CN112882059B - Unmanned ship inland river obstacle sensing method based on laser radar - Google Patents
Unmanned ship inland river obstacle sensing method based on laser radar Download PDFInfo
- Publication number
- CN112882059B CN112882059B CN202110026459.9A CN202110026459A CN112882059B CN 112882059 B CN112882059 B CN 112882059B CN 202110026459 A CN202110026459 A CN 202110026459A CN 112882059 B CN112882059 B CN 112882059B
- Authority
- CN
- China
- Prior art keywords
- grid
- target
- point cloud
- information
- ship
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
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
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/10—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration
- G01C21/12—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning
- G01C21/16—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation
- G01C21/165—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation combined with non-inertial navigation instruments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/50—Depth or shape recovery
- G06T7/521—Depth or shape recovery from laser ranging, e.g. using interferometry; from the projection of structured light
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10028—Range image; Depth image; 3D point clouds
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Networks & Wireless Communication (AREA)
- Electromagnetism (AREA)
- Theoretical Computer Science (AREA)
- Optics & Photonics (AREA)
- Automation & Control Theory (AREA)
- Optical Radar Systems And Details Thereof (AREA)
- Radar Systems Or Details Thereof (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention relates to a laser radar-based unmanned ship inland river obstacle sensing method, which comprises the following steps: 1. constructing a polygonal geographic electronic fence area of a task water area in real time on line by taking the position of the ship as a reference point; 2. carrying out distortion correction on the 360-degree point cloud generated by the laser radar, and filtering the processed point cloud information by combining the ship position according to the geo-fence area in the step 1; 3, carrying out horizontal plane grid projection dimensionality reduction processing on the corrected point cloud, counting point cloud multi-dimensional feature information belonging to each grid, and calculating the target confidence coefficient of each grid according to the feature values; 4, carrying out noise removal, target grid segmentation and clustering processing on the grid map, extracting the position and scale information of each target, and obtaining a target two-dimensional rectangular surrounding frame; and 5, extracting the position coordinates and the length-width ratio information of the bounding box of the obstacle target according to the raster maps of the front frame and the rear frame of the laser radar, performing related matching of the clustering targets of the front frame and the rear frame, and extracting the position and speed information of the target. The invention realizes the accurate acquisition of the position and the size of the target.
Description
Technical Field
The invention belongs to the field of target sensing and detection of unmanned intelligent ships, and particularly relates to an unmanned ship inland river obstacle sensing method based on a laser radar.
Background
Compared with the open sea, the inland waterway has the characteristics of narrow and tortuous navigation paths, small size, various types and the like for typical targets such as wharfs, swimming people, fishing net cages, floating balls, aquatic plants and the like. The main sensor navigation radar relied on by unmanned ship perception is influenced by the embankments on two sides and the land reflection clutter in a narrow inland river, the obstacle detection and tracking performance is obviously reduced, and the requirement of avoiding the unmanned ship inland river navigation obstacles can not be met.
By utilizing the characteristics of high resolution, accurate distance and scale information and short detection period of the laser radar, different types of water surface targets can be detected more quickly and accurately, and the method is an important means for sensing the obstacles of the unmanned surface ship. However, due to the characteristics of disorder, high sparsity and large data volume of the laser radar point cloud data, the obstacle perception research of the inland river unmanned ship based on the laser radar is relatively less at present.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides the unmanned ship inland river obstacle sensing method based on the laser radar, which can realize sensing detection of inland waterway obstacles and accurately obtain the target position and size.
The above object of the present invention is achieved by the following technical solutions:
an unmanned ship inland river obstacle sensing method based on a laser radar is characterized by comprising the following steps:
step 1, constructing a profile of a navigable area, which specifically comprises the following steps: constructing a polygonal geographic electronic fence area of a task navigable water area on line in real time by using longitude and latitude coordinates of the navigable area of the inland river, which are acquired from known information such as a map, a chart and the like, and taking the position of a ship as a reference point;
step 2, receiving real-time pose information by using a differential GNSS and a shipborne IMU, and compensating X, Y and Z installation angle errors obtained after the laser radar and the IMU are calibrated; carrying out distortion correction on 360-degree point cloud generated by the laser radar by utilizing IMU attitude information, and converting a point cloud spherical coordinate system into a geodetic Cartesian coordinate system; finally, combining the processed point cloud information with the position of the ship, filtering according to the geo-fencing area in the step 1, and removing point clouds outside the navigable area;
step 3, performing horizontal plane grid projection dimensionality reduction on the point cloud corrected in the step 2 to form a two-dimensional projection grid, counting point cloud multi-dimensional feature information belonging to each grid, and calculating the target confidence coefficient of each grid according to the multi-dimensional feature value;
step 4, performing obstacle grid filtering on the grid map formed by the two-dimensional projection grid in the step 3, including noise removal, target grid segmentation and clustering processing, acquiring a target two-dimensional rectangular surrounding frame, and extracting the position and scale information of each target;
and 5, extracting the position coordinates and the length-width ratio information of the bounding box of the obstacle target according to the raster maps of the front frame and the rear frame of the laser radar, performing the correlation matching of the clustering targets of the front frame and the rear frame, and extracting the position and speed information of the target.
And further: in step 2, the distortion correction adopts the following mode: bilinear interpolation is carried out on the acquired pose information, distortion correction is realized by acquiring the interpolated pose information according to timestamps corresponding to 360-degree different sectors of the laser radar point cloud, and distortion of the point cloud caused by ship body swinging and movement is reduced.
Further: in the step 3, the two-dimensional projection grid takes the laser radar as the center, is parallel to the horizontal plane, adopts a square grid, the grid division density is determined according to the minimum resolution required by the obstacle to be detected, and the side length of each grid is 0.5 times of the minimum resolution of the obstacle.
Further: in step 3, the grid point cloud multi-dimensional feature information comprises the number of point clouds, the maximum and minimum height difference, the minimum height and the average reflection intensity.
Further: in step 4, the clustering process mainly adopts Euclidean distance clustering, and comprises the following substeps:
step 4.1, dividing different thresholds in the ship and wake flow region, and extracting the obstacle by using the threshold H thre The following settings are set:
within the normal threshold region, H w The wave height during navigation is determined by sea conditions(ii) a In a high threshold value area, the ship speed v of the ship is set to be in direct proportion, and a specific proportionality coefficient is related to ship characteristics;
the barrier grid filtering algorithm provided by the invention specifically comprises the following steps: traversing the grid map which is extracted in the whole step 3 and contains point cloud multi-dimensional characteristic information, and sliding a window to remove noise influence generated by water surface waves, wherein the judgment of the target grid simultaneously meets the following conditions:
N＞2
H min ＞1.5H vessel
(H max -H min )＞H thre
intensity＞10
wherein N is the number of point clouds falling within the grid coordinate interval, H max Is the maximum height of point cloud in the grid, H min Is the minimum height, H, of the point cloud in the grid vessel The height of the highest point of the ship from the water surface and intensity are average reflection intensity (in the example, the range is 0-255).
Step 4.2, traversing each target grid in the grid map, skipping if the point is processed, otherwise defining a seed grid queue and adding a seed grid, taking an eight-neighborhood grid of the grid as a clustering radius, and marking the target grid in the eight-neighborhood as a processed grid by using a rapid expansion marking method to cluster the target grids into the same target;
and 4.3, setting the minimum clustering grid number of the single target to be 2, and further reducing false alarms. And constructing a two-dimensional rectangular bounding box with the minimum area to represent the position and the scale of the water surface barrier according to the grid clustering result.
Further, in step 5, the correlation matching algorithm uses the first moment of the centroid of the grid map and the shape feature distance to perform Hungarian minimum distance cost matching method, and the method comprises the following substeps:
step 5.1, calculating the mass center and the length-width ratio of each clustering object of the current frame, and performing correlation matching on obstacle information in a grid map constructed by front and rear frames of the laser radar point cloud by using a bipartite graph optimization algorithm according to appearance characteristics such as a mass center grid coordinate and a bounding box length-width ratio;
and 5.2, updating target batch number matching according to the matching result: the problem of unmatched targets is solved by one-to-many matching of the targets in the previous frame and the new targets. The processing method comprises the following steps: calculating the offset distance of the absolute coordinate of the centroid of the target according to the speed of the ship, judging whether the target is a moving target or not, taking an offset threshold value of 2.0m, and distributing a new target batch number to the moving target;
and 5.3, updating the grid map according to the matching result: if one grid has point clouds in a new frame, the confidence coefficient is reset to 1.0, if no point clouds exist, the confidence coefficient is attenuated along with time, in the embodiment, each period is decreased by 0.04, and finally, the confidence coefficient is decreased to 0. And calculating the relative movement speed and direction information of the target according to the geometrical distance of the front frame and the back frame and the measuring period of the laser radar.
The invention has the advantages and positive effects that:
1. the method gives full play to the characteristics of the laser radar, utilizes centimeter-level precision point cloud information to perform target segmentation, clustering and extraction, and can accurately acquire the position and size information of the target.
2. The method uses the laser radar which is slightly influenced by illumination conditions and has a high detection period (up to 10 Hz), and is more suitable for detecting the complex obstacles in the inland waterway through the application of the geo-electric fence.
Drawings
FIG. 1 is a flow chart of an obstacle extraction algorithm of the present invention;
FIG. 2 is a grid map representation of the present invention;
FIG. 3 is a flow chart of the object segmentation clustering of the present invention;
FIG. 4 is a flow chart of the point inter-frame target association of the present invention.
Detailed Description
The present invention will be described in more detail below with reference to the following embodiments, which are provided by way of illustration only and are not intended to limit the scope of the present invention.
The invention discloses a method for sensing obstacles in an unmanned ship based on a laser radar, which is based on an unmanned ship platform provided with a three-dimensional laser radar, and has the following invention points: the method comprises the following steps:
step 1, constructing a profile of a navigable area, which specifically comprises the following steps: and constructing a polygonal geographic electronic fence area of the task navigable water area on line in real time by using longitude and latitude coordinates of the navigable area of the inland river, which are acquired from known information such as a map, a chart and the like, and taking the position of the ship as a reference point. The geo-electric fence is a polygonal area formed by longitude and latitude of boundary points which are obtained according to different curvatures of a channel of an unmanned ship at two sides of the channel, and the number of the boundary points is in direct proportion to the curvatures. The boundary points can be obtained from known information such as maps and river diagrams. When a polygonal geo-electric fence area is constructed, the real-time position of the ship is taken as a reference point, longitude and latitude coordinates in formats such as WGS-84 of a navigable area are dynamically converted into horizontal plane polygonal coordinates by adopting the ink card tray projection transformation, and related division calculation is simplified.
And 2, receiving real-time pose information by using a differential GNSS (global navigation satellite system) and a shipborne IMU (inertial measurement unit), and compensating X, Y and Z installation angle errors obtained after the laser radar and the IMU are calibrated. And carrying out distortion correction on the 360-degree point cloud generated by the laser radar by utilizing IMU attitude information, and converting a point cloud spherical coordinate system into a geodetic Cartesian coordinate system. Finally, combining the processed point cloud information with the position of the ship, filtering according to the geo-fencing area in the step 1, and removing the point cloud outside the navigable area;
the distortion correction is performed by the following method: and performing bilinear interpolation on longitudinal and transverse rolling, speed and bow data in the acquired pose information, and acquiring interpolated pose information according to timestamps corresponding to 360-degree different sectors of the laser radar point cloud to realize distortion correction, thereby reducing distortion of the point cloud caused by ship swinging and motion.
The specific method for converting the point cloud spherical coordinates into the geodetic Cartesian coordinate system comprises the following steps: and converting the point cloud into the point cloud in the PointXYZI format under the standard Cartesian rectangular coordinate system containing the intensity information according to the spherical coordinate information of the laser radar point cloud and the course information of the ship.
The specific method for filtering the processed point cloud information comprises the following steps:
according to the real-time online constructed polygonal geo-electronic fence area in the step 1, converting the latitude and longitude geodetic coordinates of the outline polygon into plane rectangular coordinates with a laser radar as an origin according to the real-time latitude and longitude of the ship, and calculating whether each point of the laser radar point cloud falls into the polygon of the navigable area by constructing an outline mapping table and applying a ray method to remove target point clouds, such as a bank, surrounding buildings, trees and the like, outside the navigable area.
And 3, performing horizontal plane grid projection dimensionality reduction treatment on the point cloud corrected in the step 2 to form a two-dimensional projection grid, counting point cloud multi-dimensional feature information belonging to each grid, and calculating the target confidence coefficient of each grid according to the feature value.
As shown in fig. 2, when the point cloud two-dimensional projection is implemented, the main method is to reduce the dimension of the z-axis and project the point cloud to the square grid map in the x and y directions of the plane. The size of the projection grid is determined according to the minimum resolution precision required by the obstacle to be detected, and the side length of each grid is 0.5 time of the detection precision of the obstacle. In this embodiment, a square grid with a side length of 0.5m is used to perform grid division on a 200 × 200m area within the measurement range of the laser radar, so as to obtain an 800 × 800 grid map. And extracting features according to the point cloud data of the corresponding area of each grid, and storing the point cloud number, the maximum and minimum height difference, the minimum height, the average reflection intensity and the confidence coefficient of the target grid calculated based on the information in the corresponding grid. The original point cloud is subjected to feature extraction, and a two-dimensional index table is established, so that the point cloud is efficiently searched.
And 4, performing obstacle grid filtering on the grid map formed by the two-dimensional projection grid in the step 3, specifically including noise removal, target grid segmentation and clustering processing, extracting the position and scale information of each target, and acquiring a target two-dimensional rectangular surrounding frame.
The clustering process mainly adopts Euclidean distance clustering and comprises the following substeps:
step 4.1, dividing different thresholds in the ship and wake flow region, and extracting the obstacle by using the threshold H thre The following settings are set:
within the normal threshold region, H w The wave height during navigation is determined by sea conditions; in a high threshold value area, the ship speed v of the ship is set to be in direct proportion, and a specific proportional coefficient is related to ship characteristics;
the barrier grid filtering algorithm provided by the invention specifically comprises the following steps: traversing the grid map containing the point cloud multi-dimensional feature information extracted in the whole step 3, and removing noise influence generated by water surface waves by sliding a window; the judgment of the grid as the effective target simultaneously meets the following conditions:
N＞2
H min ＞1.5H vessel
(H max -H min )＞H thre
intensity＞10
where N is the number of point clouds falling within the grid coordinate interval, H max Is the maximum height of point cloud in the grid, H min Is the minimum height, H, of the point cloud in the grid vessel The height of the highest point of the ship from the water surface and intensity are average reflection intensity (in the example, the range is 0-255).
Step 4.2, traversing each target grid in the grid map, skipping if the point is processed, otherwise defining a seed grid queue and adding a seed grid, taking an eight-neighborhood grid of the grid as a clustering radius, and marking the target grid in the eight-neighborhood as a processed grid by using a rapid expansion marking method to cluster the target grids into the same target;
and 4.3, setting the minimum clustering grid number of the single target to be 2 to further reduce false alarms, and constructing a two-dimensional rectangular bounding box with a minimum area to represent the position and the scale of the water surface barrier according to the grid clustering result.
And 5, extracting the position coordinates and the length-width ratio information of the bounding box of the obstacle target according to the raster maps of the front frame and the rear frame of the laser radar, performing the correlation matching of the clustering targets of the front frame and the rear frame, and extracting the position and speed information of the target.
The method for matching Hungarian minimum distance cost by using the first moment of the centroid of the grid map and the shape feature distance through the association matching algorithm comprises the following substeps:
step 5.1, calculating the mass center and the length-width ratio of each clustering object of the current frame, and performing correlation matching on obstacle information in a grid map constructed by front and rear frames of the laser radar point cloud by using a bipartite graph optimization algorithm according to appearance characteristics such as a mass center grid coordinate and a bounding box length-width ratio;
and 5.2, updating target batch number matching according to the matching result. The problem of unmatched targets is solved by one-to-many matching of the targets in the previous frame and the new targets. The processing method comprises the following steps: calculating the offset distance of the absolute coordinate of the centroid of the target according to the speed of the ship, judging whether the target is a moving target or not, taking an offset threshold value of 2.0m, and distributing a new target batch number to the moving target;
and 5.3, updating the grid map according to the matching result. If a grid has a point cloud in a new frame, the confidence coefficient is reset to 1.0, if no point cloud exists, the confidence coefficient decays along with time, in the embodiment, each period is decreased by 0.04, and finally, the confidence coefficient is decreased to 0. And calculating the relative movement speed and direction information of the target according to the geometrical distance of the front frame and the back frame and the measuring period of the laser radar.
Although the embodiments of the present invention and the accompanying drawings are disclosed for illustrative purposes, those skilled in the art will appreciate that: various substitutions, changes and modifications are possible without departing from the spirit and scope of the invention and the appended claims, and therefore the scope of the invention is not limited to the disclosure of the embodiments and the accompanying drawings.
Claims (5)
1. An unmanned ship inland river obstacle sensing method based on a laser radar is characterized by comprising the following steps:
step 1, constructing a profile of a navigable area, which specifically comprises the following steps: constructing a polygonal geographic electronic fence area of the task navigable water area on line in real time by using longitude and latitude coordinates of the navigable area of the inland river, which are acquired from known information of a map and a chart, and taking the position of the ship as a reference point;
step 2, receiving real-time pose information by using a differential GNSS and a shipborne IMU, and compensating X, Y and Z installation angle errors obtained after the laser radar and the IMU are calibrated; carrying out distortion correction on 360-degree point cloud generated by the laser radar by utilizing IMU attitude information, and converting a point cloud spherical coordinate system into a geodetic Cartesian coordinate system; finally, combining the processed point cloud information with the position of the ship, filtering according to the geo-fencing area in the step 1, and removing the point cloud outside the navigable area;
step 3, performing horizontal plane grid projection dimensionality reduction on the point cloud corrected in the step 2 to form two-dimensional projection grids, counting point cloud multi-dimensional feature information belonging to each grid, and calculating the target confidence coefficient of each grid according to multi-dimensional feature values;
step 4, performing obstacle grid filtering on the grid map formed by the two-dimensional projection grid in the step 3, including noise removal, target grid segmentation and clustering processing, acquiring a target two-dimensional rectangular surrounding frame, and extracting the position and scale information of each target;
step 5, extracting position coordinates and aspect ratio information of an enclosure frame of the obstacle target according to the raster maps of front and rear frames of the laser radar, performing clustering target association matching of the front and rear frames, and extracting target position and speed information;
in step 4, the clustering process mainly adopts Euclidean distance clustering, and comprises the following substeps:
step 4.1, dividing different thresholds in the ship and wake flow region, and extracting the obstacle by using the threshold H thre The setting is as follows:
within the normal threshold region, H w The wave height during navigation is determined by the sea condition; in a high threshold value area, the ship speed v of the ship is set to be in direct proportion, and a specific proportional coefficient is related to ship characteristics;
the obstacle grid filtering algorithm specifically comprises the following steps: traversing the grid map containing the point cloud multi-dimensional feature information extracted in the whole step 3, and removing noise influence generated by water surface waves by sliding a window, wherein the judgment of the target grid simultaneously meets the following conditions:
N＞2
H min ＞1.5H vessel
(H max -H min )＞H thre
intensity＞10
where N is the number of point clouds falling within the grid coordinate interval, H max Is the maximum height of point cloud in the grid, H min Is the minimum height of point cloud in the grid, H vessel The height from the highest point of the ship to the water surface and intensity are the average reflection intensity;
step 4.2, traversing each target grid in the grid map, skipping if the target grid is processed, otherwise defining a seed grid queue and adding a seed grid, taking the eight-neighborhood grid of the grid as a clustering radius, and marking the target grid in the eight-neighborhood as a processed grid by using a rapid expansion marking method to cluster the target grid into the same target;
and 4.3, setting the minimum clustering grid number of the single target to be 2, further reducing false alarms, and constructing a two-dimensional rectangular bounding box with a minimum area to represent the position and the scale of the water surface barrier according to the grid clustering result.
2. The method for unmanned marine river obstacle sensing based on lidar according to claim 1, wherein: in step 2, the distortion correction adopts the following mode: and performing bilinear interpolation on the acquired pose information, and acquiring the interpolated pose information according to timestamps corresponding to 360-degree different sectors of the laser radar point cloud to realize distortion correction, thereby reducing distortion of the point cloud caused by ship swing and motion.
3. The method for unmanned marine river obstacle sensing based on lidar according to claim 1, wherein: the two-dimensional projection grid takes a laser radar as a center, is parallel to a horizontal plane, adopts a square grid, the grid division density is determined according to the minimum resolution required by the obstacle to be detected, and the side length of each grid is 0.5 times of the minimum resolution of the obstacle.
4. The method for unmanned marine river obstacle sensing based on lidar according to claim 1, wherein: the grid point cloud multi-dimensional feature information comprises the number of point clouds, the maximum and minimum height difference, the minimum height and the average reflection intensity.
5. The unmanned endoship obstacle sensing method based on lidar according to claim 1, wherein in step 5, the correlation matching algorithm uses the first moment of the centroid of the grid map and the shape feature distance to perform Hungarian minimum distance cost matching method, comprising the following sub-steps:
step 5.1, calculating the mass center and the length-width ratio of each clustering object of the current frame, and performing correlation matching on the obstacle information in the grid map constructed by the front frame and the rear frame of the laser radar point cloud by using a bipartite graph optimization algorithm according to the mass center grid coordinate and the outline characteristics of the length-width ratio of the bounding frame;
and 5.2, updating target batch number matching according to the matching result: aiming at the problem of unmatched targets generated by one-to-many matching of a previous frame target and a new target; the processing method comprises the following steps: calculating the offset distance of the absolute coordinate of the centroid of the target according to the speed of the ship, judging whether the target is a moving target or not, taking the offset threshold value as 2.0m, and distributing a new target batch number to the moving target;
and 5.3, updating the grid map according to the matching result: if one grid has point clouds in a new frame, the confidence coefficient is reset to 1.0, if no point clouds exist, the confidence coefficient is attenuated along with time, the confidence coefficient is decreased by 0.04 every period, and finally the confidence coefficient is decreased to 0; and calculating the relative movement speed and direction information of the target according to the geometrical distance of the front frame and the back frame and the measuring period of the laser radar.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110026459.9A CN112882059B (en) | 2021-01-08 | 2021-01-08 | Unmanned ship inland river obstacle sensing method based on laser radar |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110026459.9A CN112882059B (en) | 2021-01-08 | 2021-01-08 | Unmanned ship inland river obstacle sensing method based on laser radar |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112882059A CN112882059A (en) | 2021-06-01 |
| CN112882059B true CN112882059B (en) | 2023-01-17 |
Family
ID=76046162
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110026459.9A Active CN112882059B (en) | 2021-01-08 | 2021-01-08 | Unmanned ship inland river obstacle sensing method based on laser radar |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112882059B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113534171B (en) * | 2021-06-22 | 2023-04-11 | 贵州电网有限责任公司 | Induction radar electronic fence suitable for transformer substation and positioning and tracking method |
| CN113778099A (en) * | 2021-09-16 | 2021-12-10 | 浙江大学湖州研究院 | Unmanned ship path planning method based on NDT algorithm and Hybrid A algorithm |
| CN114241211B (en) * | 2021-11-26 | 2023-02-03 | 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 |
| CN114387585B (en) * | 2022-03-22 | 2022-07-05 | 新石器慧通（北京）科技有限公司 | Obstacle detection method, detection device, and travel device |
| CN114397654B (en) * | 2022-03-24 | 2022-06-24 | 陕西欧卡电子智能科技有限公司 | Unmanned ship obstacle avoidance method based on multi-radar sensing |
| CN114879685B (en) * | 2022-05-25 | 2023-04-28 | 合肥工业大学 | River shoreline detection and autonomous cruising method for unmanned ship |
| CN114820392B (en) * | 2022-06-28 | 2022-10-18 | 新石器慧通（北京）科技有限公司 | Laser radar detection moving target distortion compensation method, device and storage medium |
| CN115267827B (en) * | 2022-08-11 | 2024-07-09 | 中国船舶集团有限公司第七一六研究所 | Laser radar harbor area obstacle sensing method based on high density screening |
| CN115656982B (en) * | 2022-12-13 | 2023-06-09 | 中国南方电网有限责任公司超高压输电公司广州局 | Water surface clutter removal method, device, computer equipment and storage medium |
| CN117830676B (en) * | 2024-03-06 | 2024-06-04 | 国网湖北省电力有限公司 | Unmanned aerial vehicle-based power transmission line construction risk identification method and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105374224A (en) * | 2015-10-29 | 2016-03-02 | 深圳市元征科技股份有限公司 | Positioning data processing method and vehicle-mounted terminal |
| CN110246159A (en) * | 2019-06-14 | 2019-09-17 | 湖南大学 | The 3D target motion analysis method of view-based access control model and radar information fusion |
| CN111239717A (en) * | 2020-01-22 | 2020-06-05 | 南京甄视智能科技有限公司 | Water surface obstacle detection method based on X-band radar |
| CN111381248A (en) * | 2020-03-23 | 2020-07-07 | 湖南大学 | Obstacle detection method and system considering vehicle bump |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN110865394A (en) * | 2019-09-24 | 2020-03-06 | 中国船舶重工集团公司第七0七研究所 | Target classification system based on laser radar data and data processing method thereof |
| CN110850403B (en) * | 2019-11-18 | 2022-07-26 | 中国船舶重工集团公司第七0七研究所 | Multi-sensor decision-level fused intelligent ship water surface target feeling knowledge identification method |
-
2021
- 2021-01-08 CN CN202110026459.9A patent/CN112882059B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105374224A (en) * | 2015-10-29 | 2016-03-02 | 深圳市元征科技股份有限公司 | Positioning data processing method and vehicle-mounted terminal |
| CN110246159A (en) * | 2019-06-14 | 2019-09-17 | 湖南大学 | The 3D target motion analysis method of view-based access control model and radar information fusion |
| CN111239717A (en) * | 2020-01-22 | 2020-06-05 | 南京甄视智能科技有限公司 | Water surface obstacle detection method based on X-band radar |
| CN111381248A (en) * | 2020-03-23 | 2020-07-07 | 湖南大学 | Obstacle detection method and system considering vehicle bump |
Non-Patent Citations (4)
| Title |
|---|
| 一种融合密度聚类与区域生长算法的快速障碍物检测方法;李炯等;《机器人》(第01期);全文 * |
| 基于相对形状上下文与概率松弛标记法的点模式匹配算法;赵键等;《信号处理》;20110525(第05期);全文 * |
| 基于遥感图像的舰船目标检测与参数估计方法;董凯旋等;《电子科技》;20150215(第02期);全文 * |
| 机载激光雷达技术在北方河道整治中的应用;李兵等;《北京测绘》;20180325(第03期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112882059A (en) | 2021-06-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112882059B (en) | Unmanned ship inland river obstacle sensing method based on laser radar | |
| CN113340295B (en) | Unmanned ship near-shore real-time positioning and mapping method with multiple ranging sensors | |
| CN108562913B (en) | Unmanned ship false target detection method based on three-dimensional laser radar | |
| CN109239709B (en) | Autonomous construction method for local environment map of unmanned ship | |
| CN107025654B (en) | SAR image self-adaptive ship detection method based on global iterative inspection | |
| CN113177593B (en) | Fusion method of radar point cloud and image data in water traffic environment | |
| CN110132284B (en) | Global positioning method based on depth information | |
| CN112394726B (en) | Unmanned ship obstacle fusion detection method based on evidence theory | |
| CN107942329B (en) | Method for detecting sea surface ship target by maneuvering platform single-channel SAR | |
| CN115761550A (en) | Water surface target detection method based on laser radar point cloud and camera image fusion | |
| CN107862271B (en) | Detection method of ship target | |
| US20230188696A1 (en) | Systems And Methods For Generating And/Or Using 3-Dimensional Information With Camera Arrays | |
| Han et al. | GPS-less coastal navigation using marine radar for USV operation | |
| CN110596728A (en) | Water surface small target detection method based on laser radar | |
| CN114241211A (en) | Laser radar point cloud feature-based shoreline extraction method | |
| CN113487631A (en) | Adjustable large-angle detection sensing and control method based on LEGO-LOAM | |
| CN111089580B (en) | Unmanned war chariot simultaneous positioning and map construction method based on covariance intersection | |
| CN110927765B (en) | Laser radar and satellite navigation fused target online positioning method | |
| CN108985292A (en) | A kind of SAR image CFAR object detection method and system based on multi-scale division | |
| CN115267827B (en) | Laser radar harbor area obstacle sensing method based on high density screening | |
| CN116071694B (en) | Ship detection method, device and computer readable storage medium | |
| CN113052117B (en) | AIS data-assisted Rayleigh CFAR detection method for SAR image | |
| CN113888589A (en) | Water surface obstacle detection and multi-target tracking method based on laser radar | |
| Deng et al. | Obstacle detection of unmanned surface vehicle based on LiDAR point cloud data | |
| CN114089376A (en) | Single laser radar-based negative obstacle detection method |
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 |