CN112149237A - Real-time ship collision avoidance method and system - Google Patents
Real-time ship collision avoidance method and system Download PDFInfo
- Publication number
- CN112149237A CN112149237A CN202011102937.1A CN202011102937A CN112149237A CN 112149237 A CN112149237 A CN 112149237A CN 202011102937 A CN202011102937 A CN 202011102937A CN 112149237 A CN112149237 A CN 112149237A
- Authority
- CN
- China
- Prior art keywords
- ship
- data
- target
- processed
- vessel
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 73
- 238000013528 artificial neural network Methods 0.000 claims abstract description 49
- 238000012545 processing Methods 0.000 claims abstract description 34
- 238000012549 training Methods 0.000 claims description 45
- 230000033001 locomotion Effects 0.000 claims description 27
- 238000012216 screening Methods 0.000 claims description 16
- 238000007781 pre-processing Methods 0.000 claims description 11
- 238000010606 normalization Methods 0.000 claims description 9
- 238000007637 random forest analysis Methods 0.000 claims description 9
- 238000011478 gradient descent method Methods 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 3
- 239000000295 fuel oil Substances 0.000 abstract 1
- 230000008569 process Effects 0.000 description 16
- 238000004364 calculation method Methods 0.000 description 11
- 230000006872 improvement Effects 0.000 description 11
- 238000003066 decision tree Methods 0.000 description 10
- 210000002569 neuron Anatomy 0.000 description 9
- 239000013598 vector Substances 0.000 description 8
- 230000006870 function Effects 0.000 description 7
- 230000002159 abnormal effect Effects 0.000 description 6
- 238000012217 deletion Methods 0.000 description 6
- 230000037430 deletion Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 239000000446 fuel Substances 0.000 description 5
- 230000004913 activation Effects 0.000 description 4
- 238000004140 cleaning Methods 0.000 description 4
- 230000001537 neural effect Effects 0.000 description 4
- 101100119135 Mus musculus Esrrb gene Proteins 0.000 description 2
- 230000009471 action Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000013501 data transformation Methods 0.000 description 2
- 230000003247 decreasing effect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000009499 grossing Methods 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000012821 model calculation Methods 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 239000000463 material Substances 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/15—Vehicle, aircraft or watercraft design
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Computer Hardware Design (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Computational Mathematics (AREA)
- Aviation & Aerospace Engineering (AREA)
- Automation & Control Theory (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a real-time ship collision avoidance method, which comprises the following steps: acquiring ship navigation data to be processed, wherein the ship navigation data comprises the ship navigation data to be processed and target ship navigation data to be processed; processing the ship navigation data to be processed to obtain processed data, wherein the processed data comprises ship navigation characteristics; and identifying the ship navigation characteristics through a neural network, and determining the steering angle of the ship. The invention also provides a real-time ship collision avoidance system. The steering angle is predicted in real time through the neural network, compared with the traditional geometric model, the method is simpler in processing, the steering angle closer to actual use can be obtained, and unnecessary fuel oil loss is reduced.
Description
Technical Field
The invention relates to the technical field of ships, in particular to a real-time ship collision avoidance method and system.
Background
In the prior art, when a marine vessel judges whether a collision danger exists between the marine vessel and a target vessel, corresponding avoidance time and avoidance amplitude are generally obtained and re-voyage time is predicted according to a geometric model, and finally, avoidance ship operation is executed according to a calculation result. The method needs to stipulate the initial steering angle in advance, the problem that the given result is reasonable exists, the steering amplitude is large easily, and the method is not suitable for practical situations. For large ships, the large steering amplitude causes a lot of unnecessary fuel consumption.
Disclosure of Invention
In order to solve the above problems, the present invention provides a real-time collision avoidance method and system for a ship, which uses a neural network to predict a steering angle in real time, thereby reducing unnecessary fuel consumption.
The invention provides a real-time ship collision avoidance method, which comprises the following steps:
acquiring ship navigation data to be processed, wherein the ship navigation data comprises the ship navigation data to be processed and target ship navigation data to be processed;
processing the ship navigation data to be processed to obtain processed data, wherein the processed data comprises ship navigation characteristics;
and identifying the ship navigation characteristics through a neural network, and determining the steering angle of the ship.
As a further improvement of the present invention, the processing the ship navigation data to be processed to obtain processed data includes:
preliminarily screening the ship navigation data to be processed to obtain first data;
normalizing the first data to obtain second data;
preprocessing the second data to obtain third data;
and selecting characteristics of the third data to obtain the processed data, wherein the processed data comprises ship navigation characteristics.
As a further improvement of the present invention, the feature selection is performed on the third data to obtain the processed data, and the processing is implemented by a random forest.
As a further improvement of the invention, the ship navigation characteristics comprise the relative distance between the ship and the target ship, the speed of the target ship, the course of the target ship, the relative azimuth between the ship and the target ship, the meeting attribute of the target ship, the relative course between the ship and the target ship, the relative speed between the ship and the target ship, the speed ratio between the ship and the target meeting characteristics.
As a further improvement of the present invention, the neural network adopts a back propagation neural network, the neural network includes 10 input nodes, at least 2 hidden layers and 1 output node, wherein the 10 input nodes are used for inputting the relative distance between the own ship and the target ship, the speed of the own ship, the speed of the target ship, the course of the target ship, the relative orientation between the own ship and the target ship, the meeting attribute of the target ship, the relative course between the own ship and the target ship, the relative speed between the own ship and the target ship, the speed ratio between the own ship and the target meeting characteristic, and the 1 output node is used for outputting the steering angle of the own ship.
As a further improvement of the present invention, the method further comprises: training the neural network through a training data set, wherein the neural network is trained using a gradient descent method.
As a further improvement of the present invention, the method further comprises: obtaining a raw data set, obtaining the training data set from the raw data set,
wherein the raw data set comprises: each target vessel motion data, own vessel motion data, and corresponding decision data, wherein the corresponding decision data comprises a steering angle of the own vessel,
wherein the obtaining the training data set from the raw data set comprises: and carrying out primary screening, normalization, pretreatment and feature selection on the data in the data set to obtain the training data set.
The invention also provides a real-time ship collision avoidance system, which comprises:
the system comprises a data acquisition module, a processing module and a processing module, wherein the data acquisition module is used for acquiring ship navigation data to be processed, and the ship navigation data comprises ship navigation data to be processed and target ship navigation data to be processed;
the data processing module is used for processing the ship navigation data to be processed to obtain processed data, wherein the processed data comprises ship navigation characteristics;
and the characteristic identification module is used for identifying the ship navigation characteristics through a neural network and determining the steering angle of the ship.
As a further refinement of the invention, the data processing module is configured to:
preliminarily screening the ship navigation data to be processed to obtain first data;
normalizing the first data to obtain second data;
preprocessing the second data to obtain third data;
and selecting characteristics of the third data to obtain the processed data, wherein the processed data comprises ship navigation characteristics.
As a further improvement of the present invention, the feature selection is performed on the third data to obtain the processed data, and the processing is implemented by a random forest.
As a further improvement of the invention, the ship navigation characteristics comprise the relative distance between the ship and the target ship, the speed of the target ship, the course of the target ship, the relative azimuth between the ship and the target ship, the meeting attribute of the target ship, the relative course between the ship and the target ship, the relative speed between the ship and the target ship, the speed ratio between the ship and the target meeting characteristics.
As a further improvement of the present invention, the neural network adopts a back propagation neural network, the neural network includes 10 input nodes, at least 2 hidden layers and 1 output node, wherein the 10 input nodes are used for inputting the relative distance between the own ship and the target ship, the speed of the own ship, the speed of the target ship, the course of the target ship, the relative orientation between the own ship and the target ship, the meeting attribute of the target ship, the relative course between the own ship and the target ship, the relative speed between the own ship and the target ship, the speed ratio between the own ship and the target meeting characteristic, and the 1 output node is used for outputting the steering angle of the own ship.
As a further improvement of the present invention, the system further comprises:
a training module for training the neural network through a training data set, wherein the neural network is trained using a gradient descent method.
As a further improvement of the present invention, the system further comprises:
a data set acquisition module for acquiring a raw data set from which the training data set is acquired,
wherein the raw data set comprises: each target vessel motion data, own vessel motion data, and corresponding decision data, wherein the corresponding decision data comprises a steering angle of the own vessel,
wherein the obtaining the training data set from the raw data set comprises: and carrying out primary screening, normalization, pretreatment and feature selection on the data in the data set to obtain the training data set.
The invention also provides an electronic device comprising a memory and a processor, the memory storing one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the method.
The invention also provides a computer-readable storage medium having stored thereon a computer program for execution by a processor to perform the method.
The invention has the beneficial effects that: the steering angle is predicted in real time through the neural network, compared with a traditional geometric model, the steering angle prediction method is simpler in processing, can obtain the steering angle closer to actual use, and reduces unnecessary fuel consumption.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. It is obvious that the drawings in the following description are only some embodiments of the invention, and that for a person skilled in the art, other drawings can be derived from them without inventive effort.
Fig. 1 is a schematic flow chart of a real-time collision avoidance method for a ship according to an exemplary embodiment of the present invention;
FIG. 2 is a schematic flow chart of a ship collision avoidance method in the prior art;
FIG. 3 is a diagram illustrating a prior art method for calculating a steering angle by a geometric model;
FIG. 4 is a schematic diagram of a neural network in accordance with an exemplary embodiment of the present invention;
FIG. 5 is a schematic diagram of the structure of each neuron according to an exemplary embodiment of the invention;
fig. 6 is a schematic diagram illustrating the determination of the target ship rendezvous feature according to the embodiment of the invention;
fig. 7 is a schematic diagram of the encounter property of the target ship according to the embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that, if directional indications (such as up, down, left, right, front, and back … …) are involved in the embodiment of the present invention, the directional indications are only used to explain the relative positional relationship between the components, the movement situation, and the like in a specific posture (as shown in the drawing), and if the specific posture is changed, the directional indications are changed accordingly.
In addition, in the description of the present invention, the terms used are for illustrative purposes only and are not intended to limit the scope of the present invention. The terms "comprises" and/or "comprising" are used to specify the presence of stated elements, steps, operations, and/or components, but do not preclude the presence or addition of one or more other elements, steps, operations, and/or components. The terms "first," "second," and the like may be used to describe various elements, not necessarily order, and not necessarily limit the elements. In addition, in the description of the present invention, "a plurality" means two or more unless otherwise specified. These terms are only used to distinguish one element from another. These and/or other aspects will become apparent to those of ordinary skill in the art in view of the following drawings, and the description of the embodiments of the present invention will be more readily understood by those of ordinary skill in the art. The drawings are only for purposes of illustrating the described embodiments of the invention. One skilled in the art will readily recognize from the following description that alternative embodiments of the structures and methods illustrated in the present application may be employed without departing from the principles described in the present application.
As shown in fig. 1, a real-time collision avoidance method for a ship according to an embodiment of the present invention includes:
s1, acquiring ship navigation data to be processed, wherein the ship navigation data comprises the ship navigation data to be processed and target ship navigation data to be processed;
s2, processing the ship navigation data to be processed to obtain processed data, wherein the processed data comprises ship navigation characteristics;
and S3, identifying the ship navigation characteristics through a neural network, and determining the steering angle of the ship.
In the prior art, as shown in fig. 2, dynamic navigation data of a ship is generally acquired, the situation that the two ships meet is judged after the data are analyzed, whether collision danger exists between the two ships is judged, corresponding avoidance time and avoidance amplitude are obtained according to a geometric model, the re-navigation time is predicted, and finally an avoidance ship operation action is executed according to a calculation result. As shown in fig. 3, when the calculation is performed by a geometric model, this method requires that an initial steering angle is specified in advance, and avoidance timing, avoidance width, and the like are calculated according to the type of the target ship. The method mostly depends on rules, artificially established angles and the like, the steering range refers to international maritime collision avoidance rules or custom practices of sea workers, and no clear avoidance angle regulation exists. Therefore, when the geometric model calculation is performed based on the predetermined initial steering angle, there is a problem that the steering angle result is reasonable. In general collision avoidance decisions, a large avoidance angle is selected in consideration of the safety of the crew. The feedback result of the real ship crew after use is that the steering amplitude is large, and the real ship crew is not suitable for the actual situation. For large ships, the large steering amplitude causes a lot of unnecessary fuel consumption. The method provided by the invention does not adopt a traditional geometric algorithm model, collects steering angle decision data in the real ship collision avoidance process, adopts a neural network to predict, and the predicted steering angle is closer to practical application.
In an optional embodiment, the processing the ship navigation data to be processed to obtain processed data includes:
preliminarily screening the ship navigation data to be processed to obtain first data;
normalizing the first data to obtain second data;
preprocessing the second data to obtain third data;
and selecting characteristics of the third data to obtain the processed data, wherein the processed data comprises ship navigation characteristics.
Wherein, the ship navigation data to be processed can be collected by using equipment or platforms such as AIS, RADAR/ARPA and the like.
The preliminary screening is to roughly process the ship navigation data to be processed, delete irrelevant data and repeated data in the ship navigation data, and preliminarily select and remove features which are weakly associated with a neural network, such as atmospheric pressure, water surface temperature and the like, so that redundant features can be reduced, and data dimensionality is reduced.
The preprocessing is to perform operations such as data cleaning, data integration, data transformation, data reduction and the like on the normalized data, so that the data quality of the input network is improved, and the data can be better adapted to the network. Data cleaning is mainly to smooth noise data and process missing values, abnormal values and the like.
Wherein, (1) noise smoothing, processing abnormal value for example adopt the following method: s1, binning: smoothly ordered data values by looking at the "neighbors" (i.e., surrounding values) of the data; s2, clustering: organizing similar data values into groups or "clusters", treating data values outside the cluster set as outliers, and retaining data values within the cluster set; s3, regression: the data was smoothed by fitting a function to the data. (2) Missing value processing: when the loss rate is low (less than or equal to 5%) and the importance of the attribute is low, if the attribute is numerical data, the filling may be simple according to the data distribution, for example: if the data are uniformly distributed, filling the data by using the average value; if the data distribution is skewed, the number of bits is used for padding. When the deletion rate is high (> 95%) and the importance of an attribute is low, the attribute can be deleted directly. When the deletion rate is high (> 95%) and the attribute degree is high, interpolation and modeling can be used because direct deletion of the attribute will have a bad effect on the result of the network.
It can be understood that a large amount of incomplete, inconsistent, repeated and abnormal data may exist in the massive raw data, which may affect the result output by the network, so that the recognition result is biased. Therefore, the data in the training data set also needs to be obtained through the above-mentioned preliminary screening, normalization, preprocessing and feature selection.
In an optional implementation manner, the feature selection is performed on the third data to obtain the processed data, and the processing is implemented by a random forest.
In an alternative embodiment, the vessel navigation characteristics include a relative distance between the own vessel and the target vessel, a speed of the own vessel, a speed of the target vessel, a heading of the target vessel, a relative azimuth between the own vessel and the target vessel, a meeting attribute of the target vessel, a relative heading between the own vessel and the target vessel, a relative speed between the own vessel and the target vessel, a speed ratio of the own vessel and the target vessel, and a target meeting characteristic. It is to be understood that the above listed ship navigation features may be adaptively selected, and the above features may be appropriately added or subtracted depending on the network configuration, the use environment, the type of ship, and the like, when the feature selection is performed.
Wherein feature selection is distinguished from feature extraction by selecting a subset of the original data set to improve the performance of the network. The process of feature selection is a process of evaluating features, and which feature has a large influence on the output variable is selected. The invention adopts random forest to measure the importance of the features and selects the features with higher importance. The following methods can be employed:
1. feature importance measure, calculating the importance of a certain feature X:
1) and selecting corresponding off-bag data for each decision tree to calculate an off-bag data error, which is recorded as err1, wherein the off-bag data means that one data obtained by repeated sampling is used for training the decision tree when the decision tree is built each time, and at the moment, the data of about 1/3 is not utilized and does not participate in the building of the decision tree. The data can be used for evaluating the performance of the decision tree and calculating the prediction error rate of the model, namely the error of the data outside the bag;
2) randomly adding noise interference to the characteristic X of all samples of the data outside the bag (the value of the sample at the characteristic X can be randomly changed), and calculating the error of the data outside the bag again and recording the error as err 2;
3) assuming there are N trees in a forest, the importance of feature X ∑ (err2-errB 1)/N.
2. Selecting characteristics:
1) calculating the importance of each feature and sorting the features in descending order;
2) determining the proportion to be eliminated, eliminating the characteristics of the corresponding proportion according to the importance of the characteristics to obtain a new characteristic set
3) Repeating the process by using the new feature set until m features (m is a preset value) are left;
4) and selecting the characteristic set with the lowest error rate outside the bag according to the characteristic sets obtained in the process and the error rates outside the bag corresponding to the characteristic sets. Such as the relative distance between the own ship and the target ship, the speed of the own ship, the speed of the target ship, the heading of the target ship, the relative azimuth between the own ship and the target ship, the meeting attribute of the target ship, the relative heading between the own ship and the target ship, the relative speed between the own ship and the target ship, the speed ratio of the own ship and the target ship and the characteristic set of the target meeting characteristic. As described above, in the feature selection, the above-described features may be appropriately added or subtracted according to a network structure, a use environment, a ship type, and the like.
In an alternative embodiment, the neural network employs a back propagation neural network, as shown in fig. 4, the neural network includes 10 input nodes, at least 2 hidden layers and 1 output node, and each hidden layer employs, for example, 60 nodes. It can be understood that the neural network may further employ a plurality of hidden layers other than two hidden layers, each hidden layer may further employ a plurality of nodes other than 60 nodes, and the number of nodes employed by the hidden layer and the hidden layer of the neural network is not particularly limited. The 10 input nodes are used for inputting the relative distance between the ship and the target ship, the speed of the target ship, the course of the target ship, the relative azimuth between the ship and the target ship, the meeting attribute of the target ship, the relative course between the ship and the target ship, the relative speed between the ship and the target ship, the speed ratio between the ship and the target meeting characteristic, and the 1 output node is used for outputting the steering angle of the ship. The connections of the nodes of adjacent layers are all assigned weights.
Wherein each neuron structure is shown in fig. 5. Wherein x isiRepresenting an input vector, wiRepresenting a vector xiThe corresponding weights, y, represent the output vector. Setting an activation value a corresponding to an input layer for each sample x in the training data setl. All weights in the network are randomly assigned, forward propagation is used after the network inputs data, and the total error of the output nodes is calculated and propagated back to the network through back propagation to calculate the gradient. The neural network may also adopt other network structures, for example, and the present invention is not particularly limited. It can be understood that, when the network structure changes, the ship navigation features corresponding to the input nodes of the network can be adaptively selected according to the network structure, and the features can be appropriately increased or decreased.
In an alternative embodiment, the method further comprises: training the neural network through a training data set, wherein the neural network is trained using a gradient descent method.
Wherein, the forward propagation: z is a radical ofl＝wlal-1+bl，al＝σ(zl)
Calculating the error generated by the output node:
back propagation error:l＝((wl+1)T l+1)⊙σ′(zl)
using the gradient descent method, the training parameters:
in the formula, wlWeight, z, representing the connection of neurons of layer l-1 to neurons of layer llRepresenting the neuronal input of layer l, alRepresenting the neuronal output of layer l, blRepresents the bias of layer i neurons, σ represents the activation function, C is a cost function for calculating the error between the output value and the actual value,
x denotes the input sample, y denotes the actual classification, aLIndicating the output of the prediction, L indicates the maximum number of layers of the neural network, L indicates a Hadamard product for point-to-point multiplication between matrices or vectors,
representing the training parameters.
In an alternative embodiment, the method further comprises: obtaining a raw data set, obtaining the training data set from the raw data set,
wherein the raw data set comprises: each target vessel motion data, own vessel motion data, and corresponding decision data, wherein the corresponding decision data comprises a steering angle of the own vessel,
wherein the obtaining the training data set from the raw data set comprises: and carrying out primary screening, normalization, pretreatment and feature selection on the data in the data set to obtain the training data set.
It will be appreciated that vessel dynamic data is collected via a device or platform such as AIS, RADAR/ARPA, etc. When a target ship appears, the dynamic data of the ship and the target ship are subjected to target motion factor calculation. The target motion element calculation can be understood as analyzing the motion element information of the target ship from the collected dynamic data, and the analyzed data may include, for example, the azimuth (OB) of the target ship relative to the target ship, the azimuth (TB) of the target ship relative to the target ship, the relative motion heading (Cr) of the two ships, the relative motion speed (VR) of the two ships, the true motion heading (Ct) and the motion speed (VT) of the target ship, and the like. And after the calculation, judging the intersection characteristics of the target ship, judging the collision danger according to the intersection characteristics of the target ship, and recording the motion data and the corresponding decision data of the target ship and the ship when the collision danger exists so as to form the original data set.
The determining of the target ship encounter characteristics and the determining of the target ship encounter attributes may be performed by, for example, a method shown in fig. 6, which includes: and determining the type of the target ship and the intersection characteristic value of the target ship according to the navigation data of the target ship and the ship. When judging whether the two ships have collision danger or not, judging whether the ship and the target ship have collision danger or not according to a danger judgment threshold value. For example, if the target ship engagement feature value is greater than the risk determination threshold value, it is determined that there is a risk of collision, whereas there is no risk of collision.
The target ship encounter attribute is shown in fig. 7, for example, and the determining of the target ship encounter characteristic may include the following steps:
(1) when the coming ship is the overtaking ship, determining that the target ship is an E-type ship, and determining that the target ship rendezvous characteristic value TEP is 1;
(2) when the target ship crosses the starboard of the ship, determining that the target ship is a starboard crossing A-type ship, and determining that the target ship meets the characteristic value TEP to be 2;
(3) when the target ship meets the ship, determining that the target ship is a class A ship, and determining that the target ship meets the characteristic value TEP to be 6;
(4) when the target ship intersects with the starboard of the ship at a large angle, determining that the target ship is a B-type ship, and determining that the target ship meets a characteristic value TEP of 3 at the moment;
(5) when the target ship is a overtaking ship, determining that the target ship is a C-type ship when the relative azimuth of the target ship and the ship is less than 210 degrees, determining that the target ship is a D-type ship when the relative azimuth of the target ship and the ship is more than or equal to 210 degrees, and determining that the intersection characteristic value TEP of the target ship is 4;
(6) when the relative orientation of the target ship and the ship is between (247.5 degrees and 354 degrees) and the target ship and the port of the ship are crossed, the target ship is determined to be a D-type ship, and the intersection characteristic value TEP of the target ship is determined to be 5.
The above-mentioned exemplary judgment methods can be adapted according to the ship type, and the present invention is not limited in particular.
When training the neural network, a training data set is required for training. After a series of processing is performed on the original data set, a training data set is obtained. A series of processes includes preliminary screening, normalization, pre-processing, and feature selection. As mentioned above, feature selection may be performed using, for example, random forests, which are not described in detail herein.
The embodiment of the invention provides a real-time ship collision avoidance system, which comprises:
the system comprises a data acquisition module, a processing module and a processing module, wherein the data acquisition module is used for acquiring ship navigation data to be processed, and the ship navigation data comprises ship navigation data to be processed and target ship navigation data to be processed;
the data processing module is used for processing the ship navigation data to be processed to obtain processed data, wherein the processed data comprises ship navigation characteristics;
and the characteristic identification module is used for identifying the ship navigation characteristics through a neural network and determining the steering angle of the ship.
In the prior art, as shown in fig. 2, dynamic navigation data of a ship is generally acquired, the situation that the two ships meet is judged after the data are analyzed, whether collision danger exists between the two ships is judged, corresponding avoidance time and avoidance amplitude are obtained according to a geometric model, the re-navigation time is predicted, and finally an avoidance ship operation action is executed according to a calculation result. As shown in fig. 3, when the calculation is performed by a geometric model, this method requires that an initial steering angle is specified in advance, and avoidance timing, avoidance width, and the like are calculated according to the type of the target ship. The method mostly depends on rules, artificially established angles and the like, the steering range refers to international maritime collision avoidance rules or custom practices of sea workers, and no clear avoidance angle regulation exists. Therefore, when the geometric model calculation is performed based on the predetermined initial steering angle, there is a problem that the steering angle result is reasonable. In general collision avoidance decisions, a large avoidance angle is selected in consideration of the safety of the crew. The feedback result of the real ship crew after use is that the steering amplitude is large, and the real ship crew is not suitable for the actual situation. For large ships, the large steering amplitude causes a lot of unnecessary fuel consumption. The system of the invention does not adopt a traditional geometric algorithm model, collects the steering angle decision data in the collision avoidance process of the real ship, adopts the neural network to predict, and the predicted steering angle is closer to the practical application.
The preliminary screening is to roughly process the ship navigation data to be processed, delete irrelevant data and repeated data in the ship navigation data, and preliminarily select and remove features which are weakly associated with a neural network, such as atmospheric pressure, water surface temperature and the like, so that redundant features can be reduced, and data dimensionality is reduced.
The preprocessing is to perform operations such as data cleaning, data integration, data transformation, data reduction and the like on the normalized data, so that the data quality of the input network is improved, and the data can be better adapted to the network. Data cleaning is mainly to smooth noise data and process missing values, abnormal values and the like.
Wherein, (1) noise smoothing, processing abnormal value for example adopt the following method: s1, binning: smoothly ordered data values by looking at the "neighbors" (i.e., surrounding values) of the data; s2, clustering: organizing similar data values into groups or "clusters", treating data values outside the cluster set as outliers, and retaining data values within the cluster set; s3, regression: the data was smoothed by fitting a function to the data. (2) Missing value processing: when the loss rate is low (less than or equal to 5%) and the importance of the attribute is low, if the attribute is numerical data, the filling may be simple according to the data distribution, for example: if the data are uniformly distributed, filling the data by using the average value; if the data distribution is skewed, the number of bits is used for padding. When the deletion rate is high (> 95%) and the importance of an attribute is low, the attribute can be deleted directly. When the deletion rate is high (> 95%) and the attribute degree is high, interpolation and modeling can be used because direct deletion of the attribute will have a bad effect on the result of the network.
It can be understood that a large amount of incomplete, inconsistent, repeated and abnormal data may exist in the massive raw data, which may affect the result output by the network, so that the recognition result is biased. Therefore, the data in the training data set also needs to be obtained through the above-mentioned preliminary screening, normalization, preprocessing and feature selection.
In an alternative embodiment, the data processing module is configured to:
preliminarily screening the ship navigation data to be processed to obtain first data;
normalizing the first data to obtain second data;
preprocessing the second data to obtain third data;
and selecting characteristics of the third data to obtain the processed data, wherein the processed data comprises ship navigation characteristics.
Wherein, the ship navigation data to be processed can be collected by using equipment or platforms such as AIS, RADAR/ARPA and the like.
In an optional implementation manner, the feature selection is performed on the third data to obtain the processed data, and the processing is implemented by a random forest.
In an alternative embodiment, the vessel navigation characteristics include a relative distance between the own vessel and the target vessel, a speed of the own vessel, a speed of the target vessel, a heading of the target vessel, a relative azimuth between the own vessel and the target vessel, a meeting attribute of the target vessel, a relative heading between the own vessel and the target vessel, a relative speed between the own vessel and the target vessel, a speed ratio of the own vessel and the target vessel, and a target meeting characteristic. It is to be understood that the above listed ship navigation features may be adaptively selected, and the above features may be appropriately added or subtracted depending on the network configuration, the use environment, the type of ship, and the like, when the feature selection is performed.
Wherein feature selection is distinguished from feature extraction by selecting a subset of the original data set to improve the performance of the network. The process of feature selection is a process of evaluating features, and which feature has a large influence on the output variable is selected. The invention adopts random forest to measure the importance of the features and selects the features with higher importance. The following methods can be employed:
1. feature importance measure, calculating the importance of a certain feature X:
1) and selecting corresponding off-bag data for each decision tree to calculate an off-bag data error, which is recorded as err1, wherein the off-bag data means that one data obtained by repeated sampling is used for training the decision tree when the decision tree is built each time, and at the moment, the data of about 1/3 is not utilized and does not participate in the building of the decision tree. The data can be used for evaluating the performance of the decision tree and calculating the prediction error rate of the model, namely the error of the data outside the bag;
2) randomly adding noise interference to the characteristic X of all samples of the data outside the bag (the value of the sample at the characteristic X can be randomly changed), and calculating the error of the data outside the bag again and recording the error as err 2;
3) assuming there are N trees in a forest, the importance of feature X ∑ (err2-errB 1)/N.
2. Selecting characteristics:
1) calculating the importance of each feature and sorting the features in descending order;
2) determining the proportion to be eliminated, eliminating the characteristics of the corresponding proportion according to the importance of the characteristics to obtain a new characteristic set
3) Repeating the process by using the new feature set until m features (m is a preset value) are left;
4) and selecting the characteristic set with the lowest error rate outside the bag according to the characteristic sets obtained in the process and the error rates outside the bag corresponding to the characteristic sets. Such as the relative distance between the own ship and the target ship, the speed of the own ship, the speed of the target ship, the heading of the target ship, the relative azimuth between the own ship and the target ship, the meeting attribute of the target ship, the relative heading between the own ship and the target ship, the relative speed between the own ship and the target ship, the speed ratio of the own ship and the target ship and the characteristic set of the target meeting characteristic. As described above, in the feature selection, the above-described features may be appropriately added or subtracted according to a network structure, a use environment, a ship type, and the like.
In an alternative embodiment, the neural network employs a back propagation neural network, which includes 10 input nodes, at least 2 hidden layers and 1 output node, and each hidden layer employs, for example, 60 nodes. It can be understood that the neural network may further employ a plurality of hidden layers other than two hidden layers, each hidden layer may further employ a plurality of nodes other than 60 nodes, and the number of nodes employed by the hidden layer and the hidden layer of the neural network is not particularly limited. The 10 input nodes are used for inputting the relative distance between the ship and the target ship, the speed of the target ship, the course of the target ship, the relative azimuth between the ship and the target ship, the meeting attribute of the target ship, the relative course between the ship and the target ship, the relative speed between the ship and the target ship, the speed ratio between the ship and the target meeting characteristic, and the 1 output node is used for outputting the steering angle of the ship.
Wherein each neuron structure is shown in fig. 5. Wherein x isiRepresenting an input vector, wiRepresenting a vector xiThe corresponding weights, y, represent the output vector. Setting an activation value a corresponding to an input layer for each sample x in the training data setl. All weights in the network are randomly assigned, forward propagation is used after the network inputs data, and the total error of the output nodes is calculated and propagated back to the network through back propagation to calculate the gradient. The neural network may also adopt other network structures, for example, and the present invention is not particularly limited. It can be understood that when the network structure changes, the input node of the network is rightThe ship navigation characteristics can be adaptively selected according to the network structure, and the characteristics can be appropriately increased or decreased.
In an alternative embodiment, the system further comprises:
a training module for training the neural network through a training data set, wherein the neural network is trained using a gradient descent method.
Wherein, the forward propagation: z is a radical ofl＝wlal-1+bl，al＝σ(zl)
Calculating the error generated by the output node:
back propagation error:l＝((wl+1)T l+1)⊙σ′(zl)
using the gradient descent method, the training parameters:
in the formula, wlWeight, z, representing the connection of neurons of layer l-1 to neurons of layer llRepresenting the neuronal input of layer l, alRepresenting the neuronal output of layer l, blRepresents the bias of layer i neurons, σ represents the activation function, C is a cost function for calculating the error between the output value and the actual value,
x denotes the input sample, y denotes the actual classification, aLIndicating the output of the prediction, L indicates the maximum number of layers of the neural network, L indicates a Hadamard product for point-to-point multiplication between matrices or vectors,
representing the training parameters.
In an alternative embodiment, the system further comprises:
a data set acquisition module for acquiring a raw data set from which the training data set is acquired,
wherein the raw data set comprises: each target vessel motion data, own vessel motion data, and corresponding decision data, wherein the corresponding decision data comprises a steering angle of the own vessel,
wherein the obtaining the training data set from the raw data set comprises: and carrying out primary screening, normalization, pretreatment and feature selection on the data in the data set to obtain the training data set.
It will be appreciated that vessel dynamic data is collected via a device or platform such as AIS, RADAR/ARPA, etc. When a target ship appears, the dynamic data of the ship and the target ship are subjected to target motion factor calculation. The target motion element calculation can be understood as analyzing the motion element information of the target ship from the collected dynamic data, and the analyzed data may include, for example, the azimuth (OB) of the target ship relative to the target ship, the azimuth (TB) of the target ship relative to the target ship, the relative motion heading (Cr) of the two ships, the relative motion speed (VR) of the two ships, the true motion heading (Ct) and the motion speed (VT) of the target ship, and the like. And after the calculation, judging the intersection characteristics of the target ship, judging the collision danger according to the intersection characteristics of the target ship, and recording the motion data and the corresponding decision data of the target ship and the ship when the collision danger exists so as to form the original data set.
The determining of the target ship encounter characteristics and the determining of the target ship encounter attributes may be performed by, for example, a method shown in fig. 6, which includes: and determining the type of the target ship and the intersection characteristic value of the target ship according to the navigation data of the target ship and the ship. When judging whether the two ships have collision danger or not, judging whether the ship and the target ship have collision danger or not according to a danger judgment threshold value. For example, if the target ship engagement feature value is greater than the risk determination threshold value, it is determined that there is a risk of collision, whereas there is no risk of collision.
The target ship encounter attribute is shown in fig. 7, for example, and the determining of the target ship encounter characteristic may include the following steps:
(1) when the coming ship is the overtaking ship, determining that the target ship is an E-type ship, and determining that the target ship rendezvous characteristic value TEP is 1;
(2) when the target ship crosses the starboard of the ship, determining that the target ship is a starboard crossing A-type ship, and determining that the target ship meets the characteristic value TEP to be 2;
(3) when the target ship meets the ship, determining that the target ship is a class A ship, and determining that the target ship meets the characteristic value TEP to be 6;
(4) when the target ship intersects with the starboard of the ship at a large angle, determining that the target ship is a B-type ship, and determining that the target ship meets a characteristic value TEP of 3 at the moment;
(5) when the target ship is a overtaking ship, determining that the target ship is a C-type ship when the relative azimuth of the target ship and the ship is less than 210 degrees, determining that the target ship is a D-type ship when the relative azimuth of the target ship and the ship is more than or equal to 210 degrees, and determining that the intersection characteristic value TEP of the target ship is 4;
(6) when the relative orientation of the target ship and the ship is between (247.5 degrees and 354 degrees) and the target ship and the port of the ship are crossed, the target ship is determined to be a D-type ship, and the intersection characteristic value TEP of the target ship is determined to be 5.
The above-mentioned exemplary judgment methods can be adapted according to the ship type, and the present invention is not limited in particular.
When training the neural network, a training data set is required for training. After a series of processing is performed on the original data set, a training data set is obtained. A series of processes includes preliminary screening, normalization, pre-processing, and feature selection. As mentioned above, feature selection may be performed using, for example, random forests, which are not described in detail herein.
The disclosure also relates to an electronic device comprising a server, a terminal and the like. The electronic device includes: at least one processor; a memory communicatively coupled to the at least one processor; and a communication component communicatively coupled to the storage medium, the communication component receiving and transmitting data under control of the processor; wherein the memory stores instructions executable by the at least one processor to implement the method of the above embodiments.
In an alternative embodiment, the memory is used as a non-volatile computer-readable storage medium for storing non-volatile software programs, non-volatile computer-executable programs, and modules. The processor executes various functional applications of the device and data processing, i.e., implements the method, by executing nonvolatile software programs, instructions, and modules stored in the memory.
The memory may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store a list of options, etc. Further, the memory may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some embodiments, the memory optionally includes memory located remotely from the processor, and such remote memory may be connected to the external device via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
One or more modules are stored in the memory and, when executed by the one or more processors, perform the methods of any of the method embodiments described above.
The product can execute the method provided by the embodiment of the application, has corresponding functional modules and beneficial effects of the execution method, and can refer to the method provided by the embodiment of the application without detailed technical details in the embodiment.
The present disclosure also relates to a computer-readable storage medium for storing a computer-readable program for causing a computer to perform some or all of the above-described method embodiments.
That is, as can be understood by those skilled in the art, all or part of the steps in the method for implementing the embodiments described above may be implemented by a program instructing related hardware, where the program is stored in a storage medium and includes several instructions to enable a device (which may be a single chip, a chip, or the like) or a processor (processor) to execute all or part of the steps of the method described in the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
In the description provided herein, numerous specific details are set forth. It is understood, however, that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.
Furthermore, those of ordinary skill in the art will appreciate that while some embodiments described herein include some features included in other embodiments, rather than other features, combinations of features of different embodiments are meant to be within the scope of the invention and form different embodiments. For example, in the claims, any of the claimed embodiments may be used in any combination.
It will be understood by those skilled in the art that while the present invention has been described with reference to exemplary embodiments, various changes may be made and equivalents may be substituted for elements thereof without departing from the scope of the invention. In addition, many modifications may be made to adapt a particular situation or material to the teachings of the invention without departing from the essential scope thereof. Therefore, it is intended that the invention not be limited to the particular embodiment disclosed, but that the invention will include all embodiments falling within the scope of the appended claims.
Claims (10)
1. A real-time ship collision avoidance method is characterized by comprising the following steps:
acquiring ship navigation data to be processed, wherein the ship navigation data comprises the ship navigation data to be processed and target ship navigation data to be processed;
processing the ship navigation data to be processed to obtain processed data, wherein the processed data comprises ship navigation characteristics;
and identifying the ship navigation characteristics through a neural network, and determining the steering angle of the ship.
2. The method of claim 1, wherein the processing the vessel voyage data to be processed to obtain processed data comprises:
preliminarily screening the ship navigation data to be processed to obtain first data;
normalizing the first data to obtain second data;
preprocessing the second data to obtain third data;
and selecting characteristics of the third data to obtain the processed data, wherein the processed data comprises ship navigation characteristics.
3. A method as claimed in claim 2, wherein said selecting features from said third data to obtain said processed data is performed by random forest.
4. The method according to any one of claims 1 to 3, wherein the vessel voyage characteristics include a relative distance between the own vessel and the target vessel, an own vessel speed, a target vessel heading, a relative azimuth between the own vessel and the target vessel, a target vessel encounter property, a relative heading between the own vessel and the target vessel, a relative speed between the own vessel and the target vessel, a speed ratio of the own vessel and the target vessel, and a target encounter characteristic.
5. The method of claim 1, wherein the neural network employs a back propagation neural network, the neural network comprising 10 input nodes, at least 2 hidden layers and 1 output node, wherein the 10 input nodes are used for inputting a relative distance between the own ship and the target ship, a speed of the own ship, a speed of the target ship, a heading of the target ship, a relative azimuth between the own ship and the target ship, a meeting property of the target ship, a relative heading between the own ship and the target ship, a relative speed between the own ship and the target ship, a speed ratio of the own ship and the target ship and a target meeting characteristic, and the 1 output node is used for outputting a steering angle of the own ship.
6. The method of claim 1, wherein the method further comprises: training the neural network through a training data set, wherein the neural network is trained using a gradient descent method.
7. The method of claim 6, wherein the method further comprises: obtaining a raw data set, obtaining the training data set from the raw data set,
wherein the raw data set comprises: each target vessel motion data, own vessel motion data, and corresponding decision data, wherein the corresponding decision data comprises a steering angle of the own vessel,
wherein the obtaining the training data set from the raw data set comprises: and carrying out primary screening, normalization, pretreatment and feature selection on the data in the data set to obtain the training data set.
8. A real-time collision avoidance system for a ship, the system comprising:
the system comprises a data acquisition module, a processing module and a processing module, wherein the data acquisition module is used for acquiring ship navigation data to be processed, and the ship navigation data comprises ship navigation data to be processed and target ship navigation data to be processed;
the data processing module is used for processing the ship navigation data to be processed to obtain processed data, wherein the processed data comprises ship navigation characteristics;
and the characteristic identification module is used for identifying the ship navigation characteristics through a neural network and determining the steering angle of the ship.
9. An electronic device comprising a memory and a processor, wherein the memory is configured to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the method of any one of claims 1-7.
10. A computer-readable storage medium, on which a computer program is stored, the computer program being executable by a processor for implementing the method according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011102937.1A CN112149237B (en) | 2020-10-15 | 2020-10-15 | Ship real-time collision prevention method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011102937.1A CN112149237B (en) | 2020-10-15 | 2020-10-15 | Ship real-time collision prevention method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112149237A true CN112149237A (en) | 2020-12-29 |
| CN112149237B CN112149237B (en) | 2024-07-16 |
Family
ID=73952038
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011102937.1A Active CN112149237B (en) | 2020-10-15 | 2020-10-15 | Ship real-time collision prevention method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112149237B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113232795A (en) * | 2021-06-01 | 2021-08-10 | 福州海联星信息科技有限公司 | Ship collision avoidance method and terminal |
| CN114860606A (en) * | 2022-05-25 | 2022-08-05 | 大连海事大学 | Method for testing stability of intelligent ship collision avoidance algorithm |
| CN115273555A (en) * | 2022-06-23 | 2022-11-01 | 集美大学 | Ship collision avoidance decision method for channel intersection area |
| CN116610125A (en) * | 2023-05-26 | 2023-08-18 | 北鲲睿航科技(上海)有限公司 | Collision prevention method and system for intelligent ship active collision avoidance system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103177290A (en) * | 2013-04-03 | 2013-06-26 | 大连海事大学 | Identification method for model of ship domain based on online self-organization neural network |
| CN108820157A (en) * | 2018-04-25 | 2018-11-16 | 武汉理工大学 | A kind of Ship Intelligent Collision Avoidance method based on intensified learning |
| CN110400491A (en) * | 2019-06-10 | 2019-11-01 | 北京海兰信数据科技股份有限公司 | A kind of Open sea area multiple target auxiliary Decision of Collision Avoidance method and decision system |
-
2020
- 2020-10-15 CN CN202011102937.1A patent/CN112149237B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103177290A (en) * | 2013-04-03 | 2013-06-26 | 大连海事大学 | Identification method for model of ship domain based on online self-organization neural network |
| CN108820157A (en) * | 2018-04-25 | 2018-11-16 | 武汉理工大学 | A kind of Ship Intelligent Collision Avoidance method based on intensified learning |
| CN110400491A (en) * | 2019-06-10 | 2019-11-01 | 北京海兰信数据科技股份有限公司 | A kind of Open sea area multiple target auxiliary Decision of Collision Avoidance method and decision system |
Non-Patent Citations (2)
| Title |
|---|
| 欧阳庆, 赵德鹏: "电子海图***中船舶避碰的神经网络方法的研究", 大连海事大学学报, no. 03 * |
| 王则胜;施朝健;: "基于改进的神经网络的船舶碰撞危险度的模型", 中国航海, no. 01 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113232795A (en) * | 2021-06-01 | 2021-08-10 | 福州海联星信息科技有限公司 | Ship collision avoidance method and terminal |
| CN114860606A (en) * | 2022-05-25 | 2022-08-05 | 大连海事大学 | Method for testing stability of intelligent ship collision avoidance algorithm |
| CN114860606B (en) * | 2022-05-25 | 2024-07-23 | 大连海事大学 | Method for testing stability of intelligent ship collision avoidance algorithm |
| CN115273555A (en) * | 2022-06-23 | 2022-11-01 | 集美大学 | Ship collision avoidance decision method for channel intersection area |
| CN115273555B (en) * | 2022-06-23 | 2024-04-09 | 集美大学 | Ship collision avoidance decision method for navigation channel intersection area |
| CN116610125A (en) * | 2023-05-26 | 2023-08-18 | 北鲲睿航科技(上海)有限公司 | Collision prevention method and system for intelligent ship active collision avoidance system |
| CN116610125B (en) * | 2023-05-26 | 2024-01-30 | 北鲲睿航科技(上海)有限公司 | Collision prevention method and system for intelligent ship active collision avoidance system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112149237B (en) | 2024-07-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112149237A (en) | Real-time ship collision avoidance method and system | |
| Nowlan | Maximum likelihood competitive learning | |
| Abraham et al. | Hybrid intelligent systems for stock market analysis | |
| CN112287468B (en) | Ship collision risk degree judging method and system | |
| US20150006444A1 (en) | Method and system for obtaining improved structure of a target neural network | |
| CN112598046B (en) | Target tactical intent recognition method in multi-machine cooperative air combat | |
| KR102042356B1 (en) | Method and system for estimating ship motion performance using artificial intelligence | |
| WO2008016109A1 (en) | Learning data set optimization method for signal identification device and signal identification device capable of optimizing the learning data set | |
| CN112668688B (en) | Intrusion detection method, system, equipment and readable storage medium | |
| CN110335466B (en) | Traffic flow prediction method and apparatus | |
| US20180137409A1 (en) | Method of constructing an artifical intelligence super deep layer learning model, device, mobile terminal, and software program of the same | |
| CN110161853A (en) | A kind of novel ship craft integrated automated driving system with real-time | |
| CN114239671A (en) | Apparatus and method for training a scale invariant convolutional neural network | |
| CN113762468A (en) | Classification model generation method based on missing data | |
| CN115409099A (en) | Internet of things flow anomaly detection model establishing method and detection method | |
| Salmond et al. | Mixture reduction algorithms for uncertain tracking | |
| CN113436125B (en) | Side-scan sonar simulation image generation method, device and equipment based on style migration | |
| Sharma | An exploratory study of chaos in human-machine system dynamics | |
| CN111582446B (en) | System for neural network pruning and neural network pruning processing method | |
| CN104063591B (en) | One-dimensional range profile identification method for non-library target based unified model | |
| Farrokhrooz et al. | A new method for spread value estimation in multi-spread PNN and its application in ship noise classification | |
| CN109726761B (en) | CNN evolution method, CNN-based AUV cluster working method, CNN evolution device and CNN-based AUV cluster working device and storage medium | |
| CN115659282B (en) | GA-KNN-based intelligent extraction method and system for effective motion characteristics of early warning target | |
| CN113327601B (en) | Method, device, computer equipment and storage medium for identifying harmful voice | |
| Popkov | Iterative Methods with Self-Learning for Solving Nonlinear Equations |
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 |