CN108562913A - A kind of unmanned boat decoy detection method based on three-dimensional laser radar - Google Patents
A kind of unmanned boat decoy detection method based on three-dimensional laser radar Download PDFInfo
- Publication number
- CN108562913A CN108562913A CN201810353709.8A CN201810353709A CN108562913A CN 108562913 A CN108562913 A CN 108562913A CN 201810353709 A CN201810353709 A CN 201810353709A CN 108562913 A CN108562913 A CN 108562913A
- Authority
- CN
- China
- Prior art keywords
- target
- laser radar
- grid
- decoy
- unmanned boat
- 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
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/88—Lidar systems specially adapted for specific applications
- G01S17/93—Lidar systems specially adapted for specific applications for anti-collision purposes
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Electromagnetism (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Optical Radar Systems And Details Thereof (AREA)
- Traffic Control Systems (AREA)
Abstract
The present invention proposes a kind of unmanned boat decoy detection method based on three-dimensional laser radar, and using the obstacle target for not influencing unmanned boat navigation as decoy, three-dimensional laser radar, difference GNSS receiver and attitude angle transducer are installed on unmanned boat;Three-dimensional laser radar data are pre-processed, real-time displacement status data, is corrected laser point cloud obtained by unmanned boat real-time attitude angular data and difference GNSS receiver obtained by attitude angle transducer;Decoy detection is carried out with three-dimensional laser radar, including dividing to obtain obstacle target to grid, multinomial feature extraction is carried out, using each single item feature as the evidence for judging target type, target type discrimination frame is established, according to each burnt first Credibility judgement decoy.The present invention individually considers three-dimensional laser radar decoy as detection target using the high three-dimensional laser radar of reliability, enables not interfered by decoy when unmanned boat avoidance, improve the accuracy of target detection.
Description
Technical field
The invention belongs to obstacle targets to identify field, while be related to the information processing of three-dimensional laser radar, specially one
Unmanned boat decoy detection method of the kind based on three-dimensional laser radar.
Background technology
The automatic obstacle-avoiding of unmanned boat is to realize the intelligentized key technology difficulty of ship.Laser radar is surveyed as a kind of active
Quantity sensor has stronger adaptive capacity to environment and higher range accuracy, essence can be provided for unmanned boat in complex condition
Really reliable Environment Obstacles object information.Laser radar for avoidance generally can be divided into two-dimensional laser radar and three-dimensional laser thunder
It reaches.Three-dimensional laser radar run-down can obtain the distance and angle value of multiple planar points simultaneously, vertical direction have compared with
Big field angle, also can guarantee when unmanned boat rocks can get obstacle target.And two-dimensional laser radar is in unmanned boat
Often the water surface or sky are arrived in scanning when rocking, and no target information is caused to return.Therefore, three-dimensional laser radar is kept away in unmanned boat
Barrier field has obtained more concerns.
Obstacle target detection is carried out using laser radar, using relatively broad in unmanned vehicle, and is achieved preferably
Effect.For the navigation environment of unmanned boat complexity, laser radar has also gradually carried out some applications in recent years, mainly sharp
The information such as position, speed, the size for obtaining the barriers such as target ship, bank base, bridge with laser radar.However, laser thunder
Laser beam up to transmitting encounters the targets such as ship overtaking wave (containing many bubbles in water), planktonic organism (seaweed etc.), floating refuse
Will produce stronger reflection, generate laser point cloud data, and these targets have little effect unmanned boat navigation, thus by this
A little targets are referred to as decoy.If these decoys detected, unmanned boat can navigation area treat as unnavigability
Region, to influence the normal/cruise of unmanned boat.Therefore, it is necessary to which it is identified and is examined using the feature of these decoys
It surveys, to ensure that unmanned boat can safely and reliably complete automatic obstacle-avoiding process.
Invention content
The technical problem to be solved by the present invention is to propose a kind of unmanned boat decoy detection side based on three-dimensional laser radar
Method can be realized to the accurate detection of the decoys such as ship overtaking wave, floating algae, obtain the information such as position and the size of decoy.
The technical solution taken by the invention to solve the above technical problem is a kind of nobody based on three-dimensional laser radar
Ship decoy detection method, using do not influence unmanned boat navigation obstacle target as decoy, detection process includes the following steps,
Step 1, three-dimensional laser radar, difference GNSS receiver and attitude angle transducer, support pair are installed on unmanned boat
The detection of unmanned boat target obstacle;
Step 2, three-dimensional laser radar data are pre-processed, including coordinate is converted, point cloud corrects, extraordinary noise point is gone
It removes, grid indicates;Described cloud correction is to utilize unmanned boat real-time attitude angular data and difference GNSS obtained by attitude angle transducer
Real-time displacement status data, is corrected laser point cloud obtained by receiver;The grid indicates, is according to laser radar
Detection range establishes two-dimensional grid map, and unmanned boat is always the origin of grating map, grid attribute include grid coordinate, whether there is or not
Reentry point quantity, average height value and the maximum height difference that laser reentry point, grid include;
Step 3, decoy detection, including following sub-step are carried out using three-dimensional laser radar,
Step 3.1, it carries out obtaining obstacle target to the segmentation of grid,
Step 3.2, multinomial feature extraction is carried out to step 3.1 gained obstacle target according to step 2 gained grid attribute;
Step 3.3, target classification includes using obstacle target each single item feature that step 3.2 is extracted as sentencing
The evidence of disconnected target type, establishes target type discrimination frame, each single item evidence respectively divides target type confidence level
Match, calculates the confidence level of each burnt member in identification framework;According to each burnt first confidence level, pass through predetermined threshold value disturbance in judgement object
Whether target is decoy.
Moreover, in step 1, three-dimensional laser radar is mounted on unmanned boat, keeps radar scanning plane parallel with horizontal plane,
Obtain real-time three-dimensional point cloud data.
Moreover, the coordinate conversion realization method of step 2 is, if laser radar launch point is spheric coordinate system on the plane of scanning motion
Origin, plane of scanning motion itself is the reference planes of direction angle alpha and elevation angle ω, and three-dimensional laser radar real time scan gets mesh
The spherical coordinates (ρ, α, ω) that punctuate returns, is converted to the attached body coordinate (x, y, z) centered on unmanned boat.
Moreover, decoy includes ship overtaking wave, planktonic organism and floating refuse.
Moreover, the obstacle target feature that step 3.2 is extracted include grid quantity, grid average height, grid height it is poor,
Point cloud degree of rarefication, the evidence that this 4 kinds of features judge respectively as target type.
Moreover, obstacle target feature, i.e. grid quantity NumGrid, barrier grid average height
AveHeightGrid, barrier grid maximum height difference DiffHeightGrid, point cloud degree of rarefication SparsityPoint, respectively
It is defined as:
DiffHeightGrid=HeightGrid_max-HeightGrid_min,
Wherein, AveHeight [i] indicates that the height value of i-th of grid, HeightGrid_max are all for obstacle target
The maximum value of grid height, HeightGrid_min are the minimum value of all grid heights of obstacle target, and NumGrid indicates barrier
The grid quantity that object target includes, PointNum [i] is hindered to indicate that i-th of grid reentry point quantity, PointNorm indicate ideal feelings
Each grid reentry point quantity of condition.
Moreover, in step 3.3, target type discrimination frame is established based on Dempster-Shafer evidence theories, it will be each
After item evidence is respectively allocated target type confidence level, each in identification framework is calculated using Dempster composition rules
The confidence level of burnt member.
Moreover, in step 3.3, target type is divided into real goal A, decoy B, Unknown Subject C, corresponding identification framework
It is defined as Θ={ { A }, { B }, { C }, { A, C }, { B, C } }, according to the concrete numerical value of 4 evidences, respectively in identification framework
Each burnt member carries out belief assignment and synthesis.
Moreover, whether being decoy by predetermined threshold value disturbance in judgement object target, include the following steps,
Step (1), if the decoy confidence level m (B) after synthesis is more than 0.7, disturbance in judgement object target is decoy;
Otherwise, it enters step (2);
Step (2), if the real goal confidence level m (A) after synthesis is more than 0.7, disturbance in judgement object target is true
Target;Otherwise, target type is indefinite.
Beneficial effects of the present invention are:The present invention is using the high three-dimensional laser radar of reliability perceptually means, to obtain
Unmanned boat Environment Obstacles object target information is taken, obstacle distance information and maritime affairs can not be obtained by effectively compensating for visual sensor
Radar has the deficiency of measurement blind area;The present invention individually considers three-dimensional laser radar decoy as detection target, makes unmanned boat
It can not be interfered by decoy when avoidance;The present invention is proposed using D-S evidence theory come dyssynthesis object target different attribute
Judgement to target classification result, and allow mixed type occur in classification results, improve the accuracy of target detection.
Description of the drawings
Fig. 1 is the three-dimensional laser radar scheme of installation of the embodiment of the present invention.
Fig. 2 is the three-dimensional laser radar decoy overhaul flow chart of the embodiment of the present invention.
Fig. 3 is that the three-dimensional laser radar grid of the embodiment of the present invention indicates schematic diagram.
Fig. 4 is the obstacle target schematic diagram after the three-dimensional laser radar grid segmentation of the embodiment of the present invention.
Fig. 5 is the evidence theory target classification schematic diagram of the embodiment of the present invention.
In figure:1- unmanned boats, 2- data processing units, 3- attitude transducers, 4- three-dimensional laser radars, the 5- waters surface, 6- swash
The optical radar plane of scanning motion, 7- vertical scanning visual angle, 8- difference GNSS receivers.
Specific implementation mode
With reference to specific embodiments and the drawings, the present invention will be further described.
The embodiment of the present invention provides a kind of unmanned boat decoy detection method based on three-dimensional laser radar, and decoy is specific
For ship overtaking wave, planktonic organism, floating refuse these do not influence unmanned boat navigation obstacle target.As shown in Fig. 2, embodiment
Flow is as follows:
(1) three-dimensional laser radar, difference GNSS receiver, attitude angle transducer are utilized, supports to realize to unmanned boat target
The detection of barrier:
When it is implemented, three-dimensional laser radar is horizontally installed at the top of unmanned boat, ensure laser when laser radar scanning
Beam is not blocked by hull;Installation posture sensor can obtain unmanned boat attitude data in real time.
In embodiment, three-dimensional laser radar scheme of installation is as shown in Figure 1, three-dimensional laser radar 4, difference GNSS are received
Machine 8, attitude angle transducer 3, data processing unit 2 realize that obstacle target detects on unmanned boat 1, three-dimensional laser thunder
Data are exported to data processing unit 2 and are handled in real time up to 4, difference GNSS receiver 8, attitude angle transducer 3, wherein answering
Ensure that the three-dimensional laser radar plane of scanning motion 6 is parallel with horizontal plane 5 to avoid laser radar scanning to the water surface as far as possible, and radar is swept
Visual angle 7 is retouched by hull 1 not stopped.
(2) three-dimensional laser radar point cloud data is pre-processed, including coordinate conversion, the correction of point cloud, extraordinary noise point
Removal, grid indicate, wherein point cloud correction realized using real-time attitude angle and real-time displacement status data;
The removal of extraordinary noise point is realized using median filter method;Grid attribute includes grid coordinate during grid indicates, whether there is or not laser
Reentry point quantity, average height value and the maximum height difference that reentry point, grid include.
Embodiment includes following sub-step:
S1, coordinate conversion.
If laser radar launch point is the origin of spheric coordinate system on the plane of scanning motion 6, the plane of scanning motion 6 is direction angle alpha in itself
With the reference planes of elevation angle ω.
Three-dimensional laser radar real time scan gets the spherical coordinates (ρ, α, ω) of target point return, and wherein ρ, α and ω distinguishes
For the sphere diameter of spherical coordinates, azimuth and elevation angle.Spherical coordinates is converted into the attached body coordinate (x, y, z) centered on unmanned boat,
Wherein x, y and z are respectively the coordinate value of x-axis, y-axis and z-axis, as shown in Figure 1.In coordinate frame attached to a body, x-axis is bow to side
To y-axis is hull horizontal direction, and z-axis is vertically upward.
Conversion formula is as follows:
S2, point cloud correction.Using unmanned boat real-time attitude angular data (φ, θ, ψ) (being obtained by attitude angle transducer 3) and
Real-time displacement status data (u, v, w) (being obtained by difference GNSS receiver 8), is corrected laser point cloud, obtains
Coordinate (X, Y, Z) of the target reentry point under earth coordinates.Wherein, φ, θ and ψ be respectively unmanned boat Angle of Heel, Angle of Trim and
Azimuth;U, v and w is respectively the speed in unmanned boat coordinate system x-axis, y-axis and z-axis direction;X, Y and Z is respectively point cloud in the earth
The coordinate of coordinate system X-axis, Y-axis and Z-direction.
Earth coordinates (X, Y, Z) and the transformational relation of unmanned boat coordinate system are:
S3, the removal of extraordinary noise point.Using the z values of point cloud data point as the gray value of pixel in image, it is assumed that certain number
The height average of strong point (x, y, z) neighborhood point is zm, setting medium filtering threshold value is Th, if | z-zm| > Th, the then data
Point is noise spot, and height value z is by zmIt substitutes;Otherwise, which is signaling point, and height value remains unchanged.
S4, grid indicate.Two-dimensional grid map is established according to laser radar detection range, unmanned boat is always grating map
Origin.Grid attribute includes grid coordinate, includes whether there is or not laser reentry point, grid reentry point quantity, average height value and
Maximum height difference.Grid coordinate attribute definition is { (xG, yG), S, PointNum, AveHeight, MaxHeight }, wherein (xG,
YG it is) coordinate value of the grid in coordinate frame attached to a body, S indicates that whether there is or not laser reentry point (1- has, 0- without), PointNum tables in grid
Show that the reentry point quantity that grid includes, AveHeight indicate average height difference (all reentry point z coordinates of reentry point in grid
The average value of value), MaxHeight indicates the maximum height (maximum values of all reentry point z coordinate values) of reentry point in grid.
Grid indicate as shown in figure 3, wherein R be three-dimensional laser radar scanning range, such as grid attribute { (xG,yG),
S, PointNum, AveHeight, MaxHeight } value be { (2,1), 1,20,5,3 }.
(3) decoy detection is carried out using three-dimensional laser radar, the present invention proposes following manner:
Feature extraction is carried out to the obstacle target that unmanned boat three-dimensional laser radar obtains;
Using the obstacle target each single item feature extracted as judging the evidence of target type, and it is based on
Dempster-Shafer (D-S) evidence theory establishes target type discrimination frame
Each single item evidence is respectively allocated target type confidence level, is calculated and is identified using Dempster composition rules
The confidence level of each burnt member in frame;
Whether each burnt first confidence level obtained according to synthesis, be decoy by threshold method disturbance in judgement object target
It is realized using following sub-step in embodiment:
S5, grid segmentation.The segmentation to grid is realized using the fast area labelling method of eight neighborhood expansion, will be connected together
Grid as one group, be identified as an obstacle target, and number to it.The obstacle target obtained after segmentation such as Fig. 4 institutes
Show, wherein the grid labeled as 1,2,3 respectively belongs to an obstacle target respectively.
S6, feature extraction.Feature extraction, the grid attribute obtained according to step S4 are carried out to the obstacle target after segmentation
({(xG, yG), S, PointNum, AveHeight, MaxHeight) and the obtained object segmentation results of step S5 extract barrier
Hinder object feature, including grid quantity, grid average height, grid height are poor, point cloud degree of rarefication etc..Therefore, by some barrier
Objective attribute target attribute may be defined as { NumGrid, AveHeightGrid, DiffHeightGrid, SparsityPoint }, wherein
NumGrid indicates that grid quantity, AveHeightGrid indicate the average height of all grids of obstacle target,
DiffHeightGrid indicates that all grid maximum height absolute value of the difference of obstacle target, SparsityPoint indicate point cloud
Degree of rarefication.Wherein, AveHeightGrid indicates that a certain obstacle target includes the average value of all grid heights;
DiffHeightGrid indicates that a certain obstacle target includes the maximum value of the difference of the height of all grids;SparsityPoint
The sparse degree for describing return target point, for the decoys such as ship overtaking wave, floating algae, the big portion of laser beam of laser transmitting
It is to be only absorbed by the water to divide, and small part is just reflected, therefore the point cloud of decoy is more sparse.For example, certain obstacle target attribute takes
Value is { 2,0.5,0.3,50% }.
Obstacle target grid height average value AveHeightGrid is defined as:
Wherein, AveHeight [i] indicates the height value of i-th of grid.
All grid maximum height absolute value of the difference DiffHeightGrid of obstacle target are defined as:
DiffHeightGrid=HeightGrid_max-HeightGrid_min (4),
Wherein, HeightGrid_max is the maximum value of all grid heights of obstacle target, and HeightGrid_min is
The minimum value of all grid heights of obstacle target.
The sparse degree SparsityPoint of target point is defined as:
Wherein, PointNum [i] indicates that i-th of grid reentry point quantity, PointNorm indicate each grid of ideal situation
Lattice reentry point quantity.
S7, target classification.The classification to decoy is realized using Dempster-Shafer (D-S) evidence theory, and flow is such as
Shown in Fig. 5.For decoy classification problem, target is divided into 3 classes, i.e. A (real goal), B (decoy), C (Unknown Subject).
Therefore, the identification framework of evidence theory can be defined as
Θ1={ { A }, { B }, { C }, { A, B }, { A, C }, { B, C }, { A, B, C }, { φ } }.
Wherein, the probability very little that proposition { A, B }, { A, B, C } and { φ } occurs.It is calculated to simplify, identification framework is changed
For Θ2={ { A }, { B }, { C }, { A, C }, { B, C } }.By defined in step S6 4 attribute of obstacle target NumGrid,
AveHeightGrid, DiffHeightGrid, SparsityPoint } respectively as 4 evidences, obtain identification framework Θ2Often
A kind of confidence value of classification passes through the type of the final disturbance in judgement object target of threshold value.Fig. 5 lists that evidence 1~4 is corresponding can
Certainty value, shown in table specific as follows.
In table, NumGrid indicates that the grid quantity that barrier includes, AveHeightGrid indicate that obstacle target is all
The average height of grid, DiffHeighGrid indicate all grid maximum height absolute value of the difference of obstacle target,
SparsityPoint indicates point cloud degree of rarefication.mi(X) indicate i-th of evidence to the degree of belief of proposition X, i=1,2,3,4, m (X)
The degree of belief of proposition X after expression combining evidences.
According to the concrete numerical value of 4 evidences, belief assignment and synthesis are carried out to burnt member each of in identification framework respectively.
The present invention proposes that the belief assignment using 4 kinds of different evidences of Dempster methods pair synthesizes, and obtains each burnt member and closes
Confidence level after, then the concrete type by threshold method disturbance in judgement object target.
The confidence value after 4 evidence fusions is obtained using Dempster composition rules, specially:
Wherein,For evidence conflict coefficient, and K ≠ 1.
Y1、Y2、Y3And Y4Correspond to any proposition of evidence 1, evidence 2, evidence 3 and evidence 4 respectively, i.e., { A }, { B }, { C },
{ A, C } and { B, C }.
It obtains after combining evidences after the degree of belief m (A) of proposition A, combining evidences after the degree of belief m (B) of proposition B, according to Fig. 5
Shown in flow come whether disturbance in judgement object target is decoy, when specific implementation, can be sentenced according to preset judgment threshold
It is disconnected.It is 0.7 that embodiment, which is preferably provided with judgment threshold,:
Step a can determine whether that obstacle target is decoy if the decoy confidence level m (B) after synthesis is more than 0.7;
Otherwise, a is entered step;
Step b can determine whether that obstacle target is true if the real goal confidence level m (A) after synthesis is more than 0.7
Target;Otherwise, target type is indefinite.
When it is implemented, the target classification result that will determine that can be output to subsequent unmanned boat avoidance decision system
System ensures that unmanned boat being capable of safe avoiding obstacles for calculating collision prevention path according to the obstacle target information of offer.
When it is implemented, the automatic running of computer software technology implementation process can be used.
Be described in above-described embodiment illustrate the present invention, though text in illustrated by specific term, not
Protection scope of the present invention can be limited with this, be familiar with this technical field personage can understand the present invention spirit with it is right after principle
It changes or changes and reaches equivalent purpose, and this equivalent change and modification, should all be covered by right institute circle
Determine in scope.
Claims (9)
1. a kind of unmanned boat decoy detection method based on three-dimensional laser radar, it is characterised in that:Not influence unmanned boat boat
Capable obstacle target is decoy, and detection process includes the following steps,
Step 1, three-dimensional laser radar, difference GNSS receiver and attitude angle transducer are installed on unmanned boat, supported to nobody
The detection of ship target obstacle;
Step 2, three-dimensional laser radar data are pre-processed, including coordinate conversion, point cloud correct, extraordinary noise point removes,
Grid indicates;Described cloud correction, is connect using unmanned boat real-time attitude angular data and difference GNSS obtained by attitude angle transducer
Real-time displacement status data, is corrected laser point cloud obtained by receipts machine;The grid indicates, is examined according to laser radar
It surveys range and establishes two-dimensional grid map, unmanned boat is always the origin of grating map, and grid attribute includes grid coordinate, whether there is or not swash
Reentry point quantity, average height value and the maximum height difference that light reentry point, grid include;
Step 3, decoy detection, including following sub-step are carried out using three-dimensional laser radar,
Step 3.1, it carries out obtaining obstacle target to the segmentation of grid,
Step 3.2, multinomial feature extraction is carried out to step 3.1 gained obstacle target according to step 2 gained grid attribute;
Step 3.3, target classification includes using obstacle target each single item feature that step 3.2 is extracted as judging mesh
The evidence for marking type, establishes target type discrimination frame, each single item evidence is respectively allocated target type confidence level, counts
Calculate the confidence level of each burnt member in identification framework;According to each burnt first confidence level, pass through predetermined threshold value disturbance in judgement object target
Whether it is decoy.
2. the unmanned boat decoy detection method according to claim 1 based on three-dimensional laser radar, it is characterised in that:Step
In rapid 1, three-dimensional laser radar is mounted on unmanned boat, keeps radar scanning plane parallel with horizontal plane, obtains real-time three-dimensional point
Cloud data.
3. the unmanned boat decoy detection method according to claim 1 based on three-dimensional laser radar, it is characterised in that:Step
Rapid 2 coordinate conversion realization method is, if laser radar launch point is the origin of spheric coordinate system, the plane of scanning motion on the plane of scanning motion
Itself it is the reference planes of direction angle alpha and elevation angle ω, the ball that three-dimensional laser radar real time scan gets target point return is sat
It marks (ρ, α, ω), is converted to the attached body coordinate (x, y, z) centered on unmanned boat.
4. the unmanned boat decoy detection method according to claim 1 based on three-dimensional laser radar, it is characterised in that:It is false
Target includes ship overtaking wave, planktonic organism and floating refuse.
5. the unmanned boat decoy detection method according to claim 1 based on three-dimensional laser radar, it is characterised in that:Step
The obstacle target feature of rapid 3.2 extraction includes grid quantity, grid average height, grid height is poor, puts cloud degree of rarefication, this 4
The evidence that kind feature judges respectively as target type.
6. the unmanned boat decoy detection method based on three-dimensional laser radar described in claim 5, it is characterised in that:Barrier
Target signature, i.e. grid quantity NumGrid, barrier grid average height AveHeightGrid, barrier grid maximum height
Poor DiffHeightGrid, point cloud degree of rarefication SparsityPoint, are respectively defined as:
DiffHeightGrid=HeightGrid_max-HeightGrid_min,
Wherein, AveHeight [i] indicates that the height value of i-th of grid, HeightGrid_max are all grids of obstacle target
The maximum value of height, HeightGrid_min are the minimum value of all grid heights of obstacle target, and NumGrid indicates barrier
The grid quantity that target includes, PointNum [i] indicate that i-th of grid reentry point quantity, PointNorm indicate that ideal situation is every
One grid reentry point quantity.
7. the unmanned boat decoy detection method according to claim 1 based on three-dimensional laser radar, it is characterised in that:Step
In rapid 3.3, target type discrimination frame is established based on Dempster-Shafer evidence theories, by each single item evidence respectively to mesh
After mark type confidence level is allocated, the confidence level of each burnt member in identification framework is calculated using Dempster composition rules.
8. the unmanned boat decoy detection method according to claim 7 based on three-dimensional laser radar, it is characterised in that:Step
In rapid 3.3, target type is divided into real goal A, decoy B, Unknown Subject C, corresponding identification framework be defined as Θ={ A },
{ B }, { C }, { A, C }, { B, C } }, according to the concrete numerical value of 4 evidences, burnt member each of in identification framework is carried out respectively credible
Degree distribution and synthesis.
9. the unmanned boat decoy detection method according to claim 8 based on three-dimensional laser radar, it is characterised in that:It is logical
It crosses whether predetermined threshold value disturbance in judgement object target is decoy, includes the following steps,
Step (1), if the decoy confidence level m (B) after synthesis is more than 0.7, disturbance in judgement object target is decoy;It is no
Then, it enters step (2);
Step (2), if the real goal confidence level m (A) after synthesis is more than 0.7, disturbance in judgement object target is real goal;
Otherwise, target type is indefinite.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810353709.8A CN108562913B (en) | 2018-04-19 | 2018-04-19 | Unmanned ship false target detection method based on three-dimensional laser radar |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810353709.8A CN108562913B (en) | 2018-04-19 | 2018-04-19 | Unmanned ship false target detection method based on three-dimensional laser radar |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108562913A true CN108562913A (en) | 2018-09-21 |
CN108562913B CN108562913B (en) | 2021-12-17 |
Family
ID=63535843
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810353709.8A Active CN108562913B (en) | 2018-04-19 | 2018-04-19 | Unmanned ship false target detection method based on three-dimensional laser radar |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108562913B (en) |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109725331A (en) * | 2019-03-18 | 2019-05-07 | 燕山大学 | A kind of unmanned boat barrier-avoiding method based on laser radar |
CN110175186A (en) * | 2019-05-15 | 2019-08-27 | 中国舰船研究设计中心 | A kind of intelligent ship environmental threat target apperception system and method |
CN110580045A (en) * | 2019-08-30 | 2019-12-17 | 安徽科微智能科技有限公司 | obstacle avoidance method for unmanned ship |
CN110675418A (en) * | 2019-09-26 | 2020-01-10 | 深圳市唯特视科技有限公司 | Target track optimization method based on DS evidence theory |
CN110737271A (en) * | 2019-10-23 | 2020-01-31 | 西南科技大学 | Autonomous cruise system and method for water surface robots |
WO2020082947A1 (en) * | 2018-10-22 | 2020-04-30 | 科沃斯机器人股份有限公司 | Travel control method, device, and storage medium |
CN111323418A (en) * | 2020-02-26 | 2020-06-23 | 天津大学 | Sea surface oil stain monitoring system based on unmanned ship |
CN111337948A (en) * | 2020-02-25 | 2020-06-26 | 达闼科技成都有限公司 | Obstacle detection method, radar data generation device, and storage medium |
CN111708038A (en) * | 2020-06-23 | 2020-09-25 | 上海埃威航空电子有限公司 | Unmanned ship laser radar point cloud data correction method based on attitude sensor and GNSS |
CN113144264A (en) * | 2021-03-18 | 2021-07-23 | 武汉联一合立技术有限公司 | Intelligent killing system and method |
CN113970922A (en) * | 2020-07-22 | 2022-01-25 | 商汤集团有限公司 | Point cloud data processing method and intelligent driving control method and device |
CN114241211A (en) * | 2021-11-26 | 2022-03-25 | 中国船舶重工集团公司第七0四研究所 | Laser radar point cloud feature-based shoreline extraction method |
CN114578312A (en) * | 2022-05-09 | 2022-06-03 | 中国电子科技集团公司信息科学研究院 | False target suppression method and device based on random elimination |
RU2796811C1 (en) * | 2022-04-11 | 2023-05-29 | Федеральное государственное казенное военное образовательное учреждение высшего образования "Военный учебно-научный центр Военно-воздушных сил "Военно-воздушная академия имени профессора Н.Е. Жуковского и Ю.А. Гагарина" (г. Воронеж) Министерства обороны Российской Федерации | Method of simulation of an optoelectronic device |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104297743A (en) * | 2014-10-11 | 2015-01-21 | 中国林业科学研究院资源信息研究所 | Method and device for eliminating distance measuring ambiguity of high repetition frequency airborne laser radar system |
CN106355194A (en) * | 2016-08-22 | 2017-01-25 | 广东华中科技大学工业技术研究院 | Treatment method for surface target of unmanned ship based on laser imaging radar |
CN106997049A (en) * | 2017-03-14 | 2017-08-01 | 奇瑞汽车股份有限公司 | A kind of method and apparatus of the detection barrier based on laser point cloud data |
CN107092020A (en) * | 2017-04-19 | 2017-08-25 | 北京大学 | Merge the surface evenness monitoring method of unmanned plane LiDAR and high score image |
CN107121980A (en) * | 2017-03-17 | 2017-09-01 | 北京理工大学 | A kind of automatic driving vehicle paths planning method based on virtual constraint |
CN107330925A (en) * | 2017-05-11 | 2017-11-07 | 北京交通大学 | A kind of multi-obstacle avoidance detect and track method based on laser radar depth image |
CN104764457B (en) * | 2015-04-21 | 2017-11-17 | 北京理工大学 | A kind of urban environment patterning process for unmanned vehicle |
CN107807365A (en) * | 2017-10-20 | 2018-03-16 | 国家***昆明勘察设计院 | Small-sized digital photography there-dimensional laser scanning device for the unmanned airborne vehicle in low latitude |
CN107870334A (en) * | 2017-10-27 | 2018-04-03 | 西安电子科技大学昆山创新研究院 | Single pixel laser infrared radar imaging device and imaging method based on embedded gpu |
-
2018
- 2018-04-19 CN CN201810353709.8A patent/CN108562913B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104297743A (en) * | 2014-10-11 | 2015-01-21 | 中国林业科学研究院资源信息研究所 | Method and device for eliminating distance measuring ambiguity of high repetition frequency airborne laser radar system |
CN104764457B (en) * | 2015-04-21 | 2017-11-17 | 北京理工大学 | A kind of urban environment patterning process for unmanned vehicle |
CN106355194A (en) * | 2016-08-22 | 2017-01-25 | 广东华中科技大学工业技术研究院 | Treatment method for surface target of unmanned ship based on laser imaging radar |
CN106997049A (en) * | 2017-03-14 | 2017-08-01 | 奇瑞汽车股份有限公司 | A kind of method and apparatus of the detection barrier based on laser point cloud data |
CN107121980A (en) * | 2017-03-17 | 2017-09-01 | 北京理工大学 | A kind of automatic driving vehicle paths planning method based on virtual constraint |
CN107092020A (en) * | 2017-04-19 | 2017-08-25 | 北京大学 | Merge the surface evenness monitoring method of unmanned plane LiDAR and high score image |
CN107330925A (en) * | 2017-05-11 | 2017-11-07 | 北京交通大学 | A kind of multi-obstacle avoidance detect and track method based on laser radar depth image |
CN107807365A (en) * | 2017-10-20 | 2018-03-16 | 国家***昆明勘察设计院 | Small-sized digital photography there-dimensional laser scanning device for the unmanned airborne vehicle in low latitude |
CN107870334A (en) * | 2017-10-27 | 2018-04-03 | 西安电子科技大学昆山创新研究院 | Single pixel laser infrared radar imaging device and imaging method based on embedded gpu |
Cited By (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11487297B2 (en) | 2018-10-22 | 2022-11-01 | Ecovacs Robotics Co., Ltd. | Method of travel control, device and storage medium |
WO2020082947A1 (en) * | 2018-10-22 | 2020-04-30 | 科沃斯机器人股份有限公司 | Travel control method, device, and storage medium |
CN109725331A (en) * | 2019-03-18 | 2019-05-07 | 燕山大学 | A kind of unmanned boat barrier-avoiding method based on laser radar |
CN109725331B (en) * | 2019-03-18 | 2023-04-07 | 燕山大学 | Unmanned ship obstacle avoidance method based on laser radar |
CN110175186A (en) * | 2019-05-15 | 2019-08-27 | 中国舰船研究设计中心 | A kind of intelligent ship environmental threat target apperception system and method |
CN110175186B (en) * | 2019-05-15 | 2023-09-26 | 中国舰船研究设计中心 | Intelligent ship environment threat target sensing system and method |
CN110580045A (en) * | 2019-08-30 | 2019-12-17 | 安徽科微智能科技有限公司 | obstacle avoidance method for unmanned ship |
CN110675418A (en) * | 2019-09-26 | 2020-01-10 | 深圳市唯特视科技有限公司 | Target track optimization method based on DS evidence theory |
CN110737271A (en) * | 2019-10-23 | 2020-01-31 | 西南科技大学 | Autonomous cruise system and method for water surface robots |
CN111337948A (en) * | 2020-02-25 | 2020-06-26 | 达闼科技成都有限公司 | Obstacle detection method, radar data generation device, and storage medium |
CN111323418A (en) * | 2020-02-26 | 2020-06-23 | 天津大学 | Sea surface oil stain monitoring system based on unmanned ship |
CN111708038A (en) * | 2020-06-23 | 2020-09-25 | 上海埃威航空电子有限公司 | Unmanned ship laser radar point cloud data correction method based on attitude sensor and GNSS |
WO2022017131A1 (en) * | 2020-07-22 | 2022-01-27 | 商汤集团有限公司 | Point cloud data processing method and device, and intelligent driving control method and device |
CN113970922A (en) * | 2020-07-22 | 2022-01-25 | 商汤集团有限公司 | Point cloud data processing method and intelligent driving control method and device |
CN113144264A (en) * | 2021-03-18 | 2021-07-23 | 武汉联一合立技术有限公司 | Intelligent killing system and method |
CN114241211A (en) * | 2021-11-26 | 2022-03-25 | 中国船舶重工集团公司第七0四研究所 | Laser radar point cloud feature-based shoreline extraction method |
RU2796811C1 (en) * | 2022-04-11 | 2023-05-29 | Федеральное государственное казенное военное образовательное учреждение высшего образования "Военный учебно-научный центр Военно-воздушных сил "Военно-воздушная академия имени профессора Н.Е. Жуковского и Ю.А. Гагарина" (г. Воронеж) Министерства обороны Российской Федерации | Method of simulation of an optoelectronic device |
CN114578312B (en) * | 2022-05-09 | 2022-08-02 | 中国电子科技集团公司信息科学研究院 | False target suppression method and device based on random elimination |
CN114578312A (en) * | 2022-05-09 | 2022-06-03 | 中国电子科技集团公司信息科学研究院 | False target suppression method and device based on random elimination |
RU2813678C1 (en) * | 2023-07-21 | 2024-02-15 | Федеральное государственное казенное военное образовательное учреждение высшего образования "Военный учебно-научный центр Военно-воздушных сил "Военно-воздушная академия имени профессора Н.Е. Жуковского и Ю.А. Гагарина" (г. Воронеж) Министерства обороны Российской Федерации | Method for imitating reflection surfaces of optoelectronic device |
Also Published As
Publication number | Publication date |
---|---|
CN108562913B (en) | 2021-12-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108562913A (en) | A kind of unmanned boat decoy detection method based on three-dimensional laser radar | |
CN111352112B (en) | Target detection method based on vision, laser radar and millimeter wave radar | |
CN109670411B (en) | Ship point cloud depth image processing method and system based on generation countermeasure network | |
CN110275153B (en) | Water surface target detection and tracking method based on laser radar | |
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 | |
CN112882059B (en) | Unmanned ship inland river obstacle sensing method based on laser radar | |
CN110411366B (en) | Road water depth detection method and electronic equipment | |
CN107884767A (en) | A kind of method of binocular vision system measurement ship distance and height | |
CN111598952B (en) | Multi-scale cooperative target design and online detection identification method and system | |
CN107192994A (en) | Multi-line laser radar mass cloud data is quickly effectively extracted and vehicle, lane line characteristic recognition method | |
CN111126335B (en) | SAR ship identification method and system combining significance and neural network | |
CN112394726B (en) | Unmanned ship obstacle fusion detection method based on evidence theory | |
Ruiz et al. | A short-range ship navigation system based on ladar imaging and target tracking for improved safety and efficiency | |
KR102530847B1 (en) | Method and device for monitoring harbor and ship | |
CN109213204A (en) | AUV sub-sea floor targets based on data-driven search navigation system and method | |
CN114740493A (en) | Road edge detection method based on multi-line laser radar | |
Zhang et al. | Vessel detection and classification fusing radar and vision data | |
CN114384492A (en) | Point cloud processing method and device for laser radar and storage medium | |
CN102998001B (en) | A kind of object detection system | |
Thompson | Maritime object detection, tracking, and classification using lidar and vision-based sensor fusion | |
CN113933828A (en) | Unmanned ship environment self-adaptive multi-scale target detection method and system | |
CN113790718A (en) | Unmanned ship-based information sharing system | |
CN116071694B (en) | Ship detection method, device and computer readable storage medium | |
CN111105419B (en) | Vehicle and ship detection method and device based on polarized SAR image |
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 |