CN109657698A - A kind of magnetic-levitation obstacle detection method based on cloud - Google Patents
A kind of magnetic-levitation obstacle detection method based on cloud Download PDFInfo
- Publication number
- CN109657698A CN109657698A CN201811386577.5A CN201811386577A CN109657698A CN 109657698 A CN109657698 A CN 109657698A CN 201811386577 A CN201811386577 A CN 201811386577A CN 109657698 A CN109657698 A CN 109657698A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- magnetic
- levitation
- point
- cloud
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
Abstract
The present invention relates to a kind of magnetic-levitation obstacle detection methods based on cloud, comprising the following steps: 1) builds the vehicle-mounted motion scan system of magnetic-levitation;2) magnetic suspension rail level reference points of origin cloud and point cloud data to be detected are obtained by motion scan;3) K-D tree index is established, carries out characteristic matching acquired disturbance object point cloud with point cloud data to be detected;4) after carrying out European clustering processing to obstacle object point cloud, acquired disturbance object point cloud cluster set;5) each independent barrier profile and border and position are finally exported, the automatic detection to magnetic-levitation face barrier is realized.Compared with prior art, the present invention has many advantages, such as stronger environment sensing ability, and Informational support is played the role of safely in the operation maintenance that can be floated to magnetic.
Description
Technical field
The present invention relates to a kind of magnetic-levitation obstacle detection methods, more particularly, to a kind of magnetic suspension based on cloud
Track obstacle detection method.
Background technique
Magnetic suspension traffic makes train suspend in orbit " nap of the earth flight " by electromagnetic force, fast, the low in energy consumption, nothing with speed
Friction, noise is small, the service life is long etc., and advantages are paid close attention to by more and more countries as a kind of novel tip technology of track traffic.
With the raising required magnetic suspension desin speed, detection of obstacles is carried out to magnetic-levitation face, is excluded in time, ability
It avoids in energization, high-tension current causes barrier to be powered, to cause centainly to damage to orbital plane etc.;Just it is avoided that column
Vehicle bumps against caused risk with barrier in the process of moving, it is ensured that traffic route safety.
It is less for the research of magnetic-levitation face detection of obstacles at present, it is concentrated mainly on the gauge to track, smooth
Property etc., and mainly check magnetic suspension travel route that there is large labor intensity, low efficiency etc. by the way of artificial
Disadvantage.The method that reference pair railroad track, highway communication etc. carry out detection of obstacles has: the detection of obstacles based on stereoscopic vision
Method, the detection of obstacles based on multiple-sensor integration, is based on optical flow method obstacle quality testing at the detection of obstacles based on laser radar
Survey etc..(1) detection of obstacles based on binocular stereo vision is the mainstream detection mode to barrier at present, passes through binocular camera
After obtaining barrier left images, obstacle target region is obtained using the methods of Threshold segmentation, progress characteristic point detection later,
Extraction, Stereo matching etc., acquired disturbance object profile and position, depth information have many advantages, such as strong flexibility, at low cost, but its
By weather, such as illumination, sleet complex environment are affected;(2) detection method based on laser radar passes through Laser emission
Device goes out Laser emission, then receives laser and encounters the reflected light after body surface, is emitted between reception by measurement
Time difference carries out ranging, the much informations such as the emission characteristics, distance and speed of target can be obtained, finally by barrier and road surface
The detection barrier such as segmentation, cluster, has the characteristics such as principle simple, precision is high, strong interference immunity;(3) melted based on multisensor
The detection method of conjunction, for example obtain barrier image first with visual sensor, by barrier edge and feature extraction etc.,
After acquired disturbance object Position Approximate, using laser sensor, infrared sensor or ultrasonic sensor etc., the letter measured using it
It number is combined with visual sensor and environment is described, small, strong robustness is influenced by environmental change, the information of perception is reliable, quasi-
Exactness is high, overcomes the disadvantage of single-sensor information deficiency, but different sensors to be perfectly combined algorithm difficulty big, and cost
It is high;(4) it is to obtain ambient condition information in different moments using single camera based on optical flow method, light is estimated by sequence image
The characteristics of stream using barrier in light stream is a mutation is split optical flow field, to realize detection of obstacles, but light stream
Method is computationally intensive, processing time height etc..
Summary of the invention
It is an object of the present invention to overcome the above-mentioned drawbacks of the prior art and provide a kind of magnetic based on cloud
Suspension railway obstacle detection method.
The purpose of the present invention can be achieved through the following technical solutions:
A kind of magnetic-levitation obstacle detection method based on cloud, comprising the following steps:
1) the vehicle-mounted motion scan system of magnetic-levitation is built;
2) magnetic suspension rail level reference points of origin cloud and point cloud data to be detected are obtained by motion scan;
3) K-D tree index is established, carries out characteristic matching acquired disturbance object point cloud with point cloud data to be detected;
4) after carrying out European clustering processing to obstacle object point cloud, acquired disturbance object point cloud cluster set;
5) finally export each independent barrier profile and border and position, realize to magnetic-levitation face barrier from
Dynamicization detection.
Preferably, the vehicle-mounted motion scan system of the magnetic-levitation includes Z+F laser profile scanning instrument, two sets of SICK
Profile scanning instrument, Inertial Measurement Unit IMU and matched GNSS receiver antenna, GPS clock, 360 ° of panorama cameras;
Wherein the Z+F laser profile scanning instrument is used to obtain the point cloud data above magnetic levitation track, and described two sets
SICK profile scanning instrument is used to obtain the point cloud data of magnetic levitation track two sides;The Inertial Measurement Unit IMU and matched
GNSS receiver antenna is for providing Position and orientation parameters;The GPS clock is used for time synchronizing;Described 360 °
Panorama camera is used to obtain the visualization display of magnetic-levitation image.
Preferably, the step 3) establishes K-D tree index specifically:
Recycle each dimension (X, Y), (X, Z) and (Y, Z) at strong point of sequentially fetching, distribution feelings of the correlation data point in each dimension
Condition is distributed more disperses if the variance of a certain latitude coordinates value the big, and the smaller distribution of variance is more concentrated;The big dimension of variance is made
For cutting dimension, fetch strong point the dimension intermediate value as bisecting plane, coordinate of the plane perpendicular to current partition dimension
Axis, and the data point on the left of intermediate value is hung over into its left subtree, the data point on the right side of intermediate value is hung over into its right subtree;Recursion process its
Subtree establishes original reference point cloud data until all data point carries finish and completes K-D tree index.
Preferably, characteristic matching acquired disturbance object point cloud tool is carried out with point cloud data to be detected in the step 3)
Body are as follows:
Measuring point cloud to be checked gives query point and inquires the threshold value of distance, is by the threshold value in the center of circle, inquiry distance of query point
Radius determines ball domain;By binary search, the sequence or backtracking range searching in path are carried out to left and right subspace, from original number
All data for being less than threshold value with query point distance are found out according to concentrating;Judge whether there is point cloud data to fall into ball in original point cloud
In, if not having, illustrate query point in set distance threshold value not in reference points of origin cloud spatial dimension, as obstacle object point
Cloud;Index is finally established, obstacle object point all in measuring point cloud to be checked is obtained and converges.
It is preferably, described that European clustering processing is carried out to obstacle object point cloud specifically:
(1) K-D tree index is established to obstacle object point cloud;
(2) since obstacle object point cloud any point p, point set cluster is added in p;Centered on p, D is radius,
It is scanned in the K-D tree index of foundation, cluster is added in the point in distance range D;
(3) step (2) are successively repeated as center p using the point being newly added in cluster, until not new point is added to
In cluster, and by setting index tab, classified cloud is filtered, search efficiency is increased with this;
(4) start a new cluster, (1)~(3) step is repeated from point remaining in obstacle object point cloud, until all
Point was all clustered.
Preferably, the automatic detection to magnetic-levitation face barrier specifically:
After completing obstacle object point cloud cluster, filtering points are less than the cluster of given threshold, calculate each independent barrier
Maximum magnitude determines its position, so that the detection of barrier is completed, and the image obtained from panorama camera is checked,
For verifying the correctness of testing result and confirming to the entity type of barrier.
Compared with prior art, the present invention attempts vehicle-carried mobile laser scanner technique being applied to magnetic-levitation barrier
Detection, overcomes heavy workload, the poor efficiency of artificial detection, overcomes the restricted of visual sensor etc., not only can be under
It is detected in the adverse circumstances such as rainy day gas, night, and can efficiently obtain the environmental information of magnetic levitation track, it is floating to magnetic to carry out
Comprehensive monitoring.By that can check the operation of track to the detection data and original reference date comprision obtained every time
Situation takes full advantage of the structural information of data such as in detection of obstacles, accelerates characteristic matching by establishing indexed search
Efficiency finally utilizes neighbour domain clustering processing etc., size, shape even depth information and the location information of available barrier
Deng, have stronger environment sensing ability, can to magnetic float operation maintenance play the role of Informational support safely.
Detailed description of the invention
Fig. 1 is detection of obstacles flow chart of the present invention;
Fig. 2 is reference points of origin cloud schematic diagram;
Fig. 3 is measuring point cloud schematic diagram to be checked;
Fig. 4 is three-dimensional space K-D building and space division schematic diagram;
Fig. 5 is the European cluster schematic diagram of obstacle object point cloud;
Fig. 6 is the detection of obstacles result schematic diagram in measuring point cloud to be checked;
Fig. 7 is each independent barrier schematic diagram.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiment is a part of the embodiments of the present invention, rather than whole embodiments.Based on this hair
Embodiment in bright, those of ordinary skill in the art's every other reality obtained without making creative work
Example is applied, all should belong to the scope of protection of the invention.
The present invention devises the Vehicle-borne Laser Scanning system of a set of multiple-sensor integration for magnetic-levitation, and laser is swept
It retouches instrument and passes through record range-to-go, angle, strength information etc., it, can be with motion scan by being handled with multiple-sensor integration
Three-dimensional laser point cloud data of the magnetic-levitation face under local coordinate system is obtained, so as to quickly rebuild magnetic-levitation mould
Type.After obtaining magnetic suspension rail level reference points of origin cloud and point cloud data to be detected by motion scan, by the original ginseng of acquisition
Examination point cloud, establishes K-D tree index, characteristic matching acquired disturbance object point cloud is carried out with point cloud data to be detected, to obstacle object point
After cloud carries out the processing such as European cluster, so that acquired disturbance object point cloud cluster set, finally exports each independent barrier profile side
Boundary and position, realize the automatic detection to magnetic-levitation face barrier, and invention process flow is as shown in Figure 1.
It is based below main involved in Vehicle-borne Laser Scanning system progress magnetic-levitation face obstacle detection method
Algorithm and key link are introduced.
1. the design of the vehicle-mounted laser scanning system of magnetic-levitation
To carry out laser point cloud data acquisition to magnetic-levitation, it is disconnected to be integrated with Z+F laser profile scanning instrument, two sets of SICK
Surface scan instrument, Inertial Measurement Unit (Inertial Measurement Unit, IMU) and matched GNSS receiver antenna,
GPS clock, router, 360 ° of panorama cameras etc..Point cloud data, the IMU above magnetic levitation track with two sides are obtained by 3 scanners
It is combined with GNSS antenna and provides Position and orientation parameters, GPS clock for time synchronizing, panorama camera for obtaining magnetcisuspension
The visualization display etc. of floating orbit imagery.By integrated, the calibration of multisensor, time synchronization, space are carried out to multi-source data
Synchronous, point cloud resolves, after filtering processing, three-dimensional continuity point cloud of the available magnetic-levitation under local coordinate system.Here
The track face data mainly obtained using the Z+F scanner at vehicle-mounted top.
2. reference points of origin cloud establishes K-D tree index
To carry out detection of obstacles to magnetic-levitation face, it is necessary first to be defined to barrier, i.e., relatively primitive rail
The extra object of road environment.Therefore, this programme obtains two sets of point cloud datas altogether, and first set is the magnetic-levitation of clear
Reference points of origin cloud (as shown in Figure 2), second set is point cloud data to be detected (as shown in Figure 3).For treat test point cloud with
Reference points of origin cloud carries out characteristic matching, can use data set itself and shows the structural informations such as the cluster shape of tufted, first
Data directory is established, then carries out Rapid matching again, search efficiency can be greatly speeded up.
K-D tree (k-dimensional tree, K-D) is a kind of data structure for dividing k dimension data space, main to apply
In the search of hyperspace critical data.For the three-dimensional point cloud number of the original reference of the magnetic levitation track face clear of acquisition
According to constructing K-D tree to it first: circulation is sequentially fetched each dimension (X, Y) at strong point, (X, Z), (Y, Z), and correlation data point is each
The distribution situation of dimension is distributed more disperses if the variance of a certain latitude coordinates value the big, and the smaller distribution of variance is more concentrated;It will be square
The big dimension of difference is used as cutting dimension, and access strong point is used as bisecting plane in the intermediate value of the dimension, and the plane is perpendicular to currently stroke
The reference axis of fractional dimension, and the data point on the left of intermediate value is hung over into its left subtree, the data point on the right side of intermediate value is hung over into its right son
Tree;Its subtree of Recursion process establishes original reference point cloud data until all data point carries finish and completes K-D tree rope
Draw, as shown in Figure 4.
3. measuring point cloud to be checked carries out characteristic matching
It after reference points of origin cloud establishes index, treats test point cloud and carries out range searching in K-D tree, with this by original point
Cloud and measuring point cloud to be checked carry out characteristic matching, thus acquired disturbance object point cloud data.Detailed process are as follows: measuring point cloud to be checked is given to be looked into
The threshold value for asking point and inquiry distance, is that radius determines ball domain using query point as the center of circle, distance threshold;By binary search, to a left side
Right subspace carries out the sequence or backtracking range searching in path, finds out and all is less than with query point distance from initial data concentration
The data of threshold value;Judge whether there is point cloud data to fall into ball in original point cloud, if not having, illustrates query point in certain distance
Not in reference points of origin cloud spatial dimension, as obstacle object point cloud in threshold value;Index is finally established, institute in measuring point cloud to be checked is obtained
Some obstacle object points converge.
4. the European cluster of obstacle object point cloud
After acquired disturbance object point cloud data, needs to carry out European clustering processing to cloud, could obtain each independent
Barrier.Point cloud cluster is exactly that will belong to same category of point according to certain criteria for classifying to gather for one kind, the barrier of acquisition
The point cloud overwhelming majority is distributed in settlement, apart there is certain distance, the European cluster of point cloud that the present invention uses between settlement
It is exactly according to the Euclidean distance between coordinate that different settlements are separated, if obstacle object point cloud to be split is P={ Pi
(Xi,Yi) | i=1,2,3 ... }, segmentation distance parameter is D, and the specific method is as follows:
(1) K-D tree index is established to obstacle object point cloud
(2) since obstacle object point cloud any point p, point set cluster is added in p;Centered on p, D is radius,
It is scanned in the K-D tree index of foundation, cluster is added in the point in distance range D, such as Fig. 5-a;
(3) (2) step such as Fig. 5-b is successively repeated as center p using the point being newly added in cluster, until not new point
It is added in cluster, and by setting index tab, filters classified cloud, search efficiency is increased with this, such as Fig. 5-c;
(4) start a new cluster, (1)~(3) step is repeated from point remaining in obstacle object point cloud, until all
Point was all clustered, such as Fig. 5-d.
5. detection of obstacles
After completing obstacle object point cloud cluster, the very few cluster of filtering points is reduced because of instrument scanning, characteristic matching noise etc.
Bring detection error calculates the maximum magnitude of each independent barrier, its position is determined, to complete the inspection of barrier
It surveys, the image obtained from panorama camera is checked, the correctness of testing result and the entity type to barrier can be verified
Confirmed.It is illustrated in figure 6 the size and distribution of barrier in the relatively primitive reference point clouds of measuring point cloud to be checked, is hindered according to detection
The size for hindering object is classified as person (p) and two class of bird (b), and it is numbered;Fig. 7 is each independent barrier
Block.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can readily occur in various equivalent modifications or replace
It changes, these modifications or substitutions should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with right
It is required that protection scope subject to.
Claims (6)
1. a kind of magnetic-levitation obstacle detection method based on cloud, which comprises the following steps:
1) the vehicle-mounted motion scan system of magnetic-levitation is built;
2) magnetic suspension rail level reference points of origin cloud and point cloud data to be detected are obtained by motion scan;
3) K-D tree index is established, carries out characteristic matching acquired disturbance object point cloud with point cloud data to be detected;
4) after carrying out European clustering processing to obstacle object point cloud, acquired disturbance object point cloud cluster set;
5) each independent barrier profile and border and position are finally exported, the automation to magnetic-levitation face barrier is realized
Detection.
2. a kind of magnetic-levitation obstacle detection method based on cloud according to claim 1, which is characterized in that institute
The vehicle-mounted motion scan system of the magnetic-levitation stated includes Z+F laser profile scanning instrument, two sets of SICK profile scanning instrument, inertia survey
Measure unit IMU and matched GNSS receiver antenna, GPS clock, 360 ° of panorama cameras;
Wherein the Z+F laser profile scanning instrument is used to obtain the point cloud data above magnetic levitation track, two sets of SICK
Profile scanning instrument is used to obtain the point cloud data of magnetic levitation track two sides;The Inertial Measurement Unit IMU and matched GNSS connect
Receipts machine antenna is for providing Position and orientation parameters;The GPS clock is used for time synchronizing;360 ° of panorama phases
Machine is used to obtain the visualization display of magnetic-levitation image.
3. a kind of magnetic-levitation obstacle detection method based on cloud according to claim 1, which is characterized in that institute
The step 3) stated establishes K-D tree index specifically:
Circulation sequentially fetch strong point each dimension (X, Y), (X, Z) and (Y, Z), correlation data point each dimension distribution situation,
It is distributed more disperses if the variance of a certain latitude coordinates value the big, the smaller distribution of variance is more concentrated;Using the big dimension of variance as
Cutting dimension, access strong point the dimension intermediate value as bisecting plane, the plane perpendicular to current partition dimension reference axis,
And the data point on the left of intermediate value is hung over into its left subtree, the data point on the right side of intermediate value is hung over into its right subtree;Its son of Recursion process
Tree establishes original reference point cloud data until all data point carries finish and completes K-D tree index.
4. a kind of magnetic-levitation obstacle detection method based on cloud according to claim 3, which is characterized in that institute
Characteristic matching acquired disturbance object point cloud is carried out with point cloud data to be detected in the step 3) stated specifically:
Measuring point cloud to be checked gives query point and inquires the threshold value of distance, is radius by the threshold value in the center of circle, inquiry distance of query point
Determine ball domain;By binary search, the sequence or backtracking range searching in path are carried out to left and right subspace, from raw data set
In find out the data that all distances with query point are less than threshold value;Judge whether there is point cloud data to fall into ball in original point cloud, if
No, then illustrate query point in set distance threshold value not in reference points of origin cloud spatial dimension, as obstacle object point cloud;Finally
Index is established, obstacle object point all in measuring point cloud to be checked is obtained and converges.
5. a kind of magnetic-levitation obstacle detection method based on cloud according to claim 1, which is characterized in that institute
That states carries out European clustering processing to obstacle object point cloud specifically:
(1) K-D tree index is established to obstacle object point cloud;
(2) since obstacle object point cloud any point p, point set cluster is added in p;Centered on p, D is radius, is being established
K-D tree index in scan for, by distance range D point be added cluster;
(3) step (2) are successively repeated as center p using the point being newly added in cluster, until not new point is added to
In cluster, and by setting index tab, classified cloud is filtered, search efficiency is increased with this;
(4) start a new cluster, from point remaining in obstacle object point cloud repeat (1)~(3) step, until all the points all
It was clustered.
6. a kind of magnetic-levitation obstacle detection method based on cloud according to claim 1, which is characterized in that institute
The automatic detection to magnetic-levitation face barrier stated specifically:
After completing obstacle object point cloud cluster, filtering points are less than the cluster of given threshold, calculate the maximum of each independent barrier
Range determines its position, to complete the detection of barrier, and the image obtained from panorama camera is checked, is used for
It verifies the correctness of testing result and the entity type of barrier is confirmed.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811386577.5A CN109657698B (en) | 2018-11-20 | 2018-11-20 | Magnetic suspension track obstacle detection method based on point cloud |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811386577.5A CN109657698B (en) | 2018-11-20 | 2018-11-20 | Magnetic suspension track obstacle detection method based on point cloud |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109657698A true CN109657698A (en) | 2019-04-19 |
CN109657698B CN109657698B (en) | 2021-09-03 |
Family
ID=66112102
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811386577.5A Active CN109657698B (en) | 2018-11-20 | 2018-11-20 | Magnetic suspension track obstacle detection method based on point cloud |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109657698B (en) |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110471085A (en) * | 2019-09-04 | 2019-11-19 | 深圳市镭神智能系统有限公司 | A kind of rail detection system |
CN110501719A (en) * | 2019-08-27 | 2019-11-26 | 王玉娇 | A kind of train obstacle detection method based on laser radar |
CN110889350A (en) * | 2019-11-18 | 2020-03-17 | 四川西南交大铁路发展股份有限公司 | Line obstacle monitoring and alarming system and method based on three-dimensional imaging |
CN110889871A (en) * | 2019-12-03 | 2020-03-17 | 广东利元亨智能装备股份有限公司 | Robot running method and device and robot |
CN111178413A (en) * | 2019-12-20 | 2020-05-19 | 浙江欣奕华智能科技有限公司 | 3D point cloud semantic segmentation method, device and system |
CN111289998A (en) * | 2020-02-05 | 2020-06-16 | 北京汽车集团有限公司 | Obstacle detection method, obstacle detection device, storage medium, and vehicle |
CN111583257A (en) * | 2020-05-28 | 2020-08-25 | 中国铁道科学研究院集团有限公司 | Railway clearance foreign matter intrusion detection method, device and system |
CN112581521A (en) * | 2020-12-22 | 2021-03-30 | 同济大学 | Method for extracting central line of magnetic suspension track |
CN112750498A (en) * | 2020-12-30 | 2021-05-04 | 同济大学 | Method for inhibiting HIV virus replication by targeting reverse transcription primer binding site |
CN112857254A (en) * | 2021-02-02 | 2021-05-28 | 北京大成国测科技有限公司 | Parameter measurement method and device based on unmanned aerial vehicle data and electronic equipment |
WO2021143778A1 (en) * | 2020-01-14 | 2021-07-22 | 长沙智能驾驶研究院有限公司 | Positioning method based on laser radar |
CN113313137A (en) * | 2021-04-01 | 2021-08-27 | 杭州兰亭视觉科技有限责任公司 | Gluing obstacle recognition and removal method |
CN113406105A (en) * | 2021-06-30 | 2021-09-17 | 湖南凌翔磁浮科技有限责任公司 | Obstacle detection method and system for magnetic levitation detection trolley |
CN114022760A (en) * | 2021-10-14 | 2022-02-08 | 湖南北斗微芯数据科技有限公司 | Railway tunnel barrier monitoring and early warning method, system, equipment and storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104931977A (en) * | 2015-06-11 | 2015-09-23 | 同济大学 | Obstacle identification method for smart vehicles |
CN106406338A (en) * | 2016-04-14 | 2017-02-15 | 中山大学 | Omnidirectional mobile robot autonomous navigation apparatus and method based on laser range finder |
CN108416257A (en) * | 2018-01-19 | 2018-08-17 | 北京交通大学 | Merge the underground railway track obstacle detection method of vision and laser radar data feature |
WO2018166747A1 (en) * | 2017-03-15 | 2018-09-20 | Jaguar Land Rover Limited | Improvements in vehicle control |
-
2018
- 2018-11-20 CN CN201811386577.5A patent/CN109657698B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104931977A (en) * | 2015-06-11 | 2015-09-23 | 同济大学 | Obstacle identification method for smart vehicles |
CN106406338A (en) * | 2016-04-14 | 2017-02-15 | 中山大学 | Omnidirectional mobile robot autonomous navigation apparatus and method based on laser range finder |
WO2018166747A1 (en) * | 2017-03-15 | 2018-09-20 | Jaguar Land Rover Limited | Improvements in vehicle control |
CN108416257A (en) * | 2018-01-19 | 2018-08-17 | 北京交通大学 | Merge the underground railway track obstacle detection method of vision and laser radar data feature |
Non-Patent Citations (2)
Title |
---|
FEI GAO 等: "《Online Quadrotor Trajectory Generation and Autonomous Navigation on Point Clouds》", 《2016 IEEE INTERNATIONAL SYMPOSIUM ON SAFETY,SECURITY,AND RESCUE ROBOTICS》 * |
王东敏 等: "《视觉与激光点云融合的深度图像获取方法》", 《军事交通学院学报》 * |
Cited By (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110501719A (en) * | 2019-08-27 | 2019-11-26 | 王玉娇 | A kind of train obstacle detection method based on laser radar |
CN110501719B (en) * | 2019-08-27 | 2024-02-09 | 湖南九域同创高分子新材料有限责任公司 | Laser radar-based train obstacle detection method |
CN110471085A (en) * | 2019-09-04 | 2019-11-19 | 深圳市镭神智能系统有限公司 | A kind of rail detection system |
CN110889350A (en) * | 2019-11-18 | 2020-03-17 | 四川西南交大铁路发展股份有限公司 | Line obstacle monitoring and alarming system and method based on three-dimensional imaging |
CN110889871A (en) * | 2019-12-03 | 2020-03-17 | 广东利元亨智能装备股份有限公司 | Robot running method and device and robot |
CN110889871B (en) * | 2019-12-03 | 2021-03-23 | 广东利元亨智能装备股份有限公司 | Robot running method and device and robot |
CN111178413A (en) * | 2019-12-20 | 2020-05-19 | 浙江欣奕华智能科技有限公司 | 3D point cloud semantic segmentation method, device and system |
WO2021143778A1 (en) * | 2020-01-14 | 2021-07-22 | 长沙智能驾驶研究院有限公司 | Positioning method based on laser radar |
CN111289998A (en) * | 2020-02-05 | 2020-06-16 | 北京汽车集团有限公司 | Obstacle detection method, obstacle detection device, storage medium, and vehicle |
CN111583257A (en) * | 2020-05-28 | 2020-08-25 | 中国铁道科学研究院集团有限公司 | Railway clearance foreign matter intrusion detection method, device and system |
CN112581521A (en) * | 2020-12-22 | 2021-03-30 | 同济大学 | Method for extracting central line of magnetic suspension track |
CN112581521B (en) * | 2020-12-22 | 2023-04-07 | 同济大学 | Method for extracting central line of magnetic suspension track |
CN112750498A (en) * | 2020-12-30 | 2021-05-04 | 同济大学 | Method for inhibiting HIV virus replication by targeting reverse transcription primer binding site |
CN112857254A (en) * | 2021-02-02 | 2021-05-28 | 北京大成国测科技有限公司 | Parameter measurement method and device based on unmanned aerial vehicle data and electronic equipment |
CN113313137A (en) * | 2021-04-01 | 2021-08-27 | 杭州兰亭视觉科技有限责任公司 | Gluing obstacle recognition and removal method |
CN113406105A (en) * | 2021-06-30 | 2021-09-17 | 湖南凌翔磁浮科技有限责任公司 | Obstacle detection method and system for magnetic levitation detection trolley |
CN114022760A (en) * | 2021-10-14 | 2022-02-08 | 湖南北斗微芯数据科技有限公司 | Railway tunnel barrier monitoring and early warning method, system, equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN109657698B (en) | 2021-09-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109657698A (en) | A kind of magnetic-levitation obstacle detection method based on cloud | |
CN111551958B (en) | Mining area unmanned high-precision map manufacturing method | |
US11573090B2 (en) | LIDAR and rem localization | |
CN106584451B (en) | automatic transformer substation composition robot and method based on visual navigation | |
CN104005325B (en) | Based on pavement crack checkout gear and the method for the degree of depth and gray level image | |
US20190265038A1 (en) | Vehicle localisation using the ground surface with an event camera | |
Manandhar et al. | Auto-extraction of urban features from vehicle-borne laser data | |
CN108460416A (en) | A kind of structured road feasible zone extracting method based on three-dimensional laser radar | |
CN108171131B (en) | Improved MeanShift-based method for extracting Lidar point cloud data road marking line | |
Schlichting et al. | Localization using automotive laser scanners and local pattern matching | |
US11566902B2 (en) | Localization of autonomous vehicles via ground image recognition | |
CN107463918A (en) | Lane line extracting method based on laser point cloud and image data fusion | |
Abuhadrous et al. | Digitizing and 3D modeling of urban environments and roads using vehicle-borne laser scanner system | |
CN108647646A (en) | The optimizing detection method and device of low obstructions based on low harness radar | |
WO2016146641A1 (en) | Guided geometry extraction for localization of a device | |
Zhang et al. | Background filtering and vehicle detection with roadside lidar based on point association | |
CN110308729A (en) | The AGV combined navigation locating method of view-based access control model and IMU or odometer | |
Józsa et al. | Towards 4D virtual city reconstruction from Lidar point cloud sequences | |
Ibrahim et al. | High definition LiDAR mapping of perth CBD | |
CN114897777A (en) | Full-automatic extraction method of laser point cloud of overhead line system supporting facility considering spatial relationship | |
Shi et al. | Towards automatic road mapping by fusing vehicle-borne multi-sensor data | |
WO2020118623A1 (en) | Method and system for generating an environment model for positioning | |
US11834065B2 (en) | System, method, and computer program product for detecting road marking points from LiDAR data | |
Dai et al. | Accurate Ranging Based on Transmission Line Channel Monitoring Image and Point Cloud Data Mapping | |
Wu et al. | Voxel-based marked neighborhood searching method for identifying street trees using vehicle-borne laser scanning data |
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 | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20210104 Address after: 200092 Siping Road 1239, Shanghai, Yangpu District Applicant after: TONGJI University Applicant after: CRRC QINGDAO SIFANG Co.,Ltd. Address before: 200092 Siping Road 1239, Shanghai, Yangpu District Applicant before: TONGJI University |
|
TA01 | Transfer of patent application right | ||
GR01 | Patent grant | ||
GR01 | Patent grant |