route 1604 of a vessel used in subsequent iterations of methods 1500, 300 to perform motion compensation against the acquired radar scan 200. The speed and route of a ship may be estimated from a radar scan using any suitable image feature extraction technique known in the art, for example, a suitable technique may be disclosed in Y. Linder and V. Taranukha, "Deep Learning Method for Extracting Information from Radar Images of Marine Objects," 4th IEEE International Conference on Advanced Trends in Information Theory (ATIT) 2022, Kyiv, Ukraine, 2022, pp. 381–384, doi:10.1109 / ATIT58178.2022.10024204. Other examples include, as is well known in the art, libraries, ORB, BRISK, FAST, KAZ, and MINEIGEN.

[0108] Some embodiments of system 1600 include the apparatus 1700 shown in Figure 17.

[0109] The device 1700 may include a standard computer system or a system specifically designed to perform a standard procedure. The device 1700 may include a standalone system or be integrated into the bridge deck along with other navigation equipment.

[0110] Figure 17A schematically shows an example of an apparatus 1700 according to several embodiments of the disclosed technology, which may be used to carry out exemplary embodiments of the disclosed technology. The apparatus 1700 may include a general-purpose computer adapted to run software such as the software shown in Figure 17B or computer code 1716, which is appropriately stored in the device's memory 1702, in some embodiments.

[0111] An exemplary embodiment of the apparatus 1700 may comprise one or more processors or processing circuits 1704 and / or one or more controllers and / or control circuits 1708. The computer code 1716 may include, for example, exemplary embodiments of the pseudocode described above, which can be loaded from memory and executed by one or more processors or processing circuits 1704 to carry out an exemplary embodiment of any of the methods 300, 1200, 1300, or 1500 described above.

[0112] As shown in Figure 17A, the device 1700 also includes a suitable power supply 1710, although in some embodiments, if the device is integrated into an equipment console, the power may be supplied to the equipment console via the power supply. In some embodiments, a backup power supply, such as a battery, may also be provided. As shown in the exemplary embodiment of the device 1700 shown in Figure 17A, a suitable data input and output interface 1706, such as one or more data ports, is also provided to receive electronic chart information data 1610, which may in some embodiments be stored in memory 1702, and / or the device 1700 may include a user interface configured to receive user-input initial position information and / or compass information from an electronic source, or to manually input heading information. The data I / O 1706 may be an air interface for wireless communication in some embodiments, which also include a suitable wireless receiver / transmitter and antenna equipment shown as TX / RX 1712 in Figure 17A.

[0113] Figure 17B shows computer code 1716. In some embodiments, and as schematically shown in Figure 17B, the computer code may include several different modules or functions that can be called to implement one or more or all embodiments of methods 300, 1200, 1300, and 1500 according to the disclosed technology. For example, computer code represented by one or more or all of modules M300, M1200, M1300, and M1500 may be stored in memory 1702. When each method module code is loaded from memory 1702, it causes the device to execute one embodiment of the corresponding method 300, 1200, 1300, and 1500, respectively, thereby enabling the device 1700 to implement at least partially an exemplary embodiment of method 1500 shown in Figure 15.

[0114] Some embodiments of the disclosed technology include a computer program product that includes computer program code, which is loaded from memory and executed by one or more processors or processing circuits of the device, causing the device to carry out a method according to one or more of the method embodiments described above.

[0115] In some embodiments, the computer program code may include one or more modules, which can be represented by the following pseudocode example, of a function for a method of radar positioning a ship and estimating the accuracy resulting from the ship's position derived from the radar. Function to obtain speed and route from radar scan: Extract features from the first scan. Extract features from the second scan. Match features and obtain the offset as a 2D vector. Return speed and route End of function Function: Radar Chart Matcher Retrieve coastlines from a chart database. For each location within the search area: Directions from 0 to 360 degrees Find the distance to the nearest coastline. If a coastline is required Calculate the radial gradient of the radar echo over distance. Sum the results and store them. end The system stores the current position along with the sum of the gradients. end When the maximum gradient is determined, it is the best fit. End of function Estimate the accuracy of a function. Retrieve coastlines from a chart database. Directions from 0 to 360 degrees Find the distance to the nearest coastline. If a coastline is required For example, use multivariate techniques to calculate accuracy. Store accuracy and / or confidence. end end Return: Accuracy at the specified confidence level End of function #The following is the main method: Obtain the initial position from the user or a reference system. Initialize the ship model using its position and, if applicable, its speed and route. While the application is running Get two final pre-processed radar scans. Call "Get speed and route from radar scan". Obtain the estimated position (extrapolated or predicted to the current time) from the ship model. Call the "Radar Chart Matcher" with the last radar scan and estimated position. Call "Estimate accuracy" Update ship models with new estimates of speed, route, and position. Report position, speed, and route.

[0116] In some embodiments of the code, a suitable multivariate method may be used to determine the confidence level of the ship's position. Predictive accuracy is calculated in all directions from the ship's position relative to the coastline. To achieve this, a suitable multivariate method may be used, combining accuracy from individual directions and radar ranges. If the confidence level is determined to be above a threshold, a quality assessment is performed, which may form the output of system 1600 shown in Figure 16.

[0117] However, it will be apparent to those skilled in the art that the present invention is not limited to the specific code module structure for the exemplary implementation of the pseudocode shown in Figure 17B and described above.

[0118] The terms used herein are intended to describe only specific aspects and are not intended to limit the disclosure. The singular forms “a,” “an,” and “the” used herein are intended to include the plural form unless the context clearly indicates otherwise.

[0119] As used herein, the term "and / or" includes any combination of one or more of the related enumerated items, and may be abbreviated as " / ".

[0120] It will be further understood that, when used herein, the terms “comprises,” “comprising,” “includes,” and / or “including” specify the presence of the described features, integers, steps, actions, elements, and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, actions, elements, components, and / or groups thereof.

[0121] In this specification, terms such as "first," "second," etc., may be used to describe various elements, but it should be understood that these elements should not be limited by these terms. These terms are used solely to distinguish one element from another. For example, without departing from the scope of this disclosure, a first element may be called a second element, and similarly, a second element may be called a first element.

[0122] Relative terms such as “down,” “up,” “top,” “bottom,” “horizontal,” or “vertical” may be used herein to describe the relationship of one element to another, as shown in the figures. These terms and those mentioned above will be understood to encompass different orientations of the device, in addition to the orientation depicted in the figures. When one element is said to be “connected” or “joined” to another element, it will be understood that it may be directly connected or joined to the other element, or there may be an intervening element. In contrast, when one element is said to be “directly connected” or “directly joined” to another element, there is no intervening element.

[0123] Unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as those generally understood by those skilled in the art to which this disclosure belongs. Terms used herein should be construed to have meanings consistent with their meanings in the context of this specification and related art, and it will be further understood that they should not be construed in an idealized or overly formal sense unless expressly defined herein.

[0124] While some embodiments are described in the context of the apparatus, it is clear that these embodiments also represent a description of the corresponding method, where a block or device corresponds to a method step or a feature of a method step. Similarly, embodiments described in the context of a method step also represent a description of the corresponding block, item, or feature of the corresponding apparatus.

[0125] It should be understood that this disclosure is not limited to the embodiments described above and shown in the drawings, and rather, those skilled in the art will recognize that many changes and modifications may be made within the scope of this disclosure and the appended claims. There are disclosed embodiments in the drawings and specification that are for illustrative purposes only and not limiting purposes, and the scope of the concept of the present invention is described in the following claims.

Claims

1. A computer implementation method (300, 1500) for determining the geographical position of a ship using radar, wherein the method is Acquire at least two time-series radar scan images (S302), The radar image is processed by at least compensating for the movement of the vessel between radar scan images (S306) and suppressing radar echoes from moving objects (S308) (S304), (S310) Deriving a composite radar scan image from the processed radar scan image, Enter the ship's position, For a plurality of estimated ship positions within the search area including the input ship position, the composite radar scan image is matched with the coastline position of one or more nautical charts closest to each of the plurality of estimated ship positions (S312, S314, S316), The geographical location of the vessel is determined based on the estimated vessel position that yields the best match between the composite radar scan image and the coastline position of one or more nautical charts (S318) Methods that include...

2. The coastline position closest to each estimated ship position is, Based on the first possible ship position, obtain coastline chart data. From the acquired coastline chart data, identify the position of the coastline within the line of sight of each possible ship position within the search area. The method according to claim 1, obtained by...

3. The method of claim 2, wherein in each iteration of method (300), the matching includes projecting the nearest coastline position data of the electronic chart onto the composite radar scan in order to determine the best fit for all possible ship positions.

4. The method according to claim 3, wherein the best fit is determined by finding the possible ship position within the search area having the maximum sum of radial gradients in amplitude radar echo image data at the predicted position of the coastline data closest to the possible ship position.

5. The method according to any one of claims 1 to 4, wherein the processing of the radar image (S304) results in a blurred representation of one or more or all of moving objects, clutter, noise, and interference in the composite radar scan having a lower radial gradient of echo signal intensity along the line of sight from the vessel.

6. The method according to any one of claims 1 to 5, wherein the processing further comprises one or both of the following: enhancing the signal-to-noise ratio of the radar echo signal in each motion-compensated acquired scan, or filtering out noise from each motion-compensated acquired scan.

7. The method according to any one of claims 1 to 6, wherein the nearest coastline data also includes other stationary objects shown in the electronic chart data from which radar scan echoes can be generated.

8. To generate information on the speed and route of a ship, Using the generated ship speed and route information, input the estimated starting ship position in a subsequent iteration of the method according to any one of claims 1 to 7. The method according to any one of claims 1 to 7, further comprising:

9. To generate information on the speed and route of a ship, Using the generated ship speed and route information, motion compensation is performed in subsequent iterations of the method according to any one of claims 1 to 7. The method according to any one of claims 1 to 8, further comprising:

10. The method further includes generating an estimated value of the accuracy of the estimated ship position, Acquiring chart coverage area data for the estimated ship position (S1302), Identifying coastline data from the aforementioned coverage area (S1304), Identifying the coastline data closest to the estimated ship position (S1306), To obtain at least one of the radar range accuracy and azimuth accuracy for the position of the nearest coastline, In S1310, the estimated radar range accuracy and azimuth accuracy are combined for each of the nearest coastlines across all directions. Outputting a ship position accuracy indicator and The method according to any one of claims 1 to 9, including the method described in any one of claims 1 to 9.

11. A method for estimating the accuracy of a ship position obtained using the method of any one of claims 1 to 10, wherein the method for estimating the accuracy is Acquiring chart coverage area data for the estimated ship position (S1302), Identifying coastline data from the aforementioned coverage area (S1304), Identifying the coastline data closest to the estimated ship position (S1306), To obtain at least one of the radar range accuracy and azimuth accuracy for the position on the nearest coastline (s1308), Combining the estimated radar range accuracy and azimuth accuracy for each of the nearest coastlines across all directions (S1310), Outputting a ship position accuracy indicator (S1312) Methods that include...

12. The aforementioned ship position accuracy indicator, Accuracy auditory or visual value metrics, Visual shape contours superimposed on synthetic radar scan images The method according to claim 11, comprising one or more of the above.

13. at least, Memory and One or more processors or processing circuits, Computer code and A device that includes, When the computer code is stored in the memory and loaded and executed by one or more processors or processing circuits, the apparatus is made to perform the method according to any one of claims 1 to 10 and / or 11 to 12. Device.

14. A computer-readable medium containing computer code, wherein the computer code is configured such that, when loaded from memory and executed by one or more processors or processing circuits, it causes a device to carry out the method according to any one of claims 1 to 10 and / or 11 to 12.

15. A vessel having an unmanned bridge, comprising a device (1700) configured to carry out the method described in any one of claims 1 to 10 and / or 11 to 12.

16. The vessel according to claim 15, wherein the vessel includes an autonomous vessel or a remotely controlled vessel.

Images