CN111126225A - Multiline laser radar ground segmentation method, vehicle and computer readable medium - Google Patents
Multiline laser radar ground segmentation method, vehicle and computer readable medium Download PDFInfo
- Publication number
- CN111126225A CN111126225A CN201911301701.8A CN201911301701A CN111126225A CN 111126225 A CN111126225 A CN 111126225A CN 201911301701 A CN201911301701 A CN 201911301701A CN 111126225 A CN111126225 A CN 111126225A
- Authority
- CN
- China
- Prior art keywords
- ground
- scanning point
- segmentation
- point
- mth
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- 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
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Optical Radar Systems And Details Thereof (AREA)
Abstract
The invention discloses a multiline laser radar ground segmentation method, a vehicle and a computer readable medium, wherein the multiline laser radar ground segmentation method is applied to automatic driving of the vehicle and comprises the following steps: acquiring a preprocessing result of ground detection data of the multi-line laser radar; acquiring preliminary segmentation data of ground segmentation according to a preprocessing result; and acquiring final segmentation data of the ground segmentation according to the preliminary segmentation data, and realizing the ground segmentation according to the final segmentation data. The multi-line laser radar ground segmentation method disclosed by the invention can enable the automatic driving vehicle to adapt to the driving environment with severe road surface fluctuation and complex road, and has very strong adaptability; the ground feature data of the next wire harness is segmented through the recorded actual elevation included angle between the adjacent wire harnesses, and the ground feature segmentation is more accurate and reliable; in addition, by confirming the ground feature division state for a plurality of times and further performing ground feature division screening, the possibility of false recognition of the environment perception of the vehicle can be greatly reduced.
Description
Technical Field
The invention relates to the technical field of automatic driving, in particular to a multiline laser radar ground segmentation method, a vehicle and a computer readable medium.
Background
As one of the key technologies of the automatic driving, accurate output of the environmental perception determines whether the automatic driving function is implemented or not. In the current automotive industry, multiline lidar has become a preferred environmental sensing device for all homes. The ground segmentation processing is the most front-end processing flow in the multi-line laser radar algorithm.
The existing ground segmentation processing method still has certain defects, for example, in the prior art, the top point and the lowest point of point cloud data in a polar coordinate grid are extracted by constructing a polar coordinate grid map, and are compared and judged with a set threshold value to realize the ground point segmentation in a non-edge grid, and then the ground points in an edge grid are further segmented by adopting a 3 sigma criterion to realize the ground segmentation processing; or the method comprises the steps of rasterizing all points in the region of interest under a polar coordinate grid, then calculating the maximum height difference and the average height in each grid, and finally realizing data segmentation of the ground and non-ground by comparing with a set threshold value. In the prior art, the ground segmentation processing method has poor environmental adaptability, and the ground under-segmentation or over-segmentation phenomenon can occur when the vehicle motion amplitude is large, so that the error processing of subsequent environment perception can be easily caused, and the automatic driving function is further seriously influenced.
Disclosure of Invention
Technical problem to be solved
The invention provides a multiline laser radar ground segmentation method, a vehicle and a computer readable medium, and aims to solve the technical problems that in the prior art, the environment adaptability is poor, the ground under-segmentation or over-segmentation phenomenon can occur when the vehicle motion amplitude is large, the subsequent error processing of environment perception is easily caused, and the automatic driving function is further seriously influenced.
(II) technical scheme
One aspect of the invention discloses a multiline laser radar ground segmentation method, which is applied to automatic driving of vehicles and comprises the following steps: acquiring a preprocessing result of ground detection data of the multi-line laser radar; acquiring preliminary segmentation data of ground segmentation according to a preprocessing result; and acquiring final segmentation data of the ground segmentation according to the preliminary segmentation data, and realizing the ground segmentation according to the final segmentation data.
As an embodiment of the present invention, the preprocessing result for obtaining the ground detection data of the multiline laser radar includes: analyzing the acquired ground detection data of the multi-line laser radar to a radar coordinate system of the multi-line laser radar, and sequencing the ground detection data according to the radar coordinate system according to the working mode of the multi-line laser radar to acquire a preprocessing result; the working mode is the beam sequence of the multi-line laser radar when the multi-line laser beam is used for detection.
As an embodiment of the present invention, acquiring preliminary wire harness data of ground segmentation according to a preprocessing result includes: acquiring the horizontal distance difference delta d between the m-th scanning point on the n-th beam and the m-th scanning point on the (n + 1) -th beam adjacent to the m-th scanning pointmnIncluded angle delta theta with elevationmn(ii) a And according to the difference of horizontal distance DeltadmnWith a predetermined distance threshold dwJudging the ground segmentation characteristics of the mth scanning point on the nth wire harness and the mth scanning point on the (n + 1) th wire harness adjacent to the mth scanning point according to the relationship between the mth scanning point and the mth wire harness; or according to elevation angle delta thetamnThreshold value theta of included angle with preset valuewJudging the ground segmentation characteristics of the mth scanning point on the nth wire harness and the mth scanning point on the (n + 1) th wire harness adjacent to the mth scanning point according to the relationship between the mth scanning point and the mth wire harness; the ground segmentation features comprise temporary ground points, ground points and non-ground points, n and m are positive integers, n is not less than 1, m is not less than 1, w is n, and n-1 is not less than 1.
As an embodiment of the present invention, a horizontal distance difference Δ d between an mth scanning point on an nth beam and an mth scanning point on an adjacent (n + 1) th beam is obtainedmnThe method comprises the following steps: acquiring the horizontal distance d of the m scanning point on the n beammnThe following formula is satisfied:
wherein (x)mn，ymn) The horizontal plane coordinate of the mth scanning point on the corresponding nth wire harness is obtained; acquiring the horizontal distance d of the mth scanning point corresponding to the adjacent nth wire harness on the (n + 1) th wire harnessm(n+1)The following formula is satisfied:
wherein (x)m(n+1)，ym(n+1)) The horizontal plane coordinate of the m-th scanning point on the n +1 th beam and the horizontal distance d of the m-th scanning point on the n-th beammnAnd a horizontal distance d from the m-th scanning point on the (n + 1) -th beam adjacent theretom(n+1)Difference of horizontal distance Δ d therebetweenmnThe following formula is satisfied:
Δdmn＝dm(n+1)－dmn
wherein n is a positive integer, and n is more than or equal to 1.
As an embodiment of the invention, an elevation included angle delta theta between the mth scanning point on the nth beam and the mth scanning point on the adjacent (n + 1) th beam is obtainedmnThe method comprises the following steps: according to the difference of horizontal distance DeltadmnObtaining elevation included angle delta thetamnThe following formula is satisfied:
wherein z ismnIs the vertical coordinate, z, of the m-th scanning point on the n-th beamm(n+1)Is the vertical coordinate of the m-th scanning point on the (n + 1) -th beam corresponding to the adjacent beam, n is a positive integer, and n is more than or equal to 1.
According to an embodiment of the present invention, the horizontal distance difference Δ d is usedmnWith a predetermined distance threshold dwThe ground segmentation characteristics of the mth scanning point on the nth wire harness and the mth scanning point on the (n + 1) th wire harness adjacent to the mth scanning point are judged according to the size relationship between the mth scanning point and the mth wire harness; or according to elevation angle delta thetamnThreshold value theta of included angle with preset valuewThe size relationship between the ground segmentation features and the ground segmentation features of the mth scanning point on the nth beam and the mth scanning point on the (n + 1) th beam adjacent to the mth scanning point is judged, and the ground segmentation features comprise the following steps: when n is 1, the ground segmentation characteristics of the mth scanning point on the 1 st beam and the mth scanning point on the 2 nd beam adjacent to the mth scanning point meet the following conditions: when Δ dm1<d1Or | Δ θm1|>θ1If so, the m-th scanning point on the 1 st wire harness and the m-th scanning point on the 2 nd wire harness are non-ground points; when Δ dm1≥d1Or | Δ θm1|≤θ1When the number of the scanning points is larger than the preset value, the m-th scanning point on the 1 st wire harness is a ground point, and the m-th scanning point on the 2 nd wire harness is a temporary ground point; when n is larger than or equal to 2 and the mth scanning point on the nth wire harness is a temporary ground point, the ground segmentation characteristics of the mth scanning point on the nth wire harness meet the following conditions: when Δ dmn<dwWhen the beam is not ground, the m-th scanning point on the n-th beam is a non-ground point, and when the beam is delta dmn≥dwAnd in the process, the m-th scanning point on the n-th wiring harness is a ground point.
According to an embodiment of the present invention, the horizontal distance difference Δ d is usedmnWith a predetermined distance threshold dwThe ground segmentation characteristics of the mth scanning point on the nth wire harness and the mth scanning point on the (n + 1) th wire harness adjacent to the mth scanning point are judged according to the size relationship between the mth scanning point and the mth wire harness; or according to elevation angle delta thetamnThreshold value theta of included angle with preset valuewThe size relationship between the ground segmentation features and the ground segmentation features of the mth scanning point on the nth beam and the mth scanning point on the (n + 1) th beam adjacent to the mth scanning point is judged, and the ground segmentation features comprise the following steps: when n is larger than or equal to 1 and the mth scanning point on the nth wire harness is a ground point, the ground segmentation characteristics of the mth scanning point on the (n + 1) th wire harness meet the following conditions: when Δ dmn<dwOr | Δ θmn|>θwIf so, the m scanning point on the (n + 1) th wiring harness is a non-ground point; when Δ dmn≥dwOr | Δ θmn|≤θwIn the process, the m-th scanning point on the (n + 1) -th wiring harness is a ground point; and when n is larger than or equal to 1 and the mth scanning point on the nth wire harness is a non-ground point, the ground segmentation characteristic of the mth scanning point on the (n + 1) th wire harness meets the following conditions: when Δ dmn<dwOr | Δ θmn|>θwThen m on the (n + 1) th wire harnessThe scanning points are non-ground points; when | Δ θmn|≤θwAnd in the meantime, the m-th scanning point on the (n + 1) -th wiring harness is a ground point.
According to an embodiment of the present invention, the horizontal distance difference Δ d is usedmnWith a predetermined distance threshold dwThe ground segmentation characteristics of the mth scanning point on the nth wire harness and the mth scanning point on the (n + 1) th wire harness adjacent to the mth scanning point are judged according to the size relationship between the mth scanning point and the mth wire harness; or according to elevation angle delta thetamnThreshold value theta of included angle with preset valuewThe size relationship between the m scanning point and the ground segmentation feature of the m scanning point on the n beam and the adjacent n +1 beam is judged, and the method further comprises the following steps: when n is larger than or equal to 2, the ground segmentation characteristic of the mth scanning point on the (n + 1) th wire harness meets the following conditions: when | Δ θmn|＞|Δθrecord|+θwIf so, the m scanning point on the (n + 1) th wiring harness is a non-ground point; when | Δ θmn|≤|Δθrecord|+θwIn the process, the m-th scanning point on the (n + 1) -th wiring harness is a ground point; wherein, the elevation angle delta theta is recordedrecordEqual to the elevation angle delta theta between the m-th scanning point on the n-1 th beam and the m-th scanning point on the n-th beam adjacent to the m-th scanning pointm(n-1)。
According to an embodiment of the present invention, the horizontal distance difference Δ d is usedmnWith a predetermined distance threshold dwThe ground segmentation characteristics of the mth scanning point on the nth wire harness and the mth scanning point on the (n + 1) th wire harness adjacent to the mth scanning point are judged according to the size relationship between the mth scanning point and the mth wire harness; or according to elevation angle delta thetamnThreshold value theta of included angle with preset valuewThe size relationship between the m scanning point and the ground segmentation feature of the m scanning point on the n beam and the adjacent n +1 beam is judged, and the method further comprises the following steps: and when the ground segmentation feature of the mth scanning point on the nth wire harness is judged to be an invalid segmentation feature, taking the m +1 scanning point on the nth wire harness as an initial scanning point to continue judging the ground segmentation feature so as to obtain preliminary segmentation data of the ground segmentation.
As an embodiment of the present invention, obtaining final segmentation data of ground segmentation according to the preliminary segmentation data includes: performing surface fitting on all ground points in a certain range of the vehicle in the obtained preliminary segmentation data to obtain a ground elevation value of each ground point; and finally segmenting each ground point according to the relationship between the ground elevation value of each ground point and the corresponding measured elevation value of the ground point so as to obtain the final segmentation data of all the ground points.
The invention also discloses a vehicle, and automatic driving of the vehicle is realized according to the multiline laser radar ground segmentation method.
In another aspect of the present invention, a computer-readable medium is disclosed, which comprises a memory and a processor, wherein the memory stores executable instructions, and the instructions, when executed by the processor, implement the above-mentioned multiline lidar ground segmentation method.
(III) advantageous effects
One aspect of the invention discloses a multiline laser radar ground segmentation method, which is applied to automatic driving of vehicles and comprises the following steps: acquiring a preprocessing result of ground detection data of the multi-line laser radar; acquiring preliminary segmentation data of ground segmentation according to a preprocessing result; and acquiring final segmentation data of the ground segmentation according to the preliminary segmentation data, and realizing the ground segmentation according to the final segmentation data. Therefore, the multiline laser radar ground segmentation method disclosed by the invention can enable the automatic driving vehicle to adapt to the driving environment with severe road surface fluctuation and complex road, and has very strong adaptability; the ground feature data of the next wire harness is segmented through the recorded actual elevation included angle between the adjacent wire harnesses, and the ground feature segmentation is more accurate and reliable; in addition, by confirming the ground feature division state for a plurality of times and further performing ground feature division screening, the possibility of false recognition of the environment perception of the vehicle can be greatly reduced.
Drawings
FIG. 1 is a flow chart illustrating a multiline lidar ground segmentation method according to an embodiment of the present invention;
FIG. 2A is a schematic diagram illustrating the definition of a laser beam for a multiline lidar in accordance with an embodiment of the present invention;
FIG. 2B is a schematic diagram illustrating the definition of scanning points on a laser beam of the multi-line lidar in an embodiment of the present invention;
FIG. 2C is a schematic diagram of a radar coordinate system of the multiline lidar in an embodiment of the present invention;
FIG. 3 is a schematic diagram illustrating a coordinate relationship between an m-th scanning point on an n-th beam and an m-th scanning point on an adjacent n + 1-th beam according to an embodiment of the present invention;
fig. 4 is a schematic distribution range diagram of the surface-fitted ground point in an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to specific embodiments and the accompanying drawings.
The automatic driving technology of the vehicle relates to the environment perception function of the vehicle, and as a device for environment perception application, the multi-line laser radar can be arranged on the body of the vehicle to detect the environment around the vehicle. Particularly, the present invention is directed to a ground environment around a vehicle or a non-ground environment around the vehicle, and the present invention is mainly directed to the ground environment around the vehicle, including a driving environment with complex road conditions, such as an automatic driving of a complex road such as a mining area, a mountain area, etc.
One aspect of the present invention discloses a multiline laser radar ground segmentation method, which is applied to automatic driving of a vehicle, as shown in fig. 1, and includes:
s110, acquiring a preprocessing result of ground detection data of the multi-line laser radar;
s120, acquiring primary segmentation data of ground segmentation according to the preprocessing result; and
and S130, acquiring final segmentation data of the ground segmentation according to the preliminary segmentation data, and realizing the ground segmentation according to the final segmentation data.
Specifically, the ground environment is subjected to multi-line laser scanning detection by using the multi-line laser radar so as to obtain ground detection data of the ground environment scanned and detected by the light laser radar, and the ground detection data is preprocessed so as to obtain a preprocessing result of the ground detection data. The ground object segmentation processing of the preprocessing result is used for realizing the primary segmentation of the ground segmentation and obtaining the primary segmentation data after the process; and finally, further screening and segmenting according to the preliminary segmentation data to obtain final segmentation data of the final ground environment, and realizing ground segmentation of the ground environment by the vehicle by utilizing the final segmentation data.
Therefore, the multiline laser radar ground segmentation method disclosed by the invention can enable the automatic driving vehicle to adapt to the driving environment with severe road surface fluctuation and complex road, and has very strong adaptability; the ground feature data of the next wire harness is segmented through the recorded actual elevation included angle between the adjacent wire harnesses, and the ground feature segmentation is more accurate and reliable; in addition, by confirming the ground feature division state for a plurality of times and further performing ground feature division screening, the possibility of false recognition of the environment perception of the vehicle can be greatly reduced.
According to the embodiment of the disclosure, the preprocessing result of the ground detection data of the multi-line laser radar is obtained, which comprises the following steps: analyzing the acquired ground detection data of the multi-line laser radar to a radar coordinate system of the multi-line laser radar, and sequencing the ground detection data according to the radar coordinate system according to the working mode of the multi-line laser radar to acquire a preprocessing result; the working mode is the beam sequence of the multi-line laser radar when the multi-line laser beam is used for detection.
Specifically, as shown in fig. 2A-2C, the multiline lidar data acquisition sends multiline laser through the radar body, and then receives scanning detection imaging of the multiline laser to form detection feedback data. The multi-line laser radar adopted in the invention can be provided with a plurality of laser beam emitting sources, the laser beam emitting sources are driven by a radar motor to rotate together, the laser beam emitting sources emit laser beams once every fixed time (nanosecond level) and receive the reflected laser beams, a plurality of scanning detection processes of the laser beams can be formed within a certain time, and the distance between the environment and the vehicle can be calculated through the time difference between the emission and the reception of each scanning detection of the laser beams. After the scanning detection is completed by the multi-line laser radar, the obtained environment detection data can be transmitted, for example, the environment detection data can be transmitted to a data processing system of a vehicle for preprocessing in an ethernet mode, and a corresponding data processing system can be arranged on the multi-line laser radar for directly preprocessing.
It should be noted that the motor of the multi-line lidar drives the plurality of emission sources to rotate, wherein a plurality of laser line beams for detection and scanning are generated, and the plurality of laser line beams are defined, so that the line beam number of the multi-line lidar can be obtained, for example, the laser line beams for specific scanning detection are N, which can be defined as N line beams, each line beam is defined as the 1 st to the N th line beams from bottom to top according to the sequence in the scanning process, N is greater than or equal to 1, and N is an integer, as shown in fig. 2A. For example, if the multi-line radar is 16-line, N is 16, and the number of lines is defined as N, and the lines are numbered from bottom to top, thereby obtaining the 1 st to 16 th lines.
The motor drives the radar emission source to rotate for a circle for scanning detection, the number of the acquisition scanning points of the corresponding single scanning laser beam can be M, M is not less than 1, and M is an integer. In an embodiment of the present invention, the radar scan output is selected to be 270 degrees, as shown in FIG. 2B. M may also be determined by the number of scanning points on each laser beam, for example, 270 degrees is scanned and output by the radar, each line can output 800 scanning points, and then M is 800, the scanning points on the beam may be numbered according to the scanning point monitoring sequence, so as to obtain the 1 st to 800 th scanning points on the beam. Therefore, when the number of scanning points on each laser beam is M, the scanning points can be defined as the 1 st to mth scanning points according to the sequential scanning order of each scanning point, that is, each set of environment detection data can be stored as an N × M array.
Therefore, the working mode also comprises the scanning point sequence when the multiline laser radar carries out detection by using a single laser beam. In addition, the N multiplied by M array of the invention can store data according to a fixed scanning detection sequence, for example, the data are orderly arranged according to the number of the wire harness from small to large when storing the radar data.
Each scanning point on each laser beam is defined in the space coordinate system of the radar, and the environment detection data can be analyzed to the radar coordinate system corresponding to the (x, y, z) coordinate value on the space coordinate system of the radar. The corresponding radar coordinate system is a space coordinate system which takes the vehicle as an origin, defines the front of the vehicle as Y, the right side as X and upwards as Z.
According to the embodiment of the disclosure, the obtaining of the preliminary wire harness data of the ground segmentation according to the preprocessing result comprises:
acquiring the horizontal distance difference delta d between the m-th scanning point on the n-th beam and the m-th scanning point on the (n + 1) -th beam adjacent to the m-th scanning pointmnIncluded angle delta theta with elevationmn(ii) a And
according to the difference of horizontal distance DeltadmnWith a predetermined distance threshold dwJudging the ground segmentation characteristics of the mth scanning point on the nth wire harness and the mth scanning point on the (n + 1) th wire harness adjacent to the mth scanning point according to the relationship between the mth scanning point and the mth wire harness;
or according to elevation angle delta thetamnThreshold value theta of included angle with preset valuewJudging the ground segmentation characteristics of the mth scanning point on the nth wire harness and the mth scanning point on the (n + 1) th wire harness adjacent to the mth scanning point according to the relationship between the mth scanning point and the mth wire harness;
the ground segmentation features comprise temporary ground points, ground points and non-ground points, and when the judgment result is invalid and an invalid value is output, the ground segmentation features are invalid. n and m are positive integers, n is more than or equal to 1, m is more than or equal to 1, w is equal to n, and w is the corresponding number between adjacent wire harnesses from the 1 st wire harness to the n th wire harness, so n-1 is more than or equal to w is more than or equal to 1.
In particular, a preset distance threshold dwAnd a preset included angle threshold value thetawThe value may be a predetermined constant value or a variable value defined according to the beam characteristics between any two adjacent beams, for example, a variable value varying with the scanning distance or the number of the displayed scanning beams. . Wherein, M is more than or equal to M and more than or equal to 1, and M is the total scanning point number on the corresponding wire harness. Wherein, there is no relation between any n scanning point and m scanning point. Wherein, the m-th scanning point on the n-th beam and the m-th scanning point on the n + 1-th beam can be understood as [ n ] of two-dimensional data][m]And [ n +1 ]][m]The angle of the different wire harnesses is represented, and accurate acquisition or processing of data is facilitated. In addition, "adjacent" as expressed herein may be scan detection order adjacent or bracketing. The above definition of the scanning beam and the scanning point on the scanning beam, and the definition of the scanning point coordinate complete the preprocessing of the ground detection data.
According to the embodiment of the present disclosure, as shown in fig. 3, the coordinate of the mth scanning point on the nth beam corresponding to the radar coordinate system is (x)mn，ymn，zmn) The coordinate of the m-th scanning point on the (n + 1) -th beam is (x)m(n+1)，ym(n+1)，zm(n+1)) Corresponding to the two filled circles in fig. 3, respectively.
Acquiring the horizontal distance difference delta d between the m-th scanning point on the n-th beam and the m-th scanning point on the adjacent n + 1-th beammnThe method comprises the following steps: acquiring the horizontal distance d of the m scanning point on the n beammnThe following formula is satisfied:
wherein (x)mn，ymn) The horizontal plane (the plane corresponding to the X-Y axis) coordinate of the m-th scanning point on the corresponding n-th beam; acquiring the horizontal distance d of the mth scanning point corresponding to the adjacent nth wire harness on the (n + 1) th wire harnessm(n+1)The following formula is satisfied:
wherein (x)m(n+1)，ym(n+1)) The horizontal plane coordinate of the m-th scanning point on the n +1 th beam and the horizontal distance d of the m-th scanning point on the n-th beammnAnd a horizontal distance d from the m-th scanning point on the (n + 1) -th beam adjacent theretom(n+1)Difference of horizontal distance Δ d therebetweenmnThe following formula is satisfied:
Δdmn＝dm(n+1)－dmn
wherein n is a positive integer, and n is more than or equal to 1.
In particular, in the preliminary segmentation data acquisition process, the segmentation of the ground objects of the wire harness data is involved. The method comprises the steps of scanning a 1 st scanning point on a 1 st beam of a starting scanning beam and a corresponding 1 st scanning point on the 1 st beam, wherein n is 1, and m is 1, and scanning a 1 st scanning point on a 2 nd beam of an adjacent scanning beam and a corresponding 2 nd beam, wherein n is 2, and m is 1.
Calculating the horizontal distance d between the 1 st scanning point of the 1 st wire harness and the mounting point of the radar11Horizontal distance d between the 1 st scanning point of the 2 nd beam of the adjacent scanning line and the mounting point of the radar21The horizontal distance is the distance from the scanning point of the scanning beam to the origin of the radar coordinate system corresponding to the projection point on the horizontal plane, and the calculation formula is
Calculating the horizontal distance difference deltad between the two points11Is of the formula
Δd11＝d12－d11
The horizontal distance d between the 1 st scanning point of the 1 st beam and the mounting point of the radar is obtained according to the preprocessing result11Horizontal distance d between the 1 st scanning point of the 2 nd beam of the adjacent scanning line and the mounting point of the radar21And the difference Δ d between the horizontal distances between the two corresponding scanning points11。
According to the embodiment of the disclosure, the elevation included angle delta theta between the m-th scanning point on the n-th beam and the m-th scanning point on the adjacent n + 1-th beam is obtainedmnThe method comprises the following steps: according to the difference of horizontal distance DeltadmnObtaining elevation included angle delta thetamnThe following formula is satisfied:
wherein z ismnIs the vertical coordinate, z, of the m-th scanning point on the n-th beamm(n+1)Is the vertical coordinate of the m-th scanning point on the (n + 1) -th beam corresponding to the adjacent beam, n is a positive integer, and n is more than or equal to 1.
Specifically, the elevation included angle delta theta between the 1 st scanning point on the 1 st beam and the 1 st scanning point on the adjacent 2 nd beam is obtained11For example, the following steps are carried out:
wherein z is11Is the vertical coordinate of the 1 st scanning point on the 1 st beam, z12Is the vertical coordinate of the 1 st scanning point on the 2 nd beam adjacent to the first scanning point.
According to an embodiment of the present disclosure, the horizontal distance difference Δ d is usedmnWith a predetermined distance threshold dwThe ground segmentation characteristics of the mth scanning point on the nth wire harness and the mth scanning point on the (n + 1) th wire harness adjacent to the mth scanning point are judged according to the size relationship between the mth scanning point and the mth wire harness; or according to elevation angle delta thetamnThreshold value theta of included angle with preset valuewThe size relationship between the ground segmentation features and the ground segmentation features of the mth scanning point on the nth beam and the mth scanning point on the (n + 1) th beam adjacent to the mth scanning point is judged, and the ground segmentation features comprise the following steps:
when n is 1, the ground segmentation characteristics of the mth scanning point on the 1 st beam and the mth scanning point on the 2 nd beam adjacent to the mth scanning point meet the following conditions:
when Δ dm1<d1Or | Δ θm1|>θ1If so, the m-th scanning point on the 1 st wire harness and the m-th scanning point on the 2 nd wire harness are non-ground points;
when Δ dm1≥d1Or | Δ θm1|≤θ1When the number of the scanning points is larger than the preset value, the m-th scanning point on the 1 st wire harness is a ground point, and the m-th scanning point on the 2 nd wire harness is a temporary ground point;
specifically, taking the ground segmentation characteristics of the 1 st scanning point on the 1 st beam and the 1 st scanning point on the adjacent 2 nd beam as an example,
when Δ d11<d1Or | Δ θ11|>θ1If so, the 1 st scanning point on the 1 st wire harness and the 1 st scanning point on the 2 nd wire harness are non-ground points;
when Δ d11≥d1Or | Δ θ11|≤θ1When the first scanning point on the 1 st wire harness is a ground point, the 1 st scanning point on the 2 nd wire harness is a temporary ground point; at this time, the elevation angle value recorded in the process is recorded as delta thetarecord，Δθrecord＝Δθ11。
When n is larger than or equal to 2 and the mth scanning point on the nth wire harness is a temporary ground point, the ground segmentation characteristics of the mth scanning point on the nth wire harness meet the following conditions:
when Δ dmn<dwAnd then the m-th scanning point on the n-th beam is a non-ground point,
when Δ dmn≥dwAnd in the process, the m-th scanning point on the n-th wiring harness is a ground point.
Specifically, in the embodiment of the disclosure, the determination of the ground object segmentation characteristics is performed on the present beam according to the ground object segmentation characteristics of the previous beam, taking the ground segmentation characteristics of the 1 st scanning point on the 2 nd beam and the 1 st scanning point on the adjacent 3 rd beam as an example,
calculating the horizontal distance difference Delta d between the 1 st scanning point of the 2 nd beam and the 1 st scanning point of the 3 rd beam12Included angle delta theta with elevation12Method of calculating and the above-mentioned Δ d11And Δ θ11The same is true.
If the No. 1 point of the 2 nd wire harness is a temporary ground point, an
When Δ d12<d2And then the 1 st scanning point on the 2 nd wire harness is a non-ground point,
when Δ d12≥d2And in the process, the 1 st scanning point on the 2 nd wire harness is a ground point.
According to an embodiment of the present disclosure, the horizontal distance difference Δ d is usedmnWith a predetermined distance threshold dwThe ground segmentation characteristics of the mth scanning point on the nth wire harness and the mth scanning point on the (n + 1) th wire harness adjacent to the mth scanning point are judged according to the size relationship between the mth scanning point and the mth wire harness; or according to elevation angle delta thetamnThreshold value theta of included angle with preset valuewThe size relationship between the ground segmentation features and the ground segmentation features of the mth scanning point on the nth beam and the mth scanning point on the (n + 1) th beam adjacent to the mth scanning point is judged, and the ground segmentation features comprise the following steps:
when n is larger than or equal to 1 and the mth scanning point on the nth wire harness is a ground point, the ground segmentation characteristics of the mth scanning point on the (n + 1) th wire harness meet the following conditions:
when Δ dmn<dwOr | Δ θmn|>θwIf so, the m scanning point on the (n + 1) th wiring harness is a non-ground point;
when Δ dmn≥dwOr | Δ θmn|≤θwIn the process, the m-th scanning point on the (n + 1) -th wiring harness is a ground point;
specifically, taking the ground segmentation feature of acquiring the 1 st scanning point on the 3 rd beam as an example, the previous beam is the 2 nd beam.
Firstly, calculating the horizontal distance difference delta d between the 1 st point of the 2 nd wire harness and the 1 st point of the 3 rd wire harness12Included angle delta theta with elevation12。
The 1 st point of the 2 nd wire harness is a ground point if
When Δ d12<d2Or | Δ θ12|>θ2If so, the 1 st scanning point on the 3 rd wiring harness is a non-ground point;
when Δ d12≥d2Or | Δ θ12|≤θ2In the process, the 1 st scanning point on the 3 rd wiring harness is a ground point;
and when n is larger than or equal to 1 and the mth scanning point on the nth wire harness is a non-ground point, the ground segmentation characteristic of the mth scanning point on the (n + 1) th wire harness meets the following conditions:
when Δ dmn<dwOr | Δ θmn|>θwIf so, the m scanning point on the (n + 1) th wiring harness is a non-ground point;
when | Δ θmn|≤θwAnd in the meantime, the m-th scanning point on the (n + 1) -th wiring harness is a ground point.
Specifically, taking the ground segmentation feature of acquiring the 1 st scanning point on the 3 rd beam as an example, the previous beam is the 2 nd beam.
When the 2 nd wire harness 1 st point is a non-ground point,
when Δ d12<d2Or | Δ θ12|>θ2If so, the 1 st scanning point on the 3 rd wiring harness is a non-ground point;
when | Δ θ12|≤θ2And in the process, the 1 st scanning point on the 3 rd wiring harness is a ground point.
According to an embodiment of the present disclosure, the horizontal distance difference Δ d is usedmnWith a predetermined distance threshold dwThe ground segmentation characteristics of the mth scanning point on the nth wire harness and the mth scanning point on the (n + 1) th wire harness adjacent to the mth scanning point are judged according to the size relationship between the mth scanning point and the mth wire harness; or according to elevation angle delta thetamnThreshold value theta of included angle with preset valuewThe size relationship between the m scanning point and the ground segmentation feature of the m scanning point on the n beam and the adjacent n +1 beam is judged, and the method further comprises the following steps:
when n is larger than or equal to 2, the ground segmentation characteristic of the mth scanning point on the (n + 1) th wire harness meets the following conditions:
when | Δ θmn|＞|Δθrecord|+θwIf so, the m scanning point on the (n + 1) th wiring harness is a non-ground point;
when | Δ θmn|≤|Δθrecord|+θwIn the process, the m-th scanning point on the (n + 1) -th wiring harness is a ground point;
wherein, the elevation angle delta theta is recordedrecordEqual to the elevation angle delta theta between the m-th scanning point on the n-1 th beam and the m-th scanning point on the n-th beam adjacent to the m-th scanning pointm(n-1)I.e. the elevation angle corresponding to the adjacent previous wire harness.
Specifically, when the above other conditions cannot be satisfied, it is also necessary to record the elevation angle Δ θrecordProcessing is performed, taking the ground segmentation feature of the 1 st scanning point on the 3 rd beam as an example, and the previous beam is the 2 nd beam.
When | Δ θ12|＞|Δθrecord|+θ2If so, the 1 st scanning point on the 3 rd wiring harness is a non-ground point;
when | Δ θ12|≤|Δθrecord|+θ2And in the time, the 1 st scanning point on the 1 st wire harness is a ground point.
The final preliminary segmentation data can be obtained by analogy with the content disclosed above.
According to an embodiment of the present invention, the horizontal distance difference Δ d is usedmnWith a predetermined distance threshold dwThe ground segmentation characteristics of the mth scanning point on the nth wire harness and the mth scanning point on the (n + 1) th wire harness adjacent to the mth scanning point are judged according to the size relationship between the mth scanning point and the mth wire harness; or according to elevation angle delta thetamnThreshold value theta of included angle with preset valuewThe size relationship between the m scanning point and the ground segmentation feature of the m scanning point on the n beam and the adjacent n +1 beam is judged, and the method further comprises the following steps:
when the ground segmentation feature of the mth scanning point on the nth wire harness is judged to be an invalid segmentation feature, the m +1 scanning point on the nth wire harness is taken as an initial scanning point to continue the judgment of the ground segmentation feature, which is equivalent to screening out the mth scanning point on the nth wire harness, so as to avoid influencing the accuracy of the final preliminary segmentation data, and meanwhile, the acquisition of the ground segmentation features of other scanning points can be continuously completed, so that the algorithm efficiency and the accuracy are improved, and the preliminary segmentation data of the ground segmentation can be finally acquired.
The invalid segmentation features correspond to the features that the above-disclosed judgment basis cannot be executed, or the judgment result is an invalid value, and the invalid segmentation features cannot correspond to the ground points, non-ground points or temporary ground point features of the ground segmentation features. In the embodiment of the present invention, therefore, when the scanning point number 500 on the 1 st beam, i.e., n is 1, M is 500, when the 100 th scanning point on the 1 st beam is an invalid segmentation feature, the judgment of the ground segmentation feature is continuously completed by taking the 101 th scanning point on the 1 st beam as a starting scanning point, and storing the ground segmentation characteristics of all the scanning points from the 1 st scanning point to the 100 th scanning point on the 1 st beam, and finally after finishing the judgment of the ground segmentation characteristics from the 101 st scanning point to the 500 th scanning point on the 1 st beam, and combining the stored preliminary segmentation data of all the scanning points between the 1 st scanning point and the 100 th scanning point on the 1 st beam with the stored preliminary segmentation data of all the scanning points between the 101 st scanning point and the 500 th scanning point on the 1 st beam, wherein the number of the scanning points on the corresponding 1 st beam is M-1-499.
According to an embodiment of the present disclosure, obtaining final segmentation data of the ground segmentation from the preliminary segmentation data includes: performing surface fitting on all ground points in a certain range of the vehicle in the obtained preliminary segmentation data to obtain a ground elevation value of each ground point, wherein the ground elevation value is an elevation value, namely a distance value, from a detected ground point to a horizontal plane (an X-Y axis plane) of a radar coordinate system; and finally segmenting each ground point according to the relationship between the ground elevation value of each ground point and the corresponding measured elevation value of the ground point so as to obtain the final segmentation data of all the ground points. The measured elevation value is a distance value of a vehicle from an actual ground environment detected by a scanning point corresponding to a scanning detection line beam emitted by the multi-line laser radar (the scanning point at the moment needs to be a corresponding ground point). Through the discrete fitting of the curved surface, the ground elevation values of all ground points in a certain range of the vehicle can be obtained.
Specifically, as shown in fig. 4, in the embodiment of the present invention, the vehicle fixed range may be a fixed area F around the vehicle C, which is a range formed by the vehicle front L1 being 60m, the vehicle rear L2 being 40m, the vehicle left L3 being 40m, and the vehicle right L4 being 40m, and the area F may be a range of the area occupied by the vehicle.
And according to the fitted ground height, further processing all the scanning points which are divided into ground points in the preliminary division data, specifically, if the measured elevation value divided into the ground points is greater than the ground elevation value fitted by the ground points, further dividing the points into non-ground points, otherwise, keeping the points as the ground points. And then, calculating the height values of all the non-ground points and the fitting curved surface in the judgment result to be used as the distance values of the non-ground points to the actual ground. The ground feature data of the next wire harness is segmented through the recorded actual elevation included angle between the adjacent wire harnesses, and the ground feature segmentation is more accurate and reliable; in addition, by confirming the ground feature division state for a plurality of times and further performing ground feature division screening, the possibility of false recognition of the environment perception of the vehicle can be greatly reduced.
The invention also discloses a vehicle, and automatic driving of the vehicle is realized according to the multiline laser radar ground segmentation method. The vehicle has stronger road adaptability by applying the ground segmentation method, can well adapt to road conditions such as mining areas, mountain roads and the like, has more accurate and reliable ground object segmentation and lower environment perception false recognition rate, and is very favorable for automatic driving application of the vehicle.
In another aspect of the present invention, a computer-readable medium is disclosed, which comprises a memory and a processor, wherein the memory stores executable instructions, and the instructions, when executed by the processor, implement the above-mentioned multiline lidar ground segmentation method. The computer-readable storage medium may be embodied in the apparatus/device/system described in the above embodiments; or may exist separately and not be assembled into the device/apparatus/system. The above-mentioned computer-readable storage medium carries one or more programs which, when executed, implement the multiline lidar ground segmentation method according to an embodiment of the present disclosure.
According to embodiments of the present disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, which may include, for example but is not limited to: a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. For example, according to embodiments of the present disclosure, a computer-readable storage medium may include the ROM and/or RAM and/or one or more memories other than ROM and RAM described above.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
Those skilled in the art will appreciate that various combinations and/or combinations of features recited in the various embodiments and/or claims of the present disclosure can be made, even if such combinations or combinations are not expressly recited in the present disclosure. In particular, various combinations and/or combinations of the features recited in the various embodiments and/or claims of the present disclosure may be made without departing from the spirit or teaching of the present disclosure. All such combinations and/or associations are within the scope of the present disclosure.
Claims (12)
1. A multiline laser radar ground segmentation method is applied to automatic driving of vehicles and is characterized by comprising the following steps:
acquiring a preprocessing result of ground detection data of the multi-line laser radar;
acquiring preliminary segmentation data of the ground segmentation according to the preprocessing result;
and acquiring final segmentation data of the ground segmentation according to the preliminary segmentation data, and realizing the ground segmentation according to the final segmentation data.
2. The multiline lidar ground segmentation method of claim 1, wherein the obtaining of the preprocessing result of the ground detection data of the multiline lidar comprises:
resolving the acquired ground detection data of the multiline lidar to a radar coordinate system of the multiline lidar, an
Sequencing the ground detection data according to the radar coordinate system according to the working mode of the multi-line laser radar to obtain the preprocessing result;
the working mode is the beam sequence of the multi-line laser radar when the multi-line laser radar utilizes the multi-line laser beam to detect.
3. The multiline lidar ground segmentation method of claim 1, wherein the obtaining of the preliminary line bundle data of the ground segmentation according to the preprocessing result comprises:
acquiring the horizontal distance difference delta d between the m-th scanning point on the n-th beam and the m-th scanning point on the (n + 1) -th beam adjacent to the m-th scanning pointmnIncluded angle delta theta with elevationmn(ii) a And
according to the difference of horizontal distance DeltadmnWith a predetermined distance threshold dwJudging the ground segmentation characteristics of the mth scanning point on the nth wire harness and the mth scanning point on the (n + 1) th wire harness adjacent to the mth scanning point according to the relationship between the mth scanning point and the mth wire harness; or
According to the elevation included angle delta thetamnThreshold value theta of included angle with preset valuewJudging the ground segmentation characteristics of the mth scanning point on the nth wire harness and the mth scanning point on the (n + 1) th wire harness adjacent to the mth scanning point according to the relationship between the mth scanning point and the mth wire harness;
the ground segmentation features comprise temporary ground points, ground points and non-ground points, n and m are positive integers, n is not less than 1, m is not less than 1, w is n, and n-1 is not less than 1.
4. The multiline lidar ground segmentation method of claim 3, wherein obtaining a horizontal distance difference Δ d between an mth scanning point on an nth beam and an mth scanning point on an adjacent (n + 1) th beammnThe method comprises the following steps:
acquiring the horizontal distance d of the m scanning point on the n beammnThe following formula is satisfied:
wherein, the (x)mn，ymn) The horizontal plane coordinate of the m scanning point on the corresponding n beam;
acquiring the horizontal distance d of the mth scanning point corresponding to the adjacent nth wire harness on the (n + 1) th wire harnessm(n+1)Satisfies the following formula：
Wherein, the (x)m(n+1)，ym(n+1)) Is the horizontal plane coordinate of the m-th scanning point on the (n + 1) -th beam,
the horizontal distance d of the m-th scanning point on the n-th beammnAnd a horizontal distance d from the m-th scanning point on the (n + 1) -th beam adjacent theretom(n+1)Difference of horizontal distance Δ d therebetweenmnThe following formula is satisfied:
Δdmn＝dm(n+1)－dmn
wherein n is a positive integer, and n is more than or equal to 1.
5. The multiline lidar ground segmentation method of claim 3 wherein the elevation angle Δ θ between the mth scanning point on the nth beam and the mth scanning point on the adjacent (n + 1) th beam is obtainedmnThe method comprises the following steps:
according to the difference of horizontal distance DeltadmnObtaining the elevation included angle delta thetamnThe following formula is satisfied:
wherein, z ismnIs the vertical coordinate, z, of the m-th scanning point on the n-th beamm(n+1)Is the vertical coordinate of the m-th scanning point on the (n + 1) -th beam corresponding to the adjacent beam, n is a positive integer, and n is more than or equal to 1.
6. The multiline lidar ground segmentation method of claim 3 wherein the distance Δ d is based on the horizontal distance differencemnWith a predetermined distance threshold dwThe ground segmentation characteristics of the m-th scanning point on the n-th beam and the m-th scanning point on the n + 1-th beam adjacent to the m-th scanning point are judged according to the size relationship between the m-th scanning point and the n-th beam; or according to said elevation angle delta thetamnAnd a preset clipAngular threshold θwThe size relationship between the m scanning points and the ground segmentation characteristics of the m scanning point on the nth beam and the adjacent n +1 th beam is judged, and the ground segmentation characteristics comprise the following steps:
when n is 1, the ground segmentation characteristics of the mth scanning point on the 1 st beam and the mth scanning point on the 2 nd beam adjacent to the mth scanning point meet the following conditions:
when Δ dm1<d1Or | Δ θm1|>θ1If so, the m-th scanning point on the 1 st wire harness and the m-th scanning point on the 2 nd wire harness are non-ground points;
when Δ dm1≥d1Or | Δ θm1|≤θ1When the number of the scanning points is larger than the preset value, the m-th scanning point on the 1 st wire harness is a ground point, and the m-th scanning point on the 2 nd wire harness is a temporary ground point;
when n is larger than or equal to 2 and the mth scanning point on the nth wire harness is a temporary ground point, the ground segmentation characteristics of the mth scanning point on the nth wire harness meet the following conditions:
when Δ dmn<dwAnd then the m-th scanning point on the n-th beam is a non-ground point,
when Δ dmn≥dwAnd in the process, the m-th scanning point on the n-th wiring harness is a ground point.
7. The multiline lidar ground segmentation method of claim 3 wherein the distance Δ d is based on the horizontal distance differencemnWith a predetermined distance threshold dwThe ground segmentation characteristics of the m-th scanning point on the n-th beam and the m-th scanning point on the n + 1-th beam adjacent to the m-th scanning point are judged according to the size relationship between the m-th scanning point and the n-th beam; or according to said elevation angle delta thetamnThreshold value theta of included angle with preset valuewThe size relationship between the m scanning points and the ground segmentation characteristics of the m scanning point on the nth beam and the adjacent n +1 th beam is judged, and the ground segmentation characteristics comprise the following steps:
when n is larger than or equal to 1 and the mth scanning point on the nth wire harness is a ground point, the ground segmentation characteristic of the mth scanning point on the (n + 1) th wire harness meets the following conditions:
when Δ dmn<dwOr | Δ θmn|>θwIf so, the m scanning point on the (n + 1) th wiring harness is a non-ground point;
when Δ dmn≥dwOr | Δ θmn|≤θwIn the process, the m-th scanning point on the (n + 1) -th wiring harness is a ground point; and
when n is larger than or equal to 1 and the mth scanning point on the nth wire harness is a non-ground point, the ground segmentation characteristic of the mth scanning point on the (n + 1) th wire harness meets the following conditions:
when Δ dmn<dwOr | Δ θmn|>θwIf so, the m scanning point on the (n + 1) th wiring harness is a non-ground point;
when | Δ θmn|≤θwAnd in the meantime, the m-th scanning point on the (n + 1) -th wiring harness is a ground point.
8. The multiline lidar ground segmentation method of claim 3 wherein the distance Δ d is based on the horizontal distance differencemnWith a predetermined distance threshold dwThe ground segmentation characteristics of the m-th scanning point on the n-th beam and the m-th scanning point on the n + 1-th beam adjacent to the m-th scanning point are judged according to the size relationship between the m-th scanning point and the n-th beam; or according to said elevation angle delta thetamnThreshold value theta of included angle with preset valuewThe size relationship between the m-th scanning point and the ground segmentation characteristics of the m-th scanning point on the n-th beam and the adjacent n + 1-th beam is judged, and the method further comprises the following steps:
when n is larger than or equal to 2, the ground segmentation characteristic of the mth scanning point on the (n + 1) th wire harness meets the following condition:
when | Δ θmn|＞|Δθrecord|+θwIf so, the m scanning point on the (n + 1) th wiring harness is a non-ground point;
when | Δ θmn|≤|Δθrecord|+θwIn the process, the m-th scanning point on the (n + 1) -th wiring harness is a ground point;
wherein the elevation angle delta theta is recordedrecordEqual to the elevation angle delta theta between the m-th scanning point on the n-1 th beam and the m-th scanning point on the n-th beam adjacent to the m-th scanning pointm(n-1)。
9. The multiline lidar ground segmentation method of claim 3 wherein the distance Δ d is based on the horizontal distance differencemnWith a predetermined distance threshold dwThe ground segmentation characteristics of the m-th scanning point on the n-th beam and the m-th scanning point on the n + 1-th beam adjacent to the m-th scanning point are judged according to the size relationship between the m-th scanning point and the n-th beam; or according to said elevation angle delta thetamnThreshold value theta of included angle with preset valuewThe size relationship between the m-th scanning point and the ground segmentation characteristics of the m-th scanning point on the n-th beam and the adjacent n + 1-th beam is judged, and the method further comprises the following steps:
and when the ground segmentation feature of the mth scanning point on the nth wire harness is judged to be an invalid segmentation feature, taking the m +1 scanning point on the nth wire harness as an initial scanning point to continue judging the ground segmentation feature so as to obtain the preliminary segmentation data of the ground segmentation.
10. The multiline lidar ground segmentation method of claim 3 wherein the obtaining final segmentation data for the ground segmentation from the preliminary segmentation data comprises:
performing surface fitting on all ground points in the acquired preliminary segmentation data within a certain range of the vehicle to acquire a ground elevation value of each ground point;
and finally segmenting each ground point according to the relationship between the ground elevation value of each ground point and the corresponding measured elevation value of the ground point so as to obtain the final segmentation data of all the ground points.
11. A vehicle implementing automated driving of the vehicle according to the multiline lidar ground segmentation method of any one of claims 1-10.
12. A computer readable medium comprising a memory and a processor, the memory storing executable instructions that when executed by the processor implement a multiline lidar ground segmentation method according to any of claims 1-10.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911301701.8A CN111126225B (en) | 2019-12-17 | 2019-12-17 | Multi-line laser radar ground segmentation method, vehicle and computer readable medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911301701.8A CN111126225B (en) | 2019-12-17 | 2019-12-17 | Multi-line laser radar ground segmentation method, vehicle and computer readable medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111126225A true CN111126225A (en) | 2020-05-08 |
| CN111126225B CN111126225B (en) | 2023-08-04 |
Family
ID=70499431
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911301701.8A Active CN111126225B (en) | 2019-12-17 | 2019-12-17 | Multi-line laser radar ground segmentation method, vehicle and computer readable medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111126225B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113253294A (en) * | 2021-06-16 | 2021-08-13 | 北京海天瑞声科技股份有限公司 | Method, apparatus and medium relating to ground point detection in 3D radar point cloud data |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107272019A (en) * | 2017-05-09 | 2017-10-20 | 深圳市速腾聚创科技有限公司 | Curb detection method based on Laser Radar Scanning |
| CN108596860A (en) * | 2018-05-10 | 2018-09-28 | 芜湖航飞科技股份有限公司 | A kind of ground point cloud dividing method based on three-dimensional laser radar |
| CN108828621A (en) * | 2018-04-20 | 2018-11-16 | 武汉理工大学 | Obstacle detection and road surface partitioning algorithm based on three-dimensional laser radar |
| CN109738910A (en) * | 2019-01-28 | 2019-05-10 | 重庆邮电大学 | A kind of curb detection method based on three-dimensional laser radar |
| US20190178989A1 (en) * | 2017-12-11 | 2019-06-13 | Automotive Research & Testing Center | Dynamic road surface detecting method based on three-dimensional sensor |
| CN110163871A (en) * | 2019-05-07 | 2019-08-23 | 北京易控智驾科技有限公司 | A kind of ground dividing method of multi-line laser radar |
-
2019
- 2019-12-17 CN CN201911301701.8A patent/CN111126225B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107272019A (en) * | 2017-05-09 | 2017-10-20 | 深圳市速腾聚创科技有限公司 | Curb detection method based on Laser Radar Scanning |
| US20190178989A1 (en) * | 2017-12-11 | 2019-06-13 | Automotive Research & Testing Center | Dynamic road surface detecting method based on three-dimensional sensor |
| CN108828621A (en) * | 2018-04-20 | 2018-11-16 | 武汉理工大学 | Obstacle detection and road surface partitioning algorithm based on three-dimensional laser radar |
| CN108596860A (en) * | 2018-05-10 | 2018-09-28 | 芜湖航飞科技股份有限公司 | A kind of ground point cloud dividing method based on three-dimensional laser radar |
| CN109738910A (en) * | 2019-01-28 | 2019-05-10 | 重庆邮电大学 | A kind of curb detection method based on three-dimensional laser radar |
| CN110163871A (en) * | 2019-05-07 | 2019-08-23 | 北京易控智驾科技有限公司 | A kind of ground dividing method of multi-line laser radar |
Non-Patent Citations (3)
| Title |
|---|
| MILA NIKOLOVA等: "Segmentation of a road from a vehicle-mounted radar and accuracy of the estimation" * |
| 石庭敏等: "基于双多线激光雷达的低矮道边检测" * |
| 苏致远;徐友春;李永乐;: "基于三维激光雷达的车辆目标检测方法" * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113253294A (en) * | 2021-06-16 | 2021-08-13 | 北京海天瑞声科技股份有限公司 | Method, apparatus and medium relating to ground point detection in 3D radar point cloud data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111126225B (en) | 2023-08-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10823855B2 (en) | Traffic recognition and adaptive ground removal based on LIDAR point cloud statistics | |
| CN110033457B (en) | Target point cloud segmentation method | |
| US10132919B2 (en) | Object detecting device, radar device, and object detection method | |
| CN110879994A (en) | Three-dimensional visual inspection detection method, system and device based on shape attention mechanism | |
| CN110163871B (en) | Ground segmentation method and device for multi-line laser radar | |
| CN110674705A (en) | Small-sized obstacle detection method and device based on multi-line laser radar | |
| CN111179152A (en) | Road sign identification method and device, medium and terminal | |
| EP3040909A1 (en) | Method of providing an envelope with respect to a cluster | |
| US10444398B2 (en) | Method of processing 3D sensor data to provide terrain segmentation | |
| CN111427028B (en) | Parameter monitoring method, device, equipment and storage medium | |
| CN110554409B (en) | Concave obstacle detection method and system | |
| CN109633686B (en) | Method and system for detecting ground obstacle based on laser radar | |
| CN114384491B (en) | Point cloud processing method and device for laser radar and storage medium | |
| CN114494075A (en) | Obstacle identification method based on three-dimensional point cloud, electronic device and storage medium | |
| CN115018879A (en) | Target detection method, computer-readable storage medium, and driving apparatus | |
| CN114384492B (en) | Point cloud processing method and device for laser radar and storage medium | |
| CN114692734A (en) | Mining area dust filtering method, device, equipment and medium based on point cloud | |
| CN115151954A (en) | Method and device for detecting a drivable region | |
| WO2018180081A1 (en) | Deteriorated ground-based object identification device, deteriorated ground-based object identification system, deteriorated ground-based object identification method, deteriorated ground-based object identification program and computer-readable recording medium having deteriorated ground-based object identification program recorded therein | |
| CN111126225A (en) | Multiline laser radar ground segmentation method, vehicle and computer readable medium | |
| CN114581371A (en) | Tunnel deformation detection method, system, equipment and storage medium | |
| CN113376643A (en) | Distance detection method and device and electronic equipment | |
| US11428782B2 (en) | Neural network-based object surface estimation in radar system | |
| CN110058263B (en) | Object positioning method in vehicle driving process | |
| CN115128637A (en) | Method for determining the range of a lidar sensor |
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 |