Ship tracking device, automatic ship control system, ship tracking method, and ship tracking program

Abstract

To track a target even in a small area in the vicinity of an own ship.SOLUTION: A ship tracking device 10 includes an acquisition unit 20, an area setting unit 30, a selection unit 50, and a tracking unit 60. The acquisition unit 20 acquires point cloud data comprising a plurality of feature points obtained by ranging a surrounding environment including another ship with respect to an own ship. The area setting unit 30 sets individual areas for the plurality of feature points based on a distribution of positions of the plurality of feature points. The selection unit 50 selects a target tracking ship based on the size of the individual areas. The tracking unit 60 generates tracking data based on a time-series change of the positions of the target tracking ship.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a technique for tracking other ships. [Background technology]

[0002] Patent Document 1 describes a technique for tracking targets such as other ships using a radar. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2012-042343 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, radar cannot detect targets in small areas near the ship, such as in harbors or narrow waterways, and cannot track targets.

[0005] Therefore, an object of the present invention is to provide a technique capable of tracking a target even in a small area near the ship. [Means for solving the problem]

[0006] The ship tracking device of the present invention comprises an acquisition unit, an area setting unit, a selection unit, and a tracking unit. The acquisition unit acquires point cloud data composed of a plurality of characteristic points obtained by measuring the distance of the surrounding environment including other ships with respect to the own ship. The area setting unit sets individual areas for the plurality of characteristic points based on the distribution of the positions of the plurality of characteristic points. The selection unit selects a ship to be tracked based on the size of the individual area. The tracking unit generates tracking data based on the time series change in the position of the ship to be tracked.

[0007] In this configuration, the ship tracking device can detect targets such as land and ships at close ranges that radar cannot. Furthermore, the ship tracking device can select a ship (tracking target) from the targets. And, by detecting changes in the ship's position, the ship tracking device can generate ship tracking data. This allows the ship tracking device to track a ship (target) even in a small area near the ship itself. Furthermore, in the ship tracking device of the present invention, the characteristic points are represented by three-dimensional position coordinates.

[0008] This configuration shows, for example, an aspect using LiDAR, and each object surface including a ship (target) can be detected in three-dimensional position, improving detection performance.

[0009] In addition, in the ship tracking device of the present invention, the area setting unit sets the multiple individual areas using three-dimensional position coordinates, and the tracking unit generates tracking data based on two-dimensional position coordinates generated by converting the three-dimensional position coordinates into two dimensions.

[0010] With this configuration, the load of the tracking process can be reduced without reducing the tracking accuracy of the ship to be tracked.

[0011] In addition, in the ship tracking device of the present invention, the region setting unit includes a storage unit, a positioning unit, and a setting unit. The storage unit stores point cloud data at multiple times. The positioning unit uses SLAM processing to perform positioning of feature points at multiple times included in the point cloud data at multiple times. The setting unit sets individual regions based on the aligned feature points at multiple times.

[0012] With this configuration, the vessel tracking device can align the characteristic points at multiple times with high accuracy, thereby enabling the vessel tracking device to set the individual regions with high accuracy.

[0013] The ship tracking device of the present invention further comprises a classification unit that classifies the types of the individual regions. The selection unit selects a ship to be tracked based on the classified types.

[0014] In this configuration, a ship to be tracked can be selected from a number of individual regions before tracking is performed, allowing the ship tracking device to perform tracking processing only on ships that require tracking.

[0015] Furthermore, in the ship tracking device of the present invention, the positioning unit selects whether or not to use SLAM processing depending on the proportion of individual areas classified as land in the detection area.

[0016] In this configuration, the vessel tracking device can determine whether SLAM processing is enabled or disabled, which makes it possible to set a different process when SLAM processing is disabled.

[0017] Furthermore, in the ship tracking device of the present invention, the position alignment unit performs position alignment without using SLAM processing when the ratio is equal to or less than the ratio threshold value.

[0018] In this configuration, the vessel tracking device can be aligned to a given accuracy even when SLAM processing is not enabled.

[0019] The vessel tracking device of the present invention also includes a vessel information measuring unit that measures the vessel's position coordinates and attitude. The positioning unit performs positioning using the vessel's position and attitude when SLAM processing is not used.

[0020] In this configuration, the vessel tracking device can align with high accuracy even when SLAM processing is not enabled.

[0021] Also, in the ship tracking device of the present invention, the region setting section sets an individual region for each set of multiple feature points where the distance between adjacent feature points at the same time is equal to or less than the region setting threshold value.

[0022] In this configuration, an individual region is set by a plurality of adjacent feature points, which allows the vessel tracking device to appropriately set the individual region.

[0023] In addition, in the ship tracking device of the present invention, the classification unit classifies the type of the individual area into land if the size of the individual area is equal to or larger than the classification threshold, and into ship if the size of the individual area is less than the classification threshold.

[0024] In this configuration, the vessel tracking device can properly classify land and vessels.

[0025] In addition, in the ship tracking device of the present invention, if there is an individual area classified as land between the individual area to be classified and the ship, the classification unit classifies the type of the individual area to be classified as land.

[0026] In this configuration, the vessel tracking device is able to properly classify land.

[0027] Furthermore, in the ship tracking device of the present invention, if the number of feature points constituting an individual region is less than the noise classification threshold, the classification unit classifies the individual region as noise.

[0028] This configuration allows proper classification of land, tracked ships, and noise.

[0029] In addition, in the ship tracking device of the present invention, the classification unit converts the three-dimensional positions of feature points of individual areas that are not classified as noise into two-dimensional positions, and classifies the type of individual area into ship or land based on the converted individual area.

[0030] With this configuration, noise can be properly classified into land and tracked ships, and further, land and tracked ships can be properly classified while suppressing a decrease in classification accuracy.

[0031] In the vessel tracking device of the present invention, the tracking unit further includes a determination unit that determines whether the individual regions at the multiple times are due to the same target based on the position coordinates at the multiple times. The tracking unit calculates time-series changes in the position coordinates of the same target.

[0032] In this configuration, the vessel tracking device can accurately determine individual areas at multiple times that correspond to one target, and can suppress erroneous tracking.

[0033] Also, in the ship tracking device of the present invention, the tracking unit includes a shape calculation unit and a change amount calculation unit. The shape calculation unit calculates the vertical and / or horizontal lengths of individual areas at multiple times that have been determined to be the same target. The change amount calculation unit calculates the amount of change in the area at multiple times. If the amount of change in the vertical and / or horizontal lengths at multiple times is equal to or greater than a change amount threshold, the tracking unit does not calculate the time series change in the position coordinates of the same target.

[0034] In this configuration, the vessel tracking device will stop tracking when the shape (length and / or width) of the tracked object changes significantly, which allows the vessel tracking device to suppress suspicious tracking.

[0035] In addition, in the ship tracking device of the present invention, the tracking unit does not generate tracking data for a tracked ship corresponding to an individual area that includes a feature point included in the farthest area of ​​a detection area based on the own ship, or a feature point the number of times of acquisition of which is equal to or less than a count threshold.

[0036] In this configuration, the ship tracking device does not use feature points with low reliability, thereby improving tracking accuracy.

[0037] The vessel tracking device of the present invention also includes a display data generating unit that generates display data based on the tracking data.

[0038] In this configuration, the vessel tracking device can provide the results of tracking of the vessel (target) (including, for example, heading, wake, etc.) in a displayable form.

[0039] The vessel tracking device of the present invention also includes a display device that displays the display data.

[0040] In this configuration, the vessel tracking device can display the results of tracking the vessel (target) (including, for example, the direction of travel and the wake, etc.), allowing the user to easily visually confirm the results of tracking the vessel (target).

[0041] In addition, in the ship tracking device of the present invention, the display data generation unit generates display data in which the display form is changed depending on the state of the tracked ship.

[0042] In this configuration, the vessel tracking device can display the results of vessel (target) tracking so that the user can easily view them.

[0043] The present invention also provides an automatic ship maneuvering system including a ship tracking device and a navigation control device having any of the above-mentioned configurations. The navigation control device performs automatic navigation control to follow the tracked ship based on the tracking data, or to avoid collision with the tracked ship.

[0044] In this configuration, the automatic ship steering system can track the tracked ship with high accuracy using the ship tracking device. Therefore, the navigation control device can track the tracked ship with high accuracy or more reliably avoid collision with the tracked ship. [Brief description of the drawings]

[0045] [Figure 1] FIG. 1 is a functional block diagram of a vessel tracking device according to a first embodiment of the present invention. [Diagram 2] FIG. 2 is a functional block diagram of the region setting unit according to the first embodiment of the present invention. [Diagram 3] FIG. 3(A) is a plan view showing an example of sea conditions in a narrow waterway including the ship, and FIG. 3(B) is a plan view showing an example of the distribution of feature points in the situation of FIG. 3(A). [Figure 4] FIG. 4 is a diagram showing an example of region division. [Diagram 5] FIG. 5 is a functional block diagram of the classification unit according to the first embodiment of the present invention. [Figure 6] FIG. 6 is a diagram illustrating an example of classification. [Figure 7] FIG. 7 is a functional block diagram of a tracking unit according to the first embodiment of the present invention. [Figure 8] FIG. 8(A) is a diagram showing an example of determining whether the ships are the same, and FIG. 8(B) is a diagram showing an example of determining whether the ships are different. [Figure 9] FIG. 9(A) is a diagram showing an example of continuing tracking, and FIG. 9(B) is a diagram showing an example of stopping tracking. [Figure 10] FIG. 10A is a diagram showing an example of sea conditions that serve as a criterion for determining whether or not SLAM processing is performed, and FIG. 10B is a diagram showing an example of sea conditions that serve as a criterion for determining whether or not SLAM processing is performed. [Figure 11] FIG. 11 is a flowchart showing an example of a vessel tracking method according to the first embodiment of the present invention. [Figure 12] FIG. 12 is a flowchart showing an example of a region division method. [Figure 13] FIG. 13 is a flowchart showing an example of a registration method. [Figure 14] FIG. 14 is a flowchart showing an example of a classification method. [Figure 15] FIG. 15 is a flow chart illustrating an example of a tracking method. [Figure 16] FIG. 16 is a flowchart showing an example of a method for determining a tracking target. [Figure 17] FIG. 17 is a flowchart showing an example of a method for selecting whether to continue tracking or stop tracking. [Figure 18] FIG. 18 is a functional block diagram of a vessel tracking device according to the second embodiment of the present invention. [Figure 19] FIG. 19 is a diagram showing an example of the display. [Figure 20] FIG. 20 is a functional block diagram of a tracking unit in a vessel tracking device according to a third embodiment of the present invention. [Figure 21] FIG. 21 is a functional block diagram of an automatic ship steering system according to the fourth embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0046] [First embodiment] A ship tracking technology according to a first embodiment of the present invention will be described with reference to the drawings.

[0047] (General configuration and general process of the ship tracking device 10) Fig. 1 is a functional block diagram of a ship tracking device according to a first embodiment of the present invention. As shown in Fig. 1, the ship tracking device 10 includes a LiDAR 101, a ship information measurement unit 102, an acquisition unit 20, a region setting unit 30, a classification unit 40, a selection unit 50, and a tracking unit 60.

[0048] The LiDAR 101 and the own ship information measurement unit 102 are functionally included in the ship tracking device 10, but the LiDAR 101 and the own ship information measurement unit 102 are separate from the functional units from the acquisition unit 20 onwards. The functional units from the acquisition unit 20 onwards are configured, for example, by arithmetic processing devices such as various types of computer devices. Furthermore, the ship tracking device 10 can be configured not to include the LiDAR 101 and the own ship information measurement unit 102, so long as it is configured to acquire point cloud data and own ship information from outside.

[0049] The LiDAR 101 is a device that performs light detection and ranging. The LiDAR 101 irradiates a detection area around the ship while scanning the detection laser light in two dimensions. The LiDAR 101 receives the reflected light of the detection laser light reflected by an object. The LiDAR 101 measures the distance and direction of an object from the device by using the irradiation time of the laser light, the reception time of the reflected light, and the scan angle.

[0050] The LiDAR 101 detects points where the reception intensity of reflected light is equal to or greater than a threshold value, and sets the points as feature points. The LiDAR 101 outputs the reception intensity (reflected light intensity) of the reflected light from the feature point (reflection point) and the three-dimensional position coordinates of the feature point as feature point data. The LiDAR 101 generates and outputs feature point data for multiple feature points within the detection area.

[0051] A collection of a plurality of feature point data is point cloud data, and the LiDAR 101 outputs a collection of a plurality of feature point data obtained by one laser light scan as point cloud data for one timing.

[0052] The ship information measurement unit 102 includes a positioning sensor and an attitude sensor, detects the position and attitude of the ship, and outputs the three-dimensional position coordinates and attitude of the ship as ship information.

[0053] The acquisition unit 20 acquires point cloud data from the LiDAR 101, and acquires own ship information (position coordinates and attitude) from the own ship information measurement unit 102. The acquisition unit 20 acquires the point cloud data and the own ship information at multiple times. The acquisition unit 20 outputs the point cloud data and the own ship information at multiple times to the area setting unit 30.

[0054] The area setting unit 30 divides the feature point data included in the point cloud data into a plurality of individual areas based on the distribution of position coordinates of the feature points included in the point cloud data. An individual area is a partial area that is set by dividing the entire detection area. In other words, the area setting unit 30 sets each of the feature point data so that it is included in one of the plurality of individual areas.

[0055] The area setting section 30 outputs the divided individual areas to the classification section 40. Information contained in the individual areas includes a plurality of feature point data set (assigned) to each individual area.

[0056] The classification unit 40 calculates the size of each individual region and classifies the type of the individual region based on the size. The size of an individual region is the area of ​​the individual region projected onto a horizontal plane. The classification unit 40 sets a classification threshold, compares the area of ​​the individual region with the classification threshold, and classifies the type of the individual region based on the comparison result. For example, there are three types of classification: land, ship, and noise.

[0057] The classification section 40 outputs the individual regions and their classifications to the selection section 50. The classifications of the individual regions are fed back to the region setting section 30.

[0058] The selection unit 50 selects the individual areas classified as ships.

[0059] The tracking unit 60 generates tracking data based on time-series changes in the position coordinates of the tracked ship, with the individual areas classified as ships as the tracked ships. At this time, the tracking unit 60 converts the position coordinates in the three-dimensional coordinate system into a two-dimensional coordinate system parallel to the horizontal plane to generate the tracking data.

[0060] In this way, the vessel tracking device 10 can detect a vessel (tracking target) at multiple times using short-distance optical ranging technology. This allows the vessel tracking device 10 to track a vessel (tracking target) even in a small area near the vessel. Note that the tracking target is not limited to a vessel, and may be an obstacle on the water.

[0061] (Specific configurations and processing of each part of the ship tracking device 10) (Area setting unit 30) FIG. 2 is a functional block diagram of the region setting unit according to the first embodiment of the present invention.

[0062] Fig. 3(A) is a plan view showing an example of sea conditions in a narrow waterway including the ship, and Fig. 3(B) is a plan view showing an example of the distribution of characteristic points in the situation of Fig. 3(A). Note that in the following drawings including Fig. 3(B), characteristic points are indicated by circles, but for ease of visibility and understanding, they are drawn enlarged and thinned out appropriately to provide wider intervals.

[0063] Fig. 4 is a diagram showing an example of region division, in the sea state shown in Fig. 3(A).

[0064] As shown in FIG. 2, the region setting unit 30 includes a storage unit 31, a position adjustment unit 32, and a setting unit 33.

[0065] The accumulation unit 31 receives point cloud data from a plurality of times and own ship information (position coordinates and attitude of the own ship) from a plurality of times.

[0066] For example, in the case shown in Figure 3(A), the own ship VSLo is navigating a narrow waterway NC0. Lands LD1 and LD2 exist in the vicinity of the own ship VSLo, and in the narrow waterway NC0 sandwiched between the land LD1 and LD2, there exist ships (other ships) VSL1, VSL2, and VSL3 different from the own ship VSLo.

[0067] In the case of Fig. 3(A), as shown in Fig. 3(B), approximately point-like reflected light is obtained from the edges of the land LD1 and LD2, buildings on the land LD1, and the ships VSL1, VSL2, and VSL3. The points that produce this approximately point-like reflected light are characteristic points, and are distributed according to the shapes of the land LD1 and LD2, and the shapes and attitudes of the ships VSL1, VSL2, and VSL3.

[0068] Point cloud data configured with feature points distributed in this manner (feature point data: received light intensity of reflected light, three-dimensional position coordinates) is input to the storage unit 31.

[0069] The storage unit 31 stores the point cloud data and the ship information in that order. At this time, the storage unit 31 stores the point cloud data and the ship information of the same time in association with each other.

[0070] The storage unit 31 outputs the stored point cloud data for multiple times and the ship information to the positioning unit 32.

[0071] The alignment unit 32 corrects the position coordinates (three-dimensional position coordinates) of the feature point data at multiple times that each make up the point cloud data at multiple times, using the ship's own information (the ship's own position coordinates and attitude) at the time each feature point data was acquired.

[0072] Furthermore, if the situation is such that it is effective to use SLAM processing, the alignment unit 32 aligns feature point data at multiple times using Simulaneous Localization and Mapping (SLAM) processing.

[0073] If the situation does not make it effective to use SLAM processing, the positioning unit 32 performs positioning of feature point data at multiple times using only the ship information and chart information without using SLAM processing. The ship information includes the ship's position and attitude. The ship's position and attitude are measured by a ship information measurement unit (not shown) using, for example, a GPS or the like.

[0074] The specific method for selecting the SLAM process will be described later.

[0075] The position alignment unit 32 outputs the feature point data at multiple times after the position alignment to the setting unit 33.

[0076] The setting unit 33 performs division setting into a plurality of individual regions based on the aligned feature point data at a plurality of times.

[0077] More specifically, the setting unit 33 calculates the distance between adjacent feature points for a plurality of feature point data for the same time. The setting unit 33 stores a region setting threshold value in advance.

[0078] The setting unit 33 sets an individual area for each set of multiple feature point data in which the distance between adjacent feature points is equal to or less than a threshold for setting an area.

[0079] For example, in the case of FIG. 4, the setting unit 33 detects a group in which the distance between adjacent feature points is equal to or smaller than the region setting threshold, and sets a plurality of individual regions RG1-RG9.

[0080] The setting unit 33 outputs the set individual regions to the classification unit 40.

[0081] (Classification section 40) FIG. 5 is a functional block diagram of the classification unit according to the first embodiment of the present invention.

[0082] As shown in FIG. 5, the classification unit 40 includes a noise classification unit 41, a conversion unit 42, and a type discrimination unit 43.

[0083] The noise classification unit 41 measures the number of feature points constituting each individual region for each of the individual regions, and sets a noise classification threshold for the number of feature points.

[0084] If the number of feature points constituting an individual region is less than a noise classification threshold, the noise classification unit 41 classifies the individual region as noise.

[0085] If the number of feature points constituting an individual region is equal to or greater than a noise classification threshold, the noise classification unit 41 classifies this individual region as not being noise.

[0086] The noise classification unit 41 outputs the individual regions classified as not being noise to the conversion unit 42.

[0087] The conversion unit 42 converts the three-dimensional position coordinates of the multiple feature point data constituting the input individual area into two-dimensional position coordinates. More specifically, the conversion unit 42 projects the multiple feature point data represented in the three-dimensional coordinate system constituting the individual area onto a horizontal plane and converts it into a two-dimensional coordinate system parallel to the horizontal plane. The conversion unit 42 outputs the individual area converted into the two-dimensional position coordinates to the type discrimination unit 43.

[0088] The type discrimination unit 43 calculates the area (size) of the multiple individual regions that are not classified as noise. Since the individual regions are converted into two-dimensional position coordinates, the area of ​​the multiple individual regions is the area of ​​the multiple individual regions projected onto a horizontal plane.

[0089] The type discrimination unit 43 stores a classification threshold value. The classification threshold value is set, for example, so as to classify the individual areas into "land" and "ship" in descending order of size.

[0090] The type discrimination unit 43 compares the area of ​​the multiple individual regions not classified as noise with a classification threshold. If the area is less than the classification threshold, the type discrimination unit 43 classifies the type of this individual region as "ship." If the area is equal to or greater than the classification threshold, the type discrimination unit 43 classifies the type of this individual region as "land."

[0091] Furthermore, if there is an individual area classified as "land" between the individual area to be classified and the ship, the type discrimination unit 43 classifies the type of the individual area to be classified this time as "land".

[0092] Fig. 6 is a diagram showing an example of classification, taken in the sea state shown in Fig. 3(A).

[0093] 6, for example, the type discrimination unit 43 calculates the areas S1-S9 of the multiple individual regions RG1-RG9 that are not classified as noise. The type discrimination unit 43 classifies the multiple individual regions RG1-RG9 into "land" or "ship" based on the areas S1-S9.

[0094] More specifically, the type discrimination unit 43 compares the areas S1-S9 with the classification threshold Sth1. The type discrimination unit 43 detects that the areas S1, S2, S3, S5, S6, and S7 are greater than the classification threshold Sth1, and classifies the individual regions RG1, RG2, RG3, RG5, RG6, and RG7 as "land." The type discrimination unit 43 detects that the areas S4, S8, and S9 are smaller than the classification threshold Sth1, and classifies the individual regions RG4, RG8, and RG9 as "ships."

[0095] In this case, an individual area RG5 classified as "land" exists between the individual areas RG6, RG7 and the own ship VSLo. The type discrimination unit 43 detects that the individual area RG5 exists between the individual areas RG6, RG7 and the own ship VSLo from the position coordinates of the individual areas RG5, RG6, RG7 (for example, the position coordinates of the center points of each individual area) and the position coordinates of the own ship VSLo.

[0096] In this case, the type discrimination unit 43 classifies the multiple individual regions RG6 and RG7 as "land" even if the areas S6 and S7 of the multiple individual regions RG6 and RG7 are smaller than the main classification threshold Sth1.

[0097] By carrying out such processing, the classification unit 40 can classify a plurality of individual regions with high accuracy.

[0098] The classification unit 40 outputs the individual areas classified as "land" or "ship" to the selection unit 50. The individual areas output from the classification unit 40 include the area and the classified type.

[0099] (Selection unit 50) The selection unit 50 selects the individual areas classified as “ships” and outputs them to the tracking unit 60 .

[0100] (Tracking Section 60) FIG. 7 is a functional block diagram of a tracking unit according to the first embodiment of the present invention.

[0101] The tracking unit 60 includes a determination unit 61 , a tracking data generation unit 62 , a shape calculation unit 63 , and a change amount calculation unit 64 .

[0102] The determination unit 61 determines whether or not individual regions at multiple times are due to the same target, based on position coordinates at multiple times.

[0103] More specifically, the determination unit 61 calculates position coordinates of individual regions at multiple times used for the determination. The position coordinates of an individual region are, for example, the position coordinates of the center of the smallest outer shape that contains the multiple feature point data constituting the individual region. The outer shape is, for example, a rectangle. The position coordinates of the center can be geometrically calculated from the position coordinates of the multiple feature point data.

[0104] The determination unit 61 calculates the amount of change in the position coordinates of the individual regions calculated at multiple times (difference in position coordinates in a two-dimensional coordinate system).

[0105] The determination unit 61 stores a threshold value for determining whether or not the same target object exists. The threshold value for determining whether or not the same target object exists is set based on the upper limit of the normal navigation speed of a ship in a narrow waterway.

[0106] The determination unit 61 determines that the individual regions at the multiple times for which the amount of change in position coordinates was calculated are the same target (ship) if the amount of change in position coordinates is equal to or less than the threshold for determining the same target. The determination unit 61 determines that the individual regions at the multiple times for which the amount of change in position coordinates was calculated are different targets (ships) if the amount of change in position coordinates is greater than the threshold for determining the same target.

[0107] FIG. 8(A) is a diagram showing an example of determining whether the ships are the same, and FIG. 8(B) is a diagram showing an example of determining whether the ships are different.

[0108] 8(A) and 8(B), for example, the individual area RG at the first time (ni) and the individual area RG at the second time (n) are targeted. The determination unit 61 calculates the amount of change ΔP(n) between the position coordinate P(ni) of the individual area RG at the first time (ni) and the position coordinate P(n) of the individual area RG at the second time (n).

[0109] 8A, the amount of change ΔP(n) is smaller than the threshold value ΔPth for determining whether the target is the same. Therefore, the determination unit 61 determines that the individual region RG at the first time (ni) and the individual region RG at the second time (n) are the same target (ship).

[0110] 8B, the amount of change ΔP(n) is greater than the threshold value ΔPth for determining whether the target is the same as the target, and the determination unit 61 therefore determines that the individual area RG at the first time (ni) and the individual area RG at the second time (n) are different targets (ships).

[0111] The determination unit 61 outputs the individual areas at multiple times that have been determined to be the same target to the tracking data generation unit 62 and the shape calculation unit 63. Note that the determination unit 61, for example, assigns identification information that enables the individual areas at multiple times that have been determined to be the same target to be recognized as the same target.

[0112] The tracking data generating unit 62 calculates the velocity vector of the tracked ship by treating the individual areas (ships) at multiple times that are determined to be the same target as the tracked ship. The tracking data generating unit 62 generates tracking data including the identification information, position coordinates, and velocity vector of the tracked ship.

[0113] With this configuration, the tracking unit 60 can properly select and track the ship to be tracked.

[0114] In this case, the tracking unit 60 uses an individual area represented by two-dimensional position coordinates. Here, since the ship to be tracked moves on a horizontal plane, even if the ship to be tracked is tracked in two dimensions, the same accuracy as when the ship to be tracked is tracked in three dimensions can be ensured. On the other hand, processing in two dimensions has a lower processing load than processing in three dimensions because there is one less dimension. This allows the tracking unit 60 to reduce the processing load without reducing the tracking accuracy of the ship to be tracked. Therefore, the tracking unit 60 can generate tracking data by simpler and faster processing than when using position coordinates in a three-dimensional coordinate system.

[0115] In addition, since the display of the tracked ship (described later) is two-dimensional, even if the tracking is performed in two dimensions, the deterioration of the accuracy of the display of the tracked ship can be suppressed. Therefore, the ship tracking device can suppress the deterioration of the display accuracy of the tracked ship while reducing the processing load.

[0116] The shape calculation unit 63 calculates the vertical length and / or horizontal length of each of the individual regions at multiple times that are determined to be the same target.

[0117] More specifically, the shape calculation unit 63 calculates the vertical length and / or horizontal length of the smallest outer shape that contains the multiple feature point data that constitute an individual region. The outer shape is, for example, a rectangle that is set to have sides parallel to two orthogonal axes that constitute a two-dimensional coordinate system. The vertical length and / or horizontal length can be geometrically calculated from the position coordinates of the multiple feature point data. In this case, by setting the individual region using a rectangle with sides parallel to the two orthogonal axes, the vertical length and / or horizontal length of the individual region can be calculated by a simple calculation.

[0118] The shape calculation section 63 outputs the calculated vertical length and / or horizontal length of each individual region at each time to the change amount calculation section 64.

[0119] The change amount calculation unit 64 calculates the change amount of the vertical length and / or the horizontal length of the individual areas at multiple times that are determined to be the same target. Specifically, for example, the change amount calculation unit 64 calculates the absolute value of the change amount of the vertical length and / or the horizontal length.

[0120] The change amount calculation unit 64 outputs the change amount of the vertical length and / or the horizontal length to the tracking data generation unit 62.

[0121] The tracking data generating unit 62 stores a threshold value for the amount of change.

[0122] If the change in the vertical length and the horizontal length is less than the threshold for the change, the tracking data generation unit 62 calculates the speed vector of the individual area (tracked ship) of this identification information and generates tracking data. In other words, the tracking unit 60 continues tracking the individual area (ship).

[0123] If the change in the vertical length and / or horizontal length is equal to or greater than the threshold for the change, the tracking data generating unit 62 does not calculate the velocity vector of the individual area (tracked ship) of this identification information and stops generating tracking data. In other words, the tracking unit 60 stops tracking the individual area (tracked ship).

[0124] FIG. 9(A) is a diagram showing an example of continuing tracking, and FIG. 9(B) is a diagram showing an example of stopping tracking.

[0125] For example, in the case of Figures 9(A) and 9(B), the individual areas RG at the first time (ni) and the individual areas RG at the second time (n) that are determined to be the same target by the determination unit 61 are targeted.

[0126] The shape calculation unit 63 calculates the vertical length Y(ni) and horizontal length X(ni) of the individual area RG at a first time (ni), and the vertical length Y(n) and horizontal length X(n) of the individual area RG at a second time (n).

[0127] The change amount calculation unit 64 calculates the change amount ABS(Yn) between the vertical length Y(ni) and the vertical length Y(n). The change amount calculation unit 64 calculates the change amount ABS(Xn) between the horizontal length X(ni) and the horizontal length X(n).

[0128] 9A, the amount of change ABS(ΔYn) and the amount of change ABS(Xn) are smaller than the amount of change threshold Δth, so the tracking data generator 62 generates tracking data and continues tracking.

[0129] In the case of FIG. 9B, the amount of change ABS (ΔYn) is greater than the change amount threshold value Δth. Therefore, the tracking data generation unit 62 stops generating tracking data and stops tracking. The tracking data generation unit 62 also stops generating tracking data and stops tracking when the amount of change ABS (ΔXn) is greater than the change amount threshold value Δth.

[0130] With this configuration, the tracking unit 60 can suppress erroneous tracking.

[0131] The tracking section 60 may use the amount of change in the area of ​​an individual region, instead of the amount of change in the vertical length and / or horizontal length of the individual region, to determine whether to continue or stop tracking.

[0132] (Specific process for selecting whether or not to perform SLAM processing) The selection of whether or not to perform the SLAM processing by the above-mentioned registration unit 32 is made depending on the proportion of individual areas classified as land by the classification unit 40.

[0133] The classification unit 40 outputs the classification result to the selection unit 50 and also feeds it back to the region setting unit 30 .

[0134] Fig. 10(A) is a diagram showing an example of sea conditions that are the criterion for judgment with SLAM processing, and Fig. 10(B) is a diagram showing an example of sea conditions that are the criterion for judgment without SLAM processing. Fig. 10(A) and Fig. 10(B) are expressed using feature points.

[0135] 10A, land and ships are mixed in the entire detection area of ​​the LiDAR 101, and land dominates. Therefore, land occupies a large proportion of the entire detection area.

[0136] 10B, there are many ships and few land areas in the entire detection area of ​​the LiDAR 101. Therefore, the proportion of land areas in the entire detection area is small.

[0137] The positioning unit 32 acquires the area used when the classification unit 40 classifies land and ship. The positioning unit 32 may calculate the area of ​​land and the area of ​​ship depending on whether or not SLAM processing is performed. The positioning unit 32 stores the total area Sall of the detection region in advance.

[0138] The position adjustment unit 32 calculates the total land area Sln. The position adjustment unit 32 calculates the ratio of the total land area Sln to the total area Sall.

[0139] The position alignment unit 32 stores the ratio threshold value Sth. The position alignment unit 32 compares the ratio (Sln / Sall) with the ratio threshold value Sth.

[0140] If the ratio (Sln / Sall) is greater than the ratio threshold Sth, the alignment unit 32 performs SLAM processing (SLAM processing is performed). For example, in the case of FIG. 10(A), the total land area Sln is S1+S2+S3+S5+S6+S7, and the ratio (Sln / Sall) is large. Therefore, the ratio (Sln / Sall) is greater than the ratio threshold Sth, and SLAM processing is performed.

[0141] If the ratio (Sln / Sall) is equal to or less than the ratio threshold Sth, the alignment unit 32 does not perform SLAM processing (no SLAM processing). For example, in the case of FIG. 10B, the total land area Sln is S11+S12, and the ratio (Sln / Sall) is small. Therefore, the ratio (Sln / Sall) is equal to or less than the ratio threshold Sth, and SLAM processing is not performed.

[0142] SLAM processing can be expected to improve alignment accuracy if there are many targets whose positions are invariant, such as land, whose absolute position coordinates (position coordinates based on the earth) do not change. However, if there are few targets whose positions are invariant, alignment accuracy decreases.

[0143] Therefore, by not performing SLAM processing when the proportion of land is small and performing SLAM processing when the proportion of land is large, the alignment unit 32 can improve the alignment accuracy through SLAM processing when the proportion of land is large, and can suppress the decrease in accuracy due to performing SLAM processing when the proportion of land is small.

[0144] The total area Sall of the detection area may be the total area detected by the LiDAR 101, or may be the total area of ​​the detected individual areas. In other words, the total area Sall may be the total area of ​​the land and the total area of ​​the ship.

[0145] The position alignment unit 32 may also determine whether or not to use SLAM processing based on the number of classified land areas. For example, the position alignment unit 32 performs SLAM processing if the number of land areas is equal to or greater than a threshold for determining whether SLAM is applied, and does not perform SLAM processing if the number of land areas is less than the threshold for determining whether SLAM is applied.

[0146] (Ship tracking method) FIG. 11 is a flowchart showing an example of a ship tracking method according to the first embodiment of the present invention. FIG. 12 is a flowchart showing an example of an area setting method. FIG. 13 is a flowchart showing an example of an alignment method. FIG. 14 is a flowchart showing an example of a classification method. FIG. 15 is a flowchart showing an example of a tracking method. FIG. 16 is a flowchart showing an example of a method of determining a tracking target. FIG. 17 is a flowchart showing an example of a method of selecting whether to continue tracking or stop tracking.

[0147] The vessel tracking method is, for example, programmed (vessel tracking program) and stored in a storage medium, etc. The arithmetic processing device reads out the vessel tracking program from the storage medium, etc., and executes it. In this way, the vessel tracking method is realized.

[0148] Note that the specific contents of each process shown in FIG. 11 to FIG. 17 have already been explained in the above description of the configuration and process, and therefore explanations will be omitted except for points where additional explanation is necessary.

[0149] (Overall processing: Figure 11) As shown in Fig. 11, the arithmetic processing device acquires point cloud data (S10). The arithmetic processing device sets individual regions of the point cloud data (S20). The arithmetic processing device classifies the individual regions (S30). Specifically, the arithmetic processing device classifies the types of the individual regions into land and ship.

[0150] The arithmetic processing device selects a ship to be tracked (S40).The arithmetic processing device tracks the ship to be tracked (S50).

[0151] (Area setting process: Figure 12) 12, the arithmetic processing device accumulates point cloud data at multiple times (S21). The arithmetic processing device aligns feature point data at multiple times included in the point cloud data at multiple times (S22).

[0152] The calculation processing device sets individual regions based on the aligned feature point data at multiple times (S23).

[0153] (Alignment process: Figure 13) As shown in FIG. 13, if classification information is present (S221: YES), the calculation processing device calculates the land ratio (S222).

[0154] If the land ratio is not equal to or less than the ratio threshold (S223: NO), the calculation processing device performs alignment using SLAM processing (S224).

[0155] If the land ratio is equal to or less than the ratio threshold (S223: YES), the calculation processing device performs alignment without using SLAM processing (S225).

[0156] If there is no classification information (S221: NO), the calculation processing device first performs alignment using SLAM processing (S224).

[0157] (Classification process: Figure 14) As shown in Fig. 14, the arithmetic processing device counts the number of feature points that make up an individual region (S31). The arithmetic processing device compares the number of feature points with a noise classification threshold. If the number of feature points is less than the noise classification threshold (S32: YES), the arithmetic processing device classifies the individual region as noise (S301).

[0158] If the score is equal to or greater than the noise classification threshold (S32: NO), the arithmetic processing device determines that this individual region is not noise. The arithmetic processing device converts the three-dimensional position coordinates of the feature point data constituting the individual region determined not to be noise into two-dimensional position coordinates (S33).

[0159] The arithmetic processing device calculates the area of ​​each two-dimensionally converted individual region (S34). If the area is equal to or greater than the classification threshold (S35: YES), the arithmetic processing device classifies the individual region as "land" (S302).

[0160] If the area of ​​the individual area to be classified is less than the classification threshold (S35: NO) and the individual area classified as "land" is between the individual area to be classified and the ship (S36: YES), the calculation processing device classifies the individual area as "land" (S302).

[0161] If the area of ​​the individual area to be classified is less than the classification threshold (S35: NO) and the individual area classified as "land" is not between the individual area to be classified and the own ship (S36: NO), the calculation processing device classifies the individual area as a "ship" (S303).

[0162] (Tracking process: Figure 15) As shown in Fig. 15, the arithmetic processing device determines the ship to be tracked (S41). At this time, the arithmetic processing device specifically uses the process shown in Fig. 16. The arithmetic processing device generates tracking data based on the time-series changes in the position coordinates of the ship to be tracked (S42).

[0163] The arithmetic processing device determines whether to stop tracking the vessel (S43) using the process shown in Fig. 17. At this time, the arithmetic processing device specifically uses the process shown in Fig. 17. If the tracking stop condition is met (S44: YES), the arithmetic processing device stops tracking (S45). If the tracking stop condition is not met (S44: NO), the arithmetic processing device continues tracking (S46).

[0164] (Processing to determine the target to be tracked: Figure 16) 16, the arithmetic processing device acquires position coordinates of individual areas at multiple times (S421). The arithmetic processing device calculates differences in position coordinates of individual areas at multiple times (S422).

[0165] If the difference is equal to or less than the same target determination threshold (S423: YES), the arithmetic processing device determines that the target is the same target (tracking target) (S424). If the difference is not equal to or less than the same target determination threshold (S423: NO), the arithmetic processing device determines that the target is a different target (S425).

[0166] (Selection process for continuing or canceling tracking: Figure 17) 17, the arithmetic processing device calculates the vertical and / or horizontal lengths of individual areas of a tracking target at multiple times (S441). The arithmetic processing device calculates the amount of change in the vertical and / or horizontal lengths (S442).

[0167] If the amount of change is equal to or greater than the change amount threshold (S443: YES), the arithmetic processing device stops tracking (S444). If the amount of change is not equal to or greater than the change amount threshold (S443: NO), the arithmetic processing device continues tracking (S445).

[0168] [Second embodiment] A vessel tracking technique according to a second embodiment of the present invention will be described with reference to Fig. 18. Fig. 18 is a functional block diagram of a vessel tracking device according to the second embodiment of the present invention.

[0169] As shown in Fig. 18, the vessel tracking device 10A according to the second embodiment differs from the vessel tracking device 10 according to the first embodiment in that it has a display function. Other configurations of the vessel tracking device 10A according to the second embodiment are similar to those of the vessel tracking device 10 according to the first embodiment, and a description of similar parts will be omitted.

[0170] The vessel tracking device 10A includes a display data generating unit 70 and a display 700. Note that the display 700 may be separate from (constituted separately from) the vessel tracking device 10A.

[0171] The display data generating unit 70 generates display data based on the tracking data of the tracked ship output by the tracking unit 60. The display data includes a current position display mark and a trail display mark. The current position display mark and the trail display mark are generated based on the position coordinates and the velocity vector included in the tracking data.

[0172] For example, the current position mark is a simplified ship mark, and its position and orientation are set based on the position coordinates and the directional component of the velocity vector, while the wake mark is a bar-shaped mark, and its length is set based on the speed component of the velocity vector.

[0173] The display data generating unit 70 outputs the display data to the display device 700. The display device 700 displays the display data.

[0174] Fig. 19 is a diagram showing an example of a display, in which tracking data is generated in the sea state shown in the first embodiment, and display data is generated based on this tracking data.

[0175] As shown in FIG. 19, the display screen of the display device 700 displays current position display marks DVSL1, DVSL2, DVSL3 and trail display marks DW1, DW2, DW3 of a plurality of tracked ships based on the display data.

[0176] This allows the user to visually and easily grasp the current position, sailing speed, and sailing direction of the tracked ship. At this time, as shown in Fig. 19, for example, a ship position mark DVSLo is displayed based on ship information, so that the user can visually and easily grasp the current position, sailing speed, and sailing direction of the tracked ship relative to the ship. Furthermore, as shown in Fig. 19, for example, quay (land) marks DLD1 and DLD2 are displayed based on chart information, so that the user 90 can visually and easily grasp the current position, sailing speed, and sailing direction of the tracked ship relative to the ship in the narrow waterway NC0.

[0177] The display data generating unit 70 may generate display data in which the display mode is changed according to the state of the tracked ship. The state of the tracked ship includes at least one of the distance, direction, and speed from the ship itself. The display mode includes at least one of the color and size.

[0178] The display data generating unit 70 uses a different display color for each tracking target ship. The display data generating unit 70 also displays tracking target ships that are close to the ship in a different color from tracking target ships that are far away. The display data generating unit 70 also displays tracking target ships traveling at or above a predetermined speed in a different color from tracking target ships traveling at or below the predetermined speed.

[0179] Furthermore, the vessel tracking device 10A is provided with an operator that accepts an operation from the user, and the display data generating unit 70 causes the display of the tracking target ship selected by the operator to differ from other tracking target ships.

[0180] [Third embodiment] A vessel tracking technique according to a third embodiment of the present invention will be described with reference to Fig. 20. Fig. 20 is a functional block diagram of a tracking unit in a vessel tracking device according to the third embodiment of the present invention.

[0181] As shown in Fig. 20, the vessel tracking device according to the third embodiment differs from the vessel tracking device 10 according to the first embodiment in the configuration of the tracking unit 60B. Other configurations of the vessel tracking device according to the third embodiment are similar to those of the vessel tracking device 10 according to the first embodiment, and a description of similar parts will be omitted.

[0182] The tracking unit 60B differs from the tracking unit 60 according to the first embodiment in that it further includes a data generation determination unit 65 and in additional processing performed by a tracking data generation unit 62B. Other configurations and processing of the tracking unit 60B are similar to those of the tracking unit 60, and descriptions of similar parts will be omitted.

[0183] The data generation determination unit 65 determines, based on the position coordinates of the individual areas, that a tracking target ship corresponding to an individual area including feature point data included in the farthest area within the range in which point cloud data can be acquired, as a tracking stop target ship.

[0184] In addition, the data generation determination unit 65 determines that a tracking target ship corresponding to an individual area including characteristic point data whose number of acquisitions is equal to or less than the number-of-acquisition threshold is a tracking stop target ship.

[0185] The data generation determination unit 65 outputs information about the ship to be tracked (for example, the position coordinates of the individual area) to the tracking data generation unit 62B.

[0186] The tracking data generation unit 62B does not generate tracking data for ships that are subject to tracking discontinuation.

[0187] This allows the vessel tracking device to exclude vessels based on unreliable feature point data from tracking targets, thereby enabling the vessel tracking device to generate highly reliable tracking data.

[0188] [Fourth embodiment] An automatic ship steering technique according to a fourth embodiment of the present invention will be described with reference to the drawings. Fig. 21 is a functional block diagram of an automatic ship steering system according to the fourth embodiment of the present invention.

[0189] As shown in Fig. 21, the ship automatic steering system 1 includes a ship tracking device 10 and a navigation control device 2. The ship tracking device 10 has the configuration according to the first embodiment. Note that the ship tracking device 10 can also adopt the configurations according to the second and third embodiments described above.

[0190] The navigation control device 2 acquires tracking data from the ship tracking device 10. Based on the tracking data, the navigation control device 2 performs automatic navigation control so as to follow the tracked ship. Based on the tracking data, the navigation control device 2 also performs automatic navigation control so as to avoid collision with the tracked ship. The navigation control device 2 controls the rudder angle of the rudder 3 and the output of the thrust generator 4 by the automatic navigation control.

[0191] With this configuration, the automatic ship steering system 1 can track the tracking target ship with high accuracy based on the highly accurate tracking data of the tracking target ship. Also, the automatic ship steering system 1 can more reliably avoid collision with the tracking target ship.

[0192] <1> an acquisition unit that acquires point cloud data composed of a plurality of feature points obtained by measuring the distance of the surrounding environment including other ships based on the own ship; a region setting unit that sets individual regions for the plurality of feature points based on a distribution of positions of the plurality of feature points; A selection unit that selects a ship to be tracked based on the size of the individual area; a tracking unit that generates tracking data based on time-series changes in the position of the tracked ship; A vessel tracking device comprising:

[0193] <2> <1> A vessel tracking device comprising: The feature points are represented by three-dimensional position coordinates. Ship tracking device.

[0194] <3> <2> A vessel tracking device comprising: the region setting unit sets the plurality of individual regions using the three-dimensional position coordinates; The tracking unit generates the tracking data based on two-dimensional position coordinates generated by converting the three-dimensional position coordinates into two dimensions. Ship tracking device.

[0195] <4> <1> ~ <3> Any one of the vessel tracking devices described above, The region setting unit is A storage unit that stores the point cloud data at multiple times; a position alignment unit that aligns feature points at multiple times included in the point cloud data at multiple times using SLAM processing; a setting unit that sets the individual regions based on the feature points at the multiple time points that have been aligned; Equipped with Ship tracking device.

[0196] <5> <1> ~ <4> Any one of the vessel tracking devices described above, A classification unit that classifies the types of the individual regions, The selection unit selects a ship to be tracked based on the classified type. Ship tracking device.

[0197] <6> <4> A vessel tracking device comprising: The positioning unit selects whether to use SLAM processing depending on the ratio of individual areas classified as land to the detection area. Ship tracking device.

[0198] <7> <6> A vessel tracking device comprising: When the ratio is equal to or less than a ratio threshold, the alignment unit performs the alignment without using the SLAM process. Ship tracking device.

[0199] <8> <7> A vessel tracking device comprising: a ship information measuring unit for measuring a position and an attitude of the ship; When the SLAM processing is not used, the alignment unit performs the alignment using the position coordinates and attitude of the ship. Ship tracking device.

[0200] <9> <5> ~ <8> Any one of the vessel tracking devices described above, the region setting unit sets the individual region for each set of the plurality of feature points in which a distance between adjacent feature points at the same time is equal to or less than a region setting threshold; Ship tracking device.

[0201] <10> <5> ~ <9> Any one of the vessel tracking devices described above, The classification unit includes: If the size of the individual area is equal to or greater than a classification threshold, classify the type of the individual area as land; If the size of the individual region is less than a classification threshold, classify the type of the individual region as ship. Ship tracking device.

[0202] <11> <10> A vessel tracking device comprising: the classification unit classifies the type of the individual area to be classified as land when an individual area classified as land is present between the individual area to be classified and the ship; Ship tracking device.

[0203] <12> <5> ~ <11> Any one of the vessel tracking devices described above, The classification unit includes: If the number of feature points constituting the individual region is less than a noise classification threshold, the individual region is classified as noise. Ship tracking device.

[0204] <13> <12> A vessel tracking device comprising: The classification unit includes: converting the three-dimensional positions of the feature points of the individual regions not classified as noise into two-dimensional positions, and classifying the types of the individual regions into ship or land based on the individual regions after the conversion; Ship tracking device.

[0205] <14> <1> ~ <13> Any one of the vessel tracking devices described above, The tracking unit is A determination unit is further provided for determining whether or not individual regions at multiple times are due to the same target based on position coordinates at multiple times, Calculating a time series change in the position coordinates of the same target; Ship tracking device.

[0206] <15> <14> A vessel tracking device comprising: The tracking unit is a shape calculation unit that calculates the vertical or horizontal length of the individual regions at the multiple times that are determined to be the same target; A change amount calculation unit that calculates a change amount of the vertical or horizontal length at the multiple times; Equipped with If the amount of change in the vertical or horizontal length at the multiple times is equal to or greater than a threshold for the amount of change, calculation of the time series change in the position coordinates of the same target is not performed. Ship tracking device.

[0207] <16> <1> ~ <15> Any one of the vessel tracking devices described above, the tracking unit does not generate the tracking data for the tracked ship corresponding to the individual area including the feature point included in the farthest area of ​​a detection area based on the ship itself, or the feature point whose acquisition count is equal to or less than a count threshold; Ship tracking device.

[0208] <17> <1> ~ <16> Any one of the vessel tracking devices described above, a display data generating unit that generates display data based on the tracking data; Ship tracking device.

[0209] <18> <17> A vessel tracking device comprising: A display device for displaying the display data is provided. Ship tracking device.

[0210] <19> <17> or <18> A vessel tracking device comprising: The display data generating unit generates the display data in a different display mode depending on the state of the tracking target ship. Ship tracking device.

[0211] <20> <1> ~ <19> Any of the following vessel tracking devices: a navigation control device that performs automatic navigation control to follow the tracked ship or to avoid collision with the tracked ship based on the tracking data; An automatic ship steering system. [Explanation of symbols]

[0212] 1: Automatic ship steering system 2: Navigation control device 4: Thrust generator 10, 10A: Ship tracking device 20: Acquisition part 30: Area setting section 31: Storage section 32: Alignment section 33: Setting section 40: Classification section 41: Noise classification unit 42: Conversion section 43: Type discrimination section 50: Selection section 60, 60B: Tracking unit 61: Judgment section 62, 62B: Tracking data generation unit 63: Shape calculation section 64: Change amount calculation unit 65: Data generation determination unit 70: Display data generation unit 90: User 101: LiDAR 102: Ship information measurement section 700: Display

Claims

1. an acquisition unit that acquires point cloud data composed of a plurality of feature points obtained by measuring the distance of the surrounding environment including other ships based on the own ship; a region setting unit that sets individual regions for the plurality of feature points based on a distribution of positions of the plurality of feature points; A selection unit that selects a ship to be tracked based on the size of the individual area; a tracking unit that generates tracking data based on time-series changes in the position of the tracked ship; A vessel tracking device comprising:

2. 2. The vessel tracking device according to claim 1, The feature points are represented by three-dimensional position coordinates. Ship tracking device.

3. 3. The vessel tracking device according to claim 2, the region setting unit sets the plurality of individual regions using the three-dimensional position coordinates; The tracking unit generates the tracking data based on two-dimensional position coordinates generated by converting the three-dimensional position coordinates into two dimensions. Ship tracking device.

4. 2. The vessel tracking device according to claim 1, The region setting unit is A storage unit that stores the point cloud data at multiple times; a position alignment unit that aligns feature points at multiple times included in the point cloud data at multiple times using a SLAM process; a setting unit that sets the individual regions based on the feature points at the multiple time points that have been aligned; Equipped with Ship tracking device.

5. 2. The vessel tracking device according to claim 1, A classification unit that classifies the types of the individual regions, The selection unit selects a ship to be tracked based on the classified type. Ship tracking device.

6. The vessel tracking device according to claim 4, The positioning unit selects whether or not to use SLAM processing depending on the ratio of individual areas classified as land to the detection area. Ship tracking device.

7. 7. The vessel tracking device according to claim 6, When the ratio is equal to or less than a ratio threshold, the alignment unit performs the alignment without using the SLAM process. Ship tracking device.

8. 8. The vessel tracking device according to claim 7, a ship information measuring unit for measuring a position and an attitude of the ship; When the SLAM processing is not used, the alignment unit performs the alignment using position coordinates and attitude of the ship. Ship tracking device.

9. 6. The vessel tracking device according to claim 5, the region setting unit sets the individual region for each set of the plurality of feature points in which a distance between adjacent feature points at the same time is equal to or less than a region setting threshold; Ship tracking device.

10. 6. The vessel tracking device according to claim 5, The classification unit includes: If the size of the individual area is equal to or greater than a classification threshold, classify the type of the individual area as land; If the size of the individual region is less than a classification threshold, classify the type of the individual region as ship. Ship tracking device.

11. 11. The vessel tracking device according to claim 10, the classification unit classifies the type of the individual area to be classified as land when an individual area classified as land is present between the individual area to be classified and the ship; Ship tracking device.

12. 6. The vessel tracking device according to claim 5, The classification unit includes: If the number of feature points constituting the individual region is less than a noise classification threshold, the individual region is classified as noise. Ship tracking device.

13. 13. The vessel tracking device according to claim 12, The classification unit includes: converting the three-dimensional positions of the feature points of the individual regions not classified as noise into two-dimensional positions, and classifying the types of the individual regions into ship or land based on the individual regions after the conversion; Ship tracking device.

14. 2. The vessel tracking device according to claim 1, The tracking unit is A determination unit is further provided for determining whether or not individual regions at multiple times are caused by the same target based on position coordinates at multiple times, Calculating a time series change in the position coordinates of the same target; Ship tracking device.

15. 15. The vessel tracking device of claim 14, The tracking unit is a shape calculation unit that calculates the vertical or horizontal length of the individual regions at the multiple times that are determined to be the same target; A change amount calculation unit that calculates a change amount of the vertical or horizontal length at the multiple times; Equipped with If the amount of change in the vertical or horizontal length at the multiple times is equal to or greater than a threshold for the amount of change, calculation of the time series change in the position coordinates of the same target is not performed. Ship tracking device.

16. 2. The vessel tracking device according to claim 1, the tracking unit does not generate the tracking data for the tracked ship corresponding to the individual area including the feature point included in the farthest area of ​​a detection area based on the ship itself, or the feature point whose acquisition count is equal to or less than a count threshold; Ship tracking device.

17. 2. The vessel tracking device according to claim 1, a display data generating unit that generates display data based on the tracking data; Ship tracking device.

18. 18. The vessel tracking device of claim 17, A display device for displaying the display data is provided. Ship tracking device.

19. 18. The vessel tracking device of claim 17, The display data generating unit generates the display data in a different display mode depending on the state of the tracking target ship. Ship tracking device.

20. The vessel tracking device according to claim 1 ; a navigation control device that performs automatic navigation control to follow the tracked ship or to avoid collision with the tracked ship based on the tracking data; An automatic ship steering system.

21. Obtain point cloud data consisting of multiple characteristic points obtained by measuring the distance of the surrounding environment, including other ships, based on the ship itself, and setting the plurality of feature points as individual regions based on a distribution of the positions of the plurality of feature points; Selecting a ship to be tracked based on the size of the individual region; generating tracking data based on time series changes in the position of the tracked vessel; Ship tracking methods.

22. Obtain point cloud data consisting of multiple characteristic points obtained by measuring the distance of the surrounding environment, including other ships, based on the ship itself, and setting the plurality of feature points as individual regions based on a distribution of the positions of the plurality of feature points; Selecting a ship to be tracked based on the size of the individual region; generating tracking data based on time series changes in the position of the tracked vessel; A ship tracking program that causes a processing unit to execute processing.

Images