CN110554405A - normal scanning registration method and system based on cluster combination - Google Patents
normal scanning registration method and system based on cluster combination Download PDFInfo
- Publication number
- CN110554405A CN110554405A CN201910794043.4A CN201910794043A CN110554405A CN 110554405 A CN110554405 A CN 110554405A CN 201910794043 A CN201910794043 A CN 201910794043A CN 110554405 A CN110554405 A CN 110554405A
- Authority
- CN
- China
- Prior art keywords
- classification
- laser radar
- reference frame
- module
- radar point
- 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
- G01S17/89—Lidar systems specially adapted for specific applications for mapping or imaging
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Probability & Statistics with Applications (AREA)
- Computer Networks & Wireless Communication (AREA)
- Electromagnetism (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Optical Radar Systems And Details Thereof (AREA)
Abstract
The invention discloses a normal scanning registration method based on combination clustering, which comprises the following steps: the method comprises the steps of acquiring two continuous laser radar point cloud data from a mobile robot in real time, using the former laser radar point cloud data as a reference frame and the latter laser radar point cloud data as a current frame, processing the reference frame by using a density-based spatial clustering algorithm to obtain an initial classification result of the reference frame, judging whether the ratio of the maximum two singular values of each classification in the initial classification result of the reference frame is larger than a set threshold value or not according to each classification, processing the classification by using a K-means algorithm to obtain K subdivision classifications, adding the K subdivision classifications into a final classification set of the reference frame, and aiming at each laser radar point in the current frame. The invention can solve the technical problem that the accuracy of the finally obtained environment map is poor due to the fact that the registration result has errors caused by the adoption of grid division in the conventional normal scanning registration method.
Description
Technical Field
the invention belongs to the technical field of mobile robots, and particularly relates to a normal scanning registration method and system based on combination clustering.
background
in an unknown environment, synchronous positioning and mapping are a foundation for realizing autonomy, intellectualization and more complex tasks of the mobile robot, wherein scanning registration is a prerequisite for realizing two functions of synchronous positioning and mapping.
the most common scanning registration method at present is a Normal-distribution registration (NDT) method, which obtains a series of point cloud data in an environment through a sensing sensor (such as a laser radar and a depth camera), optimizes and solves an inter-frame coordinate transformation relation by establishing a registration relation between two point cloud data before and after, and finally obtains an environment map of a mobile robot operating area.
However, for the NDT method, in the process of constructing the registration relationship between the two point cloud data, the laser radar points are classified at equal intervals by using grid division, so that the local geometric characteristics of the point cloud data cannot be sufficiently expressed by the classification result of the reference frame, and therefore, the established registration result has errors, and the finally obtained environment map has poor accuracy.
disclosure of Invention
Aiming at the defects or improvement requirements of the prior art, the invention provides a normal scanning registration method and system based on combined clustering, and aims to solve the technical problem that the registration result has errors due to the adoption of grid division in the conventional normal scanning registration method, so that the finally obtained environment map has poor precision.
According to one aspect of the invention, a normal scan registration method based on cluster grouping is provided, which comprises the following steps:
(1) Acquiring two continuous laser radar point cloud data in real time from the mobile robot, wherein the former laser radar point cloud data is used as a reference frame, and the latter laser radar point cloud data is used as a current frame;
(2) processing the reference frame obtained in the step (1) by using a density-based spatial clustering algorithm to obtain an initial classification result of the reference frame;
(3) And (3) judging whether the ratio of the maximum two singular values of each classification in the initial classification result of the reference frame obtained in the step (2) is greater than a set threshold value, if so, entering the step (4), and otherwise, entering the step (5).
(4) Processing the classification by using a K-means algorithm to obtain K subdivision classifications, adding the K subdivision classifications into a final classification set of a reference frame, and then turning to the step (6);
(5) The classification is added to the final classification set of the reference frame.
(6) aiming at each laser radar point in the current frame, obtaining the distance between the laser radar point and each classification in the final classification set of the reference frame according to the coordinate of the laser radar point and the coordinate of each classification in the final classification set of the reference frame, and selecting the classification corresponding to the minimum distance value from the distances as the pairing classification of the laser radar point, wherein the classified coordinate is equal to the coordinate mean value of all the laser radar points included in the classification;
(7) aiming at each laser radar point in the current frame, constructing an error equation of the laser radar point and the pairing classification of the laser radar point;
(8) and summing all the laser radar points in the current frame with the error equations classified by matching the laser radar points with the laser radar points to obtain an overall error equation, and optimizing the overall error equation by adopting a Gaussian-Newton method to obtain the position variation delta x of the mobile robot between the reference frame and the current frame.
in the spatial clustering processing procedure in the step (2) and the K-means clustering processing procedure in the step (4), the distance alpha between two laser radar points is calculated by the following formula:
Where ρ i and a represent the polar coordinates of the ith lidar point, and Δ θ represents the angular increment of the ith lidar point relative to the (i-1) th lidar point, respectively.
The value of K in step (4) is the ratio of the largest two singular values of the classification in step (3).
The error equation of the classification of the laser radar point and the pairing thereof is as follows:
Where x represents the coordinates of the lidar point and μ and σ represent the normal distribution mean and variance, respectively, of the pairing class of the lidar point.
According to another aspect of the present invention, there is provided a combined cluster-based normal scan registration system, including:
The system comprises a first module, a second module and a third module, wherein the first module is used for acquiring two continuous laser radar point cloud data from the mobile robot in real time, the former laser radar point cloud data is used as a reference frame, and the latter laser radar point cloud data is used as a current frame;
the second module is used for processing the reference frame obtained by the first module by using a density-based spatial clustering algorithm to obtain an initial classification result of the reference frame;
A third module, configured to determine, for each classification in the initial classification result of the reference frame obtained by the second module, whether a ratio of the maximum two singular values of the classification result is greater than a set threshold, if so, enter the fourth module, otherwise, enter the fifth module;
the fourth module is used for processing the classification by utilizing a K-means algorithm to obtain K subdivision classifications, adding the K subdivision classifications into a final classification set of the reference frame, and then switching to the sixth module;
A fifth module for adding the classification to a final classification set of the reference frame;
a sixth module, configured to, for each lidar point in the current frame, obtain, according to the coordinate of the lidar point and the coordinate of each classification in the final classification set of the reference frame, a distance between the lidar point and each classification in the final classification set of the reference frame, and select, from the distances, a classification corresponding to a minimum distance value as a pairing classification of the lidar point, where the classified coordinate is equal to a coordinate mean value of all the lidar points included in the classification;
The seventh module is used for constructing an error equation of pairing and classification of each laser radar point in the current frame;
and the eighth module is used for summing all the laser radar points in the current frame with the error equations classified by matching the laser radar points with the error equations to obtain an overall error equation, and optimizing the overall error equation by adopting a Gaussian-Newton method to obtain the position variation delta x of the mobile robot between the reference frame and the current frame.
in general, compared with the prior art, the above technical solution contemplated by the present invention can achieve the following beneficial effects:
(1) According to the method, the steps (2) to (5) are adopted, and the density-based spatial clustering method and the K-means method can realize accurate classification of the laser radar points, so that the technical problem that the accuracy of the finally obtained environment map is poor due to the fact that the existing NDT cannot sufficiently express the local geometric characteristics of point cloud data and errors exist in the established registration result is solved.
(2) because the step (3) is adopted, the results after the density-based spatial clustering processing are screened by utilizing singular value decomposition, unreasonable classification results are secondarily subdivided by adopting a K-means algorithm, and the ratio of two maximum singular values is used as a parameter for K-means processing, so that the final classification result is more accurate.
(3) because the distance measurement method based on the trigonometric function is adopted in both the spatial clustering and the K-means clustering, the influence of the radiation characteristic (the distance between sampling points close to the laser radar is small, and the distance between sampling points far away from the laser radar is large) of the laser radar on the distance measurement can be reduced, and the normalized distance value can be obtained.
Drawings
FIG. 1 is a schematic diagram of a reference frame obtained in step (1) of the method of the present invention;
FIG. 2 is a schematic diagram of a current frame obtained in step (1) of the method of the present invention;
FIG. 3 is a schematic representation of the initial classification results obtained in step (2) of the method of the present invention;
FIGS. 4(a) and 4(b) are environment maps obtained by a conventional NDT method and the method of the present invention, respectively;
FIGS. 5(a), 5(b) and 5(c) are respectively the error between the registration result and the offset value obtained after the randomly selected lidar point cloud data is subjected to offset along the horizontal, vertical and rotational directions, respectively, and processed by the NDT method and the invention;
FIG. 6 is a flow chart of the normal scan registration method based on cluster grouping of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
As shown in fig. 6, the normal scan registration method based on cluster grouping of the present invention includes the following steps:
(1) Acquiring two continuous laser radar point cloud data in real time from a mobile robot, wherein the former laser radar point cloud data is used as a Reference frame (Reference frame), and the latter laser radar point cloud data is used as a current frame (Currentframe);
As shown in fig. 1, which shows an acquired reference frame; as shown in fig. 2, which shows the current frame acquired.
(2) processing the reference frame obtained in the step (1) by using a Density-based spatial clustering (Density-based spatial clustering) algorithm to obtain an initial classification result of the reference frame;
in the spatial clustering process, the distance alpha between two laser radar points is not an Euclidean distance, but is calculated by the following formula:
Where ρ i and a represent the polar coordinates of the ith lidar point, and Δ θ represents the angular increment of the ith lidar point relative to the (i-1) th lidar point, respectively.
In the laser radar point cloud data, because the obtained radar points are continuous long straight areas after discrete sampling is performed on boundary areas such as walls, and the like, in order to extract the continuous areas, spatial clustering based on density needs to be performed on all the radar points.
As shown in fig. 3, which shows the initial classification results obtained in this step, it can be seen that all the long straight regions have been aggregated into different classes, represented by different boxes in the figure.
(3) Judging whether the ratio of the maximum two singular values of each classification in the initial classification result of the reference frame obtained in the step (2) is larger than a set threshold value, if so, entering the step (4), otherwise, entering the step (5);
specifically, the threshold set in this step is a natural number of 3 or more.
(4) Processing the classification by using a K-means algorithm to obtain K subdivision classifications, adding the K subdivision classifications into a final classification set of a reference frame, and then turning to the step (6);
And (3) in the K-means clustering process, calculating the distance alpha between two laser radar points through the formula in the step (2).
Specifically, the value of K is the ratio of the maximum two singular values of the classification in step (3);
(5) adding the classification into a final classification set of the reference frame;
(6) aiming at each laser radar point in the current frame, obtaining the distance between the laser radar point and each classification in the final classification set of the reference frame according to the coordinate of the laser radar point and the coordinate of each classification in the final classification set of the reference frame, and selecting the classification corresponding to the minimum distance value from the distances as the pairing classification of the laser radar point, wherein the classified coordinate is equal to the coordinate mean value of all the laser radar points included in the classification;
(7) aiming at each laser radar point in the current frame, an error equation of the laser radar point and the pairing classification of the laser radar point is established, which specifically comprises the following steps:
Where x represents the coordinates of the lidar point and μ and σ represent the normal distribution mean and variance, respectively, of the pairing class of the lidar point.
(8) And summing all the laser radar points in the current frame with the error equations of the paired classification of the laser radar points to obtain an overall error equation, and optimizing the overall error equation by adopting a Gauss-Newton Method to obtain the position variation delta x of the mobile robot between the reference frame and the current frame.
performance comparison
the method of the present invention is compared with a Normal distribution transformation method (NDT) in the following.
fig. 4(a) and 4(b) are environment maps obtained after the NDT method and the present invention method are processed, respectively. As can be seen from the figure, the mobile robot can construct an environment map with clearer boundaries than an NDT method under the condition of only using the laser radar.
fig. 5(a), 5(b) and 5(c) are errors between the registration result and the offset value obtained after the randomly selected lidar point cloud data is respectively subjected to offset along the horizontal direction, the vertical direction and the rotating direction and processed by the NDT method and the invention, wherein a black line is the invention, and a gray line is the existing NDT method.
As can be seen from the three figures, the error of the invention is smaller.
table 1 below shows the results obtained after statistics of fig. 5(a), 5(b) and 5(c), where X corresponds to fig. 5(a), Y corresponds to fig. 5(b), θ corresponds to fig. 5(c), and CCNDT represents the method of the present invention.
TABLE 1
as can be seen from the above Table 1, the mean, median and variance of the method of the present invention are all smaller than those of the NDT method, which shows that the method of the present invention has better stability and stronger robustness.
It will be understood by those skilled in the art that the foregoing is only a preferred embodiment of the present invention, and is not intended to limit the invention, and that any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (5)
1. A normal scanning registration method based on combination clustering is characterized by comprising the following steps:
(1) acquiring two continuous laser radar point cloud data in real time from the mobile robot, wherein the former laser radar point cloud data is used as a reference frame, and the latter laser radar point cloud data is used as a current frame;
(2) Processing the reference frame obtained in the step (1) by using a density-based spatial clustering algorithm to obtain an initial classification result of the reference frame;
(3) and (3) judging whether the ratio of the maximum two singular values of each classification in the initial classification result of the reference frame obtained in the step (2) is greater than a set threshold value, if so, entering the step (4), and otherwise, entering the step (5).
(4) Processing the classification by using a K-means algorithm to obtain K subdivision classifications, adding the K subdivision classifications into a final classification set of a reference frame, and then turning to the step (6);
(5) The classification is added to the final classification set of the reference frame.
(6) Aiming at each laser radar point in the current frame, obtaining the distance between the laser radar point and each classification in the final classification set of the reference frame according to the coordinate of the laser radar point and the coordinate of each classification in the final classification set of the reference frame, and selecting the classification corresponding to the minimum distance value from the distances as the pairing classification of the laser radar point, wherein the classified coordinate is equal to the coordinate mean value of all the laser radar points included in the classification;
(7) aiming at each laser radar point in the current frame, constructing an error equation of the laser radar point and the pairing classification of the laser radar point;
(8) And summing all the laser radar points in the current frame with the error equations classified by matching the laser radar points with the laser radar points to obtain an overall error equation, and optimizing the overall error equation by adopting a Gaussian-Newton method to obtain the position variation delta x of the mobile robot between the reference frame and the current frame.
2. The normal scan registration method according to claim 1, wherein in the spatial clustering process of step (2) and the K-means clustering process of step (4), the distance α between two lidar points is calculated by the following formula:
where ρ i and a represent the polar coordinates of the ith lidar point, and Δ θ represents the angular increment of the ith lidar point relative to the (i-1) th lidar point, respectively.
3. the normal scan registration method according to claim 1 or 2, wherein the value of K in step (4) is a ratio of the largest two singular values of the classification in step (3).
4. The normal scan registration method of any of claims 1-3, wherein the error equation for the classification of lidar points and their pairs is:
where x represents the coordinates of the lidar point and μ and σ represent the normal distribution mean and variance, respectively, of the pairing class of the lidar point.
5. a normal scan registration system based on cluster grouping, comprising:
The system comprises a first module, a second module and a third module, wherein the first module is used for acquiring two continuous laser radar point cloud data from the mobile robot in real time, the former laser radar point cloud data is used as a reference frame, and the latter laser radar point cloud data is used as a current frame;
the second module is used for processing the reference frame obtained by the first module by using a density-based spatial clustering algorithm to obtain an initial classification result of the reference frame;
a third module, configured to determine, for each classification in the initial classification result of the reference frame obtained by the second module, whether a ratio of the maximum two singular values of the classification result is greater than a set threshold, if so, enter the fourth module, otherwise, enter the fifth module;
The fourth module is used for processing the classification by utilizing a K-means algorithm to obtain K subdivision classifications, adding the K subdivision classifications into a final classification set of the reference frame, and then switching to the sixth module;
a fifth module for adding the classification to a final classification set of the reference frame;
a sixth module, configured to, for each lidar point in the current frame, obtain, according to the coordinate of the lidar point and the coordinate of each classification in the final classification set of the reference frame, a distance between the lidar point and each classification in the final classification set of the reference frame, and select, from the distances, a classification corresponding to a minimum distance value as a pairing classification of the lidar point, where the classified coordinate is equal to a coordinate mean value of all the lidar points included in the classification;
the seventh module is used for constructing an error equation of pairing and classification of each laser radar point in the current frame;
and the eighth module is used for summing all the laser radar points in the current frame with the error equations classified by matching the laser radar points with the error equations to obtain an overall error equation, and optimizing the overall error equation by adopting a Gaussian-Newton method to obtain the position variation delta x of the mobile robot between the reference frame and the current frame.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910794043.4A CN110554405B (en) | 2019-08-27 | 2019-08-27 | Normal scanning registration method and system based on cluster combination |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910794043.4A CN110554405B (en) | 2019-08-27 | 2019-08-27 | Normal scanning registration method and system based on cluster combination |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110554405A true CN110554405A (en) | 2019-12-10 |
| CN110554405B CN110554405B (en) | 2021-07-30 |
Family
ID=68738382
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910794043.4A Active CN110554405B (en) | 2019-08-27 | 2019-08-27 | Normal scanning registration method and system based on cluster combination |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110554405B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113253246A (en) * | 2021-06-01 | 2021-08-13 | 奥特酷智能科技(南京)有限公司 | Calibration method for laser radar and camera |
| CN115019167A (en) * | 2022-05-26 | 2022-09-06 | 中国电信股份有限公司 | Fusion positioning method, system, equipment and storage medium based on mobile terminal |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105856230A (en) * | 2016-05-06 | 2016-08-17 | 简燕梅 | ORB key frame closed-loop detection SLAM method capable of improving consistency of position and pose of robot |
| WO2017147552A1 (en) * | 2016-02-26 | 2017-08-31 | Daniela Brunner | Multi-format, multi-domain and multi-algorithm metalearner system and method for monitoring human health, and deriving health status and trajectory |
| CN107291093A (en) * | 2017-07-04 | 2017-10-24 | 西北工业大学 | Unmanned plane Autonomous landing regional selection method under view-based access control model SLAM complex environment |
| CN108460779A (en) * | 2018-02-12 | 2018-08-28 | 浙江大学 | A kind of mobile robot image vision localization method under dynamic environment |
| CN108632625A (en) * | 2017-03-21 | 2018-10-09 | 华为技术有限公司 | A kind of method for video coding, video encoding/decoding method and relevant device |
| CN109308737A (en) * | 2018-07-11 | 2019-02-05 | 重庆邮电大学 | A kind of mobile robot V-SLAM method of three stage point cloud registration methods |
-
2019
- 2019-08-27 CN CN201910794043.4A patent/CN110554405B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017147552A1 (en) * | 2016-02-26 | 2017-08-31 | Daniela Brunner | Multi-format, multi-domain and multi-algorithm metalearner system and method for monitoring human health, and deriving health status and trajectory |
| CN105856230A (en) * | 2016-05-06 | 2016-08-17 | 简燕梅 | ORB key frame closed-loop detection SLAM method capable of improving consistency of position and pose of robot |
| CN108632625A (en) * | 2017-03-21 | 2018-10-09 | 华为技术有限公司 | A kind of method for video coding, video encoding/decoding method and relevant device |
| CN107291093A (en) * | 2017-07-04 | 2017-10-24 | 西北工业大学 | Unmanned plane Autonomous landing regional selection method under view-based access control model SLAM complex environment |
| CN108460779A (en) * | 2018-02-12 | 2018-08-28 | 浙江大学 | A kind of mobile robot image vision localization method under dynamic environment |
| CN109308737A (en) * | 2018-07-11 | 2019-02-05 | 重庆邮电大学 | A kind of mobile robot V-SLAM method of three stage point cloud registration methods |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113253246A (en) * | 2021-06-01 | 2021-08-13 | 奥特酷智能科技(南京)有限公司 | Calibration method for laser radar and camera |
| CN113253246B (en) * | 2021-06-01 | 2021-09-10 | 奥特酷智能科技(南京)有限公司 | Calibration method for laser radar and camera |
| CN115019167A (en) * | 2022-05-26 | 2022-09-06 | 中国电信股份有限公司 | Fusion positioning method, system, equipment and storage medium based on mobile terminal |
| CN115019167B (en) * | 2022-05-26 | 2023-11-07 | 中国电信股份有限公司 | Fusion positioning method, system, equipment and storage medium based on mobile terminal |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110554405B (en) | 2021-07-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kang et al. | Automatic targetless camera–lidar calibration by aligning edge with gaussian mixture model | |
| KR102204818B1 (en) | Selection of balanced-probe sites for 3-d alignment algorithms | |
| CN104040590B (en) | Method for estimating pose of object | |
| CN111060888A (en) | Mobile robot repositioning method fusing ICP and likelihood domain model | |
| CN111781608B (en) | Moving target detection method and system based on FMCW laser radar | |
| CN110533726B (en) | Laser radar scene three-dimensional attitude point normal vector estimation correction method | |
| CN103727930A (en) | Edge-matching-based relative pose calibration method of laser range finder and camera | |
| JP2014523572A (en) | Generating map data | |
| CN110554405B (en) | Normal scanning registration method and system based on cluster combination | |
| CN110908374B (en) | Mountain orchard obstacle avoidance system and method based on ROS platform | |
| CN112381862B (en) | Full-automatic registration method and device for CAD (computer-aided design) model and triangular mesh | |
| CN110853081A (en) | Ground and airborne LiDAR point cloud registration method based on single-tree segmentation | |
| CN114155265A (en) | Three-dimensional laser radar road point cloud segmentation method based on YOLACT | |
| CN112669458A (en) | Method, device and program carrier for ground filtering based on laser point cloud | |
| CN111913169A (en) | Method, equipment and storage medium for correcting laser radar internal reference and point cloud data | |
| CN113721254B (en) | Vehicle positioning method based on road fingerprint space association matrix | |
| CN107765257A (en) | A kind of laser acquisition and measuring method based on the calibration of reflected intensity accessory external | |
| CN113658203A (en) | Method and device for extracting three-dimensional outline of building and training neural network | |
| CN115953604B (en) | Real estate geographic information mapping data acquisition method | |
| Sun et al. | Automatic targetless calibration for LiDAR and camera based on instance segmentation | |
| CN117029870A (en) | Laser odometer based on road surface point cloud | |
| CN115267724B (en) | Position re-identification method of mobile robot capable of estimating pose based on laser radar | |
| JP3251840B2 (en) | Image recognition device | |
| CN116579949A (en) | Airborne point cloud ground point filtering method suitable for urban multi-noise environment | |
| CN116309817A (en) | Tray detection and positioning method based on RGB-D camera |
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 |