CN110780305A - Track cone bucket detection and target point tracking method based on multi-line laser radar - Google Patents
Track cone bucket detection and target point tracking method based on multi-line laser radar Download PDFInfo
- Publication number
- CN110780305A CN110780305A CN201910995353.2A CN201910995353A CN110780305A CN 110780305 A CN110780305 A CN 110780305A CN 201910995353 A CN201910995353 A CN 201910995353A CN 110780305 A CN110780305 A CN 110780305A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- laser radar
- cone
- point
- track
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 40
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000001914 filtration Methods 0.000 claims abstract description 23
- 238000012545 processing Methods 0.000 claims abstract description 12
- 238000012216 screening Methods 0.000 claims abstract description 10
- 238000007619 statistical method Methods 0.000 claims abstract description 8
- 238000004422 calculation algorithm Methods 0.000 claims description 25
- 238000009434 installation Methods 0.000 claims description 9
- 230000001133 acceleration Effects 0.000 claims description 5
- 238000005070 sampling Methods 0.000 claims description 5
- 230000005484 gravity Effects 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 abstract description 2
- 230000008447 perception Effects 0.000 description 7
- 238000005286 illumination Methods 0.000 description 6
- 238000011160 research Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 230000000007 visual effect Effects 0.000 description 4
- 230000004438 eyesight Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 239000010426 asphalt Substances 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000004870 electrical engineering Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 238000010408 sweeping Methods 0.000 description 1
- 230000016776 visual perception Effects 0.000 description 1
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
- G01S17/02—Systems using the reflection of electromagnetic waves other than radio waves
-
- 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
- G01S17/66—Tracking systems using electromagnetic waves other than radio waves
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0257—Control of position or course in two dimensions specially adapted to land vehicles using a radar
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Electromagnetism (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- General Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Aviation & Aerospace Engineering (AREA)
- Automation & Control Theory (AREA)
- Optical Radar Systems And Details Thereof (AREA)
Abstract
The invention discloses a track cone bucket detection and target point tracking method based on a multi-line laser radar, which comprises the following steps: 1) reading laser radar point cloud data; 2) performing through filtering on the laser radar point cloud data; 3) eliminating the interference of ground point cloud data on the detection of the cone barrel; 4) screening out point cloud clusters of the cone barrel; 5) carrying out statistical analysis on the point cloud clusters obtained by clustering, setting a maximum standard deviation threshold value according to the characteristics of the actual size of the cone barrel, and screening out the cone barrel; 6) acquiring coordinates of a point cloud cluster center point; 7) calculating the coordinates of the center points of the cone barrels at the left side and the right side of the laser radar to obtain the center point of the cone barrel as the nearest target point in the current state; 8) and circulating the steps to obtain the latest target point. According to the invention, through real-time filtering, segmentation, clustering and other processing of the laser radar point cloud, the vehicles are continuously controlled to move towards the target points, and finally the track cone bucket detection and target point tracking based on the multi-line laser radar are realized.
Description
Technical Field
The invention relates to the field of environment perception of unmanned formula racing cars, in particular to a track cone bucket detection and target point tracking method based on a multi-line laser radar.
Background
The formula of unmanned Chinese college students (English test: FSAC) is a race designed and manufactured by unmanned racing cars participated in by college motor projects or motor related specialties in school student teams. This event is known as "cradle by automotive engineers". In this event, it is common for each unmanned racing team to employ multiline lidar as an important sensor for the unmanned environmental sensing system.
In the event, the unmanned racing cars of the racing vehicle group need to complete dynamic racing items such as a linear acceleration item, an 8-shaped circling item and a high-speed tracking item. Different racetracks are marked by conical barrels with fixed sizes (20X 30 cm) according to different racetrack shapes. According to the requirements of the race rules, before the racing car carries out the dynamic race, survey and map building are not allowed to be carried out on the race track, namely, the unmanned racing car cannot acquire the race track map to be completed in advance. Therefore, the cone bucket is an important mark for the unmanned system to effectively identify the track, and the vehicle-mounted sensor is required to be fully utilized to detect the track boundary and the travelable area in real time. The track cone bucket detection and target point tracking method based on the multiline laser radar is mainly applied to cone bucket identification of the event, and can be widely applied to other similar scenes, such as parking lot cone bucket detection in an autonomous parking environment and the like.
The unmanned perception sensor comprises a camera, a laser radar, a GPS inertial navigation sensor and other sensors, and the invention mainly aims at the explanation of a method for cone barrel detection and target tracking of the laser radar sensor. The laser radars are classified into MEMS type laser radars, Flash type laser radars, phased array laser radars, and mechanical rotation type laser radars according to the scanning manner. Different types of laser radars have different scanning modes, and are different in manufacturing cost, laser data processing mode and application scene. The laser radars are classified according to the number of lines, and can be further classified into single-line machine radars and multi-line laser radars, and the single-line laser radars can only scan in a plane mode and are mainly applied to service robots such as sweeping robots; the multi-line laser radar realizes three-dimensional scanning of different laser beams according to a certain included angle according to the density degree of the number of lines, the higher the number of laser lines is, the denser the scanned laser point cloud is, the more obvious the embodied target shape and size characteristics are, the larger the data volume is, and the more expensive the price is. Compared with visual sensors such as a monocular camera and the like, the laser radar has the advantages of being capable of obtaining high-precision depth information and target three-dimensional size information, not being easily influenced by illumination conditions and the like, and in documents [ tsukudani [ unmanned automobile environment perception technology overview [ J ]. automobile and driving maintenance.2017 ] ] and [ royal art ] automatic driving automobile perception system key technology overview [ J ]. automobile electrical appliances.2016 (12) (12: 12-16 ]) and [ royal art ] automatic driving automobile perception system key technology overview [ J ] automobile electrical appliances.2016 (12) (12: 12-16 ]), due to the fact that point cloud data scanned by the laser radar is sparse, the laser radar also has the defect that visual target texture and color information cannot be obtained.
For target detection, the mainstream target detection algorithm in the industry at present is a visual identification target detection algorithm designed for a visual sensor, which includes a traditional image processing method and a target detection algorithm based on machine learning. Along with the rapid development of intelligent driving automobiles, laser radars are gradually valued and applied by numerous automatic driving practitioners, and more people are put into the research of target detection algorithms based on laser radar sensors, wherein the research comprises a method for directly carrying out data processing means such as filtering and clustering on laser point clouds so as to realize target detection and a point cloud target detection method based on deep learning. The later method needs to label and train a plurality of point cloud data sets, has higher requirements on hardware computing power for processing data and has higher realization difficulty, so the method mainly utilizes data processing methods such as point cloud filtering, clustering and the like to realize track cone bucket target detection.
For unmanned racing under the application scenarios described above, [ Thanglical. Vision-based unmanned automotive research review [ J]Manufacturing automation 2016(08) 134-.][Dhall, Ankit et al. “Real-time 3D Traffic Cone Detection for Autonomous Driving.”
2019 IEEE Intelligent Vehicles Symposium (IV)(2019): 494-501.]And [ Panagiotaki E, An effective tracking detection and Mapping System for Autonomous Driving Race car [ D]. Departmentof Information Technology and Electrical Engineering, 2017.]In the literature, it is also mentioned that obtaining accurate target position information through a sensing system under the condition of over-strong or over-weak illumination intensity is the primary premise for ensuring decision planning of an unmanned system and stable operation of vehicle control. Therefore, in order to get rid of the influence of the illumination condition on the environment perception system, the multi-line laser radar is taken as a research object, and only the laser point cloud perception data is relied on to carry out target detection and target point tracking algorithm research on the track cone bucket. The cone bucket detection method is not easily influenced by illumination conditions, does not need to carry out complicated data annotation, and can meet the requirement of real-time performance.
Disclosure of Invention
The invention aims to solve the technical problem that the unmanned racetrack cone bucket detection and target tracking are realized only by the aid of laser radar point cloud data, and the problem that a visual perception system is prone to being affected by illumination intensity and is invalid is solved.
The invention is realized by at least one of the following technical schemes.
A track cone bucket detection and target point tracking method based on a multi-line laser radar comprises the following steps:
1) reading point cloud data of a laser radar under an ROS robot operating system;
2) carrying out through filtering on the point cloud data of the laser radar according to different track scenes by adopting a through filter;
3) adopting a random sampling consistency algorithm to eliminate the interference of ground point cloud data on the detection of the cone bucket;
4) preliminarily screening out a point cloud Cluster of a cone bucket by adopting an Euclidean clustering Extraction algorithm (Euclidean Cluster Extraction);
5) carrying out statistical analysis on the point cloud clusters obtained by clustering by adopting a maximum standard deviation threshold value method, setting a maximum standard deviation threshold value in the direction of X, Y, Z according to the characteristics of the actual size of the cone bucket, and screening out the point cloud clusters meeting the conditions to be regarded as the cone bucket;
6) performing statistical analysis on the detected cone barrels, finding out one cone barrel on the left side and the right side closest to the laser radar, and acquiring the coordinates of the center points of the point cloud clusters;
7) calculating the coordinates of the center points of the cone barrels on the left side and the right side of the laser radar to obtain the center point of the nearest cone barrel on the two sides of the laser radar as a nearest target point in the current state, and controlling the trolley to move forward towards the target point;
8) and (5) circulating the steps 1) to 7), and continuously acquiring and tracking the latest target point.
Further, the step 1) specifically comprises the steps of installing an ROS (robot Operating system) robot Operating system on a computing platform provided with an ubuntu16.04 Operating system, configuring a laser radar driving package, starting a laser radar Operating node to collect point cloud data in real time, wherein the size of a cone barrel to be detected is a b according to the track characteristics, in order to enable the laser radar to scan on the cone barrel to the maximum extent, the installation position of the laser radar is located below the nose at the forefront of the racing car, and the installation height and ground clearance is b/2.
Further, the step 2) comprises the steps of presetting a track scene to be detected before starting detection, wherein the track scene comprises a 75-meter linear acceleration track, an 8-shaped encircling track and a high-speed tracking track; and setting the detection range of the laser radar aiming at different tracks, and filtering point cloud data outside the range.
Further, the step 3) comprises the steps of setting a plane filtering threshold value to be a/5 by utilizing a random sampling consistency algorithm, judging the maximum deviation distance of points in the normal direction of the point cloud cluster extracted by the algorithm, regarding the point cloud cluster with the maximum deviation distance larger than a/5 as a plane, regarding the point cloud cluster with the maximum deviation distance not larger than a/5 in the normal direction of the point cloud cluster as the same plane, and removing the point cloud cluster belonging to the plane in the current point cloud so as to achieve the aim of filtering ground point cloud data.
Further, the step 4) is specifically that through analyzing the point cloud data collected in advance, the number of points on different cone barrels of the track are collected by a laser radar, the number of the closest cone barrel points is R, and the number of the cone barrel points which can be swept to the farthest position is R, so that according to the statistical rule of the collected data for a plurality of times, the minimum clustering points are set as R points, the maximum clustering points are set as R points, the maximum distance of searching two points in the clustering process is set as L, a KD tree is used for point cloud searching, and the point clouds meeting the clustering conditions, namely the point clouds with the point in the (R, R) range are divided into a plurality of point cloud clusters; and counting the number of points of each point cloud cluster, and respectively calculating an average value of X, Y, Z coordinates of each point cloud cluster to serve as a point cloud gravity center point of the point cloud cluster for replacing the position of the point cloud cluster relative to the laser radar.
Further, the step 5) specifically includes, according to the shape and size characteristics of the cone bucket, setting a maximum standard deviation threshold value in X, Y direction to Q, setting a height of the cone bucket in Z axis to b, and setting a maximum standard deviation threshold value in Z axis to Q, where the standard deviations of the three directional coordinate values of all points X, Y, Z in the clustered point cloud cluster are counted to be smaller than corresponding threshold values, that is, when the calculated standard deviations in X and Y directions of the point cloud cluster are smaller than Q and the standard deviation in Z direction is smaller than Q, the point cloud cluster is regarded as a qualified cone bucket.
Further, the step 6) specifically comprises counting the number of the conical barrels detected in the step 5), and calculating the average Y coordinate of all points in each conical barrel; traversing all the detected cone buckets, finding the cone bucket with the maximum number of the middle points of the cone buckets with the average Y coordinate larger than 0 as the nearest cone bucket on the left side of the laser radar, and similarly, finding the cone bucket with the maximum number of the middle points of the cone buckets with the average Y coordinate smaller than 0 as the nearest cone bucket on the right side of the laser radar.
Further, the step 7) specifically comprises the steps of respectively calculating X, Y, Z coordinate average values of point cloud clusters of cone barrels nearest to the left front and the right front of the laser radar, regarding the point cloud average values as center point coordinates of left and right nearest cone barrels of the laser radar, calculating midpoint coordinates of a connecting line of the left and right cone barrels of the laser radar, and taking the midpoint coordinates as a target point in the current state.
Further, the step 8) specifically includes performing the processing of the steps 1) to 7) on the point cloud data collected by each frame of laser radar under the ROS robot operating system, outputting a moving target point of the next frame of the robot after the processing of each frame of data is finished, and controlling the robot to move towards the target point, and meanwhile, performing the processing of the steps 1) to 7) on the data collected by the next frame of laser radar, so as to achieve the purpose of detecting the cone barrel in real time and tracking the target point.
Further, the laser radar is a 16-line laser radar.
Compared with the prior art, the invention has the beneficial effects that:
1. the cone-bucket target detection method only depends on the laser radar sensor, and performs operations such as filtering, segmentation and clustering on the original data acquired by the laser radar to perform cone-bucket detection and positioning. Compared with the cone barrel detection based on stereoscopic vision, the method is not easily influenced by illumination conditions, and has high ranging precision and strong real-time performance.
2. The method for detecting and tracking the cone-barrel target can directly output the coordinate position of the target point relative to the vehicle, can adapt to different heights of laser radar installation by adjusting the filtering parameters, is applied to vehicle chassis of different types by algorithm migration, and has strong adaptability.
Drawings
Fig. 1 is a schematic flow chart illustrating a track cone bucket detection and target point tracking method based on a multi-line lidar according to an embodiment of the present invention;
FIG. 2 is a racetrack diagram of the present embodiment;
wherein: 1-cone barrel and 2-trolley.
Detailed Description
The purpose of the present invention is described in further detail below by using specific embodiments, and in order to reduce cost and facilitate experimental tests, the present invention uses an experimental cart to simulate a real application scene of a racing car, and uses the same installation position and installation height as those of the real car to simulate a data acquisition mode in which a laser radar is installed on the real racing car.
As shown in fig. 1, a method for track cone bucket detection and target point tracking based on multi-line laser radar includes the following steps:
1. reading point cloud data of a laser radar under an ROS robot operating system; specifically, reading laser radar point cloud data under the ROS robot operating system comprises the following steps: and installing an ROS robot operating system on a computing platform provided with a ubuntu16.04 operating system, configuring a laser radar driving package, and starting a laser radar operation node to acquire point cloud data in real time. A laser radar and algorithm operation platform, namely a notebook computer, is fixed on the experimental trolley, according to the characteristics of the track, the size of a cone barrel 1 to be detected is 20 x 30cm, the height of the cone barrel 1 relative to the ground is lower, in order to enable laser radar point cloud to scan on the cone barrel 1 to the maximum extent, the installation position of the laser radar is located on the frontmost plane of the experimental trolley 2, and the ground clearance of the installation height is 10 cm.
2. And performing through filtering on the laser radar point cloud data according to different track scenes by adopting a through filtering algorithm, removing redundant point cloud data at a position far away from the track, and reducing subsequent calculation amount.
Specifically, before starting the detection, the track scenes to be detected are preset, including a 75-meter linear acceleration track, an 8-shaped circling track, a high-speed tracking track, and the like. Setting a straight-through filtering range for a 75-meter straight-line acceleration track: the range of the laser radar in the X direction (0, 100) meter right ahead, the range of the laser radar in the lateral Y direction (-3, + 3) meter, and the range of the laser radar in the vertical Z direction (-0.5, 0.3) meter are filtered to remove point cloud data outside the range. Aiming at the 8-shaped circumambient track, a straight-through filtering range is set, an isosceles trapezoid which takes a Y axis as a lower bottom edge, is 6m long and 10 m high and has a left and right bevel edge included angle of 90 degrees is arranged on an XY plane of the laser radar, and point cloud data outside the range is filtered in the range of (-0.5, 0.3) meters in the vertical Z direction of the laser radar. Aiming at the high-speed tracking track, a straight-through filtering range is set, an isosceles trapezoid which takes a Y axis as a lower bottom edge, is 6m long and 20m high and has a left-right bevel edge included angle of 90 degrees is arranged on an XY plane of a laser radar, and point cloud data outside the range are filtered within a range of (-0.5, 0.3) meters in the vertical Z direction of the laser radar within a range of (0, 20) meters in the right front of the laser radar. Figure 2 is a straight accelerating racetrack.
3. And (3) adopting a random sampling consensus algorithm (RANSAC algorithm) to eliminate the interference of the ground point cloud data on the cone bucket detection.
The point cloud after the straight-through filtering is provided with a plane filtering, as the ground of the track is generally asphalt pavement and is often slightly uneven, the threshold value of the plane filtering is set to be 0.04m by utilizing a random sample consensus algorithm (RANSAC algorithm), the point cloud with the deviation distance not more than 0.04m in the same direction is considered as the same plane, and the main purpose of the step is to filter the point cloud data of the laser radar scanned on the ground and eliminate the interference of the point cloud on the cone detection by the ground. Filtering the ground point cloud is a link in the RANSAC algorithm, and the algorithm can select to only retain the plane point cloud and also can select to only filter the plane point cloud.
4. And (4) preliminarily screening point cloud clusters which are possibly cone buckets by adopting an Euclidean clustering algorithm. Specifically, by analyzing the point cloud data packet collected in advance, the laser radar used by the invention is a 16-line laser radar, 16 line beams scanned out during working are uniformly distributed within a 30-degree included angle taking a horizontal plane as a symmetrical plane, and the distance capable of being scanned out at the farthest is 150 m. The number of points swept on different cone barrels of the track changes with the distance, the number of the nearest cone barrel points is close to 120, and the number of the cone barrel points swept at the farthest position is only 2. Therefore, according to the statistical rule of data collected for many times, the minimum clustering point number is set to be 2 points, the maximum clustering point number is set to be 120 points, the maximum distance between two searched points in the clustering process is set to be 0.3m, a KD tree is adopted for point cloud searching, and the point clouds meeting the clustering conditions are divided into a plurality of point cloud clusters. And counting the number of points of each point cloud cluster, and respectively calculating an average value of X, Y, Z coordinates for each point cloud cluster to serve as a point cloud gravity center point of the point cloud cluster, wherein the gravity center point is not a geometric center point of the point cloud cluster, but can be used for approximately replacing the position of the point cloud cluster relative to the laser radar.
5. And (3) carrying out statistical analysis on the point cloud clusters obtained by clustering by adopting a maximum standard deviation threshold algorithm, setting a maximum standard deviation threshold in the direction of X, Y, Z according to the characteristics of the actual size of the cone bucket, and screening out the point cloud clusters meeting the conditions to be regarded as the cone bucket.
Specifically, the method for screening out the point cloud clusters meeting the conditions as the cone bucket comprises the following steps of performing statistical analysis on the point cloud clusters obtained by clustering by adopting a maximum standard deviation threshold algorithm, setting a maximum standard deviation threshold in the X, Y, Z direction according to the characteristics of the actual size of the cone bucket, and screening out the point cloud clusters meeting the conditions as the cone bucket, wherein the method comprises the following steps: according to the shape and size characteristics of the cone barrel, the cone barrel can be used as a rotating body around a vertical central axis, so that the characteristics are consistent in the X, Y direction, the maximum standard deviation threshold is set to be 0.08m, the height of the cone barrel is 30cm on the Z axis, the maximum standard deviation threshold is set to be 0.15m, and when the standard deviation of the coordinate values of the three directions of X, Y, Z of all the points in the clustered point cloud cluster is counted to be smaller than the corresponding threshold, the point cloud cluster is considered to be a qualified cone barrel.
6. And carrying out statistical analysis on the detected cone barrels, finding one cone barrel on the left side and the right side closest to the laser radar, and acquiring the coordinates of the center points of the point cloud clusters. Specifically, the number of points of the cone bucket detected in the previous step is counted, and an average Y coordinate Mean _ Y of all the points in each cone bucket is calculated. Traversing all the detected cone buckets, finding out the cone bucket with the average Y coordinate larger than 0, namely the cone bucket with the maximum point number in the cone bucket with Mean _ Y >0, considering the cone bucket as the closest cone bucket on the left side of the laser radar, and discarding the rest point cloud clusters. Similarly, the cone bucket with the largest number of middle points of the cone buckets with the average Y coordinate smaller than 0, namely Mean _ Y <0, is found to be the closest cone bucket on the right side of the laser radar.
7. And calculating the coordinates of the central points of the left and right conical barrels to obtain the central point of the left and right closest conical barrels as the nearest target in the current state, and controlling the trolley to move forward towards the target. Specifically, X, Y, Z coordinate average values are respectively calculated for the point cloud clusters of the closest cone barrels in the left front and the right front of the laser radar obtained in the last step, the point cloud clusters are used as the space coordinate positions of the closest cone barrels in the left front and the right front, and the coordinates of the central point of the connecting line of the point cloud clusters are calculated and used as the motion target points of the robot in the current state.
8. And circulating the steps, and continuously acquiring and tracking the latest target point. And calculating to obtain a next motion target point of the vehicle in the current state according to the steps, adjusting the orientation angle of the vehicle according to the coordinate position of the vehicle relative to the vehicle, and controlling the vehicle to move towards the target point. Meanwhile, the laser radar acquires the next frame of point cloud data, starts a new point cloud data processing step, obtains the next motion target point and controls the vehicle to continue to move forward. And continuously circulating the steps of point cloud processing and vehicle control in an ROS robot operating system to realize track cone bucket detection and target point tracking based on the multi-line laser radar.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can substitute or change the technical solution of the present invention and its inventive concept within the scope of the present invention disclosed by the present invention.
Claims (10)
1. A track cone bucket detection and target point tracking method based on a multi-line laser radar is characterized by comprising the following steps:
1) reading point cloud data of a laser radar under an ROS robot operating system;
2) carrying out through filtering on the point cloud data of the laser radar according to different track scenes by adopting a through filter;
3) adopting a random sampling consistency algorithm to eliminate the interference of ground point cloud data on the detection of the cone bucket;
4) preliminarily screening out a point cloud cluster of the cone bucket by adopting an Euclidean clustering extraction algorithm;
5) carrying out statistical analysis on the point cloud clusters obtained by clustering by adopting a maximum standard deviation threshold value method, setting a maximum standard deviation threshold value in the direction of X, Y, Z according to the characteristics of the actual size of the cone bucket, and screening out the point cloud clusters meeting the conditions to be regarded as the cone bucket;
6) performing statistical analysis on the detected cone barrels, finding out one cone barrel on the left side and the right side closest to the laser radar, and acquiring the coordinates of the center points of the point cloud clusters;
7) calculating the coordinates of the center points of the cone barrels on the left side and the right side of the laser radar to obtain the center point of the nearest cone barrel on the two sides of the laser radar as a nearest target point in the current state, and controlling the trolley to move forward towards the target point;
8) and (5) circulating the steps 1) to 7), and continuously acquiring and tracking the latest target point.
2. The method for track cone bucket detection and target point tracking based on multiline laser radar according to claim 1, wherein the step 1) specifically includes that an ROS (robot Operating system) robot Operating system is installed on a computing platform equipped with a ubuntu16.04 Operating system, a laser radar driving package is configured, a laser radar Operating node is started to conduct point cloud data real-time collection, according to track characteristics, the size of a cone bucket to be detected is a b, in order to enable the laser radar to scan on the cone bucket to the maximum extent, the installation position of the laser radar is located below the nose at the forefront of the racing car, and the installation height and ground clearance is b/2.
3. The method of claim 1, wherein the method comprises the steps of: step 2) comprises the steps of presetting a track scene to be detected before starting detection, wherein the track scene comprises a 75-meter linear acceleration track, an 8-shaped encircling track and a high-speed tracking track; and setting the detection range of the laser radar aiming at different tracks, and filtering point cloud data outside the range.
4. The method of claim 1, wherein the method comprises the steps of: and step 3) comprises the steps of setting a plane filtering threshold value as a/5 by utilizing a random sampling consistency algorithm, judging the maximum deviation distance of points in the normal direction of the point cloud cluster extracted by the algorithm, regarding the point cloud cluster with the maximum deviation distance larger than a/5 as a plane, regarding the point cloud cluster with the maximum deviation distance not larger than a/5 in the normal direction of the point cloud cluster as the same plane, and removing the point cloud cluster belonging to the plane in the current point cloud so as to achieve the aim of filtering ground point cloud data.
5. The method of claim 1, wherein the method comprises the steps of: step 4) specifically, the number of points on different cone barrels of the track is collected by adopting a laser radar through analyzing the point cloud data collected in advance, the point number of the nearest cone barrel is R, and the point number of the cone barrel which can be swept to the farthest position is R; and counting the number of points of each point cloud cluster, and respectively calculating an average value of X, Y, Z coordinates of each point cloud cluster to serve as a point cloud gravity center point of the point cloud cluster for replacing the position of the point cloud cluster relative to the laser radar.
6. The method of claim 1, wherein the method comprises the steps of: step 5) specifically comprises the steps that according to the shape and size characteristics of the cone barrel, because the cone barrel is a rotating body around a central axis in the vertical direction, the width of the cone barrel in the direction X, Y is a, the maximum standard deviation threshold value in the X, Y direction is set to be Q, the height of the cone barrel in the Z axis is b, the maximum standard deviation threshold value in the Z axis is set to be Q, and when the standard deviation of three direction coordinate values of X, Y, Z of all the points in the clustered point cloud cluster is counted to be smaller than the corresponding threshold value, namely when the standard deviation in the X direction and the Y direction of the point cloud cluster is calculated to be smaller than Q and the standard deviation in the Z direction is smaller than Q, the point cloud cluster is regarded as a cone barrel meeting the conditions.
7. The method of claim 1, wherein the method comprises the steps of: step 6) specifically comprises counting the number of the conical barrels detected in the step 5), and calculating the average Y coordinate of all points in each conical barrel; traversing all the detected cone buckets, finding the cone bucket with the maximum number of the middle points of the cone buckets with the average Y coordinate larger than 0 as the nearest cone bucket on the left side of the laser radar, and similarly, finding the cone bucket with the maximum number of the middle points of the cone buckets with the average Y coordinate smaller than 0 as the nearest cone bucket on the right side of the laser radar.
8. The method of claim 1, wherein the method comprises the steps of: and 7) specifically calculating X, Y, Z coordinate average values of cone barrel point cloud clusters closest to the left front and the right front of the laser radar respectively, regarding the calculated X, Y, Z coordinate average values as center point coordinates of left and right closest cone barrels of the laser radar, calculating midpoint coordinates of a connecting line of the left and right cone barrels of the laser radar, and taking the midpoint coordinates as a target point in the current state.
9. The method of claim 1, wherein the method comprises the steps of: and step 8) specifically comprises the steps of 1) processing to step 7) of point cloud data collected by each frame of laser radar under an ROS robot operating system, outputting a motion target point of the next frame of the robot after the data processing of each frame is finished, controlling the robot to move towards the target point, and meanwhile, processing the steps 1) to 7) of data collected by the next frame of laser radar so as to achieve the purposes of detecting the cone barrel in real time and tracking the target point.
10. The method of claim 1, wherein the method comprises the steps of: the laser radar is a 16-line laser radar.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910995353.2A CN110780305B (en) | 2019-10-18 | 2019-10-18 | Track cone detection and target point tracking method based on multi-line laser radar |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910995353.2A CN110780305B (en) | 2019-10-18 | 2019-10-18 | Track cone detection and target point tracking method based on multi-line laser radar |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110780305A true CN110780305A (en) | 2020-02-11 |
| CN110780305B CN110780305B (en) | 2023-04-21 |
Family
ID=69385978
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910995353.2A Active CN110780305B (en) | 2019-10-18 | 2019-10-18 | Track cone detection and target point tracking method based on multi-line laser radar |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110780305B (en) |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111308499A (en) * | 2020-03-09 | 2020-06-19 | 中振同辂(江苏)机器人有限公司 | Obstacle detection method based on multi-line laser radar |
| CN111487641A (en) * | 2020-03-19 | 2020-08-04 | 福瑞泰克智能系统有限公司 | Method and device for detecting object by using laser radar, electronic equipment and storage medium |
| CN111709355A (en) * | 2020-06-12 | 2020-09-25 | 北京百度网讯科技有限公司 | Method and device for identifying target area, electronic equipment and road side equipment |
| CN111723716A (en) * | 2020-06-11 | 2020-09-29 | 深圳地平线机器人科技有限公司 | Method, device, system, medium and electronic equipment for determining orientation of target object |
| CN112101128A (en) * | 2020-08-21 | 2020-12-18 | 东南大学 | Unmanned formula racing car perception planning method based on multi-sensor information fusion |
| CN112294197A (en) * | 2020-11-04 | 2021-02-02 | 深圳市普森斯科技有限公司 | Sweeping control method of sweeper, electronic device and storage medium |
| CN112415538A (en) * | 2020-11-16 | 2021-02-26 | 重庆长安汽车股份有限公司 | Planning method and system for automatic driving vehicle to detour conical cylinder and vehicle |
| CN112578673A (en) * | 2020-12-25 | 2021-03-30 | 浙江科技学院 | Perception decision and tracking control method for multi-sensor fusion of formula-free racing car |
| CN112731925A (en) * | 2020-12-21 | 2021-04-30 | 浙江科技学院 | Conical barrel identification and path planning and control method for unmanned formula racing car |
| CN112883820A (en) * | 2021-01-26 | 2021-06-01 | 上海应用技术大学 | Road target 3D detection method and system based on laser radar point cloud |
| CN112884687A (en) * | 2021-01-29 | 2021-06-01 | 郑州信大云筑工程科技有限公司 | Mapping laser radar scanning strategy control method and system based on artificial intelligence |
| CN113095324A (en) * | 2021-04-28 | 2021-07-09 | 合肥工业大学 | Classification and distance measurement method and system for cone barrel |
| CN113325428A (en) * | 2021-04-29 | 2021-08-31 | 中国科学院沈阳自动化研究所 | Distance and course measuring method based on laser radar |
| CN113343840A (en) * | 2021-06-02 | 2021-09-03 | 合肥泰瑞数创科技有限公司 | Object identification method and device based on three-dimensional point cloud |
| WO2021222279A1 (en) * | 2020-04-28 | 2021-11-04 | Raven Industries, Inc. | Object detection and tracking for automated operation of vehicles and machinery |
| CN113655498A (en) * | 2021-08-10 | 2021-11-16 | 合肥工业大学 | Method and system for extracting cone bucket information in racetrack based on laser radar |
| CN113778078A (en) * | 2021-03-05 | 2021-12-10 | 北京京东乾石科技有限公司 | Positioning information generation method and device, electronic equipment and computer readable medium |
| CN114387305A (en) * | 2021-12-30 | 2022-04-22 | 驭势科技（北京）有限公司 | Trailer queue tracking method and device, computer equipment and storage medium |
| CN115049696A (en) * | 2021-03-08 | 2022-09-13 | 北京金茂绿建科技有限公司 | Personnel monitoring method and device based on radar data |
| US11734917B2 (en) | 2017-11-13 | 2023-08-22 | Raven Industries, Inc. | Safety system for autonomous operation of off-road and agricultural vehicles using machine learning for detection and identification of obstacles |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180108146A1 (en) * | 2016-10-13 | 2018-04-19 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method and apparatus for annotating point cloud data |
| WO2018205119A1 (en) * | 2017-05-09 | 2018-11-15 | 深圳市速腾聚创科技有限公司 | Roadside detection method and system based on laser radar scanning |
| CN108932736A (en) * | 2018-05-30 | 2018-12-04 | 南昌大学 | Two-dimensional laser radar Processing Method of Point-clouds and dynamic robot pose calibration method |
| CN109188459A (en) * | 2018-08-29 | 2019-01-11 | 东南大学 | A kind of small obstacle recognition method in ramp based on multi-line laser radar |
| CN110018489A (en) * | 2019-04-25 | 2019-07-16 | 上海蔚来汽车有限公司 | Target tracking method, device and controller and storage medium based on laser radar |
| CN110275153A (en) * | 2019-07-05 | 2019-09-24 | 上海大学 | A kind of waterborne target detection and tracking based on laser radar |
-
2019
- 2019-10-18 CN CN201910995353.2A patent/CN110780305B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180108146A1 (en) * | 2016-10-13 | 2018-04-19 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method and apparatus for annotating point cloud data |
| WO2018205119A1 (en) * | 2017-05-09 | 2018-11-15 | 深圳市速腾聚创科技有限公司 | Roadside detection method and system based on laser radar scanning |
| CN108932736A (en) * | 2018-05-30 | 2018-12-04 | 南昌大学 | Two-dimensional laser radar Processing Method of Point-clouds and dynamic robot pose calibration method |
| CN109188459A (en) * | 2018-08-29 | 2019-01-11 | 东南大学 | A kind of small obstacle recognition method in ramp based on multi-line laser radar |
| CN110018489A (en) * | 2019-04-25 | 2019-07-16 | 上海蔚来汽车有限公司 | Target tracking method, device and controller and storage medium based on laser radar |
| CN110275153A (en) * | 2019-07-05 | 2019-09-24 | 上海大学 | A kind of waterborne target detection and tracking based on laser radar |
Cited By (29)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11734917B2 (en) | 2017-11-13 | 2023-08-22 | Raven Industries, Inc. | Safety system for autonomous operation of off-road and agricultural vehicles using machine learning for detection and identification of obstacles |
| CN111308499A (en) * | 2020-03-09 | 2020-06-19 | 中振同辂(江苏)机器人有限公司 | Obstacle detection method based on multi-line laser radar |
| CN111487641A (en) * | 2020-03-19 | 2020-08-04 | 福瑞泰克智能系统有限公司 | Method and device for detecting object by using laser radar, electronic equipment and storage medium |
| CN111487641B (en) * | 2020-03-19 | 2022-04-22 | 福瑞泰克智能系统有限公司 | Method and device for detecting object by using laser radar, electronic equipment and storage medium |
| WO2021222279A1 (en) * | 2020-04-28 | 2021-11-04 | Raven Industries, Inc. | Object detection and tracking for automated operation of vehicles and machinery |
| AU2021262764B2 (en) * | 2020-04-28 | 2023-11-30 | Raven Industries, Inc. | Object detection and tracking for automated operation of vehicles and machinery |
| CN111723716A (en) * | 2020-06-11 | 2020-09-29 | 深圳地平线机器人科技有限公司 | Method, device, system, medium and electronic equipment for determining orientation of target object |
| CN111723716B (en) * | 2020-06-11 | 2024-03-08 | 深圳地平线机器人科技有限公司 | Method, device, system, medium and electronic equipment for determining target object orientation |
| CN111709355B (en) * | 2020-06-12 | 2023-08-29 | 阿波罗智联(北京)科技有限公司 | Method and device for identifying target area, electronic equipment and road side equipment |
| CN111709355A (en) * | 2020-06-12 | 2020-09-25 | 北京百度网讯科技有限公司 | Method and device for identifying target area, electronic equipment and road side equipment |
| CN112101128A (en) * | 2020-08-21 | 2020-12-18 | 东南大学 | Unmanned formula racing car perception planning method based on multi-sensor information fusion |
| CN112294197A (en) * | 2020-11-04 | 2021-02-02 | 深圳市普森斯科技有限公司 | Sweeping control method of sweeper, electronic device and storage medium |
| CN112415538A (en) * | 2020-11-16 | 2021-02-26 | 重庆长安汽车股份有限公司 | Planning method and system for automatic driving vehicle to detour conical cylinder and vehicle |
| CN112731925B (en) * | 2020-12-21 | 2024-03-15 | 浙江科技学院 | Cone barrel identification and path planning and control method for formula car |
| CN112731925A (en) * | 2020-12-21 | 2021-04-30 | 浙江科技学院 | Conical barrel identification and path planning and control method for unmanned formula racing car |
| CN112578673A (en) * | 2020-12-25 | 2021-03-30 | 浙江科技学院 | Perception decision and tracking control method for multi-sensor fusion of formula-free racing car |
| CN112883820B (en) * | 2021-01-26 | 2024-04-19 | 上海应用技术大学 | Road target 3D detection method and system based on laser radar point cloud |
| CN112883820A (en) * | 2021-01-26 | 2021-06-01 | 上海应用技术大学 | Road target 3D detection method and system based on laser radar point cloud |
| CN112884687A (en) * | 2021-01-29 | 2021-06-01 | 郑州信大云筑工程科技有限公司 | Mapping laser radar scanning strategy control method and system based on artificial intelligence |
| CN112884687B (en) * | 2021-01-29 | 2022-02-11 | 郑州信大云筑工程科技有限公司 | Mapping laser radar scanning strategy control method and system based on artificial intelligence |
| CN113778078A (en) * | 2021-03-05 | 2021-12-10 | 北京京东乾石科技有限公司 | Positioning information generation method and device, electronic equipment and computer readable medium |
| CN115049696A (en) * | 2021-03-08 | 2022-09-13 | 北京金茂绿建科技有限公司 | Personnel monitoring method and device based on radar data |
| CN113095324A (en) * | 2021-04-28 | 2021-07-09 | 合肥工业大学 | Classification and distance measurement method and system for cone barrel |
| CN113325428A (en) * | 2021-04-29 | 2021-08-31 | 中国科学院沈阳自动化研究所 | Distance and course measuring method based on laser radar |
| CN113343840B (en) * | 2021-06-02 | 2022-03-08 | 合肥泰瑞数创科技有限公司 | Object identification method and device based on three-dimensional point cloud |
| CN113343840A (en) * | 2021-06-02 | 2021-09-03 | 合肥泰瑞数创科技有限公司 | Object identification method and device based on three-dimensional point cloud |
| CN113655498B (en) * | 2021-08-10 | 2023-07-18 | 合肥工业大学 | Method and system for extracting cone barrel information in racetrack based on laser radar |
| CN113655498A (en) * | 2021-08-10 | 2021-11-16 | 合肥工业大学 | Method and system for extracting cone bucket information in racetrack based on laser radar |
| CN114387305A (en) * | 2021-12-30 | 2022-04-22 | 驭势科技（北京）有限公司 | Trailer queue tracking method and device, computer equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110780305B (en) | 2023-04-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110780305A (en) | Track cone bucket detection and target point tracking method based on multi-line laser radar | |
| CN108955702B (en) | Lane-level map creation system based on three-dimensional laser and GPS inertial navigation system | |
| CN110531376B (en) | Obstacle detection and tracking method for port unmanned vehicle | |
| CN110320504B (en) | Unstructured road detection method based on laser radar point cloud statistical geometric model | |
| CN114842438B (en) | Terrain detection method, system and readable storage medium for automatic driving automobile | |
| CN109849922B (en) | Visual information and GIS information fusion-based method for intelligent vehicle | |
| CN112101092A (en) | Automatic driving environment sensing method and system | |
| CN110705458B (en) | Boundary detection method and device | |
| CN113255520B (en) | Vehicle obstacle avoidance method based on binocular vision and deep learning and electronic equipment | |
| GB2558716A (en) | Road sign recognition | |
| CN112184736B (en) | Multi-plane extraction method based on European clustering | |
| CN108280840B (en) | Road real-time segmentation method based on three-dimensional laser radar | |
| CN110673107B (en) | Road edge detection method and device based on multi-line laser radar | |
| CN112346463B (en) | Unmanned vehicle path planning method based on speed sampling | |
| CN107796373B (en) | Distance measurement method based on monocular vision of front vehicle driven by lane plane geometric model | |
| CN110674705A (en) | Small-sized obstacle detection method and device based on multi-line laser radar | |
| CN115049700A (en) | Target detection method and device | |
| CN114488194A (en) | Method for detecting and identifying targets under structured road of intelligent driving vehicle | |
| CN115861968A (en) | Dynamic obstacle removing method based on real-time point cloud data | |
| CN114821526A (en) | Obstacle three-dimensional frame detection method based on 4D millimeter wave radar point cloud | |
| CN116573017A (en) | Urban rail train running clearance foreign matter sensing method, system, device and medium | |
| CN115220449A (en) | Path planning method and device, storage medium, chip and vehicle | |
| CN111539278A (en) | Detection method and system for target vehicle | |
| CN111323026B (en) | Ground filtering method based on high-precision point cloud map | |
| CN114743179A (en) | Panoramic visible driving area detection method based on semantic segmentation |
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 |