CN114879153A - Radar parameter calibration method and device and vehicle - Google Patents
Radar parameter calibration method and device and vehicle Download PDFInfo
- Publication number
- CN114879153A CN114879153A CN202210641520.5A CN202210641520A CN114879153A CN 114879153 A CN114879153 A CN 114879153A CN 202210641520 A CN202210641520 A CN 202210641520A CN 114879153 A CN114879153 A CN 114879153A
- Authority
- CN
- China
- Prior art keywords
- target
- radar
- point set
- point
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 77
- 238000006243 chemical reaction Methods 0.000 claims abstract description 17
- 238000012545 processing Methods 0.000 claims description 24
- 230000004927 fusion Effects 0.000 claims description 14
- 238000005520 cutting process Methods 0.000 claims description 8
- 230000004044 response Effects 0.000 claims description 8
- 230000008569 process Effects 0.000 description 19
- 239000011159 matrix material Substances 0.000 description 10
- 238000013519 translation Methods 0.000 description 9
- 239000007787 solid Substances 0.000 description 7
- 238000003860 storage Methods 0.000 description 7
- 230000007547 defect Effects 0.000 description 5
- 238000000605 extraction Methods 0.000 description 4
- 230000011218 segmentation Effects 0.000 description 4
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000005034 decoration Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000003550 marker Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 239000013598 vector Substances 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
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/40—Means for monitoring or calibrating
-
- 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
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/48—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S17/00
- G01S7/497—Means for monitoring or calibrating
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention discloses a radar parameter calibration method, a radar parameter calibration device and a vehicle. Wherein, the method comprises the following steps: acquiring multi-frame point cloud data collected by a target radar installed on a target vehicle, wherein the multi-frame point cloud data is obtained by sensing a plurality of targets in front of the target vehicle by the target radar; fusing multi-frame point cloud data, and determining a first point set, wherein the first point set comprises: coordinates of center points of the multiple targets in a first coordinate system corresponding to the target radar; registering the first point set and a second point set, and determining a target calibration parameter of the target radar, wherein the second point set comprises: the calibration parameters are used for representing conversion parameters for converting the coordinates in the first coordinate system to the coordinates in the second coordinate system, and the second point set can be obtained through a high-precision means. The invention solves the technical problem of low precision when the vehicle-mounted multi-solid-state radar performs automatic calibration in the prior art.
Description
Technical Field
The invention relates to the field of intelligent automobiles, in particular to a radar parameter calibration method and device and a vehicle.
Background
In an automatic driving system, a solid-state radar is used as a high-level core sensing device to assist automatic driving for sensing judgment, so that a plurality of solid-state radars are equipped on a vehicle body of an automatic driving vehicle. The installation of the solid-state radar has certain deviation, so that a multi-radar local coordinate system needs to be uniformly converted into a vehicle body coordinate system by a calibration method, the fusion and the processing of data are facilitated, and original data are provided for a sensing module. However, when the solid-state radar is calibrated in the prior art, the efficiency is low and the acquisition precision of sampling points is low.
In view of the above problems, no effective solution has been proposed.
Disclosure of Invention
The embodiment of the invention provides a method and a device for calibrating radar parameters and a vehicle, which at least solve the technical problem of low efficiency in automatic calibration of a vehicle-mounted multi-solid-state radar in the prior art.
According to an aspect of the embodiments of the present invention, there is provided a method for calibrating radar parameters, including: acquiring multi-frame point cloud data collected by a target radar installed on a target vehicle, wherein the multi-frame point cloud data is obtained by sensing a plurality of targets in front of the target vehicle by the target radar; fusing multi-frame point cloud data, and determining a first point set, wherein the first point set comprises: coordinates of center points of the multiple targets in a first coordinate system corresponding to the target radar; registering the first point set and a second point set, and determining a target calibration parameter of the target radar, wherein the second point set comprises: and the calibration parameters are used for representing conversion parameters for converting the coordinates in the first coordinate system to the coordinates in the second coordinate system.
Optionally, registering the first point set and the second point set, and determining a target calibration parameter of the target radar, includes: registering a first coordinate in the first point set and a second coordinate in the second point set to obtain an initial calibration parameter of the target radar and a registration error corresponding to the first coordinate, wherein the first coordinate and the second coordinate correspond to a central point of the same target; in response to the fact that the number of the first coordinates in the first point set is larger than the preset number, eliminating the first coordinates with the registration errors larger than the preset errors in the first point set, and registering the first coordinates in the first point set with the second coordinates in the second point set again; and determining the initial calibration parameters as the target calibration parameters in response to the number of the first coordinates contained in the first point set being less than or equal to the preset number.
Optionally, registering a first coordinate in the first point set with a second coordinate in the second point set to obtain an initial calibration parameter of the target radar and a registration error corresponding to the first coordinate, including: determining a first coordinate and a second coordinate corresponding to the same target in the first point set and the second point set based on the relative position relation of the targets; and registering the first coordinate and the second coordinate corresponding to the same target to obtain an initial calibration parameter and a registration error.
Optionally, the method further comprises: acquiring accurate position information of central points of a plurality of targets acquired by a total station; and converting the position information into a second coordinate system to obtain a second point set.
Optionally, fusing the multi-frame point cloud data to determine a first point set, including: fusing multi-frame point cloud data to obtain a first point cloud; extracting target points belonging to a plurality of targets in the first point cloud by using an Euclidean clustering mode; processing the target points based on the attribute parameters of the targets, and determining the central points of the targets; a first set of points is generated based on coordinates of a center point of the plurality of targets in a first coordinate system.
Optionally, before extracting target points belonging to a plurality of targets in the first point cloud by using an euclidean clustering method, the method further includes: cutting the first point cloud based on the target coordinate ranges corresponding to the multiple targets to obtain a second point cloud; denoising and outlier removing are carried out on the second point cloud to obtain a third point cloud; and extracting target points belonging to a plurality of targets in the third point cloud by using an Euclidean clustering mode.
Optionally, under the condition that multi-frame point cloud data is obtained for multiple times, registering the first point set and the second point set, and determining a target calibration parameter of the target radar, including: registering the first point set and the second point set for multiple times to obtain multiple calibration parameters of the target radar; and obtaining the average value of the plurality of calibration parameters to obtain the target calibration parameter.
Optionally, when the target radar is a plurality of radars, a plurality of threads are started, the threads correspond to the radars one by one, each thread is used for fusing multi-frame point cloud data acquired by the radar corresponding to the thread, determining a point set corresponding to the radar, and registering the point set corresponding to the radar with a second point set to obtain calibration parameters of the radar.
According to another aspect of the embodiments of the present invention, there is provided a radar parameter calibration apparatus, including: the acquisition module is used for acquiring multi-frame point cloud data acquired by a target radar installed on a target vehicle, wherein the multi-frame point cloud data is obtained by sensing a plurality of targets in front of the target vehicle by the target radar; the fusion module is used for fusing multi-frame point cloud data and determining a first point set, wherein the first point set comprises: coordinates of center points of the multiple targets in a first coordinate system corresponding to the target radar; a registration module, configured to register the first point set with a second point set, and determine a target calibration parameter of the target radar, where the second point set includes: and the calibration parameters are used for representing conversion parameters for converting the coordinates in the first coordinate system to the coordinates in the second coordinate system.
According to another aspect of the embodiment of the invention, a vehicle is further provided, and the vehicle comprises the radar parameter calibration device in any one of the above embodiments.
According to another aspect of the embodiments of the present invention, a non-volatile storage medium is further provided, where the non-volatile storage medium includes a stored program, and a processor of a device is controlled to execute the method for calibrating a radar parameter in any one of the above embodiments when the program runs.
According to another aspect of the embodiments of the present invention, there is further provided a processor, where the processor is configured to execute a program, where the program executes the method for calibrating radar parameters in any one of the above embodiments.
In the embodiment of the invention, in the process of sensing and judging a target vehicle by using a target radar to assist automatic driving, an automatic calibration method is set, a plurality of targets are distributed in front of the target vehicle, the target radar is used for scanning the targets to form multi-frame point cloud data, the coordinates of the target center point under an independent coordinate system of the target radar are used as a first point set, the coordinates of a white board center point under a vehicle body coordinate system, which are measured with high precision, are used as a second point set, the same target center point in the first point set and the second point set is registered, and a calculated rotation translation matrix is used as a conversion parameter for converting the independent coordinate system of the radar into the vehicle body coordinate system, namely the target calibration parameter. It is easy to notice that the method provides a convenient and efficient automatic calibration method, and the method fuses multi-frame point cloud data obtained by scanning a target radar as a first point set, so that the point cloud density is increased, the defect of sparse point cloud of the solid-state radar is effectively overcome, and the precision of extracting the center point of a white board is ensured, thereby ensuring the validity of a registration result, improving the efficiency of automatic calibration, and further solving the technical problem that the efficiency of automatic calibration of the vehicle-mounted multi-solid-state radar in the prior art is low.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the invention without limiting the invention. In the drawings:
FIG. 1 is a flow chart of a method for calibrating radar parameters according to an embodiment of the present invention;
FIG. 2 is a schematic illustration of an alternative vehicle body multiple radar profile according to an embodiment of the present invention;
FIG. 3 is a flow chart of an alternative in-thread multi-frame point cloud calibration according to an embodiment of the present invention;
FIG. 4 is a general flow chart of an alternative calibration of a multi-frame point cloud according to an embodiment of the present invention;
fig. 5 is a schematic diagram of a radar parameter calibration apparatus according to an embodiment of the present invention.
Detailed Description
In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Example 1
In accordance with an embodiment of the present invention, there is provided an embodiment of a method for calibrating radar parameters, it is noted that the steps illustrated in the flowchart of the drawings may be performed in a computer system, such as a set of computer-executable instructions, and that while a logical order is illustrated in the flowchart, in some cases the steps illustrated or described may be performed in an order different than that described herein.
Fig. 1 is a flowchart of a method for calibrating radar parameters according to an embodiment of the present invention, as shown in fig. 1, the method includes the following steps:
step S102, obtaining multi-frame point cloud data collected by a target radar installed on a target vehicle, wherein the multi-frame point cloud data is obtained by sensing a plurality of targets in front of the target vehicle through the target radar.
The target vehicle may be an autonomous vehicle equipped with a number of solid state radars around the body, and the target radar may be one or more solid state radars on the autonomous vehicle that may be used to assist autonomous driving in the perception decision. The point cloud data refers to a set of vectors in a three-dimensional coordinate system, which is used to represent surface information of an object, for example, in the calibration of solid-state radar point cloud data, the point cloud data may be surface information of all surrounding targets scanned by solid-state radar around a vehicle body. The plurality of targets may be objects preset in front of the target vehicle, for example, may be white boards, and the number of targets may be determined according to the test requirements.
In an optional embodiment, in order to achieve the purpose of automatically calibrating the target radar, a plurality of targets may be set in front of the target vehicle, and may form one frame of point cloud data under one scan of the target radar, and multiple frames of point cloud data of the plurality of targets may be obtained by performing multiple scans through the target radar and stored in the radar folder.
Step S104, fusing multi-frame point cloud data, and determining a first point set, wherein the first point set comprises: and coordinates of the center points of the plurality of targets in a first coordinate system corresponding to the target radar.
The first coordinate system may be an independent coordinate system established with the solid-state radar as an origin, the independent coordinate system may be established with the solid-state radar as an origin O, the vehicle forward direction as an X axis, the vertical ground upward direction as a Z axis, the direction perpendicular to XOZ and left of the forward direction as a Y axis, a plurality of solid-state radars are provided around the vehicle body, and a plurality of independent coordinate systems are established with respective radar positions. The first set of points may be coordinates of the center points of the surrounding target objects in the radar-independent coordinate system.
Step S106, registering the first point set and a second point set, and determining a target calibration parameter of the target radar, wherein the second point set comprises: and the calibration parameters are used for representing conversion parameters for converting the coordinates in the first coordinate system to the coordinates in the second coordinate system.
The second coordinate system may be a vehicle body coordinate system, which uses a ground projection of a center point of a front axle as an origin O, a vehicle forward direction as an X-axis, a vertically upward facing direction as a Z-axis, and a direction perpendicular to XOZ and left of the forward direction as a Y-axis. The second point set may be a coordinate set obtained by measuring the whiteboard center point by a total station and converting the whiteboard center point into a vehicle body coordinate system. The registration may refer to matching the central points of the same whiteboard in the two sets of point sets, and the calculated rotation and translation matrix may be used as a conversion parameter.
It should be noted that, autodrive mass production model increases day by day, and a plurality of solid state radars are equipped in different positions of the vehicle body, and in order to make these solid state radars can assist autodrive to sense and judge, a set of automatic calibration method needs to be designed, and a plurality of independent coordinate systems established by using a plurality of solid state radars on the autodrive vehicle as the origin can be converted to the same vehicle body coordinate system, and the vehicle body radar independent coordinate system can be parallel to the vehicle body coordinate system, so that coordinate conversion is convenient. The automatic calibration method can convert obstacles, target objects, terrain and the like in the front of the vehicle into data which can be surveyed, perform fusion and processing, provide original data of the target objects for the sensing module, process multi-frame point cloud data acquired by all radars in the same coordinate system, and prepare for obstacle recognition and terrain sensing in the next step of automatic driving. In the embodiment of the invention, after the file data of each radar is prepared, each radar is calibrated through multithreading simultaneously, and the target calibration parameter of each radar is calculated simultaneously, so that the automatic calibration process is effectively accelerated, and the beat time of factory calibration is saved for factory automatic calibration which is necessary for minutes and seconds.
In an alternative embodiment, fig. 2 is a schematic diagram of a plurality of alternative vehicle body radar distributions according to an embodiment of the present invention, as shown in fig. 2, the autonomous vehicle body is equipped with a radar L (shown as a circle L in fig. 2), a radar R (shown as a circle R in fig. 2) and a radar H (shown as a circle H in fig. 2) around the autonomous vehicle body, and a plurality of rectangular or square white boards are arranged in front of the vehicle. Each radar scans a plurality of surrounding white boards and stores multi-frame point cloud data under respective radar folders. For example, the calibration process of the radar L is to scan a plurality of white boards around, fuse the acquired multi-frame point cloud data, extract a white board center point, and convert the center points of the plurality of white boards into coordinates in an independent coordinate system of the radar L as a first point set. Setting an original point of a coordinate system of the vehicle body (as shown by a black five-pointed star in fig. 2), accurately measuring a plurality of whiteboard center points around through a total station, converting the measured coordinates into coordinates in the coordinate system of the vehicle body as a second point set, and setting the coordinate values as truth values of the whiteboard center points. And registering the coordinates of the same whiteboard center point in the first point set and the second point set, and converting the calculated rotation translation matrix into a vehicle body coordinate system as an independent coordinate system of the radar L.
In the embodiment of the invention, in the process of sensing and judging a target vehicle by using a target radar to assist automatic driving, an automatic calibration method is set, a plurality of targets are distributed in front of the target vehicle, the target radar is used for scanning the targets to form multi-frame point cloud data, the coordinates of the target center point under an independent coordinate system of the target radar are used as a first point set, the coordinates of a white board center point under a vehicle body coordinate system, which are measured with high precision, are used as a second point set, the same target center point in the first point set and the second point set is registered, and a calculated rotation translation matrix is used as a conversion parameter for converting the independent coordinate system of the radar into the vehicle body coordinate system, namely the target calibration parameter. It is easy to notice that the method provides a convenient and efficient automatic calibration method, and the method fuses multi-frame point cloud data obtained by scanning a target radar as a first point set, so that the point cloud density is increased, the defect of sparse point cloud of the solid-state radar is effectively overcome, and the precision of extracting the center point of a white board is ensured, thereby ensuring the validity of a registration result, improving the efficiency of automatic calibration, and further solving the technical problem that the efficiency of automatic calibration of the vehicle-mounted multi-solid-state radar in the prior art is low.
Optionally, according to the method in the embodiment of the present invention, registering the first point set and the second point set, and determining a target calibration parameter of the target radar includes: registering a first coordinate in the first point set and a second coordinate in the second point set to obtain an initial calibration parameter of the target radar and a registration error corresponding to the first coordinate, wherein the first coordinate and the second coordinate correspond to a central point of the same target; in response to the fact that the number of the first coordinates in the first point set is larger than the preset number, eliminating the first coordinates with the registration errors larger than the preset errors in the first point set, and registering the first coordinates in the first point set with the second coordinates in the second point set again; and determining the initial calibration parameters as the target calibration parameters in response to the number of the first coordinates contained in the first point set being less than or equal to the preset number.
The initial calibration parameter may be a rotational-translational matrix calculated when the first point set and the second point set are subjected to the registration of the same target central point for the first time. The registration error can be the distance between two central points when the two sets of point sets are registered and the central points of the same white board are registered. The preset error may be the largest error value when registering for two sets of point sets. The preset number can be the number of points needing registration in the set point set when the central points of the white boards of the two coordinate systems are registered, and if the number of the points exceeds the preset number, the central points with registration errors larger than the preset errors need to be removed. The target calibration parameters can be finally in accordance with the conversion parameters of the radar independent coordinate system to the vehicle body coordinate system, high-efficiency automatic calibration can be realized, and the view fields of a plurality of solid-state radars are unified to the view field of the vehicle body coordinate system.
In an optional embodiment, fig. 3 is an optional in-thread multi-frame point cloud calibration flowchart according to an embodiment of the present invention, and as shown in fig. 3, NLOPT registration is performed on a whiteboard center coordinate scanned by a radar, i.e., a first point set, and a whiteboard center coordinate measured and sorted by a total station, i.e., a second point set, according to a relative position relationship of a whiteboard. When the registration is carried out for the first time, a rotation translation matrix of two sets of points, which are concentrated in the central point of the same white board, is calculated, the rotation translation matrix is used as an initial calibration parameter, the registration error of each point is obtained, and if the calculation internal accuracy is met, the optimal calibration parameter can be stored. When all the points in the two sets of point sets are registered, if the number of the points in the point sets exceeds 3, the points with the largest registration error are removed, the remaining points in the two point sets are registered again, the calibration parameters are calculated, and the operation is repeated until 3 points remain in the point sets. When all the points in the two sets of point sets are aligned, if the number of the points is equal to or less than 3, determining that the initial calibration parameter is the target calibration parameter.
It should be noted that in this process, the registration of the two sets of point sets is completed by using NLOPT, which is used to solve some complex nonlinear optimization problems. The point with the largest registration error is removed successively through multiple times of registration, so that the optimal point is selected from multiple times of registration results, and the multipoint registration accuracy is guaranteed.
Optionally, according to the method in the embodiment of the present invention, registering a first coordinate in the first point set with a second coordinate in the second point set to obtain an initial calibration parameter of the target radar and a registration error corresponding to the first coordinate, includes: determining a first coordinate and a second coordinate corresponding to the same target in the first point set and the second point set based on the relative position relation of the targets; and registering the first coordinate and the second coordinate corresponding to the same target to obtain an initial calibration parameter and a registration error.
The first coordinate may be a coordinate of the whiteboard center point in a radar-independent coordinate system, that is, a coordinate of the first point concentration point. The second coordinate may be a coordinate of the whiteboard center point in the vehicle body coordinate system, i.e. a coordinate of the second point focus point, which is accurately measured by the total station point.
In an optional embodiment, the target calibration parameters are determined to need to be subjected to multiple registration, in the process of primary registration of a first point set and a second point set, according to the relative position relationship of white boards in various shapes in front of a vehicle, coordinates in the first point set and the second point set are determined to be obtained through conversion of the same white board central point, the coordinates of the same white board central point in the first point set and the second point set are registered, when the two sets of points are registered by using an NLOPT method, a rotation and translation matrix of the central point of the same white board in the two sets of point sets is obtained through calculation and is set as the initial calibration parameters, and the distance between the central points of the same white board in the two sets of point sets is used as a registration error.
Optionally, according to the method of the above embodiment of the invention, the method further includes: acquiring accurate position information of central points of a plurality of targets acquired by a total station; and converting the position information into a second coordinate system to obtain a second point set.
The total station, namely the total station type electronic distance measuring instrument, is a high-tech measuring instrument integrating light collection, machinery and electricity, and can accurately measure target position information. The second coordinate system may be a vehicle body coordinate system having a ground projection of a center point of a front axle as an origin O, a vehicle advancing direction as an X-axis, a vertically upward facing direction as a Z-axis, and a direction perpendicular to XOZ and left of the advancing direction as a Y-axis.
In an optional embodiment, fig. 4 is a flow chart of optional multi-frame point cloud calibration according to an embodiment of the present invention, and as shown in fig. 4, by using characteristics of a total station high-precision measurement point, position information of a plurality of whiteboard center points in front of a vehicle is measured as a true value, coordinates of the whiteboard center point position information acquired by the total station in a vehicle body coordinate system are acquired to form a total station point set a, the total station point set a is sorted according to a size of a coordinate on a Y axis, and is used as a second point set to perform multi-thread processing on a plurality of radar calibration data.
Optionally, according to the method in the embodiment of the present invention, fusing the multi-frame point cloud data to determine the first point set, includes: fusing multi-frame point cloud data to obtain a first point cloud; extracting target points belonging to a plurality of targets in the first point cloud by using an Euclidean clustering mode; processing the target points based on the attribute parameters of the targets, and determining the central points of the targets; a first set of points is generated based on coordinates of a center point of the plurality of targets in a first coordinate system.
The first point cloud may be an effective point cloud obtained by fusing and processing multi-frame point cloud data. The point cloud fusion is to splice and combine the acquired multi-frame point cloud data into a large piece of point cloud, so that the point cloud density is increased, the defect of sparse solid-state radar point cloud can be effectively overcome, and meanwhile, the whiteboard extraction precision is ensured. The Euclidean clustering is a clustering algorithm based on Euclidean distance measurement, and the algorithm is to aggregate surrounding points, the distance of which from the points is smaller than a set distance threshold value, in a certain point into a set, so as to realize point cloud segmentation.
In an optional embodiment, the radar scans the surface information of a plurality of white boards in front of the vehicle, stores multi-frame point cloud data into the radar folder, and then enters a multi-frame point cloud calibration process in the radar folder. As shown in fig. 3, the front k-point multi-frame point cloud data in the radar folder is fused into one point cloud as a first point cloud. And realizing point cloud segmentation by using an Euclidean clustering mode, extracting point cloud data of each whiteboard, and registering and extracting whiteboard center points according to a set whiteboard model. And converting a plurality of whiteboard center points into a radar independent coordinate system, wherein the coordinates of the center points are used as a first point set.
Optionally, according to the method in the embodiment of the present invention, before extracting target points belonging to a plurality of targets in the first point cloud by using an euclidean clustering method, the method further includes: cutting the first point cloud based on the target coordinate ranges corresponding to the multiple targets to obtain a second point cloud; denoising and outlier removing are carried out on the second point cloud to obtain a third point cloud; and extracting target points belonging to a plurality of targets in the third point cloud by using an Euclidean clustering mode.
The second point cloud can be an effective point cloud obtained by cutting the first point cloud subjected to fusion according to the target model. The third point cloud may be an effective point cloud after denoising and outlier removing processing is performed on the second point cloud. Because the target object surface information scanned by the radar may be randomly influenced by the environment, some effective information is often ignored, so that a denoising operation is required to filter noise-containing information. Outliers are extreme points that deviate from the majority of the point cloud and are invalid points that are accidentally scanned due to some environmental or hardware factors.
In an alternative embodiment, as shown in fig. 3, in each radar thread, the first point cloud formed by fusing the point cloud data of the previous k frames is segmented. And cutting the first point cloud according to a coordinate range ROI (Region Of Interest) Of the corresponding whiteboard Region to serve as a second point cloud, wherein the Region is the key point for point cloud analysis. And denoising and separating point group removing processing are carried out on the second point cloud area data after cutting, and the processed point cloud is used as a third point cloud. The point cloud segmentation method comprises the steps of utilizing an Euclidean clustering mode to achieve point cloud segmentation, extracting a white board central point, setting the shape and the size of a white board model according to a plurality of white boards arranged in front of a vehicle, sequentially fitting the judged white board point cloud edges by using the designed white board model, and determining the accurate position of the white board point cloud according to the minimum distance from points on each edge to a template edge, so that the white board central point can be calculated.
Optionally, according to the method in the embodiment of the present invention, in a case where multi-frame point cloud data is obtained for multiple times, registering the first point set and the second point set, and determining a target calibration parameter of the target radar includes: registering the first point set and the second point set for multiple times to obtain multiple calibration parameters of the target radar; and obtaining the average value of the plurality of calibration parameters to obtain the target calibration parameter.
As shown in fig. 4, after each radar folder is input, the number of frames k to be calculated each time is input by the user, and in each radar point cloud folder, the total number of times N that each folder needs to calculate is calculated as Z/k according to the total number of frames Z, and then the total number of times to perform calibration is N. And calculating N times of average calibration parameters as target calibration parameters when the radar calibration data is processed in a multithread mode according to the first point set, the second point set and the white board range region ROI.
In an alternative embodiment, as shown in fig. 3, when the first point set and the second point set are registered to determine the target calibration parameters of the target radar, in the radar thread, the first k frames of point cloud data in the radar folder are calibrated, and the calibration parameters of the k frames of point cloud of the radar are calculated, where the process is a calibration. Since each radar folder needs to be calculated for N times, N calibration parameters are obtained after the in-thread calibration process is performed for N times. And calculating the average value of the N calibration parameters as a target calibration parameter, and outputting a Yaml result file.
Optionally, according to the method in the embodiment of the present invention, when the target radar is a plurality of radars, a plurality of threads are started, the plurality of threads correspond to the plurality of radars one to one, each thread is configured to fuse multi-frame point cloud data acquired by the radars corresponding to the thread, determine a point set corresponding to the radar, and register the point set corresponding to the radar with a second point set to obtain calibration parameters of the radar.
The thread may refer to a single sequential control flow in a process, and multiple threads may be concurrently executed in one process, each thread executing different tasks in parallel. The process of calibrating data in one radar folder is one thread, a plurality of radars correspond to a plurality of threads, and data in different radar folders are processed simultaneously.
In an alternative embodiment, a plurality of solid-state radars are arranged around the vehicle body, and a plurality of frames of point cloud data obtained by scanning a plurality of targets in front of the vehicle need to be stored in each radar folder. And starting a plurality of threads, correspondingly processing one radar folder by each thread, fusing multi-frame point cloud data in the corresponding radar folder in each thread, determining an electric set corresponding to the radar, and registering a point set corresponding to the radar with a second point set to obtain calibration parameters of the radar. And after multi-thread simultaneous calibration is carried out on the multi-frame point cloud data of each radar, the conversion parameters of each radar coordinate system to the vehicle body coordinate system can be obtained simultaneously.
It should be noted that, according to the efficient automatic calibration method, the total station can be replaced by a high-line-number mechanical radar, the dense point cloud of the white board is collected by the high-line-number mechanical radar, the distance between the midpoint of the point cloud is smaller, and the collected white board is more accurate, so that the center point of the white board is accurately obtained and used as an accurate value. The whiteboard marker can be replaced by other markers, such as checkerboards, two-dimensional code boards, round targets, high-reflection paillettes and the like, and calibration can be completed. The method has the advantages that:
1. for an automobile with a body carrying a solid-state radar, a convenient and efficient automatic calibration method is provided, the characteristics of a total station high-precision measuring point are utilized, a whiteboard center point is measured to serve as a true value, the whiteboard center point collected by the radar is extracted and is registered with the total station point, and therefore all solid-state radar point clouds are unified under an automobile body coordinate system, and automatic driving data processing and obstacle sensing processing are facilitated.
2. And multi-thread simultaneous calibration is carried out on multi-frame point cloud data of each radar, so that the automatic calibration process is effectively accelerated, and the beat time of factory calibration is saved for factory automatic calibration which is necessary for minutes and seconds.
3. And the multi-frame point cloud fusion calibration is carried out, so that the point cloud density is increased, the defect of sparse solid radar point cloud is effectively overcome, and the whiteboard extraction precision is ensured.
4. By means of the template sleeve box white board point cloud, the problem that white board corner points are incompletely extracted due to sparse solid radar point cloud and different accuracy and density of horizontal and vertical direction points in the white board point cloud acquisition process is effectively solved. The precision and accuracy of whiteboard center point extraction are effectively improved.
5. When point sets are registered, two sets of point sets with the number of matching points exceeding 3 are aligned, points with the largest registration error are removed after each registration, then the remaining points are used for continuing the matching until the matching precision is very high or only 3 points remain in the point sets, the optimal point is selected from multiple registration results, and the multipoint registration precision is guaranteed.
Example 2
According to the embodiment of the present invention, a calibration apparatus for radar parameters is further provided, where the apparatus may execute the motor control method in the foregoing embodiment, and a specific implementation manner and a preferred application scenario are the same as those in the foregoing embodiment, and are not described herein again.
Fig. 5 is a schematic diagram of a radar parameter calibration apparatus according to an embodiment of the present invention, as shown in fig. 5, the apparatus includes:
the acquisition module 52 is configured to acquire multi-frame point cloud data acquired by a target radar installed on a target vehicle, where the multi-frame point cloud data is obtained by sensing multiple targets in front of the target vehicle by the target radar;
a fusion module 54, configured to fuse the multi-frame point cloud data, and determine a first point set, where the first point set includes: coordinates of center points of the multiple targets in a first coordinate system corresponding to the target radar;
a registration module 56, configured to register the first point set with a second point set, and determine a target calibration parameter of the target radar, where the second point set includes: and the calibration parameters are used for representing conversion parameters for converting the coordinates in the first coordinate system to the coordinates in the second coordinate system.
The acquisition module, the fusion module and the registration module are all included in a target vehicle system, and the control system equipped in the target vehicle is used for realizing the functions of the functional modules and processing data. The acquisition module can be used for a radar scanning module and is used for scanning the surface information of a plurality of targets arranged in front of the vehicle by a radar, acquiring multi-frame point cloud data and storing the multi-frame point cloud data into a radar folder of the radar. The fusion module can be used for selecting front k frames of point cloud files of multiple frames of point cloud data in the folder, fusing the front k frames of point cloud files into one piece of point cloud, increasing the density of the point cloud, performing data processing, extracting and converting a central point into a radar independent coordinate system, and forming a first point set by a coordinate set of the radar independent coordinate system. The registration module may be configured to register a same target center point of the first point set and the second point set, and calculate a rotation-translation matrix of the same target center point of the two sets of point sets during registration as a conversion parameter.
For example, when a plurality of solid-state radars of a vehicle body of a target automatic driving vehicle are subjected to multi-frame point cloud automatic calibration, the acquisition module is used, namely the radars are used for scanning a plurality of white boards arranged in front of the vehicle, multi-frame point cloud data are acquired and stored into a radar folder of the radars, and therefore the radar can be used for simultaneously processing and calibrating the multi-frame point cloud data of all the radars in the follow-up process, and the automatic calibration process is effectively quickened. The method comprises the steps of selecting k frames of point cloud files in front of multi-frame point cloud data in files by using a fusion module, fusing the k frames of point cloud files into one piece of point cloud, increasing the density of the point cloud, performing data processing, performing fusion calibration on the multi-frame point cloud, increasing the density of the point cloud, effectively solving the defect of point cloud sparseness of a solid-state radar, ensuring the precision of white board extraction at the same time, extracting and converting the central point into a radar independent coordinate system, and forming a first point set by using a coordinate set of the white board. And registering the coordinate of the radar independent coordinate system and the coordinate of the whiteboard center point, namely a first point set, and the coordinate of the whiteboard center point measured by the total station in a vehicle body coordinate system, namely a second point set by using a registration module, and calculating a registered rotation and translation matrix as a conversion parameter to determine a target calibration parameter of the target radar.
Optionally, according to the above-mentioned embodiment of the invention, the registration module includes: the point set registration unit is used for registering a first coordinate in the first point set and a second coordinate in the second point set to obtain an initial calibration parameter of the target radar and a registration error corresponding to the first coordinate, wherein the first coordinate and the second coordinate correspond to a central point of the same target; the dead pixel removing unit is used for removing the first coordinates with the registration errors larger than the preset errors in the first point set in response to the fact that the number of the first coordinates in the first point set is larger than the preset number, and registering the first coordinates in the first point set with the second coordinates in the second point set again; and the parameter determining unit is used for determining the initial calibration parameter as the target calibration parameter in response to the condition that the number of the first coordinates contained in the first point set is less than or equal to the preset number.
Optionally, according to the embodiment of the present invention, the point set registering unit is further configured to determine, according to the relative position relationship between the multiple targets, a first coordinate and a second coordinate corresponding to the same target in the first point set and the second point set; and registering the first coordinate and the second coordinate corresponding to the same target to obtain an initial calibration parameter and a registration error.
Optionally, according to the above embodiment of the invention, the apparatus further includes: the acquisition module is also used for acquiring the accurate position information of the central points of the multiple targets acquired by the total station; and the conversion module is used for converting the position information into a second coordinate system to obtain a second point set.
Optionally, according to the above embodiment of the invention, the fusion module includes: the system comprises a data processing module, a data processing module and a data processing module, wherein the data processing module is used for fusing multi-frame point cloud data to obtain a first point cloud; extracting target points belonging to a plurality of targets in the first point cloud by using an Euclidean clustering mode; processing the target points based on the attribute parameters of the targets, and determining the central points of the targets; a first set of points is generated based on coordinates of a center point of the plurality of targets in a first coordinate system.
Optionally, according to the embodiment of the present invention, the fusion module further includes: the point cloud cutting unit can be used for cutting the first point cloud in a target coordinate range corresponding to the plurality of targets to obtain a second point cloud; a de-noising and outlier removing unit which is used for de-noising and outlier removing the second point cloud to obtain a third point cloud; and the Euclidean clustering unit extracts target points belonging to a plurality of targets in the third point cloud by using the Euclidean clustering mode.
Optionally, according to the embodiment of the present invention, the registration module is further configured to perform registration on the first point set and the second point set for multiple times to obtain multiple calibration parameters of the target radar under the condition that multiple frames of point cloud data are obtained for multiple times; and obtaining the average value of the plurality of calibration parameters to obtain the target calibration parameter.
Optionally, according to the embodiment of the present invention, when the target radar is a plurality of radars, a plurality of threads are started, the plurality of threads correspond to the plurality of radars one to one, each thread is configured to fuse multi-frame point cloud data acquired by the radars corresponding to the thread, determine a point set corresponding to the radar, and register the point set corresponding to the radar with a second point set to obtain calibration parameters of the radar.
Example 3
According to another aspect of the embodiment of the invention, a vehicle is further provided, and the vehicle comprises the radar parameter calibration device in any one of the above embodiments.
Example 4
According to another aspect of the embodiments of the present invention, a non-volatile storage medium is further provided, where the non-volatile storage medium includes a stored program, and a processor of a device is controlled to execute the method for calibrating a radar parameter in any one of the above embodiments when the program runs.
Example 5
According to another aspect of the embodiments of the present invention, there is further provided a processor, where the processor is configured to execute a program, where the program executes the method for calibrating radar parameters in any one of the above embodiments.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
In the above embodiments of the present invention, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.
In the embodiments provided in the present application, it should be understood that the disclosed technology can be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units may be a logical division, and in actual implementation, there may be another division, for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, units or modules, and may be in an electrical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one position, or may be distributed on a plurality of units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic or optical disk, and other various media capable of storing program codes.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.
Claims (10)
1. A calibration method of radar parameters is characterized by comprising the following steps:
acquiring multi-frame point cloud data collected by a target radar installed on a target vehicle, wherein the multi-frame point cloud data is obtained by sensing a plurality of targets in front of the target vehicle by the target radar;
fusing the multi-frame point cloud data, and determining a first point set, wherein the first point set comprises: coordinates of center points of the plurality of targets in a first coordinate system corresponding to the target radar;
registering the first point set and a second point set, and determining a target calibration parameter of the target radar, wherein the second point set comprises: coordinates of the center points of the targets in a second coordinate system corresponding to the target vehicle, wherein the calibration parameters are used for representing conversion parameters for converting the coordinates in the first coordinate system to the coordinates in the second coordinate system.
2. The method of claim 1, wherein registering the first set of points with a second set of points, determining target calibration parameters for the target radar, comprises:
registering a first coordinate in the first point set with a second coordinate in the second point set to obtain an initial calibration parameter of the target radar and a registration error corresponding to the first coordinate, wherein the first coordinate and the second coordinate correspond to a central point of the same target;
in response to the fact that the number of the first coordinates in the first point set is larger than a preset number, eliminating the first coordinates with the registration errors larger than the preset errors in the first point set, and registering the first coordinates in the first point set with the second coordinates in the second point set again;
and determining the initial calibration parameters as the target calibration parameters in response to the number of the first coordinates contained in the first point set being less than or equal to the preset number.
3. The method according to claim 2, wherein registering a first coordinate in the first point set with a second coordinate in the second point set to obtain an initial calibration parameter of the target radar and a registration error corresponding to the first coordinate comprises:
determining a first coordinate and a second coordinate corresponding to the same target in the first point set and the second point set based on the relative position relationship of the plurality of targets;
and registering the first coordinate and the second coordinate corresponding to the same target to obtain the initial calibration parameter and the registration error.
4. The method of claim 2, further comprising:
acquiring accurate position information of central points of the multiple targets acquired by a total station;
and converting the position information into the second coordinate system to obtain the second point set.
5. The method of claim 1, wherein fusing the plurality of frames of point cloud data to determine a first set of points comprises:
fusing the multi-frame point cloud data to obtain a first point cloud;
extracting target points belonging to the multiple targets in the first point cloud by using an Euclidean clustering mode;
processing the target points based on the attribute parameters of the targets, and determining the central points of the targets;
generating the first set of points based on coordinates of center points of the plurality of targets in the first coordinate system.
6. The method of claim 5, wherein before extracting target points belonging to the plurality of targets in the first point cloud by Euclidean clustering, the method further comprises:
cutting the first point cloud based on the target coordinate ranges corresponding to the targets to obtain a second point cloud;
denoising and outlier removing are carried out on the second point cloud to obtain a third point cloud;
and extracting target points belonging to the plurality of targets in the third point cloud by using the Euclidean clustering mode.
7. The method of claim 1, wherein registering the first point set with a second point set in a case where the multi-frame point cloud data is acquired multiple times, and determining target calibration parameters of the target radar comprises:
registering the first point set and the second point set for multiple times to obtain multiple calibration parameters of the target radar;
and obtaining the average value of the plurality of calibration parameters to obtain the target calibration parameter.
8. The method according to claim 1, wherein, when the target radar is a plurality of radars, a plurality of threads are started, the threads correspond to the radars one by one, and each thread is used for fusing multi-frame point cloud data collected by the radar corresponding to the thread, determining a point set corresponding to the radar, and registering the point set corresponding to the radar with a second point set to obtain calibration parameters of the radar.
9. A radar parameter calibration device is characterized by comprising:
the system comprises an acquisition module, a processing module and a processing module, wherein the acquisition module is used for acquiring multi-frame point cloud data acquired by a target radar installed on a target vehicle, and the multi-frame point cloud data is obtained by sensing a plurality of targets in front of the target vehicle by the target radar;
a fusion module, configured to fuse the multiple frames of point cloud data, and determine a first point set, where the first point set includes: coordinates of center points of the plurality of targets in a first coordinate system corresponding to the target radar;
a registration module, configured to register the first point set with a second point set, and determine a target calibration parameter of the target radar, where the second point set includes: coordinates of the center points of the targets in a second coordinate system corresponding to the target vehicle, wherein the calibration parameters are used for representing conversion parameters for converting the coordinates in the first coordinate system to the coordinates in the second coordinate system.
10. A vehicle, characterized by comprising: apparatus for calibration of radar parameters according to claim 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210641520.5A CN114879153A (en) | 2022-06-08 | 2022-06-08 | Radar parameter calibration method and device and vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210641520.5A CN114879153A (en) | 2022-06-08 | 2022-06-08 | Radar parameter calibration method and device and vehicle |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114879153A true CN114879153A (en) | 2022-08-09 |
Family
ID=82678785
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210641520.5A Pending CN114879153A (en) | 2022-06-08 | 2022-06-08 | Radar parameter calibration method and device and vehicle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114879153A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116468804A (en) * | 2023-04-21 | 2023-07-21 | 湖南佑湘网联智能科技有限公司 | Laser radar and camera external parameter calibration precision evaluation method and device |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110221603A (en) * | 2019-05-13 | 2019-09-10 | 浙江大学 | A kind of long-distance barrier object detecting method based on the fusion of laser radar multiframe point cloud |
| CN111429514A (en) * | 2020-03-11 | 2020-07-17 | 浙江大学 | Laser radar 3D real-time target detection method fusing multi-frame time sequence point clouds |
| CN111735479A (en) * | 2020-08-28 | 2020-10-02 | 中国计量大学 | Multi-sensor combined calibration device and method |
| CN112462350A (en) * | 2020-12-10 | 2021-03-09 | 苏州一径科技有限公司 | Radar calibration method and device, electronic equipment and storage medium |
| CN113156411A (en) * | 2021-05-03 | 2021-07-23 | 湖北汽车工业学院 | Vehicle-mounted laser radar calibration method |
| CN113674355A (en) * | 2021-07-06 | 2021-11-19 | 中国北方车辆研究所 | Target identification and positioning method based on camera and laser radar |
| CN114265042A (en) * | 2021-12-09 | 2022-04-01 | 上海禾赛科技有限公司 | Calibration method, calibration device, calibration system and readable storage medium |
| CN114359346A (en) * | 2021-12-31 | 2022-04-15 | 中国第一汽车股份有限公司 | Point cloud data processing method and device, nonvolatile storage medium and processor |
| CN114488094A (en) * | 2021-12-21 | 2022-05-13 | 武汉中海庭数据技术有限公司 | Vehicle-mounted multi-line laser radar and IMU external parameter automatic calibration method and device |
-
2022
- 2022-06-08 CN CN202210641520.5A patent/CN114879153A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110221603A (en) * | 2019-05-13 | 2019-09-10 | 浙江大学 | A kind of long-distance barrier object detecting method based on the fusion of laser radar multiframe point cloud |
| CN111429514A (en) * | 2020-03-11 | 2020-07-17 | 浙江大学 | Laser radar 3D real-time target detection method fusing multi-frame time sequence point clouds |
| CN111735479A (en) * | 2020-08-28 | 2020-10-02 | 中国计量大学 | Multi-sensor combined calibration device and method |
| CN112462350A (en) * | 2020-12-10 | 2021-03-09 | 苏州一径科技有限公司 | Radar calibration method and device, electronic equipment and storage medium |
| CN113156411A (en) * | 2021-05-03 | 2021-07-23 | 湖北汽车工业学院 | Vehicle-mounted laser radar calibration method |
| CN113674355A (en) * | 2021-07-06 | 2021-11-19 | 中国北方车辆研究所 | Target identification and positioning method based on camera and laser radar |
| CN114265042A (en) * | 2021-12-09 | 2022-04-01 | 上海禾赛科技有限公司 | Calibration method, calibration device, calibration system and readable storage medium |
| CN114488094A (en) * | 2021-12-21 | 2022-05-13 | 武汉中海庭数据技术有限公司 | Vehicle-mounted multi-line laser radar and IMU external parameter automatic calibration method and device |
| CN114359346A (en) * | 2021-12-31 | 2022-04-15 | 中国第一汽车股份有限公司 | Point cloud data processing method and device, nonvolatile storage medium and processor |
Non-Patent Citations (2)
| Title |
|---|
| 张艳国;李擎;: "基于惯性测量单元的激光雷达点云融合方法", 系统仿真学报, no. 11, 8 November 2018 (2018-11-08) * |
| 王开鑫;王世峰;孙琪;刘传义;陈森;: "点云分割匹配的三维重建算法", 长春理工大学学报(自然科学版), no. 04, 15 August 2020 (2020-08-15) * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116468804A (en) * | 2023-04-21 | 2023-07-21 | 湖南佑湘网联智能科技有限公司 | Laser radar and camera external parameter calibration precision evaluation method and device |
| CN116468804B (en) * | 2023-04-21 | 2024-04-02 | 湖南佑湘网联智能科技有限公司 | Laser radar and camera external parameter calibration precision evaluation method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108932736B (en) | Two-dimensional laser radar point cloud data processing method and dynamic robot pose calibration method | |
| CN107507167B (en) | Cargo tray detection method and system based on point cloud plane contour matching | |
| CN103047943B (en) | Based on the door skin geomery detection method of single projection coded structured light | |
| CN110517216B (en) | SLAM fusion method and system based on multiple types of cameras | |
| US20180018528A1 (en) | Detecting method and device of obstacles based on disparity map and automobile driving assistance system | |
| CN110879994A (en) | Three-dimensional visual inspection detection method, system and device based on shape attention mechanism | |
| CN112270713A (en) | Calibration method and device, storage medium and electronic device | |
| CN106897995B (en) | A kind of components automatic identifying method towards Automatic manual transmission process | |
| CN112562093B (en) | Object detection method, electronic medium, and computer storage medium | |
| CN116229408A (en) | Target identification method for fusing image information and laser radar point cloud information | |
| CN103500327A (en) | Vehicle type identification method of vehicles of same brand based on space position information | |
| CN114910891A (en) | Multi-laser radar external parameter calibration method based on non-overlapping fields of view | |
| CN107743201A (en) | A kind of quick pattern splicing method of power line gallery digital cameras in general use and device | |
| CN114879153A (en) | Radar parameter calibration method and device and vehicle | |
| CN115376109A (en) | Obstacle detection method, obstacle detection device, and storage medium | |
| CN114994647A (en) | Laser radar and camera combined external parameter calibration system and method | |
| CN115097419A (en) | External parameter calibration method and device for laser radar IMU | |
| CN104966283A (en) | Imaging layered registering method | |
| CN110675442A (en) | Local stereo matching method and system combined with target identification technology | |
| CN112578405B (en) | Method and system for removing ground based on laser radar point cloud data | |
| CN114428259A (en) | Automatic vehicle extraction method in laser point cloud of ground library based on map vehicle acquisition | |
| CN107392948B (en) | Image registration method of amplitude-division real-time polarization imaging system | |
| CN116862922B (en) | Target positioning method, system and medium based on image segmentation and radar information fusion | |
| Wujanz et al. | Defo Scan++: Surface based registration of terrestrial laser scans for deformation monitoring | |
| CN114152942B (en) | Millimeter wave radar and vision second-order fusion multi-classification target detection method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |