CN109213204A - AUV sub-sea floor targets based on data-driven search navigation system and method - Google Patents
AUV sub-sea floor targets based on data-driven search navigation system and method Download PDFInfo
- Publication number
- CN109213204A CN109213204A CN201811196477.6A CN201811196477A CN109213204A CN 109213204 A CN109213204 A CN 109213204A CN 201811196477 A CN201811196477 A CN 201811196477A CN 109213204 A CN109213204 A CN 109213204A
- Authority
- CN
- China
- Prior art keywords
- auv
- level
- mission
- path
- module
- 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
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/12—Target-seeking control
Abstract
The present invention discloses a kind of AUV sub-sea floor targets search navigation system and method based on data-driven, and the system comprises mission bit stream processing module, side scan sonar, target's feature-extraction module, altimeter, level-one path-generating module, second level path-generating module, three-level path-generating module, module of making decisions on one's own and AUV executing agencies；Mission bit stream processing module receives mission bit stream and handles in conjunction with AUV self-position state it, obtains dynamic grid sea chart coordinate, generates level-one path point set by level-one path-generating module；In conjunction with the sonar image that side scan sonar detects, target area coordinates will be obtained by target's feature-extraction module and send second level path-generating module generation second level path point set to；Three-level path-generating module is sent to based on the altitude information that altitude measuring obtains, and generates three-level path point set；Three kinds of point sets are finally based on by module of making decisions on one's own and obtain the path AUV, and send AUV executing agency to.The present invention may be implemented to take into account collision prevention fixed high navigation while the search to sub-sea floor targets, and the efficiency of sub-sea floor targets search can be improved.
Description
Technical field
The invention belongs to habitata fields, and in particular to a kind of AUV sub-sea floor targets search navigation based on data-driven
System and method.
Background technique
AUV executes no matter oceanographic survey task is all widely used in military, economic or scientific domain.However, mesh
Preceding to execute survey tasks using AUV, the either air route of AUV, behavior or movement is all preparatory planning or designed offline rule
It draws, the effect of AUV is the movement observations platform for being equipped with detection sensor；Moreover, marine environment is unknown, dynamic
With it is uncertain, AUV does not have the ability made decisions on one's own, will result in acquisition data quality it is low, insincere, even in vain
The phenomenon that, this is just needed to improve workload or operations number to make up, and is not only wasted the energy of AUV but also is caused oceanographic survey
Efficiency it is low, seriously affect further applying for AUV.
As shown in Figure 1, the sailing method that traditional approach covers sea chart has rebound method, side to stride method, a few word methods, helix
Method and radioactive method etc., algorithmic approach is single, and belongs to off-line algorithm, can neither carry out benefit to the sea area data of acquisition in real time
With only allowing AUV to be moved according to specified track.For example using sonar is pulled, underwater trailing portions can be by ship
It at sea slowly advances in tow, speed is usually 1 to 5 sections；Either using traditional navigation side of Autonomous Underwater Vehicle (AUV)
Method, searching principle is to scan by sonar set in seabed, but a whole day can only search and be equivalent to a football pitch size
One square kilometre of local range search may need one month, it is generally the case that it uses sonar contact first, after finding suspicious object
Go up to the water surface, dive again after video camera, provides underwater picture for search personnel in repacking.
As it can be seen that conventional method be it is offline search, the disadvantages such as that there are search times is long, searching efficiency is low can to meet AUV
In conjunction with the demand of different oceanographic survey tasks, the threat situation and navigation environment currently faced is quickly analyzed and is assessed,
Corresponding strategy, adjust automatically navigation route are formulated according to assessment result, safe autonomous navigation is realized, finally makes oceanographic survey task
More efficient, better quality, have become the important topic of underwater robot research field.
Summary of the invention
Technical problem to be solved by the present invention lies in, searches long for search time existing for conventional target method for searching
The defect of low efficiency overcomes drawbacks described above of the existing technology, proposes a kind of AUV sub-sea floor targets search based on data-driven
Navigation system and method, by realizing to the real-time of AUV navigation path to the progress of clarification of objective closeness is searched in line computation
Optimization effectively reduces the inefficient navigation being not necessarily to.
The present invention, which is that the following technical solution is employed, to be realized: the AUV sub-sea floor targets search navigation system based on data-driven
System, including mission bit stream processing module, side scan sonar, target's feature-extraction module, altimeter, level-one path-generating module, two
Grade path-generating module, three-level path-generating module, module of making decisions on one's own and AUV executing agency；
Task parameters of the mission bit stream processing module to receive target to be sought, mission bit stream processing module combine
AUV self-position state handles the received task parameters of institute, obtains dynamic grid sea chart coordinate, and pass it to one
Grade path-generating module obtains level-one path point set PI；
The side scan sonar includes that left twang and right twang are mentioned to detect target sonar image by target signature
Modulus block extracts the target area coordinates of target to be sought based on the target sonar image detected, and by the mesh of extraction
Mark area coordinate is transmitted to second level path-generating module and obtains second level path point set PII；
The altimeter is transferred to three-level path to acquire the sea floor height h in front of AUV, and by the altitude information measured
Generation module obtains three-level path point set PIII；
The module of making decisions on one's own receives level-one path point set PI, second level path point set PIIWith three-level path point set PIII, lead to
It crosses Analysis of Policy Making and obtains the path an AUV point set Po, and it is transferred to AUV executing agency, realize that the search to sub-sea floor targets controls,
This programme is by realizing and taking into account collision prevention while search to sub-sea floor targets to the progress of clarification of objective closeness is searched in line computation
Fixed high navigation, effectively improves the efficiency of sub-sea floor targets search.
Further, the altimeter is mounted on the fore body of AUV, and left twang and right twang are symmetrically mounted on AUV's
Abdomen two sides, and left twang and right twang at least have high frequency, the two different scan frequencies of low frequency, high frequency sweep imaging
Apparent, identification is more acurrate, and low-frequency sweep range is farther.
Further, the task parameters include the coordinate set P of institute's Search Areain, grid size din；Task letter
The breath current location processing module combination AUV handles task parameters, the area coordinate collection after obtaining minimum external rasterizing
Close PmissionWith mission area parameter set MissionMap, grid size dmission, AUV needs efficient, quasi- in mission area
The true target area (such as bed ripples coral reef) found in mission area:
θ0=arccos (min (Ag (max (Pin)-min(Pin))2))；
Wherein, A is transformation matrix of coordinates, θ0For best coordinates angle of transformation, MissionMap includes the length of mission area
Length, width Width and grid quantity Ngrid。
Further, current location of the mission bit stream processing module according to AUV, with the current location distance AUV geometry
Apart from nearest point P0As origin, straight line where the long side of mission area is x-axis, and straight line where the short side of mission area is y
Axis establishes rectangular coordinate system xoy, in which:
Wherein, xmissionRefer to the longitude coordinate on mission area vertex, xAUVRefer to the longitude coordinate of the current location AUV, ymission
Refer to the latitude coordinate on mission area vertex, yAUVRefer to the latitude coordinate of the current location AUV.
Further, the low-frequency sweep distance l of the level-one path-generating module combination side scan sonarLOWAnd high frequency sweep
Distance lHIGH, with 2lLOWFor unit, mission area is divided intoSub-regions generate level-one path point set PI:
Wherein (x, y) is navigation path point coordinate, and N is the positive integer greater than 0.
Further, it is located in call duration time t, left twang returns to left string target interval lL, the right string of right twang return
Target interval lR, and by left string target interval lLWith right string target interval lRIt is transmitted to target's feature-extraction module, extracts target
Provincial characteristics l describes target area feature l in xOy coordinate system and sends second level path-generating module to；Second level path is raw
At module according to the xoy coordinate system of real-time update, in conjunction with the real-time position P of AUVAUVGenerate second level path point set PIIAnd it is transmitted to
It makes decisions on one's own module, in which:
PII={ (xi,yi)|xi=PAUV,yi=ilHIGH,i∈N}
Further, the three-level path-generating module generates three-level path point set according to the sea floor height h of real-time update
PIII:
PIII=ε (hsafe-h)·f(h)+ε(h-hsafe)·hmode
Wherein, hsafeIt is safety to sea floor height, ε (hsafe-h),ε(h-hsafe) it is jump function, λ is amplitude factor,
σ is response factor, hmodeCorresponding to bottom height under side scan sonar for AUV different frequency, k is the scanning height factor.
Further, the module of making decisions on one's own is based on PI,PII,PIIITactic selection is carried out, optimal path point set is generated:
In addition the present invention also proposes that a kind of AUV sub-sea floor targets based on data-driven search sailing method, including following step
It is rapid: the task parameters information of target to be sought step S1, to be received by mission bit stream processing module, the coordinate according to institute's Search Area
Set Pin, grid size din, area coordinate set P according to the current location AUV, after obtaining minimum external rasterizingmissionWith
Mission area parameter set MissionMap:
θ0=arccos (min (A (max (Pin)-min(Pin))2))；
Wherein, A is transformation matrix of coordinates, θ0For best coordinates angle of transformation；Mission area parameter set MissionMap includes
The length L of mission area, width W and grid quantity NGRID；
Step S2, mission bit stream processing module is according to the current location of AUV, most with the current location distance AUV geometric distance
Close point P0As origin, straight line where the long side of mission area is x-axis, and straight line where the short side of mission area is y-axis foundation
Rectangular coordinate system xoy；
Wherein, xmissionRefer to the longitude coordinate on mission area vertex, xAUVRefer to the longitude coordinate of the current location AUV, ymission
Refer to the latitude coordinate on mission area vertex, yAUVRefer to the latitude coordinate of the current location AUV；
Step S3, the low-frequency sweep distance l of level-one path-generating module combination side scan sonarLOWWith high frequency sweep distance
lHIGH, with 2lLOWFor unit, Search Area is divided intoSub-regions generate level-one path point set PIAnd it passes
It is delivered to module of making decisions on one's own:
Wherein (x, y) is AUV navigation path point coordinate, and N is the positive integer greater than 0；
Step S4, the target interval set l that target's feature-extraction module is returned according to left twangLIt is returned with right twang
Target interval set lR, target area feature l is extracted, the target area feature is described in xOy coordinate system and is sent to
Second level path-generating module；
Step S5, second level path-generating module is according to the xoy coordinate system of real-time update, in conjunction with the real time position P of AUVAUVIt is raw
At second level path point set PIIAnd it is transmitted to module of making decisions on one's own: PII={ (xi,yi)|xi=PAUV,yi=ilHIGH,i∈N}；
Step S6, the sea floor height h in front of altimeter acquisition AUV, is transmitted to three-level path-generating module, and three-level path is raw
At module according to the sea floor height h of real-time update, three-level path point set P is generatedIIIAnd it is transmitted to module of making decisions on one's own:
PIII=ε (hsafe-h)·f(h)+ε(h-hsafe)·hmode
Wherein, hsafeIt is safety to sea floor height, ε (hsafe-h),ε(h-hsafe) it is jump function, λ is amplitude factor,
σ is response factor, hmodeCorresponding to bottom height under side scan sonar for AUV different frequency, k is the scanning height factor；
Step S7, by module of making decisions on one's own by PI,PII,PIIITactic selection is carried out, optimal path point set is generated:
Step S8, by PopIt is sent to AUV executing agency, which is set as path point and is gone forward side by side line trace control.
Further, in the step S6, amplitude factor λ value range is 2~10, and response factor σ value range is 20
~40, scanning height factor k value range is 0.769~0.125.
Compared with prior art, the advantages and positive effects of the present invention are:
In this programme based on data-driven when realizing the planning of second level path point, according to the figure of side scan sonar return
Picture is identified by other image processing techniques, counts the target pixel points in each grid, is then repaired to path point
Change, to achieve the purpose that make decisions on one's own, by efficient dynamic programming techniques, reliable Robot dodge strategy, by ocean number
According to key feature analyzed real-time, quickly, binding characteristic closeness optimizes detective path to improve task efficiency,
Quick detection, accurate positionin and the efficient avoidance to barrier are realized, overcomes conventional method that artificial eye is needed to identify emphasis
Target, the defects such as that there are search times is long, searching efficiency is low；
It is handled by the analysis of level-one, second level and three-level path-generating module and module of making decisions on one's own, in conjunction with local sea chart
It is planned online with navigation route of the global sea chart to AUV, after providing a piece of seabed situation unknown sea area, AUV can
Follow the sailing method in sea area less than the feature of the highest priority (such as coral reef, Watership Down) of the 30% sea area gross area
Closeness carries out reducing the inefficient navigation being not necessarily in line computation, the navigation path of real-time optimization AUV.
Detailed description of the invention
Fig. 1 is the sailing method navigation route schematic diagram of conventional offline sea chart covering；
Fig. 2 is that sub-sea floor targets described in the embodiment of the present invention search navigation system block diagram；
Fig. 3 is that sub-sea floor targets described in the embodiment of the present invention search sailing method flow diagram；
Fig. 4 is the signal of amplitude factor described in the embodiment of the present invention, response factor on collision prevention height and distance of obstacle influence
Figure；
Fig. 5 is level-one path point set schematic diagram described in the embodiment of the present invention；
Fig. 6 is the path point set schematic diagram after the embodiment of the present invention is made decisions on one's own；
Fig. 7 is that the embodiment of the present invention completes task rear region schematic diagram；
Fig. 8 is the squaring and rasterizing schematic diagram of mission bit stream processing module described in the embodiment of the present invention.
Specific embodiment
In order to which the above objects, features and advantages of the present invention is more clearly understood, with reference to the accompanying drawing and implement
The present invention will be further described for example.It should be noted that in the absence of conflict, in embodiments herein and embodiment
Feature can be combined with each other.
Embodiment 1, a kind of AUV sub-sea floor targets search navigation system based on data-driven, as shown in Fig. 2, including task
Message processing module, side scan sonar, target's feature-extraction module, altimeter, level-one path-generating module, second level coordinates measurement mould
Block, three-level path-generating module, module of making decisions on one's own and AUV executing agency；The mission bit stream processing module to receive to
The task parameters of target are searched, mission bit stream processing module combination AUV self-position state carries out the received task parameters of institute
Processing, obtains dynamic grid sea chart coordinate, and passes it to level-one path-generating module and obtain level-one path point set PI；It is described
Side scan sonar includes left twang and right twang, to detect target sonar image, is based on institute by target's feature-extraction module
The target sonar image of detection extracts the target area coordinates of target to be sought, and the target area coordinates of extraction are passed
It transports to second level path-generating module and obtains second level path point set PII；The altimeter to acquire the sea floor height h in front of AUV,
And the altitude information measured is transferred to three-level path-generating module and obtains three-level path point set PIII；The module of making decisions on one's own
Receive level-one path point set PI, second level path point set PIIWith three-level path point set PIII, a road AUV is obtained by Analysis of Policy Making
Diameter point set Pop, and it is transferred to AUV executing agency, realize that the search to sub-sea floor targets controls, wherein level-one path point set is being searched
With regard to it has been determined that second level path point set and three-level path point set then carry out in fact according to the data that sensor is returned when task determines
Shi Gengxin, by realizing to take into account while searching sub-sea floor targets and keep away to the progress of clarification of objective closeness being searched in line computation
Fixed high navigation is touched, the efficiency of sub-sea floor targets search is effectively improved.
It should be noted that the mission area refers to that entire AUV executes the sea area of task, target in the present embodiment
Region refers to the region for needing to be identified with sensor, is no more than 30%.The data-driven refers in planning second grade highway
When diameter point, used dynamic programming techniques, according to side scan sonar return image, by other image processing techniques into
Row identification, counts the target pixel points in each grid, then modifies path point, to reach the mesh made decisions on one's own
's.Feature closeness is the build-in attribute of grid in mission area, i.e., in each grid, the number of target pixel points.Target
The acquisition of pixel is to be identified in real time with certain digital image processing method to sonar image, the mesh that dynamic is united in grid
Mark the number of pixel, i.e. feature closeness.Specifically, the altimeter is mounted on the fore body of AUV, left twang and right twang
The abdomen two sides of AUV are symmetrically mounted on, and left twang and right twang at least have high frequency, the two different scannings of low frequency
Frequency, high frequency sweep imaging is apparent, and identification is more acurrate, and low-frequency sweep range is farther, and low frequency described here and high frequency are this
Field common knowledge, when providing low frequency or high frequency requirements, those skilled in the art can learn its corresponding frequency range.
In the present embodiment, the task parameters include the coordinate set P of institute's Search Areain, grid size din, wherein P
For the latitude and longitude coordinates of record；The current location mission bit stream processing module combination AUV handles task parameters, obtains minimum
Area coordinate set P after external rasterizingmissionWith mission area parameter set MissionMap, grid size dmission,
AUV needs efficiently, accurately to find the target area (such as bed ripples coral reef) in mission area in mission area:
θ0=arccos (min (A (max (Pin)-min(Pin))2))；
Wherein, the external rasterizing of minimum refers to institute Search Area coordinate set PinFor polygon vertex, obtain
The smallest boundary rectangle of area, further according to grid size dinRasterizing is carried out to rectangular area, obtains the area coordinate set, is made
For mission area, A is transformation matrix of coordinates, θ0For best coordinates angle of transformation, MissionMap includes the length of mission area
Length, width Width, grid quantity Ngrid。
Moreover, current location of the mission bit stream processing module according to AUV, nearest with the current location distance AUV geometric distance
Point P0As origin, straight line where the long side of mission area is x-axis, and straight line where the short side of mission area is that y-axis foundation is straight
Angular coordinate system xoy, in which:
Wherein, xmissionRefer to the longitude coordinate on mission area vertex, xAUVRefer to the longitude coordinate of the current location AUV, ymission
Refer to the latitude coordinate on mission area vertex, yAUVRefer to the latitude coordinate of the current location AUV, if the mission area after planning is pros
Shape, then will the current the smallest rectilinear direction of course angle of distance AUV as x-axis.
Level-one path-generating module, second level path-generating module and three-level path-generating module when generating path point set,
Especially by following means:
Level-one path-generating module: in conjunction with the low-frequency sweep distance l of side scan sonarLOWWith high frequency sweep distance lHIGH, with
2lLOWFor unit, mission area is divided intoSub-regions generate level-one path point set PI:Wherein (x, y) is navigation path point coordinate, N
For the positive integer greater than 0.
Second level path-generating module: being located in call duration time t, and left twang returns to left string target interval lL, right twang
Return to right string target interval lR, and by left string target interval lLWith right string target interval lRIt is transmitted to target's feature-extraction module, is mentioned
Target area feature l is taken out, target area feature l is described in xOy coordinate system and sends second level path-generating module to；Two
Grade path-generating module is according to the xoy coordinate system of real-time update, in conjunction with the real-time position P of AUVAUVGenerate second level path point set PII
And it is transmitted to module of making decisions on one's own, in which:
PII={ (xi,yi)|xi=PAUV,yi=ilHIGH,i∈N}。
Three-level path-generating module: according to the sea floor height h of real-time update, three-level path point set P is generatedIII:
PIII=ε (hsafe-h)·f(h)+ε(h-hsafe)·hmode
Wherein, hsafeIt is safety to sea floor height, ε (hsafe-h),ε(h-hsafe) it is jump function, λ is amplitude factor,
σ is response factor, hmodeCorresponding to bottom height under side scan sonar for AUV different frequency, k is the scanning height factor.
Finally, obtained level-one path point set, second level path point set and three-level path point set are determined in module of making decisions on one's own
Under plan analysis, optimal path point set is generated:
By AUV executing agency, which is set as path point and is gone forward side by side line trace control, and then is realized to sub-sea floor targets
Efficiently search.
Embodiment 2, based on the search navigation system that embodiment 1 proposes, the present embodiment discloses a kind of based on data-driven
AUV sub-sea floor targets search sailing method, as shown in connection with fig. 3, comprising the following steps:
Step S1, the task parameters information of target to be sought is received by mission bit stream processing module, according to institute's Search Area
Coordinate set Pin, grid size din, area coordinate set according to the current location AUV, after obtaining minimum external rasterizing
PmissionWith mission area parameter set MissionMap:
θ0=arccos (min (A (max (Pin)-min(Pin))2))；
Wherein, A is transformation matrix of coordinates, θ0For best coordinates angle of transformation；Mission area parameter set MissionMap includes
The length Length of mission area, width Width and grid quantity Ngrid；
Step S2, mission bit stream processing module is according to the current location of AUV, most with the current location distance AUV geometric distance
Close point P0As origin, straight line where the long side of mission area is x-axis, and straight line where the short side of mission area is y-axis foundation
Rectangular coordinate system xoy, as shown in Figure 8；
Wherein, xmissionRefer to the longitude coordinate on mission area vertex, the vertex of the mission area refers to rectangle mission area
Four vertex in domain, xAUVRefer to the longitude coordinate of the current location AUV, ymissionRefer to the latitude coordinate on mission area vertex, yAUVRefer to
The latitude coordinate of the current location AUV；
Step S3, the low-frequency sweep distance l of level-one path-generating module combination side scan sonarLOWWith high frequency sweep distance
lHIGH, with 2lLOWFor unit, Search Area is divided intoSub-regions generate level-one path point set PI, such as scheme
Shown in 5, and the level-one path point set obtained is transmitted to module of making decisions on one's own:
Wherein (x, y) is AUV navigation path point coordinate, and N is the positive integer greater than 0；
Step S4, the target interval set l that target's feature-extraction module is returned according to left twangLIt is returned with right twang
Target interval set lR, target area feature l is extracted, the target area feature is described in xOy coordinate system and is sent to
Second level path-generating module；
Step S5, second level path-generating module is according to the xoy coordinate system of real-time update, in conjunction with the real time position P of AUVAUVIt is raw
At second level path point set PIIAnd it is transmitted to module of making decisions on one's own: PΙΙ={ (xi,yi)|xi=PAUV,yi=ilHIGH,i∈N}；
Step S6, the sea floor height h in front of altimeter acquisition AUV, is transmitted to three-level path-generating module, as shown in figure 4,
Three-level path-generating module generates three-level path point set P according to the sea floor height h of real-time updateIIIAnd it is transmitted to mould of making decisions on one's own
Block:
PΙΙΙ=ε (hsafe-h)·f(h)+ε(h-hsafe)·hmode
Wherein, hsafeIt is safety to sea floor height, ε (hsafe-h),ε(h-hsafe) it is jump function, λ is amplitude factor,
σ is response factor, hmodeCorresponding to bottom height under side scan sonar for AUV different frequency, k is the scanning height factor；
Step S7, by module of making decisions on one's own by PI,PII,PIIITactic selection is carried out, with reference to Fig. 6, generates optimal path point
Collection:
Step S8, by PopIt is sent to AUV executing agency, which is set as path point and is gone forward side by side line trace control, and most
The scanning to mission area is realized eventually, and Fig. 7 is the areas case schematic diagram after completion task.
Moreover, in the present embodiment, the amplitude factor λ value range is 2~10, response factor σ value range is 20~
40, scanning height factor k value range is 0.769~0.125, tests by Multi simulation running, obtains the factor model for being best suitable for AUV
It encloses.
Traditional offline method for searching of "Ji" type and this method are subjected to simulation comparison, it is at the uniform velocity right with two kinds of speed change to be arranged
According to test.Under speed change, the side different to AUV sweeps working frequency and carries out different speed of a ship or plane settings, the low speed of a ship or plane of high frequency, low frequency Gao Hang
Speed.Method is assessed by comparing parameters such as path, coverage rate, discriminations, is 30.38% to a target area accounting
Sea chart planned, test that high frequency low frequency AUV route speed is identical for the first time, both compare in the case where AUV is at the uniform velocity navigated by water
Efficiency；Second of experiment compares efficiency of the two under the navigation of AUV speed change using the scheme of the low speed of a ship or plane of high frequency, the high speed of a ship or plane of low frequency,
Analyze result such as table 1.
Table 1: task overall efficiency analytical table
Wherein: 1. coverage rates refer to the ratio in the scanned region and mission area AUV；
2. discovery rate refer to scanning discovery to target area and mission area in realistic objective region ratio；
3. recognizable rate refers to realistic objective region in the inswept identifiable target area of sonar high frequency and mission area
Ratio.
As it can be seen that this programme passes through the key feature to oceanographic data by efficient planning technology, reliable Robot dodge strategy
It is analyzed real-time, quickly, binding characteristic closeness optimizes detective path to improve task efficiency, realizes to obstacle
Quick detection, accurate positionin and the efficient avoidance of object, overcome conventional method that artificial eye is needed to identify highest priority, there is search
The defects such as the time is long, searching efficiency is low realize on-line automatic identification based on feature closeness, feature closeness are independently determined
Plan, and local sea chart and global sea chart is combined to be planned that the navigation route of AUV, a piece of seabed situation is unknown when providing online
Sea area after, AUV can follow the sailing method in sea area less than the 30% sea area gross area highest priority (such as coral reef,
Watership Down etc.) feature closeness carry out in line computation, the navigation path of real-time optimization AUV, reduce It is not necessary to it is inefficient
Navigation.
The above described is only a preferred embodiment of the present invention, being not that the invention has other forms of limitations, appoint
What those skilled in the art changed or be modified as possibly also with the technology contents of the disclosure above equivalent variations etc.
It imitates embodiment and is applied to other fields, but without departing from the technical solutions of the present invention, according to the technical essence of the invention
Any simple modification, equivalent variations and remodeling to the above embodiments, still fall within the protection scope of technical solution of the present invention.
Claims (10)
1. AUV sub-sea floor targets based on data-driven search navigation system, which is characterized in that including mission bit stream processing module,
Side scan sonar, target's feature-extraction module, altimeter, level-one path-generating module, second level path-generating module, three-level path are raw
At module, module of making decisions on one's own and AUV executing agency；
Task parameters of the mission bit stream processing module to receive target to be sought, mission bit stream processing module combination AUV
Self-position state handles the received task parameters of institute, obtains dynamic grid sea chart coordinate, and pass it to level-one
Path-generating module obtains level-one path point set PI；
The side scan sonar includes left twang and right twang, to detect target sonar image, by target's feature-extraction mould
Block extracts the target area coordinates of target to be sought based on the target sonar image detected, and by the target area of extraction
Domain coordinate is transmitted to second level path-generating module and obtains second level path point set PII；
The altimeter is transferred to three-level coordinates measurement to acquire the sea floor height h in front of AUV, and by the altitude information measured
Module obtains three-level path point set PIII；
The module of making decisions on one's own receives level-one path point set PI, second level path point set PIIWith three-level path point set PIII, by certainly
Plan analysis obtains the path an AUV point set PopAnd it is transferred to AUV executing agency, realize that the search to sub-sea floor targets controls.
2. the AUV sub-sea floor targets according to claim 1 based on data-driven search navigation system, it is characterised in that: institute
The fore body that altimeter is mounted on AUV is stated, left twang and right twang are symmetrically mounted on the abdomen two sides of AUV, and left twang
At least there is high frequency, the two different scan frequencies of low frequency with right twang.
3. the AUV sub-sea floor targets according to claim 2 based on data-driven search navigation system, it is characterised in that: institute
State the coordinate set P that task parameters include institute's Search Areain, grid size din；Mission bit stream processing module combination AUV works as
Front position handles task parameters, the area coordinate set P after obtaining minimum external rasterizingmissionJoin with mission area
Manifold MissionMap, grid size dmission:
Wherein, A is transformation matrix of coordinates, θ0For coordinate transform angle, mission area parameter set MissionMap includes mission area
Length Length, width Width and grid quantity Ngrid。
4. the AUV sub-sea floor targets according to claim 3 based on data-driven search navigation system, it is characterised in that: institute
Current location of the mission bit stream processing module according to AUV is stated, with the nearest point P of the current location distance AUV geometric distance0As original
Point, straight line where the long side of mission area are x-axis, and straight line where the short side of mission area is that y-axis establishes rectangular coordinate system xOy,
Wherein:
Wherein, xmissionRefer to the longitude coordinate on mission area vertex, xAUVRefer to the longitude coordinate of the current location AUV, ymissionRefer to and appoints
The latitude coordinate on business region vertex, yAUVRefer to the latitude coordinate of the current location AUV.
5. the AUV sub-sea floor targets according to claim 4 based on data-driven search navigation system, it is characterised in that: institute
State the low-frequency sweep distance l of level-one path-generating module combination side scan sonarLOWWith high frequency sweep distance lHIGH, with 2lLOWFor list
Position, mission area is divided intoSub-regions generate level-one path point set PI:
Wherein, (xi,yi) it is navigation path point coordinate, N is the positive integer greater than 0.
6. the AUV sub-sea floor targets according to claim 5 based on data-driven search navigation system, it is characterised in that: institute
Second level path-generating module is stated according to the xOy coordinate system of real-time update, in conjunction with the real-time position P of AUVAUVGenerate second level path point
Collect PII:
PΙΙ={ (xi,yi)|xi=PAUV,yi=ilHIGH,i∈N}
7. the AUV sub-sea floor targets according to claim 6 based on data-driven search navigation system, it is characterised in that: institute
Three-level path-generating module is stated according to the sea floor height h of real-time update, generates three-level path point set PΙΙΙ:
PΙΙΙ=ε (hsafe-h)·f(h)+ε(h-hsafe)·hmode
Wherein, hsafeIt is safety to sea floor height, ε (hsafe-h),ε(h-hsafe) it is jump function, λ is amplitude factor, and σ is
Response factor, hmodeCorresponding to bottom height under side scan sonar for AUV different frequency, k is the scanning height factor.
8. the AUV sub-sea floor targets according to claim 7 based on data-driven search navigation system, it is characterised in that: institute
It states module of making decisions on one's own and is based on PΙ,PΙΙ,PΙΙΙTactic selection is carried out, optimal path point set is generated:
zi=ε (hsafe-h)·f(h)+ε(h-hsafe)·hmode,i∈N}
9. the AUV sub-sea floor targets based on data-driven search sailing method, which comprises the following steps:
Step S1, the task parameters information of target to be sought is received by mission bit stream processing module, according to the mission area searched
The coordinate set P in domainin, grid size din, area coordinate set according to the current location AUV, after obtaining minimum external rasterizing
PmissionWith mission area parameter set MissionMap:
Wherein, A is transformation matrix of coordinates, θ0For coordinate transform angle；Mission area parameter set MissionMap includes mission area
Length Length, width Width, grid quantity Ngrid；
Step S2, mission bit stream processing module is according to the current location of AUV, nearest with the current location distance AUV geometric distance
Point P0As origin, straight line where the long side of mission area is x-axis, and straight line where the short side of mission area is that y-axis establishes right angle
Coordinate system xoy；
Wherein, xmissionRefer to the longitude coordinate on mission area vertex, xAUVRefer to the longitude coordinate of the current location AUV, ymissionRefer to and appoints
The latitude coordinate on business region vertex, yAUVRefer to the latitude coordinate of the current location AUV；
Step S3, the low-frequency sweep distance l of level-one path-generating module combination side scan sonarLOWWith high frequency sweep distance lHIGH, with
2lLOWFor unit, mission area is divided intoSub-regions generate level-one path point set PIAnd it is transmitted to certainly
Main decision-making module:
Wherein (x, y) is AUV navigation path point coordinate, and N is the positive integer greater than 0；
Step S4, the target interval set l that target's feature-extraction module is returned according to left twangLThe mesh returned with right twang
Mark section set lR, extract target area feature lRegin, the target area feature is described in xOy coordinate system and sends two to
Grade path-generating module；
Step S5, second level path-generating module is according to the xOy coordinate system of real-time update, in conjunction with the real time position P of AUVAUVGenerate two
Grade path point set PIIAnd it is transmitted to module of making decisions on one's own: PII={ (xi,yi)|xi=PAUV,yi=ilHIGH,i∈N}；
Step S6, the sea floor height h in front of altimeter acquisition AUV, is transmitted to three-level path-generating module, three-level coordinates measurement mould
Root tuber generates three-level path point set P according to the sea floor height h of real-time updateIIIAnd it is transmitted to module of making decisions on one's own:
PIII=ε (hsafe-h)·f(h)+ε(h-hsafe)·hmode
Wherein, hsafeIt is safety to sea floor height, ε (hsafe-h),ε(h-hsafe) it is jump function, λ is amplitude factor, and σ is
Response factor, hmodeCorresponding to bottom height under side scan sonar for AUV different frequency, k is the scanning height factor；
Step S7, by module of making decisions on one's own by PI,PII,PIIITactic selection is carried out, optimal path point set is generated:
zi=ε (hsafe-h)·f(h)+ε(h-hsafe)·hmode,i∈N}
Step S8, by PopIt is sent to AUV executing agency, which is set as path point and is gone forward side by side line trace control.
10. the AUV sub-sea floor targets according to claim 9 based on data-driven search sailing method, it is characterised in that: institute
It states in step S6, amplitude factor λ value range is 2~10, and response factor σ value range is 20~40, and scanning height factor k takes
Being worth range is 0.769~0.125.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811196477.6A CN109213204B (en) | 2018-10-15 | 2018-10-15 | AUV (autonomous underwater vehicle) submarine target searching navigation system and method based on data driving |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811196477.6A CN109213204B (en) | 2018-10-15 | 2018-10-15 | AUV (autonomous underwater vehicle) submarine target searching navigation system and method based on data driving |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109213204A true CN109213204A (en) | 2019-01-15 |
| CN109213204B CN109213204B (en) | 2021-05-07 |
Family
ID=64980358
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811196477.6A Active CN109213204B (en) | 2018-10-15 | 2018-10-15 | AUV (autonomous underwater vehicle) submarine target searching navigation system and method based on data driving |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109213204B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109828467A (en) * | 2019-03-01 | 2019-05-31 | 大连海事大学 | A kind of the unmanned boat intensified learning controller architecture and design method of data-driven |
| CN110908404A (en) * | 2019-12-09 | 2020-03-24 | 自然资源部第二海洋研究所 | AUV intelligent observation motion method based on data driving |
| CN111721296A (en) * | 2020-06-04 | 2020-09-29 | 中国海洋大学 | Data driving path planning method for underwater unmanned vehicle |
| CN112082558A (en) * | 2020-09-14 | 2020-12-15 | 哈尔滨工程大学 | UUV submarine topography tracking path rolling generation method based on polynomial fitting |
| CN112082557A (en) * | 2020-09-14 | 2020-12-15 | 哈尔滨工程大学 | UUV submarine topography tracking path rolling generation method based on Bessel fitting |
| CN112097781A (en) * | 2019-06-17 | 2020-12-18 | 郑州宇通客车股份有限公司 | Path planning method and device based on multi-stage tentacles |
| CN113052940A (en) * | 2021-03-14 | 2021-06-29 | 西北工业大学 | Space-time associated map real-time construction method based on sonar |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105045260A (en) * | 2015-05-25 | 2015-11-11 | 湖南大学 | Mobile robot path planning method in unknown dynamic environment |
| CN107168344A (en) * | 2017-05-17 | 2017-09-15 | 哈尔滨工程大学 | A kind of UUV approaches air route generation method during seabed operation |
| CN108045530A (en) * | 2017-12-04 | 2018-05-18 | 国网山东省电力公司电力科学研究院 | A kind of submarine cable detection underwater robot and operational method |
| CN108508913A (en) * | 2018-03-29 | 2018-09-07 | 中国海洋大学 | Autonomous Underwater Vehicle seafloor path planing method based on data-driven |
-
2018
- 2018-10-15 CN CN201811196477.6A patent/CN109213204B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105045260A (en) * | 2015-05-25 | 2015-11-11 | 湖南大学 | Mobile robot path planning method in unknown dynamic environment |
| CN107168344A (en) * | 2017-05-17 | 2017-09-15 | 哈尔滨工程大学 | A kind of UUV approaches air route generation method during seabed operation |
| CN108045530A (en) * | 2017-12-04 | 2018-05-18 | 国网山东省电力公司电力科学研究院 | A kind of submarine cable detection underwater robot and operational method |
| CN108508913A (en) * | 2018-03-29 | 2018-09-07 | 中国海洋大学 | Autonomous Underwater Vehicle seafloor path planing method based on data-driven |
Non-Patent Citations (2)
| Title |
|---|
| JIE LI,MATTHEW JOHNSON-ROBERSON: "Multi-altitude Multi-sensor Fusion Framework for AUV Exploration and Survey", 《2014 OCEANS》 * |
| 李晔,姜言清,张国成,黄蜀玲,李一鸣,陈鹏云: "一种基于电子海图的欠驱动AUV区域搜索方案", 《机器人》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109828467A (en) * | 2019-03-01 | 2019-05-31 | 大连海事大学 | A kind of the unmanned boat intensified learning controller architecture and design method of data-driven |
| CN109828467B (en) * | 2019-03-01 | 2021-09-07 | 大连海事大学 | Data-driven unmanned ship reinforcement learning controller structure and design method |
| CN112097781A (en) * | 2019-06-17 | 2020-12-18 | 郑州宇通客车股份有限公司 | Path planning method and device based on multi-stage tentacles |
| CN112097781B (en) * | 2019-06-17 | 2023-05-09 | 宇通客车股份有限公司 | Path planning method and device based on multistage tentacles |
| CN110908404A (en) * | 2019-12-09 | 2020-03-24 | 自然资源部第二海洋研究所 | AUV intelligent observation motion method based on data driving |
| CN110908404B (en) * | 2019-12-09 | 2023-04-18 | 自然资源部第二海洋研究所 | AUV intelligent observation motion method based on data driving |
| CN111721296A (en) * | 2020-06-04 | 2020-09-29 | 中国海洋大学 | Data driving path planning method for underwater unmanned vehicle |
| CN112082558A (en) * | 2020-09-14 | 2020-12-15 | 哈尔滨工程大学 | UUV submarine topography tracking path rolling generation method based on polynomial fitting |
| CN112082557A (en) * | 2020-09-14 | 2020-12-15 | 哈尔滨工程大学 | UUV submarine topography tracking path rolling generation method based on Bessel fitting |
| CN113052940A (en) * | 2021-03-14 | 2021-06-29 | 西北工业大学 | Space-time associated map real-time construction method based on sonar |
| CN113052940B (en) * | 2021-03-14 | 2024-03-15 | 西北工业大学 | Space-time correlation map real-time construction method based on sonar |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109213204B (en) | 2021-05-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109213204A (en) | AUV sub-sea floor targets based on data-driven search navigation system and method | |
| CN110782481B (en) | Unmanned ship intelligent decision-making method and system | |
| CN110414396B (en) | Unmanned ship perception fusion algorithm based on deep learning | |
| CN109283538A (en) | A kind of naval target size detection method of view-based access control model and laser sensor data fusion | |
| CN103149939B (en) | A kind of unmanned plane dynamic target tracking of view-based access control model and localization method | |
| CN111968128B (en) | Unmanned aerial vehicle visual attitude and position resolving method based on image markers | |
| CN105184816A (en) | Visual inspection and water surface target tracking system based on USV and detection tracking method thereof | |
| US20220024549A1 (en) | System and method for measuring the distance to an object in water | |
| CN113627473B (en) | Multi-mode sensor-based water surface unmanned ship environment information fusion sensing method | |
| CN112487912B (en) | Arbitrary direction ship detection method based on improved YOLOv3 | |
| CN111967337A (en) | Pipeline line change detection method based on deep learning and unmanned aerial vehicle images | |
| CN114004977A (en) | Aerial photography data target positioning method and system based on deep learning | |
| Shan et al. | LiDAR-based stable navigable region detection for unmanned surface vehicles | |
| CN108871409A (en) | A kind of fault detection method and system | |
| CN114782626A (en) | Transformer substation scene mapping and positioning optimization method based on laser and vision fusion | |
| CN110427030B (en) | Unmanned ship autonomous docking recovery method based on Tiny-YolOship target detection algorithm | |
| Dos Santos et al. | Underwater sonar and aerial images data fusion for robot localization | |
| CN113933828A (en) | Unmanned ship environment self-adaptive multi-scale target detection method and system | |
| CN109584264A (en) | A kind of unmanned plane vision guide air refuelling method based on deep learning | |
| CN113790718A (en) | Unmanned ship-based information sharing system | |
| Xu et al. | An overview of robust maritime situation awareness methods | |
| Deng et al. | Obstacle Detection of Unmanned Surface Vehicle Based on Lidar Point Cloud Data | |
| Jiang et al. | A machine vision method for the evaluation of ship-to-ship collision risk | |
| Ma et al. | Research on collision avoidance method based on image recognition technology | |
| Rui et al. | Real-Time obstacle detection based on monocular vision for unmanned surface vehicles |
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 |