CN111007534A - Obstacle detection method and system using sixteen-line laser radar - Google Patents
Obstacle detection method and system using sixteen-line laser radar Download PDFInfo
- Publication number
- CN111007534A CN111007534A CN201911131844.9A CN201911131844A CN111007534A CN 111007534 A CN111007534 A CN 111007534A CN 201911131844 A CN201911131844 A CN 201911131844A CN 111007534 A CN111007534 A CN 111007534A
- Authority
- CN
- China
- Prior art keywords
- obstacle
- contour
- candidate
- obstacles
- alternative
- 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
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
- G01C21/28—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network with correlation of data from several navigational instruments
-
- 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
Abstract
The invention relates to the technical field of automatic driving, in particular to a method and a system for detecting obstacles by utilizing a sixteen-line laser radar; the method comprises the following steps: initializing, namely establishing a Kalman filtering model for carrying out speed detection and center point position updating on the obstacle; extracting an alternative contour of the obstacle and acquiring the speed and the course angle of the current vehicle; matching and tracking each obstacle candidate contour and the obstacle, and updating obstacle information; processing the alternative outlines and the obstacles on the unmatched obstacle; sorting the obstacles; outputting effective obstacles and repeating the steps; the system comprises: the system comprises a system setting module, a data acquisition module, a data matching module and a data output module; according to the embodiment of the invention, the system executes the method, and the acquired obstacle information is richer and more complete and the generated obstacle outline information is more accurate through the point cloud data of the sixteen-line laser radar and the positioning data of the combined inertial navigation.
Description
Technical Field
The invention relates to the technical field of automatic driving, in particular to a method and a system for detecting obstacles by utilizing a sixteen-line laser radar.
Background
In the automatic driving process of the vehicle, the automatic driving system needs to acquire the information of surrounding obstacles so as to achieve the aim of safely driving the vehicle on the road. The automatic driving system needs to control the vehicle so that the vehicle can run smoothly on the lane and does not collide with the obstacle. The obstacles refer to objects which influence the running of the automatic driving vehicle, such as pedestrians, motor vehicles, bicycles, cone-shaped barrels, garbage cans and the like on the road; these obstacles include both moving obstacles and stationary obstacles; how to obtain more comprehensive obstacle information is an important issue facing the automatic driving industry at present.
The traditional method is to directly acquire obstacle information containing precise outline and precise speed from a four-wire laser radar; the sixteen-line lidar on the market today is capable of representing obstacle information in a very large number of precise three-dimensional points.
The prior art has the defects that the obstacle information of the four-wire laser radar is not complete enough, the ground, leaves and the like can be used as obstacles sometimes, and the four-wire laser radar is expensive; sixteen-line lidar is unable to acquire the contour and velocity information of an obstacle.
Disclosure of Invention
The embodiment of the invention overcomes the defects of the prior art, provides the obstacle detection method and the obstacle detection system by using the sixteen-line laser radar, obtains a more complete obstacle outline, can better remove objects such as leaves on the ground and at high positions and the like which do not influence the driving of the automatic driving vehicle, and effectively reduces the scheme cost.
In one aspect, an embodiment of the present invention provides an obstacle detection method using a sixteen-line laser radar, including the following steps:
s1, initializing, and establishing a Kalman filtering model for performing speed detection and center point position update on the obstacle; establishing a vehicle body coordinate system which takes the projection point from the center of a rear axle of a vehicle to the ground as an origin in meters and takes the X-axis forward direction as the forward direction, the Y-axis forward direction as the leftward direction and the Z-axis forward direction as the upward direction, and establishing an image coordinate system which takes the X-axis forward direction as the rightward direction and the Y-axis forward direction as the backward direction and takes the pixel unit as the backward direction;
s2, extracting the candidate contour of the obstacle and obtaining the speed and the heading angle of the current vehicle; the method specifically comprises the steps of acquiring point cloud from a sixteen-line laser radar, and removing points which do not influence vehicle running on the ground and in the air; projecting the remaining point cloud to a two-dimensional plane to obtain an image with pixel width and height, and extracting the alternative obstacle outline from the image;
s3, matching each obstacle candidate contour with an obstacle according to a matching mode, tracking the matched obstacle contour, and updating obstacle information; the obstacle information includes: the center point of the obstacle, the longitudinal relative speed of the obstacle, the transverse relative speed of the obstacle, the longitudinal absolute speed of the obstacle, the transverse absolute speed of the obstacle, the outline of the obstacle, the number of tracking frames and the number of lost frames;
s4, processing the alternative contour and the obstacle which are not matched; if the candidate outlines of the obstacles which are not matched are specifically included, the candidate outlines of the obstacles are used as new obstacles to be added into an obstacle list; deleting the obstacles from the obstacle list when the number of the unmatched obstacle lost frames is larger than a set value; when the obstacle which is not matched is combined with other obstacles to form an alternative contour, deleting the obstacle from an obstacle list;
s5, sorting the obstacles; specifically, the method comprises the steps of sorting the obstacle contour points in a sequence from a plurality of points according to the number of the obstacle contour points;
s6, outputting effective barriers, waiting for the next frame data and repeating the steps S2-S5; the number of the effective barriers is greater than a set value for the number of the barrier tracking frames.
In another aspect, an embodiment of the present invention provides an obstacle detection system using a sixteen-line laser radar, including:
the system setting module is used for carrying out initialization setting and establishing a Kalman filtering model for carrying out speed detection and center point position updating on the barrier; establishing a vehicle body coordinate system which takes the projection point from the center of a rear axle of a vehicle to the ground as an origin in meters and takes the X-axis forward direction as the forward direction, the Y-axis forward direction as the leftward direction and the Z-axis forward direction as the upward direction, and establishing an image coordinate system which takes the X-axis forward direction as the rightward direction and the Y-axis forward direction as the backward direction and takes the pixel unit as the backward direction;
the data acquisition module is used for extracting the alternative contour of the obstacle and acquiring the speed and the course angle of the current vehicle; the method specifically comprises the steps of acquiring point cloud from a sixteen-line laser radar, and removing points which do not influence vehicle running on the ground and in the air; projecting the remaining point cloud to a two-dimensional plane to obtain an image with pixel width and height, and extracting the alternative obstacle outline from the image;
the data matching module is used for matching each obstacle candidate contour with an obstacle according to a matching mode, tracking the matched obstacle contour and updating obstacle information; the obstacle information includes: the center point of the obstacle, the longitudinal relative speed of the obstacle, the transverse relative speed of the obstacle, the longitudinal absolute speed of the obstacle, the transverse absolute speed of the obstacle, the outline of the obstacle, the number of tracking frames and the number of lost frames; processing the alternative contour and the obstacle which are not matched; if the candidate outlines of the obstacles which are not matched are specifically included, the candidate outlines of the obstacles are used as new obstacles to be added into an obstacle list; deleting the obstacles from the obstacle list when the number of the unmatched obstacle lost frames is larger than a set value; when the obstacle which is not matched is combined with other obstacles to form an alternative contour, deleting the obstacle from an obstacle list;
the data output module is used for sequencing the obstacles; specifically, the method comprises the steps of sorting the obstacle contour points in a sequence from a plurality of points according to the number of the obstacle contour points; outputting a valid barrier, waiting for the next frame data and repeating the steps; the number of the effective barriers is greater than a set value for the number of the barrier tracking frames.
The embodiment of the invention provides a method and a system for detecting obstacles by utilizing a sixteen-line laser radar.
Drawings
In order to more clearly illustrate the technical solution of the present invention, the drawings needed to be used in the technical description of the present invention will be briefly introduced below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without inventive labor.
Fig. 1 is a schematic flow chart of an obstacle detection method using a sixteen-line lidar according to an embodiment of the present invention;
FIG. 2 is a schematic view of a sub-flow of a method for detecting obstacles using a sixteen-line lidar according to an embodiment of the present invention;
FIG. 3 is a diagram illustrating an obstacle detection system using a sixteen-line lidar in accordance with an embodiment of the present invention
Reference numerals:
the system setting module-1 data acquisition module-2 data matching module-3 data output module-4.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, 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 some, but not all, embodiments of the present invention. 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.
Fig. 1 is a schematic flow chart of an obstacle detection method using a sixteen-line lidar according to an embodiment of the present invention; as shown in fig. 1, the method comprises the following steps:
s1, initializing, and establishing a Kalman filtering model for performing speed detection and center point position update on the obstacle; establishing a vehicle body coordinate system which takes the projection point from the center of a rear axle of a vehicle to the ground as an origin in meters and takes the X-axis forward direction as the forward direction, the Y-axis forward direction as the leftward direction and the Z-axis forward direction as the upward direction, and establishing an image coordinate system which takes the X-axis forward direction as the rightward direction and the Y-axis forward direction as the backward direction and takes the pixel unit as the backward direction;
s2, extracting the candidate contour of the obstacle and obtaining the speed and the heading angle of the current vehicle; the method specifically comprises the steps of acquiring point cloud from a sixteen-line laser radar, and removing points which do not influence vehicle running on the ground and in the air; projecting the remaining point cloud to a two-dimensional plane to obtain an image with pixel width and height, and extracting the alternative obstacle outline from the image;
s3, matching each obstacle candidate contour with an obstacle according to a matching mode, tracking the matched obstacle contour, and updating obstacle information; the obstacle information includes: the center point of the obstacle, the longitudinal relative speed of the obstacle, the transverse relative speed of the obstacle, the longitudinal absolute speed of the obstacle, the transverse absolute speed of the obstacle, the outline of the obstacle, the number of tracking frames and the number of lost frames;
s4, processing the alternative contour and the obstacle which are not matched; if the candidate outlines of the obstacles which are not matched are specifically included, the candidate outlines of the obstacles are used as new obstacles to be added into an obstacle list; deleting the obstacles from the obstacle list when the number of the unmatched obstacle lost frames is larger than a set value; when the obstacle which is not matched is combined with other obstacles to form an alternative contour, deleting the obstacle from an obstacle list;
s5, sorting the obstacles; specifically, the method comprises the steps of sorting the obstacle contour points in a sequence from a plurality of points according to the number of the obstacle contour points;
s6, outputting effective barriers, waiting for the next frame data and repeating the steps S2-S5; the number of the effective barriers is greater than a set value for the number of the barrier tracking frames.
Specifically, the initialization setting comprises the setting of the number of obstacles, the setting of tracking frame number, the setting of lost frame value, the setting of candidate outline of the obstacle and the standard for judging and matching the obstacle; the number of the obstacles is set to be 0 in the embodiment of the invention; the tracking frame number represents the number of frames for continuously tracking the obstacle, and the lost frame number is the basis for judging whether the obstacle disappears; establishing a Kalman filtering model for carrying out speed detection and center point position updating on the obstacle, wherein the Kalman filtering model expression is as follows:
wherein, (x y v)xvy)T k-1Represents the state quantity of the obstacle at time k-1, (x yvxvy)T kRepresents the state quantity of the obstacle at time k, (z)xzy)T kShowing the observed quantity of the obstacle at the time k, X is the state quantity of the X coordinate value in the image coordinate system of the center point of the obstacle, Y is the state quantity of the Y coordinate value in the image coordinate system of the center point of the obstacle, vxIs the state quantity of the X-direction velocity of the obstacle in the image coordinate system, vyIs the state quantity of the speed in the Y direction in the image coordinate system of the obstacle, zxIs an observed quantity of X coordinate value, z, of the center point of the obstacle in the image coordinate systemyIs the observed quantity of the Y coordinate value of the center point of the obstacle in the image coordinate system.
Acquiring point cloud from a sixteen-line laser radar, removing points formed by leaves and the like which do not influence the vehicle running in the air and points formed by the leaves and the like on the ground, and acquiring the course angle and the speed of the current vehicle from the combined inertial navigation; matching each obstacle candidate contour with an obstacle according to a matching mode, tracking the matched obstacle contour, and updating obstacle information; wherein the updated relative velocity of the obstacle is obtained by multiplying the old relative velocity by the transformation matrix; the updated absolute velocity of the obstacle is the updated relative velocity of the obstacle plus the current vehicle velocity.
Updating the obstacle on the unmatched; for example, defining the matchTag, where the matchTag indicates whether the obstacle obj matches the alternative contour, and setting the value of the matchTag to 0 in the initial setting of the embodiment of the present invention; and updating the center point and the contour of each obstacle with the matching mark matchTag of 0 according to the longitudinal relative speed and the transverse relative speed of the obstacle, and adding 1 to the number of lost frames of the obstacle. And if the number of the lost frames of the obstacle is more than 5, deleting the obstacle from the obstacle list.
If the obstacle which is not matched is merged with other obstacles to form an alternative contour, the obstacle is directly deleted from the obstacle list. The basis for the unmatched obstacle to merge with another obstacle as an alternative contour is that the matchTag of the obstacle is 0 and the objTag is 1; the objTag indicates whether obj is included by a certain alternative contour, and the value of the objTag is set to 0 in the initial setting of the embodiment of the present invention.
Sorting the obstacles; specifically, the method comprises the steps of sorting the obstacle contour points in a sequence from a plurality of points according to the number of the obstacle contour points; the purpose of ordering the obstacles is to only retain the obstacle with the largest number of contour points among the merged obstacles when the plurality of obstacles are merged.
Outputting a valid obstacle; judging as an effective barrier when the number of the barrier tracking frames is larger than a set value; outputting the effective obstacles to other modules of the automatic driving system for use, waiting for the next frame data, repeating the operations of the steps S2 to S5, and outputting updated effective obstacle information; the output obstacle information includes: the profile of the obstacle, the width of the obstacle, the length of the obstacle, the center point of the obstacle, the longitudinal relative velocity of the obstacle, the lateral relative velocity of the obstacle, the longitudinal absolute velocity of the obstacle, the lateral absolute velocity of the obstacle. The width of the obstacle refers to the width of a rectangular frame surrounding the obstacle outline, and the length of the obstacle refers to the length of the rectangular frame surrounding the obstacle outline; the area of the obstacle refers to the length of the obstacle multiplied by the width of the obstacle.
The embodiment of the invention provides an obstacle detection method by utilizing a sixteen-line laser radar, which is characterized in that obstacle information is acquired and obstacle contour information is generated through point cloud data of the sixteen-line laser radar and positioning data of combined inertial navigation, so that the acquired obstacle information is more abundant and complete, and the generated obstacle contour information is more accurate.
Further, the removing of the points on the ground and in the mid-air that do not affect the vehicle driving in the step S2 specifically includes converting the laser radar coordinate system of the point cloud to a vehicle body coordinate system, removing the points of the vehicle body coordinate system where the Z coordinate value is less than threshold group, and removing the points where the Z coordinate value is greater than threshold sky. In the embodiment of the invention, in the actual project, the specific value of the point with the Z coordinate value smaller than threshold group is 0.3, and the specific value of the point with the Z coordinate value larger than threshold sky is 3.0.
FIG. 2 is a schematic view of a sub-flow of a method for detecting obstacles using a sixteen-line lidar according to an embodiment of the present invention; as shown in fig. 2, the step S3 of tracking the obstacle contour specifically includes:
s31, matching according to the overlapping area of the candidate contour and the obstacle and the distance between the obstacle and the candidate contour; the method specifically comprises the steps that the alternative contour and the obstacle which cannot be matched through the overlapping area of the alternative contour and the obstacle are matched according to the distance between the obstacle and the alternative contour;
s32, updating the obstacle outline; specifically, updating the obstacle contour by the union of the alternative contours;
and S33, combining the speed and the heading angle of the current vehicle, taking the updated center point of the obstacle outline as an observed quantity, and updating the center point and the speed information of the obstacle by using the Kalman filtering model.
The matching of the overlapping area of the candidate contour and the obstacle specifically includes adding the areas of all candidate contours split by the obstacle, and when the added area is larger than half of the area of the obstacle, judging that the obstacle matches with the candidate contours.
Specifically, for each candidate contour, a contourTag is defined, the initial value of the contourTag is 0, and the contourTag indicates whether the candidate contour matches with a certain obstacle. For example, let n be the number of obstacles and m be the number of candidate contours. Note that these n obstacles are obj1,obj2,...,objnRemember the m alternative wheelsContour of contourr1,contour2,...,contourm。
Get the ith obstacle obj one by oneiDefine SplitCounti，SplitCountiRepresents objiSplit into the number of candidate profiles, split countiSet to 0; definition of SplitTagi，SplitTagiRepresents objiA list of indices of the split candidate contours; definitions of matchTagi，matchTagiRepresents objiWhether to match with the alternative contour or not, the matchTag is usediIs set to 0; definition of objTagi，objTagiRepresents objiWhether or not it is contained by an alternative contour, will objTagiIs set to 0; then, the jth candidate contour contourr is taken one by onejOnly processing the alternative contour with contourTag of 0; when objiSplitting into contoursjThen, split countiAdding 1, and adding the jth alternative contour contourrjIs added to SplitTagiIn the list; when objiIs contourrjWhen included, the objTagiThe value of (d) is set to 1.
Will objiWith contourerjArea of intersection divided by contourjThe value of (2) is denoted as ratioijWill objiWith contourerjArea of intersection divided by objiIs denoted as ratio2ij，objiSplitting into contoursjThe criterion of (1) is that condition 1 is satisfied, condition 2 is satisfied, or condition 3 is satisfied. objiIs contourrjThe criterion of inclusion is ratio2ijGreater than 0.7.
Condition 1: ratio (R)ijGreater than 0.7.
Condition 2: ratio (R)ijGreater than 0.4 and ratio2ijGreater than 0.4.
Condition 3: ratio (R)ijGreater than 0.3 and contourrjThe width of the outer surrounding rectangular frame is less than 0.5 m and constantjThe length of the outer surrounding rectangular frame is less than 0.5 m.
Condition 1 represents contourjBasic quilt objiComprising, Condition 2 represents objiCombined with other obstacles to contoursjCondition 3 denotes contourjIs a relatively small candidate profile.
ratioijThe calculation formula of (a) is as follows:
ratio2ijthe calculation formula of (a) is as follows:
wherein S (obj)i∩contourj) Represents objiWith contourerjArea of intersection, S (constant)j) Represents a contourjArea of, S (obj)i) Represents objiThe area of (a).
All the root pairs objiAddition of the areas of the split candidate contours when the area of this addition is greater than objiHalf the area, it is considered that objiMatching these alternative contours, matchTagiSet to 1, objiTrack frame number of (1) plus objiIs set to 0, the contourTag of these matching candidate contours is set to 1, and obj is updated with the union of these candidate contoursiWith updated objiThe central point of the outer surrounding rectangular frame of the outline is used as an observed quantity, and Kalman filtering is used for obtaining the updated objiA center point, a longitudinal relative velocity, and a lateral relative velocity.
Specifically, the matching according to the distance between the obstacle and the candidate contour includes calculating the obstacle closest to the candidate contour, judging that the obstacle is matched with the candidate contour when the transverse difference between the center point of the candidate contour and the center point of the obstacle is smaller than 3 times the width of the candidate contour, and the longitudinal difference is smaller than 3 times the length of the candidate contour. And matching the obstacle with the matchTag of 0 with each candidate profile with the contourTag of 0 according to the distance. After mating, Carl was usedThe barrier is updated by the Mandarin filtering, a new central point, a longitudinal relative velocity and a transverse relative velocity are obtained, and the matching mark matchTag of the barrier is set to be 1. For example, for a given jth candidate contour, the obstacle closest to this candidate contour is calculated, and it is assumed that the ith obstacle is closest to this candidate contour. When contourjCentral point of (3) and objiWhen the transverse difference of the central points is less than 3 times of the width of the bounding box outside the alternative contour, and the longitudinal difference is less than 3 times of the length of the bounding box outside the alternative contour, the obj is considerediWith contourerjAnd (4) matching. If no obstacle is matched with the alternative contour, the alternative contour is considered to be a new obstacle, the alternative contour is taken as the contour of the new obstacle, the longitudinal relative speed of the new obstacle is set to be 0, the transverse relative speed of the new obstacle is set to be 0, the longitudinal absolute speed of the new obstacle is set to be the speed of the current vehicle, the transverse absolute speed of the new obstacle is set to be 0, the tracking frame number of the new obstacle is set to be 1, the lost frame number of the new obstacle is set to be 0, the matchTag of the new obstacle is set to be 1, and the new obstacle is added into the obstacle list.
The embodiment of the invention provides an obstacle detection method using a sixteen-line laser radar, which is characterized in that obstacle information is acquired and contour information is generated through point cloud data of the sixteen-line laser radar and positioning data of combined inertial navigation, so that the acquired obstacle information is richer and more complete, and the generated contour information is more accurate; and matching and tracking the obstacles, and acquiring the center position and speed information of the obstacles by adopting a Kalman filtering model, thereby reducing the cost.
Based on the above embodiments, fig. 3 is a diagram illustrating an obstacle detection system using a sixteen-line lidar according to an embodiment of the present invention; as shown in fig. 3, includes:
the system setting module 1 is used for initializing setting and establishing a Kalman filtering model for carrying out speed detection and center point position updating on the barrier; establishing a vehicle body coordinate system which takes the projection point from the center of a rear axle of a vehicle to the ground as an origin in meters and takes the X-axis forward direction as the forward direction, the Y-axis forward direction as the leftward direction and the Z-axis forward direction as the upward direction, and establishing an image coordinate system which takes the X-axis forward direction as the rightward direction and the Y-axis forward direction as the backward direction and takes the pixel unit as the backward direction;
the data acquisition module 2 is used for extracting the alternative contour of the obstacle and acquiring the speed and the course angle of the current vehicle; the method specifically comprises the steps of acquiring point cloud from a sixteen-line laser radar, and removing points which do not influence vehicle running on the ground and in the air; projecting the remaining point cloud to a two-dimensional plane to obtain an image with pixel width and height, and extracting the alternative obstacle outline from the image;
the data matching module 3 is used for matching each obstacle candidate contour with an obstacle according to a matching mode, tracking the matched obstacle contour and updating obstacle information; the obstacle information includes: the center point of the obstacle, the longitudinal relative speed of the obstacle, the transverse relative speed of the obstacle, the longitudinal absolute speed of the obstacle, the transverse absolute speed of the obstacle, the outline of the obstacle, the number of tracking frames and the number of lost frames; processing the alternative contour and the obstacle which are not matched; if the candidate outlines of the obstacles which are not matched are specifically included, the candidate outlines of the obstacles are used as new obstacles to be added into an obstacle list; deleting the obstacles from the obstacle list when the number of the unmatched obstacle lost frames is larger than a set value; when the obstacle which is not matched is combined with other obstacles to form an alternative contour, deleting the obstacle from an obstacle list;
the data output module 4 sequences the obstacles; specifically, the method comprises the steps of sorting the obstacle contour points in a sequence from a plurality of points according to the number of the obstacle contour points; outputting a valid barrier, waiting for the next frame data and repeating the steps; the number of the effective barriers is greater than a set value for the number of the barrier tracking frames.
The embodiment of the invention provides an obstacle detection system utilizing a sixteen-line laser radar to execute the method, and the obstacle information is acquired and the contour information is generated through the point cloud data of the sixteen-line laser radar and the positioning data of the combined inertial navigation, so that the acquired obstacle information is richer and more complete, and the generated contour information is more accurate; and matching and tracking the obstacles, and acquiring the center position and speed information of the obstacles by adopting a Kalman filtering model, thereby reducing the cost.
The above-described embodiments of the apparatus are merely illustrative, and 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 place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. An obstacle detection method using a sixteen-line laser radar, characterized by comprising the steps of:
s1, initializing, and establishing a Kalman filtering model for performing speed detection and center point position update on the obstacle; establishing a vehicle body coordinate system which takes the projection point from the center of a rear axle of a vehicle to the ground as an origin in meters and takes the X-axis forward direction as the forward direction, the Y-axis forward direction as the leftward direction and the Z-axis forward direction as the upward direction, and establishing an image coordinate system which takes the X-axis forward direction as the rightward direction and the Y-axis forward direction as the backward direction and takes the pixel unit as the backward direction;
s2, extracting the candidate contour of the obstacle and obtaining the speed and the heading angle of the current vehicle; the method specifically comprises the steps of acquiring point cloud from a sixteen-line laser radar, and removing points which do not influence vehicle running on the ground and in the air; projecting the remaining point cloud to a two-dimensional plane to obtain an image with pixel width and height, and extracting the alternative obstacle outline from the image;
s3, matching each obstacle candidate contour with an obstacle according to a matching mode, tracking the matched obstacle contour, and updating obstacle information; the obstacle information includes: the center point of the obstacle, the longitudinal relative speed of the obstacle, the transverse relative speed of the obstacle, the longitudinal absolute speed of the obstacle, the transverse absolute speed of the obstacle, the outline of the obstacle, the number of tracking frames and the number of lost frames;
s4, processing the alternative contour and the obstacle which are not matched; if the candidate outlines of the obstacles which are not matched are specifically included, the candidate outlines of the obstacles are used as new obstacles to be added into an obstacle list; deleting the obstacles from the obstacle list when the number of the unmatched obstacle lost frames is larger than a set value; when the obstacle which is not matched is combined with other obstacles to form an alternative contour, deleting the obstacle from an obstacle list;
s5, sorting the obstacles; specifically, the method comprises the steps of sorting the obstacle contour points in a sequence from a plurality of points according to the number of the obstacle contour points;
s6, outputting effective barriers, waiting for the next frame data and repeating the steps S2-S5; the number of the effective barriers is greater than a set value for the number of the barrier tracking frames.
2. The obstacle detection method using a sixteen-line lidar according to claim 1, wherein said removing points that do not affect the traveling of the vehicle on the ground and in the mid-air at step S2 specifically includes converting the lidar coordinate system of the point cloud to a vehicle body coordinate system, removing points of the vehicle body coordinate system having Z-coordinate values less than threshold group, and removing points of the vehicle body coordinate system having Z-coordinate values greater than threshold sky.
3. The obstacle detection method using sixteen-line lidar according to claim 1, wherein the step of S3, tracking the obstacle profile specifically includes:
s31, matching according to the overlapping area of the candidate contour and the obstacle and the distance between the obstacle and the candidate contour; the method specifically comprises the steps that the alternative contour and the obstacle which cannot be matched through the overlapping area of the alternative contour and the obstacle are matched according to the distance between the obstacle and the alternative contour;
s32, updating the obstacle outline; specifically, updating the obstacle contour by the union of the alternative contours;
and S33, combining the speed and the heading angle of the current vehicle, taking the updated center point of the obstacle outline as an observed quantity, and updating the center point and the speed information of the obstacle by using the Kalman filtering model.
4. The obstacle detection method using sixteen-line lidar according to claim 3, wherein the matching of the candidate contours to the obstacle overlapping area specifically comprises adding the areas of all candidate contours split by the obstacle, and when the added area is greater than half of the area of the obstacle, determining that the obstacle matches with these candidate contours.
5. The obstacle detection method using a sixteen-line lidar according to claim 3, wherein the matching according to the distance between the obstacle and the candidate contour specifically comprises calculating the obstacle closest to the candidate contour, and when a lateral difference between a center point of the candidate contour and a center point of the obstacle is smaller than 3 times a width of the candidate contour, and a longitudinal difference is smaller than 3 times a length of the candidate contour, determining that the obstacle and the candidate contour match.
6. An obstacle detection system using a sixteen-line lidar, comprising:
the system setting module is used for carrying out initialization setting and establishing a Kalman filtering model for carrying out speed detection and center point position updating on the barrier; establishing a vehicle body coordinate system which takes the projection point from the center of a rear axle of a vehicle to the ground as an origin in meters and takes the X-axis forward direction as the forward direction, the Y-axis forward direction as the leftward direction and the Z-axis forward direction as the upward direction, and establishing an image coordinate system which takes the X-axis forward direction as the rightward direction and the Y-axis forward direction as the backward direction and takes the pixel unit as the backward direction;
the data acquisition module is used for extracting the alternative contour of the obstacle and acquiring the speed and the course angle of the current vehicle; the method specifically comprises the steps of acquiring point cloud from a sixteen-line laser radar, and removing points which do not influence vehicle running on the ground and in the air; projecting the remaining point cloud to a two-dimensional plane to obtain an image with pixel width and height, and extracting the alternative obstacle outline from the image;
the data matching module is used for matching each obstacle candidate contour with an obstacle according to a matching mode, tracking the matched obstacle contour and updating obstacle information; the obstacle information includes: the center point of the obstacle, the longitudinal relative speed of the obstacle, the transverse relative speed of the obstacle, the longitudinal absolute speed of the obstacle, the transverse absolute speed of the obstacle, the outline of the obstacle, the number of tracking frames and the number of lost frames; processing the alternative contour and the obstacle which are not matched; if the candidate outlines of the obstacles which are not matched are specifically included, the candidate outlines of the obstacles are used as new obstacles to be added into an obstacle list; deleting the obstacles from the obstacle list when the number of the unmatched obstacle lost frames is larger than a set value; when the obstacle which is not matched is combined with other obstacles to form an alternative contour, deleting the obstacle from an obstacle list;
the data output module is used for sequencing the obstacles; specifically, the method comprises the steps of sorting the obstacle contour points in a sequence from a plurality of points according to the number of the obstacle contour points; outputting a valid barrier, waiting for the next frame data and repeating the steps; the number of the effective barriers is greater than a set value for the number of the barrier tracking frames.
7. The obstacle detection system using a sixteen-line lidar according to claim 6, wherein said data acquisition module removing points on the ground and in mid-air that do not affect vehicle travel specifically comprises transforming the lidar coordinate system of said point cloud to a body coordinate system, removing points of said body coordinate system having Z coordinate values less than threshold group, and removing points having Z coordinate values greater than threshold sky.
8. The obstacle detection system using sixteen-line lidar according to claim 6, wherein the tracking of the obstacle profile by the data matching module specifically comprises: matching according to the overlapping area of the alternative contour and the obstacle and matching according to the distance between the obstacle and the alternative contour; the method specifically comprises the steps that the alternative contour and the obstacle which cannot be matched through the overlapping area of the alternative contour and the obstacle are matched according to the distance between the obstacle and the alternative contour; updating the obstacle profile; specifically, updating the obstacle contour by the union of the alternative contours; and updating the central point and the speed information of the obstacle by using the Kalman filtering model by combining the speed and the course angle of the current vehicle and taking the updated central point of the obstacle outline as an observed quantity.
9. The system of claim 8, wherein the matching of the overlap area between the candidate contours and the obstacle in the data matching module specifically comprises adding the areas of all candidate contours split by the obstacle, and determining that the obstacle matches the candidate contours when the added area is greater than half of the area of the obstacle.
10. The method according to claim 8, wherein the matching in the data matching module according to the distance between the obstacle and the candidate contour specifically includes calculating the obstacle closest to the candidate contour, and when the lateral difference between the center point of the candidate contour and the center point of the obstacle is smaller than 3 times the width of the candidate contour, and the longitudinal difference is smaller than 3 times the length of the candidate contour, determining that the obstacle matches the candidate contour.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911131844.9A CN111007534A (en) | 2019-11-19 | 2019-11-19 | Obstacle detection method and system using sixteen-line laser radar |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911131844.9A CN111007534A (en) | 2019-11-19 | 2019-11-19 | Obstacle detection method and system using sixteen-line laser radar |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111007534A true CN111007534A (en) | 2020-04-14 |
Family
ID=70111904
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911131844.9A Pending CN111007534A (en) | 2019-11-19 | 2019-11-19 | Obstacle detection method and system using sixteen-line laser radar |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111007534A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111665524A (en) * | 2020-04-29 | 2020-09-15 | 武汉光庭科技有限公司 | Method and system for ground rejection by utilizing multi-line laser radar |
| CN111830526A (en) * | 2020-09-17 | 2020-10-27 | 上海驭矩信息科技有限公司 | Container positioning method and device based on multi-line laser data fusion |
| CN112034488A (en) * | 2020-08-28 | 2020-12-04 | 北京海益同展信息科技有限公司 | Automatic target object labeling method and device |
| CN113110462A (en) * | 2021-04-21 | 2021-07-13 | 广州极飞科技股份有限公司 | Obstacle information processing method and device and operating equipment |
| CN113335310A (en) * | 2021-07-21 | 2021-09-03 | 新石器慧通（北京）科技有限公司 | Decision-based exercise planning method and device, electronic equipment and storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9193355B2 (en) * | 2012-09-05 | 2015-11-24 | Google Inc. | Construction zone sign detection using light detection and ranging |
| CN106199558A (en) * | 2016-08-18 | 2016-12-07 | 宁波傲视智绘光电科技有限公司 | Barrier method for quick |
| CN106570454A (en) * | 2016-10-10 | 2017-04-19 | 同济大学 | Pedestrian traffic parameter extraction method based on mobile laser scanning |
| CN107194957A (en) * | 2017-04-17 | 2017-09-22 | 武汉光庭科技有限公司 | The method that laser radar point cloud data is merged with information of vehicles in intelligent driving |
| CN107239746A (en) * | 2017-05-16 | 2017-10-10 | 东南大学 | A kind of obstacle recognition tracking towards roadside assistance security monitoring |
| CN109472831A (en) * | 2018-11-19 | 2019-03-15 | 东南大学 | Obstacle recognition range-measurement system and method towards road roller work progress |
| CN109597097A (en) * | 2018-12-06 | 2019-04-09 | 北京主线科技有限公司 | Scan-type obstacle detection method based on multi-thread laser |
| CN110210389A (en) * | 2019-05-31 | 2019-09-06 | 东南大学 | A kind of multi-targets recognition tracking towards road traffic scene |
-
2019
- 2019-11-19 CN CN201911131844.9A patent/CN111007534A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9193355B2 (en) * | 2012-09-05 | 2015-11-24 | Google Inc. | Construction zone sign detection using light detection and ranging |
| CN106199558A (en) * | 2016-08-18 | 2016-12-07 | 宁波傲视智绘光电科技有限公司 | Barrier method for quick |
| CN106570454A (en) * | 2016-10-10 | 2017-04-19 | 同济大学 | Pedestrian traffic parameter extraction method based on mobile laser scanning |
| CN107194957A (en) * | 2017-04-17 | 2017-09-22 | 武汉光庭科技有限公司 | The method that laser radar point cloud data is merged with information of vehicles in intelligent driving |
| CN107239746A (en) * | 2017-05-16 | 2017-10-10 | 东南大学 | A kind of obstacle recognition tracking towards roadside assistance security monitoring |
| CN109472831A (en) * | 2018-11-19 | 2019-03-15 | 东南大学 | Obstacle recognition range-measurement system and method towards road roller work progress |
| CN109597097A (en) * | 2018-12-06 | 2019-04-09 | 北京主线科技有限公司 | Scan-type obstacle detection method based on multi-thread laser |
| CN110210389A (en) * | 2019-05-31 | 2019-09-06 | 东南大学 | A kind of multi-targets recognition tracking towards road traffic scene |
Non-Patent Citations (2)
| Title |
|---|
| 张彩红: "城区环境下基于激光雷达的障碍物聚类和跟踪方法研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
| 邱少波等: "《车辆工程先进技术研究丛书 汽车碰撞安全工程》", 31 January 2016 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111665524A (en) * | 2020-04-29 | 2020-09-15 | 武汉光庭科技有限公司 | Method and system for ground rejection by utilizing multi-line laser radar |
| CN112034488A (en) * | 2020-08-28 | 2020-12-04 | 北京海益同展信息科技有限公司 | Automatic target object labeling method and device |
| CN112034488B (en) * | 2020-08-28 | 2023-05-02 | 京东科技信息技术有限公司 | Automatic labeling method and device for target object |
| CN111830526A (en) * | 2020-09-17 | 2020-10-27 | 上海驭矩信息科技有限公司 | Container positioning method and device based on multi-line laser data fusion |
| CN111830526B (en) * | 2020-09-17 | 2020-12-29 | 上海驭矩信息科技有限公司 | Container positioning method and device based on multi-line laser data fusion |
| WO2022057747A1 (en) * | 2020-09-17 | 2022-03-24 | 上海驭矩信息科技有限公司 | Container positioning method and apparatus based on multi-line laser data fusion |
| CN113110462A (en) * | 2021-04-21 | 2021-07-13 | 广州极飞科技股份有限公司 | Obstacle information processing method and device and operating equipment |
| CN113335310A (en) * | 2021-07-21 | 2021-09-03 | 新石器慧通（北京）科技有限公司 | Decision-based exercise planning method and device, electronic equipment and storage medium |
| CN113335310B (en) * | 2021-07-21 | 2021-11-30 | 新石器慧通（北京）科技有限公司 | Decision-based exercise planning method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111007534A (en) | Obstacle detection method and system using sixteen-line laser radar | |
| CN109521756B (en) | Obstacle motion information generation method and apparatus for unmanned vehicle | |
| CN110009718B (en) | Three-dimensional high-precision map generation method and device | |
| KR102628778B1 (en) | Method and apparatus for positioning, computing device, computer-readable storage medium and computer program stored in medium | |
| Danescu et al. | Probabilistic lane tracking in difficult road scenarios using stereovision | |
| CN102208035B (en) | Image processing system and position measuring system | |
| CN110674705B (en) | Small-sized obstacle detection method and device based on multi-line laser radar | |
| CN104517275A (en) | Object detection method and system | |
| CN111986261B (en) | Vehicle positioning method and device, electronic equipment and storage medium | |
| CN113920198B (en) | Coarse-to-fine multi-sensor fusion positioning method based on semantic edge alignment | |
| CN110705385B (en) | Method, device, equipment and medium for detecting angle of obstacle | |
| US20220398856A1 (en) | Method for reconstruction of a feature in an environmental scene of a road | |
| Kellner et al. | Road curb detection based on different elevation mapping techniques | |
| CN111681172A (en) | Method, equipment and system for cooperatively constructing point cloud map | |
| CN113835102A (en) | Lane line generation method and device | |
| CN114485698B (en) | Intersection guide line generation method and system | |
| Xiong et al. | Road-Model-Based road boundary extraction for high definition map via LIDAR | |
| CN114387576A (en) | Lane line identification method, system, medium, device and information processing terminal | |
| WO2021063756A1 (en) | Improved trajectory estimation based on ground truth | |
| Muffert et al. | Stix-fusion: A probabilistic stixel integration technique | |
| CN112651991A (en) | Visual positioning method, device and computer system | |
| JP2020098196A (en) | Estimation device, control method, program, and storage medium | |
| CN114511590A (en) | Intersection multi-guide-line construction method based on monocular vision 3D vehicle detection and tracking | |
| CN113227713A (en) | Method and system for generating environment model for positioning | |
| CN109901589B (en) | Mobile robot control method and device |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200414 |