CN118050728B - Target acquisition method and system for channel safety monitoring - Google Patents
Target acquisition method and system for channel safety monitoring Download PDFInfo
- Publication number
- CN118050728B CN118050728B CN202410452982.1A CN202410452982A CN118050728B CN 118050728 B CN118050728 B CN 118050728B CN 202410452982 A CN202410452982 A CN 202410452982A CN 118050728 B CN118050728 B CN 118050728B
- Authority
- CN
- China
- Prior art keywords
- obstacle
- point
- path
- elevation
- points
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 89
- 238000000034 method Methods 0.000 title claims abstract description 57
- 238000005070 sampling Methods 0.000 claims abstract description 93
- 230000005856 abnormality Effects 0.000 claims abstract description 44
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 22
- 238000004364 calculation method Methods 0.000 claims description 13
- 238000005259 measurement Methods 0.000 claims description 12
- 238000013507 mapping Methods 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 6
- 238000004441 surface measurement Methods 0.000 claims description 2
- 230000002159 abnormal effect Effects 0.000 abstract description 24
- 230000035945 sensitivity Effects 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 abstract 1
- 238000012545 processing Methods 0.000 description 4
- 238000010606 normalization Methods 0.000 description 3
- 239000013535 sea water Substances 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000003064 k means clustering Methods 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010835 comparative analysis Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 238000002955 isolation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000000644 propagated effect Effects 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Landscapes
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention relates to the technical field of ship anti-collision, in particular to a target acquisition method and system for channel safety monitoring. Firstly, acquiring sea surface measuring points and obstacle measuring points in a radar elevation monitoring chart at each sampling time in the ship navigation process; then, acquiring dynamic expected factors of each obstacle measuring point, and further acquiring all obstacle avoidance paths at each sampling time; and acquiring threat coefficients of each obstacle measuring point according to the difference between obstacle avoidance paths, further acquiring the degree of abnormality of each obstacle measuring point, adjusting the corresponding local reachable density, and acquiring a final radar height Cheng Jiance diagram. According to the invention, the abnormal degree of each obstacle measuring point is judged by acquiring the dynamic expected factors and the threat coefficients, and then the local reachable density is adjusted according to the abnormal degree, so that the sensitivity of a local abnormal factor algorithm to noise is reduced, all abnormal obstacle measuring points are accurately detected, and the reliability of channel safety monitoring is improved.
Description
Technical Field
The invention relates to the technical field of ship anti-collision, in particular to a target acquisition method and system for channel safety monitoring.
Background
The navigation radar is a water area safety detection technology, various obstacles such as ships, buoys, reefs and the like in the navigation water area are monitored and tracked through the navigation radar, the related information such as the size, the speed, the elevation and the like of the surrounding obstacles during the navigation of the ship can be obtained, and then a corresponding obstacle avoidance strategy is generated, so that powerful safety anti-collision guarantee is provided for the navigation of the ship. However, radar signals can be interfered by atmosphere, ocean, other radar systems and the like, so that partial radar monitoring data are abnormal, channel safety monitoring and obstacle avoidance strategy formulation can be influenced, and certain potential safety hazards are provided.
The local anomaly factor algorithm can effectively identify the anomaly data in the radar monitoring data by judging the local reachable density of the radar monitoring data in the neighborhood, but because the sensitivity degree of the local anomaly factor algorithm to noise is higher, the relative noise can more easily ignore the anomaly data, so that the anomaly of the radar monitoring data is not accurately proposed, the accuracy of a fitted radar azimuth-distance graph is low, and the judgment of subsequent obstacles and the establishment of a safety obstacle avoidance strategy are influenced.
Disclosure of Invention
In order to solve the technical problem that the reliability of channel safety monitoring is low due to the fact that abnormal radar monitoring data cannot be accurately identified by the existing local abnormal factor algorithm, the invention aims to provide a target acquisition method and a target acquisition system for channel safety monitoring, and the adopted technical scheme is as follows:
The invention provides a target acquisition method for channel safety monitoring, which comprises the following steps:
Acquiring radar elevation monitoring graphs at each sampling time in the ship navigation process, and acquiring sea surface measuring points and obstacle measuring points in each radar elevation monitoring graph;
Acquiring a dynamic expected factor of each obstacle measuring point at each sampling time according to the distribution condition of elevation differences among pixel points at the same position among radar elevation monitoring graphs corresponding to adjacent sampling times; acquiring all obstacle avoidance paths at each sampling time according to the dynamic expected factors and the distances between the sea surface measuring points and the obstacle measuring points and combining path economy factors; acquiring threat coefficients corresponding to the obstacle measuring points according to path differences among all obstacle avoidance paths corresponding to each obstacle measuring point at each sampling moment; acquiring the degree of abnormality corresponding to the obstacle measuring points in a preset range of each obstacle measuring point according to the fluctuation conditions of the threat coefficients of all the obstacle measuring points and the fluctuation conditions of the reachable distances between the rest of the obstacle measuring points and the corresponding obstacle measuring points;
And adjusting the local reachable density of each obstacle measuring point according to the abnormality degree, and acquiring a final radar elevation monitoring graph at each sampling time.
Further, the calculation formula of the dynamic anticipation factor is as follows:
; wherein, Is the firstThe following sampling timeDynamic anticipation factors for individual obstacle measure points; The sequence number of the sea surface measuring point; the sequence number of the obstacle measuring point; Is the first The distribution kurtosis of the sea surface elevation difference values corresponding to all sea surface measuring points at each sampling time; Is the first The distribution kurtosis of the obstacle elevation difference values corresponding to all obstacle measuring points at each sampling time; Is the first Remove all and the first under the sampling timeThe distribution kurtosis of all other non-target obstacle elevation differences except the obstacle elevation differences with the same target obstacle elevation difference corresponding to the obstacle measurement points; Is a preset non-zero constant.
Further, the method for acquiring the distribution kurtosis of the sea surface elevation difference value, the distribution kurtosis of the obstacle elevation difference value and the distribution kurtosis of the non-target obstacle elevation difference value comprises the following steps:
Calculating the difference value of elevation values between the sea surface measuring points at each sampling time and pixel points at the same position in the radar elevation monitoring graph corresponding to the previous adjacent sampling time, and obtaining the sea surface elevation difference value of all the sea surface measuring points; obtaining the distribution kurtosis of the sea surface elevation difference value;
Calculating the difference value of elevation values between the obstacle measuring points at each sampling time and the pixel points at the same position in the radar elevation monitoring chart corresponding to the previous adjacent sampling time, and obtaining the difference value of obstacle elevation of all the obstacle measuring points; obtaining the distribution kurtosis of the obstacle elevation difference value;
Taking any one of the obstacle measuring points as a target obstacle measuring point, taking the obstacle elevation difference value corresponding to the target obstacle measuring point as a target obstacle elevation difference value, and obtaining the distribution kurtosis of all other non-target obstacle elevation difference values except all the obstacle elevation difference values which are the same as the target obstacle elevation difference value.
Further, the method for acquiring the obstacle avoidance path comprises the following steps:
Clustering all the obstacle measuring points corresponding to the radar elevation monitoring graph at each sampling moment to obtain all obstacle areas; optionally selecting one obstacle measuring point in each obstacle region to serve as a locating point, wherein all locating points form a path obstacle point group, and obtaining all path obstacle point groups at each sampling time;
Taking the position of the ship at the sampling moment as a first path point and taking the ship destination as an end path point; acquiring a resultant force field of each sea surface measuring point under the influence of any path obstacle point group in a preset range of the first path point, and taking the sea surface measuring point with the largest resultant force field in the preset range of the first path point as a second path point; acquiring a resultant force field of each sea surface measuring point in the preset range of the second path point under the influence of the corresponding path obstacle point group, and taking the sea surface measuring point with the largest resultant force field in the preset range of the second path point as a third path point; continuously acquiring a new path point until the end point path point is within the preset range of the new path point, and stopping acquiring the path point; and connecting all the acquired path points from the first path point to the end point path point in sequence according to the acquisition order to obtain an obstacle avoidance path corresponding to the path obstacle avoidance point group.
Further, the calculation formula of the resultant field is:
; wherein, The sequence number of the path point; Is the first The following sampling timeThe first of the preset range of the path pointsA resultant field of sea surface survey points; Is a first gain factor; Is the first The lower distance of each sampling timePreset range of each path pointNearest first sea surface measuring pointThe first of the obstacle regionsDynamic anticipation factors for individual obstacle measure points; Is the first The following sampling timePreset range of each path pointSea surface measuring points and the firstWithin the obstacle regionThe distance between the obstacle measuring points; is a second gain factor; Is the number of path points; Is the first The following sampling timeThe first path pointThe distance between the path points, whereIn the time-course of which the first and second contact surfaces,Taking 0; Is the first The following sampling timePreset range of each path pointSea surface measuring points and first path pointsDistance between them.
Further, the threat coefficient acquisition method includes:
Taking any obstacle measuring point as a target obstacle measuring point, and taking any two obstacle avoidance paths in all path obstacle point groups containing the target obstacle measuring point as a path combination corresponding to all obstacle avoidance paths to obtain all path combinations; and obtaining the mean square error between the coordinates of the corresponding positions of all sea surface measuring points on two obstacle avoidance paths in each path combination, and normalizing the mean value of the mean square error corresponding to all path combinations of the target obstacle measuring points to obtain the threat coefficient of the target obstacle measuring point.
Further, the method for obtaining the abnormality degree includes:
Taking any one of the obstacle measuring points as a target obstacle measuring point, acquiring threat coefficient variances of all the obstacle measuring points corresponding to the threat coefficients in a preset range of each target obstacle measuring point, normalizing variance values of reachable distances between all the obstacle measuring points and the target obstacle measuring point, and multiplying the threat coefficient variances to obtain the degree of abnormality of each target obstacle measuring point.
Further, the method for adjusting the local reachable density comprises the following steps:
and obtaining the local reachable density of each obstacle measuring point, carrying out negative correlation mapping after normalizing the abnormality degree, adding a preset positive integer to the negative correlation mapping value, and multiplying the negative correlation mapping value by the local reachable density of the corresponding obstacle measuring point to obtain the adjusted local reachable density.
Further, the method for acquiring the sea surface measuring point and the obstacle measuring point comprises the following steps:
and dividing each radar elevation monitoring graph by adopting an Ojin threshold algorithm, wherein all pixel points corresponding to a divided area with small average elevation value in the binarized divided area are used as sea surface measuring points, and all pixel points corresponding to a divided area with large average elevation value are used as obstacle measuring points.
The invention also provides a target acquisition system for channel safety monitoring, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes the steps of the target acquisition method for channel safety monitoring when executing the computer program.
The invention has the following beneficial effects:
Firstly, acquiring sea surface measuring points and obstacle measuring points in a radar elevation monitoring chart at each sampling time in the ship navigation process; according to the distribution condition of elevation differences among pixel points at the same position among radar elevation monitoring graphs corresponding to adjacent sampling moments, acquiring a dynamic expected factor of each obstacle measuring point at each sampling moment, wherein the dynamic expected factor reflects the fluctuation synchronism of the obstacle measuring point and the sea surface degree, and if the elevation value of the obstacle measuring point does not fluctuate along with the fluctuation of the elevation value of the sea surface, the probability of the obstacle measuring point is higher; further acquiring all obstacle avoidance paths at each sampling time according to the dynamic expected factors and the distances between the sea surface measuring points and the obstacle measuring points and combining path economy factors; then, at each sampling moment, according to the path difference among all obstacle avoidance paths corresponding to each obstacle measuring point, acquiring threat coefficients corresponding to the obstacle measuring points, wherein the threat coefficients reflect the influence degree of the obstacle measuring points as path locating points; obtaining the degree of abnormality of the corresponding obstacle measuring points according to the fluctuation condition of threat coefficients of all the obstacle measuring points and the fluctuation condition of the reachable distance between the rest obstacle measuring points and the corresponding obstacle measuring points in the preset range of each obstacle measuring point, wherein the more uniform the reachable distance between the obstacle measuring points is and the larger the fluctuation of the threat coefficients is, the more abnormal the corresponding obstacle measuring points are; and adjusting the local reachable density of each obstacle measuring point according to the degree of abnormality, and reducing the local reachable density of the obstacle measuring point with larger degree of abnormality, so that the characteristic of isolated outliers is easier to detect, and a final radar elevation monitoring graph at each sampling time is obtained. The invention further judges the abnormality degree of each obstacle measuring point by acquiring the dynamic expected factors and threat coefficients, and then adjusts the local reachable density according to the abnormality degree, thereby reducing the sensitivity of a local abnormality factor algorithm to noise, accurately detecting all abnormal obstacle measuring points, and improving the accuracy and reliability of channel safety monitoring.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a method for acquiring a target for channel safety monitoring according to an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following description refers to the specific implementation, structure, characteristics and effects of the target acquisition method and system for channel safety monitoring according to the invention in combination with the accompanying drawings and the preferred embodiment. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of a target acquisition method and a target acquisition system for channel safety monitoring provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for acquiring a target for channel security monitoring according to an embodiment of the present invention is shown, where the method includes the following steps:
step S1, acquiring radar elevation monitoring graphs at each sampling time in the ship navigation process, and acquiring sea surface measuring points and obstacle measuring points in each radar elevation monitoring graph.
In order to perform safety monitoring on a ship channel, the embodiment of the invention acquires radar elevation monitoring graphs at each sampling time through a radar system loaded on the ship, further acquires sea surface measuring points and obstacle measuring points, so as to judge the abnormality degree of the obstacle measuring points according to the related information of the sea surface measuring points and the obstacle measuring points, further eliminates the abnormal obstacle measuring points, and acquires more accurate radar elevation monitoring graphs, thereby performing safety monitoring on the ship channel more accurately.
In one embodiment of the invention, a transmitter of the ship loading radar system specifically transmits electromagnetic waves of an S wave band to the periphery at a transmission frequency of one time per second, the electromagnetic waves are reflected by objects in the transmission process, after a radar receiver receives reflected electromagnetic wave signals, a three-dimensional radar elevation map of corresponding measuring points of all objects around the ship is obtained through signal processing and analysis, wherein the coordinate position of each measuring point in a two-dimensional plane constructed by an x-axis y-axis represents the position coordinate of the corresponding measuring point when the measuring point emits the electromagnetic waves relative to the ship, a z-axis represents the elevation value of the corresponding measuring point, namely the height distribution condition of the ground surface or a target object in the vertical direction, the three-dimensional distance-direction map is mapped into a two-dimensional plane, and a radar elevation monitoring map corresponding to two dimensions is obtained, wherein pixels in the radar elevation monitoring map are corresponding measuring points, and pixels in the radar elevation monitoring map are elevation values of the corresponding measuring points. It should be noted that, the method for obtaining the radar elevation monitor is a technical means well known to those skilled in the art, and will not be described herein.
Considering that the radar emission wave arrives at sea surfaces and offshore obstacles in the navigation process of the ship, the obstacles such as the ship, the cursory or the reef are key monitoring targets for influencing the safety of the navigation channel, if the related information of the monitored obstacles is abnormal data, the safety monitoring of the navigation channel is greatly threatened; therefore, the embodiment of the invention needs to acquire the sea surface measuring point and the obstacle measuring point in each radar elevation monitoring chart, further analyzes the degree of abnormality of the obstacle measuring point, eliminates the abnormality, and acquires the accurate radar elevation monitoring chart for channel safety monitoring.
Preferably, in one embodiment of the present invention, considering that the obstacle is more prominent relative to the sea surface, its corresponding elevation value is also higher than the sea surface, and the oxford thresholding algorithm can adaptively segment the image according to the intra-class variance and provide clear boundary information; based on the method, the method for acquiring the sea surface measuring point and the obstacle measuring point comprises the steps of dividing each radar height Cheng Jiance graph by adopting an Ojin threshold algorithm, taking all pixel points corresponding to a divided area with small average elevation value in a binarized divided area as the sea surface measuring point, and taking all pixel points corresponding to a divided area with large average elevation value as the obstacle measuring point. The oxford thresholding algorithm is a prior art well known to those skilled in the art and will not be described in detail herein; in other embodiments of the present invention, the practitioner may also acquire the sea surface measurement points and the obstacle measurement points using other methods, such as neural networks or adaptive thresholding.
Step S2, acquiring dynamic expected factors of each obstacle measuring point at each sampling time according to the distribution condition of elevation differences among pixel points at the same position among radar elevation monitoring graphs corresponding to adjacent sampling times; according to the dynamic expected factors and the distances between the sea surface measuring points and the obstacle measuring points, acquiring all obstacle avoidance paths at each sampling time by combining path economy factors; acquiring threat coefficients of corresponding obstacle measuring points according to path differences among all obstacle avoidance paths corresponding to each obstacle measuring point at each sampling moment; and acquiring the degree of abnormality of the corresponding obstacle measuring point according to the fluctuation of threat coefficients of all the obstacle measuring points and the fluctuation of the reachable distance between the rest obstacle measuring points and the corresponding obstacle measuring point within the preset range of each obstacle measuring point.
Considering that whether an abnormality exists at a certain obstacle measuring point can be judged by comprehensively analyzing the elevation difference among all the pixel points at the same position among radar elevation monitoring graphs corresponding to adjacent sampling moments; and the sea water is always in a fluctuation state due to the influence of factors such as wind or a marine system, so that the elevation values of the sea surface and all obstacles are changed along with the fluctuation of the sea water, the ship is still in a navigation state, and the elevation values acquired by the radar system at adjacent sampling moments have no comparative analysis value.
However, if the weather of the sea area at the adjacent sampling time is not changed severely, sea water fluctuation generated by the influence of a sea system on the sea surface is generally regular and approximately normally distributed, and because the ship is huge in volume, the analysis influence of the navigation distance between the adjacent sampling time on the elevation difference of the measuring points between radar elevation monitoring graphs is negligible, namely the distribution situation of the elevation difference between all barrier measuring points in the radar elevation monitoring graphs acquired at the adjacent time is required to be in certain connection with the sea surface fluctuation rule, namely the distribution situation of the elevation difference between the sea surface measuring points; therefore, according to the embodiment of the invention, the dynamic expected factors of each obstacle measuring point at each sampling time are obtained according to the distribution situation of the elevation difference between the pixel points at the same position between the radar elevation monitoring graphs corresponding to the adjacent sampling times, the dynamic expected factors reflect whether the obstacle measuring points can fluctuate along with sea surface fluctuation, and if the elevation values of the obstacle measuring points do not fluctuate synchronously along with the sea surface elevation values, the probability that the obstacle measuring points are abnormal obstacle measuring points is higher.
Preferably, in one embodiment of the present invention, the greater the kurtosis difference, the greater the distribution difference is accounted for in that the kurtosis can reflect the steepness of the probability distribution curve; the method for acquiring the dynamic anticipation factor comprises the following steps: calculating the difference value of elevation values between the sea surface measuring points at each sampling time and the pixel points at the same position in the radar elevation monitoring graph corresponding to the previous adjacent sampling time to obtain the sea surface elevation difference value of all the sea surface measuring points; obtaining the distribution kurtosis of the sea surface elevation difference value; calculating the difference value of elevation values between the obstacle measuring points at each sampling moment and the pixel points at the same position in the radar elevation monitoring graph corresponding to the previous adjacent sampling moment to obtain the obstacle elevation difference value of all the obstacle measuring points; obtaining the distribution kurtosis of the obstacle elevation difference value; taking any obstacle measuring point as a target obstacle measuring point, taking the obstacle elevation difference value corresponding to the target obstacle measuring point as a target obstacle elevation difference value, and obtaining the distribution kurtosis of all other non-target obstacle elevation difference values except all the obstacle elevation difference values which are the same as the target obstacle elevation difference value; acquiring a dynamic expected factor according to a calculation formula of the dynamic expected factor; it should be noted that, the calculation of the distribution kurtosis is a prior art well known to those skilled in the art, and will not be described herein.
The calculation formula of the dynamic anticipation factor is as follows:
Wherein, Is the firstThe following sampling timeDynamic anticipation factors for individual obstacle measure points; The sequence number of the sea surface measuring point; the sequence number of the obstacle measuring point; Is the first The distribution kurtosis of the sea surface elevation difference values corresponding to all sea surface measuring points at each sampling time; Is the first The distribution kurtosis of the obstacle elevation difference values corresponding to all obstacle measuring points at each sampling time; Is the first Remove all and the first under the sampling timeThe distribution kurtosis of all other non-target obstacle elevation differences except the obstacle elevation differences with the same target obstacle elevation difference corresponding to the obstacle measurement points; In order to preset a non-zero constant, in the embodiment of the invention, the preset non-zero constant is 0.1, so that the split type meaning is ensured, and an implementer can set the preset non-zero constant according to specific implementation conditions.
In the calculation formula of the dynamic anticipation factor,Reflects the distribution difference of all the rest obstacle elevation differences except all the obstacle elevation differences which are the same as the target obstacle elevation difference and the sea surface elevation difference at adjacent sampling moments,Reflecting the distribution difference of the obstacle elevation difference value relative to the sea surface elevation difference value between adjacent sampling moments, if all the obstacle elevation difference values which are the same as the target obstacle elevation difference value are removed, the distribution kurtosis of the obstacle elevation difference value relative to the sea surface elevation difference value is more approximate, namelyThe smaller the obstacle measure point corresponding to the corresponding target Cheng Chazhi is, the greater the possibility that the obstacle measure point is an abnormal obstacle measure point is; and by dividing byAnd normalizing to obtain the fluctuation synchronization coefficient of the obstacle measuring point along with the sea surface measuring point, if the ratio is smaller, the synchronism of the obstacle measuring point corresponding to the target obstacle elevation difference value and the sea surface fluctuation is lower, and adjusting the corresponding logic by subtracting the corresponding fluctuation synchronization coefficient from 1, so that the possibility that the corresponding obstacle measuring point is an abnormal obstacle measuring point is higher.
The artificial potential field algorithm is a simple, quick and wide-applicability path planning algorithm based on a physical concept, and can generate a path from a starting point to a target point and avoid an obstacle at the same time; in the path planning process, considering that the influence degree of each obstacle measuring point on the safe navigation of the channel is different, the selection of the path points can be influenced by the gravitational field caused by taking different obstacles as the path locating points, and the planned obstacle avoidance path of the channel is further influenced; when a certain obstacle measuring point is used as one positioning point of the path planning so that large differences exist among all formed obstacle avoidance paths, the more critical the obstacle measuring point is used as the positioning point, and the greater the influence on the safety planning of the channel path when the obstacle measuring point is abnormal.
Because of daily demands of oiling, wind prevention, maintenance and the like, ships need to frequently go to and from anchor places, shipyards and operation sea areas, and the sailing is influenced by factors such as natural environment, traffic flow, topography and the like, radar signals are mutually staggered and propagated, partial abnormal obstacle measuring points possibly exist in a radar elevation measuring graph, and a dynamic expected factor reflects the abnormality degree of the obstacle measuring points to a certain extent; in addition, since the channel path planning needs to consider economic cost, namely path distance, while obstacle avoidance, the embodiment of the invention obtains all obstacle avoidance paths at each sampling time according to dynamic expected factors and distances between sea surface measuring points and obstacle measuring points and path economical factors based on the artificial potential field algorithm concept, and further judges the influence, namely threat coefficient, when an obstacle measuring point is taken as a path locating point by analyzing the difference between all planned paths when the obstacle measuring point is taken as the locating point.
Preferably, in one embodiment of the present invention, the method for obtaining the obstacle avoidance path includes clustering all obstacle measurement points in the corresponding radar elevation monitoring graph at each sampling time to obtain all obstacle areas; optionally selecting one obstacle measuring point as a locating point in each obstacle region in a put-back way, wherein all locating points are a path obstacle point group, and obtaining all path obstacle point groups at each sampling time; taking the position of the ship at the sampling moment as a first path point and taking the ship destination as an end path point; acquiring a resultant force field of each sea surface measuring point under the influence of any path obstacle point group in a preset range of the first path point, and taking the sea surface measuring point with the largest resultant force field in the preset range of the first path point as a second path point; acquiring a resultant force field of each sea surface measuring point in a preset range of the second path point under the influence of a corresponding path obstacle point group, and taking the sea surface measuring point with the largest resultant force field in the preset range of the second path point as a third path point; continuously acquiring a new path point until the end point path point is within a preset range of the new path point, and stopping acquiring the path point; and connecting all the acquired path points from the first path point to the end point path points in sequence according to the acquisition order to obtain the obstacle avoidance path of the corresponding path obstacle avoidance point group.
In the embodiment of the invention, a DBSCAN clustering algorithm is specifically adopted to cluster all obstacle measuring points, the DBSCAN clustering search radius is set to be 5 meters, a plurality of obstacle areas can be obtained, one obstacle measuring point is arbitrarily selected from each obstacle area to serve as a locating point of the obstacle area, a path obstacle point group can be obtained, for example, A (a 1, a2, a 3), B (B1, B2, B3) and A, B are respectively two obstacle areas, one locating point is randomly selected from three obstacle measuring points in A (a 1, a2 and a 3) to serve as an locating point of A, a2 is selected, a locating point serving as B is also randomly selected from B (B1, B2 and B3), and B2 and B3 are path obstacle point groups; the preset range is set in a circular range with a radius of 3 meters and taking a path point as a center, wherein the path point is represented as a pixel point in the radar elevation monitoring image, so that when the distance between the obstacle measuring point and the sea surface measuring point in the preset range of the path point is acquired, the actual Euclidean distance in space is calculated through the two-dimensional coordinate position in the radar elevation monitoring image, and the distance between the corresponding preset range and the judgment pixel point is acquired in the radar elevation monitoring image; when the end point path point is in the preset range of the newly acquired path point, stopping acquiring the path point, constructing an obstacle avoidance path from the first path point to the end point path point by sequentially connecting all the path points, and marking the end point path point in a radar elevation monitoring chart so as to acquire the obstacle avoidance path. It should be noted that DBSCAN clustering is a well-known technology for those skilled in the art, and is not described herein.
The calculation formula of the resultant field is:
Wherein, The sequence number of the path point; Is the first The following sampling timeThe first of the preset range of the path pointsA resultant field of sea surface survey points; Is a first gain factor; Is the first The lower distance of each sampling timePreset range of each path pointNearest first sea surface measuring pointThe first of the obstacle regionsDynamic anticipation factors for individual obstacle measure points; Is the first The following sampling timePreset range of each path pointSea surface measuring points and the firstWithin the obstacle regionThe distance between the obstacle measuring points; is a second gain factor; Is the number of path points; Is the first The following sampling timeThe first path pointThe distance between the path points; Is the first The following sampling timePreset range of each path pointSea surface measuring points and first path pointsDistance between them. When the following is performedWhen there is only one first path point, the distance between adjacent path points cannot be obtained, i.e. whenIn the time-course of which the first and second contact surfaces,Taking 0; in an embodiment of the present invention, a first gain factorTaking 0.6, the second gain coefficientTaking 0.4, the practitioner can set according to the specific implementation.
In the calculation formula of the resultant force field,Is the firstThe gravitational field of each sea surface measuring point reflects the firstThe more distant the sea surface measuring points are from the obstacle measuring points of the obstacle avoidance path, the more distant the sea surface measuring points are from the obstacle measuring points of the obstacle region, the more distant the sea surface measuring points are from the obstacle measuring points of the obstacle avoidance pathThe larger the gravitational field of each sea surface measuring point is, the more is helpful to avoid the obstacle during path planning; the larger dynamic expected factor represents the abnormality of the corresponding obstacle measure point, and a certain danger exists for safety obstacle avoidance, so that the dynamic expected factor of the obstacle measure point is subtracted from 1 to be used as the weight of the distance, and the dynamic expected factor is reduced to the firstA gravitational field of the sea surface measuring points; Is the first The repulsive force field of each sea surface measuring point reflects the firstThe subtracted value when each sea surface measuring point is a path point of the obstacle avoidance path is obtained by accumulating the distances between all adjacent path points to obtain the total distance of the planned path, and the method is further combined with the first stepJudging the distance between the sea surface measuring point and the first path pointThe distance feasibility of the path of each sea surface measuring point as the next path point is that the shorter the distance is, the more the distance is supposed to be the path point; judging the first by combining the gravitational field and the repulsive field into a resultant force fieldThe feasibility of using the sea surface measuring point as a path point is that the larger the force field is, the more the sea surface measuring point should be used as the path point.
It should be noted that, the planning of the obstacle avoidance path is not an actual planning path when the ship sails, only the obstacle avoidance path which is planned by judging the influence of each obstacle measuring point on the channel safety for the radar elevation measuring graph obtained at each sampling moment is used for judging the degree of abnormality of each measuring point so as to eliminate and obtain an accurate radar elevation measuring graph, and then the final channel path can be planned according to the accurate radar elevation measuring graph.
After all the obstacle avoidance paths are obtained according to the method for planning the obstacle avoidance paths, at each sampling moment, whether great differences exist among paths when each obstacle measuring point is used as one path planning locating point is further judged according to the path differences among all the obstacle avoidance paths corresponding to each obstacle measuring point, so that threat coefficients corresponding to the obstacle measuring points are obtained.
Preferably, in one embodiment of the present invention, the threat coefficient obtaining method includes taking any one obstacle measuring point as a target obstacle measuring point, taking any two obstacle avoiding paths in all path obstacle point groups including the target obstacle measuring point corresponding to all obstacle avoiding paths as a path combination, and obtaining all path combinations, wherein the target obstacle measuring point is any one obstacle measuring point; and obtaining the mean square error between the coordinates of the corresponding positions of all sea surface measuring points on two obstacle avoidance paths in each path combination, and normalizing the mean value of the mean square error corresponding to all path combinations of the target obstacle measuring points to obtain the threat coefficient of the target obstacle measuring point. The threat coefficient is calculated by the following formula:
Wherein, Is the firstThreat coefficients of the target obstacle measure points; Is composed of the first Sequence numbers of corresponding path combinations of the target obstacle measuring points; Is the first The total number of the corresponding path combinations of the target obstacle measuring points; Is the first Mean square error of the individual path combinations; normalizing the function for a maximum minimum.
In the calculation formula of threat coefficient, each threat coefficient is calculated to contain a first threat coefficientThe mean square error of the path combination corresponding to each target obstacle measuring point obtains the similarity of two paths in the corresponding path combination, thereby judging the first path according to the mean value of the mean square errorThreat coefficients of the target obstacle measure points; the larger the mean value of the mean square error is, the lower the similarity among all obstacle avoidance paths planned by the target obstacle measuring point is, the larger the path difference is, and the larger the influence degree of the obstacle measuring point serving as a path locating point is; in the embodiment of the invention, the mean square error is normalized by specifically adopting maximum value and minimum value normalization, and an implementer can also select other normalization modes for processing.
It should be noted that, since the path lengths of each obstacle avoidance path may not be the same, when the similarity between the obstacle avoidance paths is evaluated, the obstacle avoidance paths in each path combination need to be aligned by using a dynamic time warping (DYNAMIC TIME WARPING, DTW) algorithm, and the DTW algorithm is a known prior art for those skilled in the art, and a detailed description of the path alignment process is omitted here.
Because the more uniform the reachable distance distribution between the data point and other data points in the neighborhood is in the local outlier (Local Outlier Factor, LOF) algorithm, the larger the density difference between the data point and other data points in the neighborhood is, the more likely the data point is an abnormal data point with local isolation; the threat coefficients of the obstacle measuring points reflect the criticality of the obstacle measuring points serving as path locating points, the greater the possibility of abnormality is, the greater the abnormal influence on path planning is, and the influence of all the obstacle measuring points in the same neighborhood range on the path planning is similar under normal conditions, and the corresponding threat coefficients are close; therefore, in the embodiment of the invention, the degree of abnormality of the corresponding obstacle measuring point is obtained according to the fluctuation condition of the threat coefficients of all the obstacle measuring points and the fluctuation condition of the reachable distance between the rest obstacle measuring points and the corresponding obstacle measuring point within the preset range of each obstacle measuring point.
Preferably, in one embodiment of the present invention, the method for obtaining the abnormality degree includes taking any one of the obstacle measurement points as a target obstacle measurement point, obtaining threat coefficient variances of threat coefficients corresponding to all the obstacle measurement points within a preset range of each target obstacle measurement point, normalizing variance values of reachable distances between all the obstacle measurement points and the target obstacle measurement point, and multiplying the normalized variance values by the threat coefficient variances to obtain the abnormality degree of each target obstacle measurement point. In the embodiment of the present invention, a circular range with a radius of 5 meters is constructed with each target obstacle measuring point as a center as a preset range of the corresponding target obstacle measuring point, and in other embodiments of the present invention, other preset ranges may be set according to specific implementation situations. It should be noted that, the obtaining of the reachable distance is well known in the art, and is not described herein.
The calculation formula of the degree of abnormality is:
Wherein, Is the firstThe degree of abnormality of each target obstacle measuring point; Is the first Target obstacle measuring pointVariance of the reachable distances among all the obstacle measuring points within the preset range of the target obstacle measuring points; Is the first Target obstacle measuring pointVariance of corresponding threat coefficients of all obstacle measuring points within a preset range of each target obstacle measuring point; is a natural constant.
In the calculation formula of the abnormality degree, the smaller the variance of the reachable distance is, the more uniform the reachable distance distribution among all the obstacle measuring points in the preset range of the target obstacle measuring point is, so that the negative correlation is mapped into an exponential function for normalization processing, and the smaller the variance is, the larger the abnormality degree of the corresponding target obstacle measuring point is; the larger the variance of the threat coefficients is, the larger the corresponding threat coefficient differences of all the obstacle measuring points in the preset range are, and the larger the corresponding abnormal degrees are.
And S3, adjusting the local reachable density of each obstacle measuring point according to the abnormality degree, and acquiring a final radar elevation monitoring graph at each sampling time.
The local reachable density is an index used for reflecting the local density degree of the data point in the neighborhood in the LOF algorithm, and the data point is represented as an abnormal condition of an isolated outlier, and the corresponding local reachable density is smaller; therefore, the embodiment of the invention adjusts the local reachable density of each obstacle measuring point according to the abnormality degree, and further adjusts the local reachable density of the obstacle measuring point with larger abnormality degree, so that the isolated outlier characteristic of the obstacle measuring point is easier to detect.
Preferably, in one embodiment of the present invention, the method for adjusting the local reachable density includes obtaining the local reachable density of each obstacle measure point, normalizing the degree of abnormality, performing negative correlation mapping, adding a preset positive integer to the negative correlation mapping value, and multiplying the positive integer by the local reachable density of the corresponding obstacle measure point to obtain the adjusted local reachable density. The method for obtaining the local reachable density is well known to those skilled in the art, and is not described herein. The adjustment formula of the local reachable density is as follows:
Wherein, Is the firstLocal reachable density after adjustment of the obstacle measuring points; Is the first Degree of abnormality of each obstacle measuring point; is a preset positive integer; Is the first Locally reachable densities of individual obstacle measuring points. In the embodiment of the invention, in order to reduce the local reachable density corresponding to the obstacle measuring point with large abnormality degree in the range of 0 to 1, a positive integer is preset to take 1, and an operator can set other values according to specific implementation conditions.
After the local reachable density of each obstacle measuring point is adjusted according to the corresponding abnormality degree, all abnormal obstacle measuring points can be further detected, and therefore an accurate radar elevation monitoring graph is obtained.
Preferably, in one embodiment of the present invention, the method for obtaining the final radar elevation monitoring map includes obtaining a local outlier factor of each obstacle measure point by using an LOF algorithm according to the adjusted local reachable density, and removing the obstacle measure points greater than a preset local outlier factor threshold from the radar elevation monitoring map; and clustering the rest obstacle measuring points and all sea surface measuring points in the radar elevation monitoring graph to obtain a final radar elevation monitoring graph. In the embodiment of the invention, the preset local outlier factor threshold is specifically set to be 1, and an implementer can set other values according to specific implementation conditions; after all abnormal obstacle measuring points are removed from the radar elevation monitoring map, clustering all obstacle measuring points again through a K-means clustering algorithm to obtain a final radar elevation monitoring map, and then converting the final radar elevation monitoring map into a three-dimensional radar map for visualization, so that an operator can conveniently formulate an obstacle avoidance strategy.
It should be noted that, the LOF algorithm and the K-means clustering algorithm are well known in the art, and are not described herein. The practitioner can judge the size, the azimuth, the speed and the like of the obstacle according to the multi-frame three-dimensional radar chart at the continuous sampling time and the artificial experience, further formulate an obstacle avoidance strategy, and also can automatically generate an obstacle avoidance path by adopting artificial intelligence so as to prevent the ship from collision.
In summary, the method firstly obtains the sea surface measuring point and the obstacle measuring point in the radar elevation monitoring chart at each sampling time in the ship navigation process; acquiring a dynamic expected factor of each obstacle measuring point at each sampling time according to the distribution condition of elevation differences among pixel points at the same position among radar elevation monitoring graphs corresponding to adjacent sampling times; further acquiring all obstacle avoidance paths at each sampling time according to the dynamic expected factors and the distances between the sea surface measuring points and the obstacle measuring points and combining path economy factors; then, at each sampling moment, acquiring threat coefficients of corresponding obstacle measuring points according to path differences among all obstacle avoidance paths corresponding to each obstacle measuring point; acquiring the abnormality degree of the corresponding obstacle measuring point according to the fluctuation condition of threat coefficients of all the obstacle measuring points and the fluctuation condition of the reachable distance between the rest obstacle measuring points and the corresponding obstacle measuring point in the preset range of each obstacle measuring point; and adjusting the local reachable density of each obstacle measuring point according to the abnormality degree, and acquiring a final radar elevation monitoring graph at each sampling time. The invention further judges the abnormality degree of each obstacle measuring point by acquiring the dynamic expected factors and threat coefficients, and then adjusts the local reachable density according to the abnormality degree, thereby reducing the sensitivity of the local abnormal factors to noise, accurately detecting all abnormal obstacle measuring points, and further improving the accuracy and reliability of channel safety monitoring.
The invention also proposes a target acquisition system for channel safety monitoring, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor executing the computer program to perform the steps of a target acquisition method for channel safety monitoring, as such.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
Claims (6)
1. A target acquisition method for channel safety monitoring, the method comprising:
Acquiring radar elevation monitoring graphs at each sampling time in the ship navigation process, and acquiring sea surface measuring points and obstacle measuring points in each radar elevation monitoring graph;
Acquiring a dynamic expected factor of each obstacle measuring point at each sampling time according to the distribution condition of elevation differences among pixel points at the same position among radar elevation monitoring graphs corresponding to adjacent sampling times; acquiring all obstacle avoidance paths at each sampling time according to the dynamic expected factors and the distances between the sea surface measuring points and the obstacle measuring points and combining path economy factors; acquiring threat coefficients corresponding to the obstacle measuring points according to path differences among all obstacle avoidance paths corresponding to each obstacle measuring point at each sampling moment; acquiring the degree of abnormality corresponding to the obstacle measuring points in a preset range of each obstacle measuring point according to the fluctuation conditions of the threat coefficients of all the obstacle measuring points and the fluctuation conditions of the reachable distances between the rest of the obstacle measuring points and the corresponding obstacle measuring points;
Adjusting the local reachable density of each obstacle measuring point according to the abnormality degree, and acquiring a final radar elevation monitoring graph at each sampling time;
The calculation formula of the dynamic expected factor is as follows:
Wherein, beta i,p is the dynamic expected factor of the p-th obstacle measuring point at the i-th sampling time; h is the sequence number of the sea surface measuring point; g is the sequence number of the obstacle measuring point; omega i,H is the distribution kurtosis of the sea surface elevation difference values corresponding to all sea surface measuring points at the ith sampling time; omega i,G is the distribution kurtosis of the obstacle elevation difference values corresponding to all obstacle measuring points at the ith sampling time; /(I) The distribution kurtosis of all other non-target obstacle elevation differences except for the obstacle elevation difference which is the same as the target obstacle elevation difference corresponding to the p-th obstacle measuring point is correspondingly removed at the ith sampling moment; gamma is a preset non-zero constant;
the method for acquiring the obstacle avoidance path comprises the following steps:
Clustering all the obstacle measuring points corresponding to the radar elevation monitoring graph at each sampling moment to obtain all obstacle areas; optionally selecting one obstacle measuring point in each obstacle region to serve as a locating point, wherein all locating points form a path obstacle point group, and obtaining all path obstacle point groups at each sampling time;
Taking the position of the ship at the sampling moment as a first path point and taking the ship destination as an end path point; acquiring a resultant force field of each sea surface measuring point under the influence of any path obstacle point group in a preset range of the first path point, and taking the sea surface measuring point with the largest resultant force field in the preset range of the first path point as a second path point; acquiring a resultant force field of each sea surface measuring point in the preset range of the second path point under the influence of the corresponding path obstacle point group, and taking the sea surface measuring point with the largest resultant force field in the preset range of the second path point as a third path point; continuously acquiring a new path point until the end point path point is within the preset range of the new path point, and stopping acquiring the path point; sequentially connecting all the acquired path points from the first path point to the end point path point according to the acquisition order to obtain an obstacle avoidance path corresponding to the path obstacle avoidance point group;
The calculation formula of the resultant field is as follows:
S is the sequence number of the path point; /(I) The method comprises the steps of setting a combined force field of a v sea surface measuring point in a preset range of a d path point at an i sampling time; μ is a first gain factor; /(I)The dynamic expected factor of the r-th obstacle measuring point in the j-th obstacle area closest to the v-th sea surface measuring point in the preset range of the s-th path point at the i-th sampling time; /(I)The distance between a v sea surface measuring point and an r obstacle measuring point in a j obstacle region in a preset range of an s-th path point at the i-th sampling time; k is a second gain factor; l v is the number of path points; /(I)Is the distance between the s-th path point and the s-1 st path point at the i-th sampling time, wherein when s=1,/>Taking 0; /(I)The distance between the v sea surface measuring point and the first path point o in the preset range of the s path point at the i-th sampling time;
the threat coefficient acquisition method comprises the following steps:
Taking any obstacle measuring point as a target obstacle measuring point, and taking any two obstacle avoidance paths in all path obstacle point groups containing the target obstacle measuring point as a path combination corresponding to all obstacle avoidance paths to obtain all path combinations; and obtaining the mean square error between the coordinates of the corresponding positions of all sea surface measuring points on two obstacle avoidance paths in each path combination, and normalizing the mean value of the mean square error corresponding to all path combinations of the target obstacle measuring points to obtain the threat coefficient of the target obstacle measuring point.
2. The target acquisition method for channel safety monitoring according to claim 1, wherein the acquisition method of the distribution kurtosis of the sea surface elevation difference value, the distribution kurtosis of the obstacle elevation difference value and the distribution kurtosis of the non-target obstacle elevation difference value comprises:
Calculating the difference value of elevation values between the sea surface measuring points at each sampling time and pixel points at the same position in the radar elevation monitoring graph corresponding to the previous adjacent sampling time, and obtaining the sea surface elevation difference value of all the sea surface measuring points; obtaining the distribution kurtosis of the sea surface elevation difference value;
Calculating the difference value of elevation values between the obstacle measuring points at each sampling time and the pixel points at the same position in the radar elevation monitoring chart corresponding to the previous adjacent sampling time, and obtaining the difference value of obstacle elevation of all the obstacle measuring points; obtaining the distribution kurtosis of the obstacle elevation difference value;
Taking any one of the obstacle measuring points as a target obstacle measuring point, taking the obstacle elevation difference value corresponding to the target obstacle measuring point as a target obstacle elevation difference value, and obtaining the distribution kurtosis of all other non-target obstacle elevation difference values except all the obstacle elevation difference values which are the same as the target obstacle elevation difference value.
3. The target acquisition method for channel safety monitoring according to claim 1, wherein the acquisition method of the degree of abnormality comprises:
Taking any one of the obstacle measuring points as a target obstacle measuring point, acquiring threat coefficient variances of all the obstacle measuring points corresponding to the threat coefficients in a preset range of each target obstacle measuring point, normalizing variance values of reachable distances between all the obstacle measuring points and the target obstacle measuring point, and multiplying the threat coefficient variances to obtain the degree of abnormality of each target obstacle measuring point.
4. The target acquisition method for channel safety monitoring according to claim 1, wherein the local reachable density adjustment method comprises:
and obtaining the local reachable density of each obstacle measuring point, carrying out negative correlation mapping after normalizing the abnormality degree, adding a preset positive integer to the negative correlation mapping value, and multiplying the negative correlation mapping value by the local reachable density of the corresponding obstacle measuring point to obtain the adjusted local reachable density.
5. The target acquisition method for channel safety monitoring according to claim 1, wherein the sea surface measurement point and the obstacle measurement point acquisition method comprises:
and dividing each radar elevation monitoring graph by adopting an Ojin threshold algorithm, wherein all pixel points corresponding to a divided area with small average elevation value in the binarized divided area are used as sea surface measuring points, and all pixel points corresponding to a divided area with large average elevation value are used as obstacle measuring points.
6. An object acquisition system for channel safety monitoring, characterized by comprising a memory, a processor and a computer program stored in said memory and executable on said processor, said processor implementing the steps of an object acquisition method for channel safety monitoring as claimed in any one of claims 1-5 when said computer program is executed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410452982.1A CN118050728B (en) | 2024-04-16 | 2024-04-16 | Target acquisition method and system for channel safety monitoring |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410452982.1A CN118050728B (en) | 2024-04-16 | 2024-04-16 | Target acquisition method and system for channel safety monitoring |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118050728A CN118050728A (en) | 2024-05-17 |
| CN118050728B true CN118050728B (en) | 2024-06-25 |
Family
ID=91045247
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410452982.1A Active CN118050728B (en) | 2024-04-16 | 2024-04-16 | Target acquisition method and system for channel safety monitoring |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118050728B (en) |
Family Cites Families (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US3737902A (en) * | 1970-08-19 | 1973-06-05 | State Street Bank & Trust Co | Collision avoidance system providing a vector signal representative of the distance and bearing between a prime vehicle and target object at a predicted closest point of approach therebetween |
| US7675458B2 (en) * | 2006-11-09 | 2010-03-09 | Raytheon Canada Limited | Dual beam radar system |
| CN103207396A (en) * | 2013-04-11 | 2013-07-17 | 浙江工业大学 | Docking system for simulating Doppler log seabed echo signals on basis of field programmable gata array (FPGA) |
| CN203465854U (en) * | 2013-07-26 | 2014-03-05 | 浙江海洋学院 | Ship collision avoidance early-warning device |
| JP5737662B2 (en) * | 2013-10-24 | 2015-06-17 | 国立研究開発法人海上技術安全研究所 | Ship jet gas supply method and jet gas control device |
| CN105548971B (en) * | 2016-01-11 | 2017-12-15 | 西安电子科技大学 | A kind of target invisible method for evaporation duct over-the-horizon detection radar |
| KR101696615B1 (en) * | 2016-08-04 | 2017-01-16 | 한국해양과학기술원 | Control system for avoiding collision of multiple ships using time series graphic interface |
| EP3674657A1 (en) * | 2016-11-14 | 2020-07-01 | SZ DJI Technology Co., Ltd. | Construction and update of elevation maps |
| CN111123913B (en) * | 2019-11-29 | 2023-05-16 | 广东工业大学 | Double-layer path planning and collaborative obstacle avoidance method for multiple unmanned boats |
| CN111580518B (en) * | 2020-05-12 | 2022-08-02 | 哈尔滨工程大学 | Unmanned ship layered obstacle avoidance method based on improved drosophila optimization and dynamic window method |
| CN113917929A (en) * | 2021-11-11 | 2022-01-11 | 中国船舶重工集团公司第七一九研究所 | Unmanned ship path optimization method and system based on hybrid particle swarm algorithm |
| US20230152092A1 (en) * | 2021-11-18 | 2023-05-18 | The Regents Of The University Of Michigan | Systems and Methods for Generating a Phase-Resolved Ocean Wave Forecasts with Ensemble Based Data Assimilation |
| CN114610046A (en) * | 2022-02-18 | 2022-06-10 | 中国船舶重工集团公司第七0七研究所 | Unmanned ship dynamic safety trajectory planning method considering dynamic water depth |
| CN115761286A (en) * | 2022-11-02 | 2023-03-07 | 中国船舶集团有限公司第七〇七研究所 | Method for detecting navigation obstacle of unmanned surface vehicle based on laser radar under complex sea condition |
-
2024
- 2024-04-16 CN CN202410452982.1A patent/CN118050728B/en active Active
Non-Patent Citations (2)
| Title |
|---|
| "卫星测高数据应用于海面变化研究的初步探索";崔树红;《中国优秀硕士论文全文数据库》;20060215;全文 * |
| "高分辨率SAR图像水体提取算法研究";庞科臣;《中国优秀硕士论文全文数据库》;20180415;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118050728A (en) | 2024-05-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10436893B2 (en) | Apparatus and method for extracting ocean wave information | |
| CN111899568B (en) | Bridge anti-collision early warning system, method and device and storage medium | |
| US8189428B2 (en) | Methods and systems to detect changes in multiple-frequency band sonar data | |
| CN111536962B (en) | Route planning method and device for intelligent ship, storage medium and computer equipment | |
| CN110221257B (en) | Sea surface target detection false alarm suppression method and device and radar system | |
| CN113177593B (en) | Fusion method of radar point cloud and image data in water traffic environment | |
| US7050652B2 (en) | Methods and arrangements to enhance correlation | |
| Vicen-Bueno et al. | Artificial neural network-based clutter reduction systems for ship size estimation in maritime radars | |
| WO2017204076A1 (en) | Signal processing device and radar device | |
| US8604969B2 (en) | System and method of using image grids in detection of discrete objects | |
| WO2020161843A1 (en) | Ship behavior analyzing device, ship behavior analyzing method, and non-transitory computer-readable medium | |
| CN118050728B (en) | Target acquisition method and system for channel safety monitoring | |
| JP2014102256A (en) | Target tracking device and target tracking method | |
| Williams | On adaptive underwater object detection | |
| CN113177097A (en) | Track initial discrimination method based on attribute clustering and space-time constraint | |
| Sun et al. | A Wave Texture Difference Method for Rainfall Detection Using X‐Band Marine Radar | |
| CN116383683A (en) | Method for UHF radar river ship extended target clustering | |
| RU2514154C1 (en) | Method for recognition of false targets caused by self-generated noise of mobile carrier | |
| CN115267827A (en) | Laser radar harbor area obstacle sensing method based on height density screening | |
| CN114236490A (en) | X-band navigation radar oil spill detection system based on water surface echo model | |
| Fefilatyev | Detection of marine vehicles in images and video of open sea | |
| RU2516000C2 (en) | Method of identifying and determining parameters of image of object on radar images | |
| CN114993262B (en) | Sea surface unmanned ship target identification method imitating reconstruction of bird prey receptive field area | |
| US11650290B2 (en) | System and method of determining target's range profiles for costal surveillance radars | |
| Huang et al. | Seafloor obstacle detection by sidescan sonar scanlines for submarine cable construction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |