CN110488226A - A kind of submarine target localization method and device - Google Patents
A kind of submarine target localization method and device Download PDFInfo
- Publication number
- CN110488226A CN110488226A CN201910784901.7A CN201910784901A CN110488226A CN 110488226 A CN110488226 A CN 110488226A CN 201910784901 A CN201910784901 A CN 201910784901A CN 110488226 A CN110488226 A CN 110488226A
- Authority
- CN
- China
- Prior art keywords
- submarine target
- target
- track
- metric data
- submarine
- 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
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/02—Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
- G01S13/50—Systems of measurement based on relative movement of target
- G01S13/58—Velocity or trajectory determination systems; Sense-of-movement determination systems
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S5/00—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations
- G01S5/18—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations using ultrasonic, sonic, or infrasonic waves
- G01S5/28—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations using ultrasonic, sonic, or infrasonic waves by co-ordinating position lines of different shape, e.g. hyperbolic, circular, elliptical or radial
Abstract
This application provides a kind of submarine target localization method and device, information at the time of receiving submarine target feedback signal based on default Hyperbolic Equation group and each base station determines the position metric data of submarine target.Based on goal-selling motion model, the motion state model of submarine target is modified.Based on goal-selling tracking gate, metric data is grouped, target metric data is obtained and clusters.The default starting track of submarine target and the interior target metric data that clusters are associated it is assumed that generating multiple relevance assumptions.Optimal relevance assumption is chosen from multiple relevance assumptions, using track corresponding to optimal relevance assumption as the true track of submarine target.Underwater robot will be used to characterize, and the metric data of environment present position and the track of underwater robot are associated hypothesis under water, in conjunction with revised motion state model, the true track of underwater robot is obtained, the specific location of underwater robot can be accurately grasped based on true track.
Description
Technical field
This application involves technical field of data processing more particularly to a kind of submarine target localization methods and device.
Background technique
Contain mining site resource and living resources abundant in ocean, it is untethered with the development and utilization to marine resources
The research of underwater robot (Autonomous UnderwaterVehicle, AUV) becomes a popular field.On the one hand,
Cableless underwater robot can explore underwater environment to avoid artificial, prevent casualties, what processing cannot manually be completed
Task.On the other hand, cableless underwater robot can arbitrarily detect underwater environment, and obtaining artificial explore can not see with detection instrument
The underwater information of survey.Cableless underwater robot may occur accident when carrying out underwater operation and lead to not work normally, and be
It can salvage in time and repair cableless underwater robot, need to be quickly and accurately positioned cableless underwater robot environment under water
Specific location.
Therefore, a kind of submarine target localization method is needed, for accurately grasping cableless underwater robot environment under water
Specific location.
Summary of the invention
This application provides a kind of submarine target localization method and devices, for accurately grasping cableless underwater robot in water
The specific location of lower environment.
To achieve the goals above, this application provides following technical schemes:
The embodiment of the present application first aspect discloses a kind of submarine target localization method, the submarine target localization method packet
It includes:
Information at the time of receiving submarine target feedback signal based on default Hyperbolic Equation group and each base station, determines institute
State the metric data of submarine target, wherein the default Hyperbolic Equation group for characterize the submarine target with it is described each
Positional relationship between base station, the metric data is for characterizing submarine target specific location locating for environment under water；
Based on goal-selling motion model, the motion state model of the submarine target is modified, the movement shape
States model is used to characterize the motion state of the submarine target；
Based on goal-selling tracking gate, the metric data is grouped, target metric data is obtained and clusters, it is described poly-
Target metric data in cluster is associated with the true track of the submarine target；
After the default starting track of the submarine target and the target metric data being associated it is assumed that combining amendment
Motion state model, generate multiple relevance assumptions, the relevance assumption for characterize the target metric data with it is described pre-
If originating the track of track composition；
Optimal relevance assumption is chosen from the multiple relevance assumption, and track corresponding to the optimal relevance assumption is made
For the true track of the submarine target.
Optionally, described to be connect based on default Hyperbolic Equation group and each base station in above-mentioned submarine target localization method
Information at the time of receiving submarine target feedback signal determines the metric data of the submarine target, comprising:
Obtain information at the time of each base station receives submarine target feedback signal；
Preset position information based on each time information and each base station, obtains the default hyperbola side
The solution of journey group, the solution of the default Hyperbolic Equation group characterize the specific location of the submarine target；
According to the specific location of the submarine target, the metric data of the submarine target is determined.
Optionally, described to be based on goal-selling motion model in above-mentioned submarine target localization method, to the underwater mesh
Target motion state model is modified, comprising:
The default initial motion status information of the submarine target is obtained, the default initial motion status information characterizes institute
State motion state locating for submarine target initial time；
According to the default initial motion status information and goal-selling motion model, the submarine target is moved
State estimation obtains the movement state information, and the movement state information is for characterizing the submarine target current time institute
The motion state at place；
According to the movement state information, the motion state model of the submarine target is corrected.
Optionally, in above-mentioned submarine target localization method, optimal relevance assumption is chosen from the multiple relevance assumption,
Using track corresponding to the optimal relevance assumption as the true track of the submarine target, comprising:
Calculate the probability of the multiple relevance assumption；
According to the probability size, optimal relevance assumption is determined, and track corresponding to the optimal relevance assumption is made
For the true track of the submarine target, wherein the probability of the optimal relevance assumption is higher than other relevance assumptions.
The embodiment of the present application second aspect discloses a kind of underwater object locating device, the underwater object locating device packet
It includes:
Determination unit is measured, for receiving submarine target feedback signal based on default Hyperbolic Equation group and each base station
At the time of information, determine the metric data of the submarine target, wherein the default Hyperbolic Equation group is for characterizing the water
Positional relationship between lower target and each base station, the metric data is for characterizing submarine target environment under water
Locating specific location；
Modifying model unit, for being based on goal-selling motion model, to the motion state model of the submarine target into
Row amendment, the motion state model are used to characterize the motion state of the submarine target；
Measurement clusters unit, for being based on goal-selling tracking gate, is grouped to the metric data, obtains aim parameter
Measured data clusters, and the interior target metric data that clusters is associated with the true track of the submarine target；
Associative cell is measured, for closing the default starting track of the submarine target and the target metric data
Connection generates multiple relevance assumptions, the relevance assumption is for characterizing the target it is assumed that in conjunction with revised motion state model
The track of metric data and the default starting track composition；
Track determination unit, for choosing optimal relevance assumption from the multiple relevance assumption, by the optimal association
Assuming that true track of the corresponding track as the submarine target.
Optionally, in above-mentioned underwater object locating device, the measurement determination unit includes:
Module is obtained, for obtaining information at the time of each base station receives submarine target feedback signal；
It solves module and obtains institute for the preset position information based on each time information and each base station
The solution of default Hyperbolic Equation group is stated, the solution of the default Hyperbolic Equation group characterizes the specific location of the submarine target；
Determining module determines the metric data of the submarine target for the specific location according to the submarine target.
Optionally, in above-mentioned underwater object locating device, the Modifying model unit includes:
Module is obtained, for obtaining the default initial motion status information of the submarine target, the default initial motion
Status information characterizes motion state locating for the submarine target initial time；
Estimation module is used for according to the default initial motion status information and goal-selling motion model, to the water
Lower target carries out state estimation, obtains the movement state information, and the movement state information is described underwater for characterizing
Motion state locating for target current time；
Correction module, for correcting the motion state model of the submarine target according to the movement state information.
Optionally, in above-mentioned underwater object locating device, the track determination unit includes:
Computing module, for calculating the probability of the multiple relevance assumption；
Determining module, for according to the probability size, determining optimal relevance assumption, and by the optimal relevance assumption institute
True track of the corresponding track as the submarine target, wherein the probability of the optimal relevance assumption is higher than other associations
Assuming that.
Submarine target localization method provided herein and device are connect based on default Hyperbolic Equation group and each base station
Information at the time of receiving submarine target feedback signal determines the metric data of submarine target.Wherein, Hyperbolic Equation group is preset to use
Positional relationship between characterization submarine target and each base station.Movement based on goal-selling motion model, to submarine target
State model is modified.Based on goal-selling tracking gate, metric data is grouped, target metric data is obtained and clusters,
Target metric data in clustering is associated with the true track of submarine target.By the default starting track and target of submarine target
Metric data is associated, and in conjunction with revised motion state model, generates multiple relevance assumptions, relevance assumption is for characterizing mesh
The track of scalar measured data and default starting track composition.Optimal relevance assumption is chosen from multiple relevance assumptions, by optimal pass
Connection assumes true track of the corresponding track as submarine target.Based on the application, by default Hyperbolic Equation group and respectively
Information at the time of a base station receives cableless underwater robot feedback signal determines the metric data of cableless underwater robot, according to
It is modified according to motion state model of the target movement model to cableless underwater robot, by the metric data and untethered underwater machine
The starting track of device people is associated, and in conjunction with revised motion state model, obtains the ring under water of cableless underwater robot
Track in border can accurately grasp the specific location of cableless underwater robot environment under water based on the track.
Detailed description of the invention
In order to illustrate the technical solutions in the embodiments of the present application or in the prior art more clearly, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of application for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is a kind of schematic diagram of submarine target localization method provided by the embodiments of the present application；
Fig. 2 is a kind of signal of the specific implementation of determining submarine target metric data provided by the embodiments of the present application
Figure；
Fig. 3 is a kind of specific implementation of motion state model for correcting submarine target provided by the embodiments of the present application
Schematic diagram；
Fig. 4 is a kind of schematic diagram to cluster provided by the embodiments of the present application；
Fig. 5 is a kind of specific implementation that optimal relevance assumption is chosen from multiple relevance assumptions provided by the embodiments of the present application
The schematic diagram of mode；
Fig. 6 is a kind of algorithm flow chart of interactive multi-model provided by the embodiments of the present application；
Fig. 7 is a kind of schematic diagram of relevance assumption tree provided by the embodiments of the present application；
Fig. 8 is a kind of structural schematic diagram of underwater object locating device provided by the embodiments of the present application；
Fig. 9 is the structural schematic diagram of another underwater object locating device provided by the embodiments of the present application；
Figure 10 is the structural schematic diagram of another underwater object locating device provided by the embodiments of the present application；
Figure 11 is the structural schematic diagram of another underwater object locating device provided by the embodiments of the present application.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present application, technical solutions in the embodiments of the present application carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of embodiments of the present application, instead of all the embodiments.It is based on
Embodiment in the application, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall in the protection scope of this application.
As shown in Figure 1, being a kind of schematic diagram of submarine target localization method provided by the embodiments of the present application, including walk as follows
It is rapid:
S101: information at the time of receiving submarine target feedback signal based on default Hyperbolic Equation group and each base station,
Determine the metric data of submarine target.
In S101, default Hyperbolic Equation group is used to characterize the positional relationship between submarine target and each base station.In advance
If in Hyperbolic Equation group including multiple Hyperbolic Equations, the quantity of Hyperbolic Equation and the quantity of base station be it is identical, arbitrarily
One Hyperbolic Equation is used to characterize the positional relationship between submarine target and any two base station.In the embodiment of the present application,
The specific location of each base station can be configured according to the actual situation by technical staff, and the quantity of base station is no less than 3.
Specifically, the specific location of first base station is B by taking 4 base stations as an example1(x1,y1), the specific location of the second base station
For B2(x2,y2), the specific location of third base station is B3(x3,y3), the specific location of the 4th base station is B4(x4,y4), submarine target
Specific location be T (x, y), first base station receive signal time be t1, the time that the second base station receives signal is t2,
The time that third base station receives signal is t3, the time that the 4th base station receives signal is t4.Wherein, Hyperbolic Equation is preset
Hyperbolic Equation in group is respectively shown in formula (1-1), (2-1) and (3-1).Simultaneous formula (1-1), (2-1) and (3-1) meter
Calculation obtains the solution of T (x, y).
In formula (1-1), (2-1) and (3-1), c characterizes the underwater spread speed of signal.
It should be noted that above-mentioned specific implementation process is used only for illustrating.
It should be noted that based on default Hyperbolic Equation group and each base station receive submarine target feedback signal when
Carve information, determine the specific implementation of submarine target metric data can be found in following Fig. 2 provided by the embodiments of the present application and
The corresponding explanation of Fig. 2.
In the embodiment of the present application, each base station is arranged on the water surface, and it is anti-to receive submarine target by sonar communication mode
Feedback signal.In each sonar scan period, each base station will record letter at the time of once receiving submarine target feedback signal
Breath.
S102: it is based on goal-selling motion model, the motion state model of submarine target is modified.
In S102, motion state model is used to characterize the motion state of submarine target, due to the movement shape of submarine target
State is motion state complicated and changeable, needing that multi-motion state model is combined to co-express submarine target.Based on goal-selling
Motion model can adaptively adjust various types motion state model, so that the motion state model of submarine target
It obtains accurately correcting and update, to accurately reflect submarine target motion state locating for environment under water.Implement in the application
In example, the motion state model of submarine target includes but is not limited to be: uniform motion model accelerates model, divertical motion
The specific type of the motion models such as model, motion state model can be configured according to the actual situation by technical staff.
It should be noted that being based on goal-selling motion model, the motion state model of submarine target is modified
Specific implementation can be found in the corresponding explanation of following Fig. 3 provided by the embodiments of the present application and Fig. 3.
S103: it is based on goal-selling tracking gate, metric data is grouped, target metric data is obtained and clusters.
In S103, the interior target metric data that clusters is associated with the true track of submarine target.Target following door refers to
: centered on the predicted position of submarine target, for reflecting that the submarine target is likely to occur the region of range.Target following
The specific range of door can be configured according to the actual situation by technical staff, and the embodiment of the present application is without limitation.
Specifically, the source according to metric data, establishes the matrix M (k) that clusters, shown in the matrix such as formula (4) that clusters.In
In formula (4), k characterizes the corresponding generation time of metric data, vjtCharacterize metric data, vjtValue characterization metric data
Source, shown in the source of metric data such as formula (5), j=1,2 ..., mk, t=0,1 ..., T, T+1 ..., T+mk, mkTable
Kth moment obtained all measurement numbers are levied, T characterizes submarine target quantity known to the kth moment, and t characterizes submarine target.
M (k)=[vjt] (4)
In formula (5), a, b, c, d characterize 4 kinds of sources of metric data respectively, and the measurement of a situation characterization serial number j is fallen
Enter in the target following door of submarine target t, the measurement that b situation characterizes serial number j is new submarine target, and c situation characterizes vj0Amount
Surveying is false-alarm or clutter, and the measurement of d situation characterization serial number j is not fallen in the target following door of submarine target t.In this Shen
Please be in embodiment, the source of target metric data is a situation.
It should be noted that above-mentioned specific implementation process be used only for for example, detailed process can refer to assume with
Track algorithm, multiple hypotheis tracking algorithm is specifically: carrying out data pass based on whole metric data obtained by multiple sonar scan periods
Connection, the result of data correlation not only relies on current period and relies on previous metric data, when data correlation clashes,
A variety of logic associations are formed to assume time to postpone to make a decision, subsequent metric data to be allowed to solve this uncertainty, change and
Yan Zhi determines the true track of the corresponding submarine target of current metric data according to subsequent metric data.
S104: the default starting track and target metric data of submarine target are associated it is assumed that in conjunction with revised
Shape model is moved, multiple relevance assumptions are generated.
In S104, relevance assumption is used to characterize the track of target metric data and default starting track composition.Aim parameter
There are three types of relationships for measured data and starting track tool, and three kinds of relationships are respectively: target metric data is the continuity for originating track, target
Metric data is the starting of new submarine target, and target metric data is false-alarm.According to target metric data and starting track it
Between three kinds of relationships, the true track of submarine target and the relationship of target metric data are associated it is assumed that generating multiple passes
Connection is it is assumed that any one relevance assumption characterizes a track.
It should be noted that default starting track includes but is not limited to be: technical staff is according to the default first of submarine target
The track of beginning position creation, alternatively, the true track that submarine target has currently been identified.
Specifically, during the relationship of true track and target metric data to submarine target is associated hypothesis,
It presets target and measures set Z (k)={ Zm(k), m=1,2 ..., MkAnd relevance assumption setWherein, k characterizes the corresponding generation time of metric data, and m characterizes the index that target measures, Mk
The total number that target measures is characterized, i characterizes the index of relevance assumption, and I (k) characterizes the total number of relevance assumption.Then, it will be associated with
Assuming that set omegak-1Set Z (k) is measured with target and carries out data correlation, obtains relevance assumption set omegakIn relevance assumption.
Above-mentioned in the specific implementation, any one target measures all in the presence of three kinds of data in target metric data set Z (k)
A possibility that a possibility that association, these three data correlations, is respectively as follows: the confirmation that target measurement is already present submarine target and navigates
The continuation of mark, target, which measures, to be the measurement of emerging submarine target and creates new track, and it is false-alarm that target, which measures,.It is being associated with
Assuming that set omegak-1During carrying out data correlation with target metric data set Z (k), the true boat of each submarine target
Mark can only be associated with a target measurement.
Specifically, for track and target metric data in cluster shown in Fig. 4, to the generating process of relevance assumption into
Row explanation:
In Fig. 4, T1 and T2 are respectively the starting track of two submarine targets, and M1, M2 and M3 are respectively three aim parameters
It surveys.Since the true track of each submarine target can only be associated with a target measurement, to the track and mesh in clustering
There are following three kinds of relevance assumptions for scalar measured data: defining track T3 (T2, M2), it is assumed that it is that track T2 continues that target, which measures M2,；
It defines N2 (M2), it is assumed that target measures the starting that M2 is the track of emerging submarine target；It defines F2 (M2), it is assumed that aim parameter
Surveying M2 is false-alarm.
It is mutual exclusion between track T3, N2 and F2 in above-mentioned three kinds of relevance assumptions.And so on, define track T4
(T1, M1), track T5 (T1, M3), track T6 (T2, M1), track T7 (T2, M3), N1 (M1), N3 (M3), F1 (M1), F1
(M3), relevance assumption set H={ H is obtainedi, i=1,2,3 ... }.
In relevance assumption set H={ Hi, i=1,2,3 ... } in:
H1:T1,T2,N1,N3；
H2:T1,T2,F1,F2,F3；
H3:T1,T4,N3；
H4:T1,T4,F3；
H5:T3,T7,N1；
H6:T3,T7,F1。
It should be noted that above-mentioned specific implementation process is merely illustrative, and in multiple hypotheis tracking algorithm, relevance assumption
It is the distribution interconnecting relation of one group of submarine target and target metric data in clustering, any one clusters false including multiple associations
If any one relevance assumption is all determined by the relationship of target metric data and the true track of submarine target.
S105: choosing optimal relevance assumption from multiple relevance assumptions, using track corresponding to optimal relevance assumption as
The true track of submarine target.
In S105, the probability of each relevance assumption is different, and the probability of relevance assumption measures number for characterizing target
According to the correlation degree with the true track of submarine target, probability is higher, then the target metric data belongs to the possibility of true track
Property is higher.Therefore, the highest relevance assumption of probability is chosen as optimal relevance assumption, by track corresponding to optimal relevance assumption
True track as submarine target.
It should be noted that the specific implementation that optimal relevance assumption is chosen from multiple relevance assumptions can be found in it is following
The corresponding explanation of Fig. 5 and Fig. 5 provided by the embodiments of the present application.
In the embodiment of the present application, submarine target feedback signal is received based on default Hyperbolic Equation group and each base station
At the time of information, determine the metric data of submarine target.Wherein, preset Hyperbolic Equation group for characterize submarine target with it is each
Positional relationship between base station.Based on goal-selling motion model, the motion state model of submarine target is modified.It is based on
Goal-selling tracking gate, is grouped metric data, obtains target metric data and clusters, cluster in target metric data with
The true track of submarine target is associated.The default starting track and target metric data of submarine target are associated it is assumed that
Generate multiple relevance assumptions.Optimal relevance assumption is chosen from multiple relevance assumptions, by track corresponding to optimal relevance assumption
True track of the true track as submarine target as submarine target.Based on the application, by presetting Hyperbolic Equation group
Information at the time of receiving cableless underwater robot feedback signal with each base station determines the measurement number of cableless underwater robot
According to being modified to the motion state model of cableless underwater robot according to target movement model, by the metric data and untethered
The starting track of underwater robot is associated, in conjunction with revised motion state model, obtain cableless underwater robot
Track in underwater environment can accurately grasp the specific location of cableless underwater robot environment under water based on the track.
Optionally, as shown in Fig. 2, being a kind of specific reality of determining submarine target metric data provided by the embodiments of the present application
The schematic diagram of existing mode, includes the following steps:
S201: information at the time of each base station receives submarine target feedback signal is obtained.
S202: the preset position information based on each time information and each base station obtains default Hyperbolic Equation group
Solution.
In S202, the specific location of the solution characterization submarine target of Hyperbolic Equation group is preset.It, can be with based on time information
Determine that base station receives the time of submarine target feedback signal.Preset position information based on base station can determine the tool of base station
Body position.
Specifically, the specific location of first base station is B by taking 4 base stations as an example1(20,30), the specific location of the second base station
For B2(40,50), the specific location of third base station are B3(50,60), the specific location of the 4th base station are B4(60,70), underwater mesh
Target specific location is T (x, y), and the time that first base station receives signal is 20S, and the time that the second base station receives signal is
30S, the time that third base station receives signal is 40S, and the time that the 4th base station receives signal is 50S.It wherein, will be each
The coordinate value of base station and each time substitute into above-mentioned formula (1-1), (2-1) and (3-1), then preset double in Hyperbolic Equation group
Curvilinear equation is respectively shown in formula (1-2), (2-2) and (3-2).T is calculated in simultaneous formula (1-2), (2-2) and (3-2)
(10,30)。
In formula (1-2), (2-2) and (3-2), c characterizes the underwater spread speed of signal,
It should be noted that above-mentioned specific implementation process is used only for illustrating.
S203: according to the specific location of submarine target, the metric data of submarine target is determined.
In S203, measurement of the specific location of submarine target as submarine target.
In the embodiment of the present application, information at the time of each base station receives submarine target feedback signal is obtained, based on each
The preset position information of a time information and each base station obtains the solution of default Hyperbolic Equation group, wherein default hyperbola side
The specific location of the solution characterization submarine target of journey group.According to the specific location of submarine target, the metric data of submarine target is determined.
Based on the application, letter at the time of receiving cableless underwater robot feedback signal based on default Hyperbolic Equation group and each base station
Breath, determines the metric data of cableless underwater robot, can guarantee accurately to grasp cableless underwater robot environment under water in real time
Specific location.
Optionally, as shown in figure 3, for one kind provided by the embodiments of the present application based on goal-selling motion model to underwater mesh
The schematic diagram for the specific implementation that target motion state model is modified, includes the following steps:
S301: the default initial motion status information of submarine target is obtained.
Wherein, motion state locating for initial motion status information characterization submarine target initial time is preset.Initial motion
Status information includes but is not limited to be: technical staff according to the preset movement state information of initial position of submarine target,
Alternatively, technical staff is current according to submarine target it has been found that the obtained movement state information of true track.
S302: according to initial motion status information and goal-selling motion model is preset, movement shape is carried out to submarine target
State estimation, obtains movement state information.
In S302, movement state information is for characterizing motion state locating for submarine target current time.
It should be noted that in the embodiment of the present application, goal-selling motion model is interactive multi-model.It is interactive more
Model refers to: using between the probability of occurrence and different motion state models of preset each motion state model
Transition probability is adaptively adjusted the probability of each motion state model, then according to the probability of each motion state model, adds
Weigh each motion state model of submarine target.According to initial motion status information is preset, the initial motion of submarine target is determined
State, initial motion state of the interactive multi-model based on submarine target, determines the motion state at submarine target current time.
It should be noted that the algorithm flow chart of interactive multi-model can be found in Fig. 6, the specific calculating of interactive multi-model
Process includes:
A1, using the motion state model at -1 moment of submarine target kth as the input of the filter at kth moment, according to public
The motion state is calculated in the probability at -1 moment of kth in formula (6), and every kind of motion state model can all correspond to a kind of filter.
In formula (6), u characterizes the motion state model of submarine target, and r characterizes the number of motion state types of models,For normalization coefficient,Specific calculating process such as formula (7) shown in, pijCharacterize motion state model i and motion state mould
Transition probability between type j.
A2, foundation formula (8) recalculate the state estimation of each motion state model at -1 moment of submarine target kth,
The covariance of the motion state model i and motion state model j at -1 moment of kth are calculated according to formula (9).
A3, foundation likelihood function determine the probability of occurrence of the motion state model at submarine target kth moment, and likelihood function is such as
Shown in formula (10), shown in the probability of occurrence such as formula (11) of the motion state model at kth moment.
In formula (10), SjFor the covariance of the motion state model probability of occurrence at kth moment, N is constant, and Z (k) is
Default measurement model.
In formula (11), c is normaliztion constant, shown in the specific calculating process such as formula (12) of c.
A4, by the motion state model at -1 moment of submarine target kth and motion state model i and motion state model j
Between covariance, substitute into the corresponding filter of motion state model j and be filtered, obtain the motion state model at kth moment
State estimation and the motion state model at k moment error covariance.Wherein, the shape of the motion state model at kth moment
State estimation is as shown in formula (13), shown in the error covariance such as formula (14) of the motion state model at kth moment.
S303: according to movement state information, the motion state model of submarine target is corrected.
In S303, revised motion state model can accurately reflect the fortune of the submarine target under current location
Dynamic state.
In the embodiment of the present application, the default initial motion status information of submarine target is obtained, according to default initial motion
Status information and goal-selling motion model carry out state estimation to submarine target, obtain movement state information.According to fortune
Dynamic status information, corrects the motion state model of submarine target.Based on the application, revised motion state model can be accurate
The motion state for reflecting cableless underwater robot environment under water, to improve the accurate of cableless underwater robot Underwater Navigation
Property.
Optionally, as shown in figure 5, choosing optimal association from multiple relevance assumptions for one kind provided by the embodiments of the present application
The schematic diagram of the specific implementation of hypothesis, includes the following steps:
S501: the probability of multiple relevance assumptions is calculated.
In S501, the calculating process of relevance assumption probability includes:
B1, it is based on default relevance assumption Θk-1,sWith correlating event θk, determine relevance assumption Θk,l.Wherein, it is false to preset association
If Θk-1,sProbability be it is known that θkCharacterization target metric data is associated with submarine target, relevance assumption Θk,lSuch as formula (15)
It is shown.
Θk,l={ Θk-1,s,θk} (15)
B2, Belize formula, calculating relevance assumption Θ are utilizedk,lPosterior probability, shown in calculating process such as formula (16).
In formula (16), k characterizes the corresponding generation time of metric data, and l, s are the index of metric data, ZkTo measure
Data, Zk-1For the metric data for belonging to default starting track, c is normaliztion constant, the calculating process of c such as formula (17) institute
Show.
C=p (Zk/Zk-1) (17)
The clutter of B3, under water environment is in the case where being uniformly distributed, if metric data ZkFor underwater clutter, then vacation is associated with
If Θk,lProbability be V-1.If metric data ZkWith default starting track association, relevance assumption Θk,lProbability function meet it is high
This distribution, and to relevance assumption Θk,lIn metric data be marked, the t that metric data is identified asi.If metric data ZkReturn
Belong to new submarine target, it is determined that each measurement is evenly distributed, then obtains being associated with vacation with formula (19) by formula (18)
If Θk,lProbability, relevance assumption Θk,lProbability such as formula (20) shown in.
In formula (19), δ (θk) characterize the true track whether sonar detects submarine target, ψ (θk) characterization association vacation
If υ (θk) the emerging submarine target of characterization.
In formula (20), μ (Φ (θk)) characterization false-alarm targets prior probability partition function, μn(υ(θk)) characterize newly to go out
The prior probability partition function of existing submarine target, mkCharacterize the number measured.It is based on formula (20) it is found that emerging underwater
The number of target is υ, and the number of false-alarm is Φ, and the number for belonging to the measurement of the true track of submarine target isIt is a.
It should be noted that when measurement i is associated with the submarine target r of motion state model j, the fortune of other submarine targets
The probability that dynamic state model is qAs shown in formula (21).
In formula (21), pkFor the k moment corresponding associated probability of hypothesis, wirCharacterization measures i in default measurement model
Observation scope in.It is rightIt is weighted, submarine target r can be obtained and measures the associated probability of iProbabilitySuch as
Shown in formula (22).
Simultaneous formula (10) and formula (22), obtain the probability of new motion state modelProbabilityMeter
Shown in calculation process such as formula (23).
It should be noted that probabilityGoal-selling motion model be will affect to the motion state model of submarine target
Amendment.In interactive multi-model, according to probabilityUpdate the end value of above-mentioned formula (11).
S502: according to probability size, optimal relevance assumption is determined, and using track corresponding to optimal relevance assumption as water
The true track of lower target.
In S502, the probability of optimal relevance assumption is higher than other relevance assumptions.
It should be noted that in the embodiment of the present application, using N-scan beta pruning method, realizing and being determined most according to probability size
This process of excellent relevance assumption.Wherein, N-scan beta pruning method is a kind of depth for controlling relevance assumption tree reduction association vacation in turn
If the algorithm of quantity.
Specifically, the realization principle of N-scan beta pruning method are as follows: the quantity for reducing relevance assumption by deferring sentence, to drop
The calculation amount of low relevance assumption probability calculation, when the depth of relevance assumption tree is N, by relevance assumption caused by the k-N moment
In k moment beta pruning.By taking the relevance assumption tree shown in Fig. 7 as an example, the k-3 moment, the k-2 moment, there are track T2 and boats there are track T1
Mark T3, the k-1 moment, there are track T7, track T8, track T9 and track T10 there are track T4, track T5 and track T6, k moment.
At the k-3 moment, the probability highest of track T8 is determined, then by track T1, track T2, track T5, track T8 institute in relevance assumption tree
The relevance assumption branch of composition is retained, other relevance assumption branches in relevance assumption tree are deleted.Accordingly, it is determined that track
Relevance assumption probability highest corresponding to T1, track T2, track T5 and track T8, thus by track T1, track T2, track T5 and
The track of track T8 composition is as true track.
It should be noted that the above process is used only for illustrating.
In the embodiment of the present application, the probability for calculating multiple relevance assumptions determines that optimal association is false according to probability size
And if wherein using track corresponding to optimal relevance assumption as the true track of submarine target, the probability of optimal relevance assumption
Higher than other relevance assumptions.The work of track corresponding to optimal relevance assumption is chosen according to the probability of relevance assumption based on the application
For true track, the specific location of cableless underwater robot environment under water can be accurately grasped based on true track.
It is corresponding with above-mentioned submarine target localization method provided by the embodiments of the present application, as shown in figure 8, implementing for the application
A kind of structural schematic diagram for underwater object locating device that example provides, the device include:
Determination unit 100 is measured, for receiving submarine target feedback based on default Hyperbolic Equation group and each base station
Information at the time of signal determines the metric data of submarine target, wherein default Hyperbolic Equation group for characterize submarine target with
Positional relationship between each base station, metric data is for characterizing submarine target specific location locating for environment under water.
Modifying model unit 200 carries out the motion state model of submarine target for being based on goal-selling motion model
Amendment, motion state model are used to characterize the motion state of submarine target.
Measurement clusters unit 300, for being based on goal-selling tracking gate, is grouped to metric data, obtains aim parameter
Measured data clusters, and the interior target metric data that clusters is associated with the true track of submarine target.
Associative cell 400 is measured, for the default starting track and target metric data of submarine target to be associated vacation
If generating multiple relevance assumptions in conjunction with revised motion state model, relevance assumption is for characterizing target metric data and pre-
If originating the track of track composition.
Track determination unit 500, for choosing optimal relevance assumption from multiple relevance assumptions, by optimal relevance assumption institute
True track of the corresponding track as submarine target.
In the embodiment of the present application, submarine target feedback signal is received based on default Hyperbolic Equation group and each base station
At the time of information, determine the metric data of submarine target.Wherein, preset Hyperbolic Equation group for characterize submarine target with it is each
Positional relationship between base station.Based on goal-selling motion model, the motion state model of submarine target is modified.It is based on
Goal-selling tracking gate, is grouped metric data, obtains target metric data and clusters, cluster in target metric data with
The true track of submarine target is associated.The default starting track and target metric data of submarine target are associated it is assumed that
Generate multiple relevance assumptions.Optimal relevance assumption is chosen from multiple relevance assumptions, by track corresponding to optimal relevance assumption
True track of the true track as submarine target as submarine target.Based on the application, by presetting Hyperbolic Equation group
Information at the time of receiving cableless underwater robot feedback signal with each base station determines the measurement number of cableless underwater robot
According to being modified to the motion state model of cableless underwater robot according to target movement model, by the metric data and untethered
The starting track of underwater robot is associated, and in conjunction with the motion state model of revised cableless underwater robot, obtains nothing
The true track in an underwater environment of cable underwater robot can accurately grasp cableless underwater robot based on the true track
The specific location of environment under water.
Optionally, as shown in figure 9, being the structural representation of another underwater object locating device provided by the embodiments of the present application
Figure, wherein measuring determination unit 100 includes:
Module 101 is obtained, for obtaining information at the time of each base station receives submarine target feedback signal.
Module 102 is solved, for the preset position information based on each time information and each base station, obtains default hyperbolic
Line solution of equations presets the specific location of the solution characterization submarine target of Hyperbolic Equation group.
Determining module 103 determines the metric data of submarine target for the specific location according to submarine target.
In the embodiment of the present application, information at the time of each base station receives submarine target feedback signal is obtained, based on each
The preset position information of a time information and each base station obtains the solution of default Hyperbolic Equation group, wherein default hyperbola side
The specific location of the solution characterization submarine target of journey group.According to the specific location of submarine target, the metric data of submarine target is determined.
Based on the application, letter at the time of receiving cableless underwater robot feedback signal based on default Hyperbolic Equation group and each base station
Breath, determines the metric data of cableless underwater robot, can guarantee accurately to grasp cableless underwater robot environment under water in real time
Specific location.
Optionally, as shown in Figure 10, the structure for another underwater object locating device provided by the embodiments of the present application is shown
It is intended to, wherein Modifying model unit 200 includes:
Module 201 is obtained, for obtaining the default initial motion status information of submarine target, presets initial motion state letter
Motion state locating for breath characterization submarine target initial time.
Estimation module 202 presets initial motion status information and goal-selling motion model for foundation, to submarine target
State estimation is carried out, obtains movement state information, movement state information is for characterizing locating for submarine target current time
Motion state.
Correction module 203, for correcting the motion state model of submarine target according to movement state information.
In the embodiment of the present application, the default initial motion status information of submarine target is obtained, according to default initial motion
Status information and goal-selling motion model carry out state estimation to submarine target, obtain movement state information.According to fortune
Dynamic status information, corrects the motion state model of submarine target.Based on the application, it is based on the application, revised motion state
Model can accurately reflect the motion state of cableless underwater robot environment under water, so that it is underwater to improve cableless underwater robot
The accuracy of positioning.
Optionally, as shown in figure 11, the structure for another underwater object locating device provided by the embodiments of the present application is shown
It is intended to, wherein track determination unit 500 includes:
Computing module 501, for calculating the probability of multiple relevance assumptions.
Determining module 502, for determining optimal relevance assumption, and will be corresponding to optimal relevance assumption according to probability size
True track of the track as submarine target, wherein the probability of optimal relevance assumption is higher than other relevance assumptions.
In the embodiment of the present application, the probability for calculating multiple association henhouses determines that optimal association is false according to probability size
And if using track corresponding to optimal relevance assumption as the true track of submarine target.Based on the application, according to relevance assumption
Probability, choose track corresponding to optimal relevance assumption as true track, can accurately be grasped based on true track untethered
The specific location of underwater robot environment under water.
If function described in the embodiment of the present application method is realized in the form of SFU software functional unit and as independent production
Product when selling or using, can store in a storage medium readable by a compute device.Based on this understanding, the application is real
The part for applying a part that contributes to existing technology or the technical solution can be embodied in the form of software products,
The software product is stored in a storage medium, including some instructions are used so that a calculating equipment (can be personal meter
Calculation machine, server, mobile computing device or network equipment etc.) execute each embodiment the method for the application whole or portion
Step by step.And storage medium above-mentioned include: USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), with
Machine accesses various Jie that can store program code such as memory (RAM, Random Access Memory), magnetic or disk
Matter.
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with it is other
The difference of embodiment, same or similar part may refer to each other between each embodiment.
The foregoing description of the disclosed embodiments makes professional and technical personnel in the field can be realized or use the application.
Various modifications to these embodiments will be readily apparent to those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the application.Therefore, the application
It is not intended to be limited to the embodiments shown herein, and is to fit to and the principles and novel features disclosed herein phase one
The widest scope of cause.
Claims (8)
1. a kind of submarine target localization method characterized by comprising
Information at the time of receiving submarine target feedback signal based on default Hyperbolic Equation group and each base station, determines the water
The metric data of lower target, wherein the default Hyperbolic Equation group is for characterizing the submarine target and each base station
Between positional relationship, the metric data is for characterizing submarine target specific location locating for environment under water；
Based on goal-selling motion model, the motion state model of the submarine target is modified, the motion state mould
Type is used to characterize the motion state of the submarine target；
Based on goal-selling tracking gate, the metric data is grouped, target metric data is obtained and clusters, it is described cluster in
Target metric data it is associated with the true track of the submarine target；
The default starting track of the submarine target and the target metric data are associated it is assumed that in conjunction with revised fortune
Dynamic state model, generates multiple relevance assumptions, and the relevance assumption is preset for characterizing the target metric data with described
The track of beginning track composition；
Optimal relevance assumption is chosen from the multiple relevance assumption, using track corresponding to the optimal relevance assumption as institute
State the true track of submarine target.
2. the method according to claim 1, wherein described connect based on default Hyperbolic Equation group and each base station
Information at the time of receiving submarine target feedback signal determines the metric data of the submarine target, comprising:
Obtain information at the time of each base station receives submarine target feedback signal；
Preset position information based on each time information and each base station obtains the default Hyperbolic Equation group
Solution, the solution of the default Hyperbolic Equation group characterizes the specific location of the submarine target；
According to the specific location of the submarine target, the metric data of the submarine target is determined.
3. the method according to claim 1, wherein described be based on goal-selling motion model, to described underwater
The motion state model of target is modified, comprising:
The default initial motion status information of the submarine target is obtained, the default initial motion status information characterizes the water
Motion state locating for lower target initial time；
According to the default initial motion status information and goal-selling motion model, motion state is carried out to the submarine target
Estimation obtains the movement state information, and the movement state information is for characterizing locating for the submarine target current time
Motion state；
According to the movement state information, the motion state model of the submarine target is corrected.
4. the method according to claim 1, wherein it is false to choose optimal association from the multiple relevance assumption
If using track corresponding to the optimal relevance assumption as the true track of the submarine target, comprising:
Calculate the probability of the multiple relevance assumption；
According to the probability size, optimal relevance assumption is determined, and using track corresponding to the optimal relevance assumption as institute
State the true track of submarine target, wherein the probability of the optimal relevance assumption is higher than other relevance assumptions.
5. a kind of underwater object locating device characterized by comprising
Measure determination unit, for based on default Hyperbolic Equation group and each base station receive submarine target feedback signal when
Information is carved, determines the metric data of the submarine target, wherein the default Hyperbolic Equation group is for characterizing the underwater mesh
Mark the positional relationship between each base station, the metric data is for characterizing the submarine target under water locating for environment
Specific location；
Modifying model unit repairs the motion state model of the submarine target for being based on goal-selling motion model
Just, the motion state model is used to characterize the motion state of the submarine target；
Measurement clusters unit, for being based on goal-selling tracking gate, is grouped to the metric data, obtains target and measure number
According to clustering, the interior target metric data that clusters is associated with the true track of the submarine target；
Associative cell is measured, for the default starting track of the submarine target and the target metric data to be associated vacation
If generating multiple relevance assumptions in conjunction with revised motion state model, the relevance assumption is measured for characterizing the target
The track of data and the default starting track composition；
Track determination unit, for choosing optimal relevance assumption from the multiple relevance assumption, by the optimal relevance assumption
True track of the corresponding track as the submarine target.
6. device according to claim 5, which is characterized in that the measurement determination unit includes:
Module is obtained, for obtaining information at the time of each base station receives submarine target feedback signal；
Module is solved to obtain described pre- for the preset position information based on each time information and each base station
If the solution of Hyperbolic Equation group, the solution of the default Hyperbolic Equation group characterizes the specific location of the submarine target；
Determining module determines the metric data of the submarine target for the specific location according to the submarine target.
7. device according to claim 5, which is characterized in that the Modifying model unit includes:
Module is obtained, for obtaining the default initial motion status information of the submarine target, the default initial motion state
Motion state locating for submarine target initial time described in information representation；
Estimation module is used for according to the default initial motion status information and goal-selling motion model, to the underwater mesh
Mark carries out state estimation, obtains the movement state information, the movement state information is for characterizing the submarine target
Motion state locating for current time；
Correction module, for correcting the motion state model of the submarine target according to the movement state information.
8. the device according to regard to claim 5, which is characterized in that the track determination unit includes:
Computing module, for calculating the probability of the multiple relevance assumption；
Determining module, for determining optimal relevance assumption, and will be corresponding to the optimal relevance assumption according to the probability size
True track of the track as the submarine target, wherein the probability of the optimal relevance assumption is higher than other relevance assumptions.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910784901.7A CN110488226B (en) | 2019-08-23 | 2019-08-23 | Underwater target positioning method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910784901.7A CN110488226B (en) | 2019-08-23 | 2019-08-23 | Underwater target positioning method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110488226A true CN110488226A (en) | 2019-11-22 |
| CN110488226B CN110488226B (en) | 2021-08-06 |
Family
ID=68553346
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910784901.7A Active CN110488226B (en) | 2019-08-23 | 2019-08-23 | Underwater target positioning method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110488226B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110927727A (en) * | 2019-11-25 | 2020-03-27 | 深圳市智慧海洋科技有限公司 | Target positioning and tracking method and device |
| CN111767639A (en) * | 2020-05-25 | 2020-10-13 | 西北工业大学 | Multi-sensor track association method |
| CN112033429A (en) * | 2020-09-14 | 2020-12-04 | 吉林大学 | Target-level multi-sensor fusion method for intelligent automobile |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1600640A (en) * | 2000-03-20 | 2005-03-30 | 弗朗索瓦·贝尔纳德 | Appts for deploying load to underwater target position with enhanced accuracy and method to control such appts. |
| CN1820210A (en) * | 2003-09-03 | 2006-08-16 | 艾利森电话股份有限公司 | Positioning method and system |
| CN101783020A (en) * | 2010-03-04 | 2010-07-21 | 湖南大学 | Video multi-target fast tracking method based on joint probability data association |
| CN103323815A (en) * | 2013-03-05 | 2013-09-25 | 上海交通大学 | Underwater acoustic locating method based on equivalent sound velocity |
| CN104833953A (en) * | 2015-05-11 | 2015-08-12 | 中国民用航空总局第二研究所 | Multipoint positioning monitoring system and multipoint positioning monitoring method in airport non-line-of-sight (NLOS) channel environment |
| CN109633589A (en) * | 2019-01-08 | 2019-04-16 | 沈阳理工大学 | The Multi-target Data Associations assumed are optimized based on multi-model more in target following |
| CN109669160A (en) * | 2019-02-21 | 2019-04-23 | 哈尔滨工程大学 | A kind of detection method of underwater transient state acoustical signal |
-
2019
- 2019-08-23 CN CN201910784901.7A patent/CN110488226B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1600640A (en) * | 2000-03-20 | 2005-03-30 | 弗朗索瓦·贝尔纳德 | Appts for deploying load to underwater target position with enhanced accuracy and method to control such appts. |
| CN1820210A (en) * | 2003-09-03 | 2006-08-16 | 艾利森电话股份有限公司 | Positioning method and system |
| CN101783020A (en) * | 2010-03-04 | 2010-07-21 | 湖南大学 | Video multi-target fast tracking method based on joint probability data association |
| CN103323815A (en) * | 2013-03-05 | 2013-09-25 | 上海交通大学 | Underwater acoustic locating method based on equivalent sound velocity |
| CN104833953A (en) * | 2015-05-11 | 2015-08-12 | 中国民用航空总局第二研究所 | Multipoint positioning monitoring system and multipoint positioning monitoring method in airport non-line-of-sight (NLOS) channel environment |
| CN109633589A (en) * | 2019-01-08 | 2019-04-16 | 沈阳理工大学 | The Multi-target Data Associations assumed are optimized based on multi-model more in target following |
| CN109669160A (en) * | 2019-02-21 | 2019-04-23 | 哈尔滨工程大学 | A kind of detection method of underwater transient state acoustical signal |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110927727A (en) * | 2019-11-25 | 2020-03-27 | 深圳市智慧海洋科技有限公司 | Target positioning and tracking method and device |
| CN110927727B (en) * | 2019-11-25 | 2021-09-03 | 深圳市智慧海洋科技有限公司 | Target positioning and tracking method and device |
| CN111767639A (en) * | 2020-05-25 | 2020-10-13 | 西北工业大学 | Multi-sensor track association method |
| CN112033429A (en) * | 2020-09-14 | 2020-12-04 | 吉林大学 | Target-level multi-sensor fusion method for intelligent automobile |
| CN112033429B (en) * | 2020-09-14 | 2022-07-19 | 吉林大学 | Target-level multi-sensor fusion method for intelligent automobile |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110488226B (en) | 2021-08-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110488226A (en) | A kind of submarine target localization method and device | |
| US10204299B2 (en) | Unsupervised matching in fine-grained datasets for single-view object reconstruction | |
| CN111445108B (en) | Data-driven power distribution network line variation relation diagnosis method, device and system | |
| Tsardoulias et al. | Critical rays scan match SLAM | |
| CN109447133A (en) | A kind of elimination method of the location information outlier based on SVR algorithm | |
| Konakoglu et al. | 2D coordinate transformation using artificial neural networks | |
| CN111982124B (en) | Deep learning-based three-dimensional laser radar navigation method and device in glass scene | |
| CN116502881B (en) | Personnel scheduling management method and system for equipment maintenance | |
| CN111707279B (en) | Matching evaluation method, medium, terminal and device for laser point cloud and map | |
| CN114427866A (en) | Path planning method, electronic device and storage medium | |
| CN115952691A (en) | Optimized station distribution method and device of multi-station passive time difference cross joint positioning system | |
| CN111192158A (en) | Transformer substation daily load curve similarity matching method based on deep learning | |
| CN109783960A (en) | A kind of Wave arrival direction estimating method based on meshing refinement | |
| CN110120066A (en) | Robust multiple targets tracking and tracking system towards monitoring system | |
| CN112797976A (en) | Positioning detection method and device, computer equipment and readable storage medium | |
| EP3536467B1 (en) | Action transfer device, action transfer method, and non-temporary computer readable medium having action transfer program stored thereon | |
| CN110135507A (en) | A kind of label distribution forecasting method and device | |
| CN114719844A (en) | Space projection-based all-celestial star map identification method, device and medium | |
| CN110927727B (en) | Target positioning and tracking method and device | |
| CN113869186A (en) | Model training method and device, electronic equipment and computer readable storage medium | |
| CN113822441A (en) | Decision model training method and device, terminal equipment and storage medium | |
| WO2021128708A1 (en) | Synchronization algorithm real-time evaluation method and system, and storage medium | |
| CN110135340A (en) | 3D hand gestures estimation method based on cloud | |
| CN114373002B (en) | Point cloud registration method and device | |
| Huang et al. | A meta-deep-learning framework for spatio-temporal underwater SSP inversion |
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 | ||
| GR01 | Patent grant |