CN110794392B - Vehicle positioning method and device, vehicle and storage medium - Google Patents
Vehicle positioning method and device, vehicle and storage medium Download PDFInfo
- Publication number
- CN110794392B CN110794392B CN201910978766.XA CN201910978766A CN110794392B CN 110794392 B CN110794392 B CN 110794392B CN 201910978766 A CN201910978766 A CN 201910978766A CN 110794392 B CN110794392 B CN 110794392B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- cloud data
- precision map
- point cloud
- determining
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 74
- 230000008569 process Effects 0.000 claims abstract description 19
- 238000004422 calculation algorithm Methods 0.000 claims description 8
- 230000004807 localization Effects 0.000 claims description 8
- 238000000605 extraction Methods 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 16
- 238000004364 calculation method Methods 0.000 description 14
- 239000011159 matrix material Substances 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 230000009466 transformation Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 239000013598 vector Substances 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/02—Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
- G01S13/06—Systems determining position data of a target
-
- 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
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
- G01S13/931—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of land vehicles
Abstract
The invention discloses a vehicle positioning method, a device, a vehicle and a storage medium, wherein the method comprises the following steps: acquiring target point cloud data acquired through millimeter wave radar arranged on a vehicle in the running process of the vehicle, and determining the position of a target object corresponding to the target point cloud data in a high-precision map according to the target point cloud data and a pre-established high-precision map, wherein the high-precision map comprises a feature descriptor determined according to the point cloud data acquired by the millimeter wave radar, and determining the position of the vehicle in the high-precision map according to the position of the target object in the high-precision map and the target point cloud data. The method and the device have the advantages that the position of the vehicle in the high-precision map is determined according to the cloud data of the target points acquired by the millimeter wave radar, and compared with the prior art, the method and the device are low in cost, high in positioning precision and high in positioning reliability.
Description
Technical Field
The embodiment of the invention relates to the field of intelligent driving, in particular to a vehicle positioning method, a vehicle positioning device, a vehicle and a storage medium.
Background
The intelligent driving positioning technology can help the automobile to realize accurate positioning, and unmanned driving is realized through accurate path planning. Meanwhile, the intelligent driving positioning technology is also a part of the artificial intelligence technology, and is the most important one for realizing intelligent driving.
At present, a vehicle can be positioned by a camera-based simultaneous positioning and map construction (Simultaneous Localization and Mapping, SLAM) technology, and the main principle is that an environment model is built in a motion process through a camera under the condition that no environment priori information exists, and meanwhile, the motion of the vehicle is estimated.
However, the above-mentioned mode too relies on the camera, when the weather condition is comparatively abominable, can't realize the location, and the reliability is lower, and simultaneously, the location precision of above-mentioned mode is also lower.
Disclosure of Invention
The invention provides a vehicle positioning method, a vehicle positioning device, a vehicle and a storage medium, which are used for solving the technical problems of low reliability and low positioning precision of the existing vehicle positioning mode.
In a first aspect, an embodiment of the present invention provides a vehicle positioning method, including:
acquiring target point cloud data acquired through millimeter wave radars arranged on a vehicle in the running process of the vehicle;
determining the position of a target object corresponding to the target point cloud data in the high-precision map according to the target point cloud data and a pre-established high-precision map; the high-precision map comprises a feature descriptor determined according to point cloud data acquired by the millimeter wave radar;
and determining the position of the vehicle in the high-precision map according to the position of the target object in the high-precision map and the target point cloud data.
In a second aspect, an embodiment of the present invention provides a vehicle positioning device, including:
the acquisition module is used for acquiring target point cloud data acquired through millimeter wave radars arranged on the vehicle in the running process of the vehicle;
the first determining module is used for determining the position of a target object corresponding to the target point cloud data in the high-precision map according to the target point cloud data and the high-precision map established in advance; the high-precision map comprises a feature descriptor determined according to point cloud data acquired by the millimeter wave radar;
and the second determining module is used for determining the position of the vehicle in the high-precision map according to the position of the target object in the high-precision map and the target point cloud data.
In a third aspect, an embodiment of the present invention further provides a vehicle, including:
one or more processors;
a memory for storing one or more programs;
one or more millimeter wave radars;
the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the vehicle locating method as provided in the first aspect.
In a fourth aspect, an embodiment of the present invention also provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the vehicle positioning method as provided in the first aspect.
The embodiment provides a vehicle positioning method, a device, a vehicle and a storage medium, wherein the method comprises the following steps: acquiring target point cloud data acquired through millimeter wave radar arranged on a vehicle in the running process of the vehicle, and determining the position of a target object corresponding to the target point cloud data in a high-precision map according to the target point cloud data and a pre-established high-precision map, wherein the high-precision map comprises a feature descriptor determined according to the point cloud data acquired by the millimeter wave radar, and determining the position of the vehicle in the high-precision map according to the position of the target object in the high-precision map and the target point cloud data. The method and the device have the advantages that the position of the vehicle in the high-precision map is determined according to the cloud data of the target points acquired by the millimeter wave radar, and compared with the prior art, the method and the device are low in cost, high in positioning precision and high in positioning reliability.
Drawings
FIG. 1A is a schematic diagram of an application scenario of an embodiment of a vehicle positioning method provided by the present invention;
FIG. 1B is a schematic illustration of a vehicle provided by the present invention;
FIG. 2 is a schematic flow chart of an embodiment of a vehicle positioning method according to the present invention;
FIG. 3 is a schematic illustration of an unclosed high-precision map constructed in an embodiment of a vehicle positioning method;
FIG. 4A is a schematic view of point cloud data before surface normal estimation in an embodiment of a vehicle positioning method;
FIG. 4B is a schematic diagram of the result of the surface normal estimation in an embodiment of the vehicle positioning method;
FIG. 5A is a schematic diagram of an area of influence in determining a point feature histogram description in an embodiment of a vehicle localization method;
FIG. 5B is a schematic diagram of a coordinate system when determining a point feature histogram description in an embodiment of a vehicle localization method;
FIG. 6 is a schematic diagram of an area of influence in determining a fast point feature histogram description in an embodiment of a vehicle localization method;
FIG. 7A is a schematic diagram of one implementation of a vehicle positioning method embodiment for determining the position of a vehicle in a high-precision map;
FIG. 7B is a schematic diagram of another implementation of a vehicle positioning method embodiment for determining the position of a vehicle in a high-precision map;
FIG. 8 is a schematic view of an embodiment of a vehicle positioning device according to the present invention;
fig. 9 is a schematic structural view of a vehicle according to the present invention.
Detailed Description
The invention is described in further detail below with reference to the drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting thereof. It should be further noted that, for convenience of description, only some, but not all of the structures related to the present invention are shown in the drawings.
Fig. 1A is a schematic diagram of an application scenario of an embodiment of a vehicle positioning method provided by the present invention. Fig. 1B is a schematic view of a vehicle according to the present invention. As shown in fig. 1B, the present invention provides a vehicle 10 on which a millimeter wave radar 11 is mounted. One millimeter wave radar 11 may be installed at the vehicle head, specifically at the midpoint of the vehicle head in the left-right direction. Alternatively, two millimeter wave radars 11 are respectively installed at both ends in the left-right direction of the vehicle head, two millimeter wave radars 11 are respectively installed at midpoints in the front-rear direction of both sides of the vehicle, two millimeter wave radars 11 are respectively installed at both ends in the left-right direction of the vehicle tail, and one millimeter wave radar 11 is installed at the midpoint in the left-right direction of the vehicle tail. As shown in fig. 1A, the vehicle may collect road condition information through a millimeter wave radar installed thereon during traveling. In this embodiment, the millimeter wave radar may collect target point cloud data corresponding to the target point 13. In the intelligent driving field, the vehicle can determine the position of the vehicle in the high-precision map according to the target point cloud data acquired by the millimeter wave radar, so that the vehicle is positioned.
Fig. 2 is a schematic flow chart of an embodiment of a vehicle positioning method according to the present invention. The embodiment is suitable for the scene of positioning the vehicle in the intelligent driving field. The present embodiment may be performed by a vehicle positioning device, which may be implemented in software and/or hardware, which may be integrated in a vehicle. As shown in fig. 2, the vehicle positioning method provided in this embodiment includes the following steps:
step 201: and acquiring target point cloud data acquired through millimeter wave radars arranged on the vehicle in the running process of the vehicle.
In particular, millimeter waves are electromagnetic waves, which consist of alternating electric and magnetic fields, and can propagate in vacuum, air and free space. The frequency band is special, the frequency is higher than the radio frequency, lower than the visible light and the infrared ray, and the frequency is approximately in the range of 10 GHz-200 GHz. In this frequency band, millimeter wave related characteristics make it very suitable for application in the field of vehicles. The millimeter wave radar frequency band in the vehicle-mounted field mainly comprises three sections: 24GHz band, 77GHz band and 79GHz band. The 24GHz frequency band is mainly used for blind spot monitoring and lane changing assistance of automobiles. The millimeter wave radar with the 24GHz frequency band can be installed in a rear bumper of a vehicle and is used for monitoring whether a lane on two sides behind the vehicle has a car or not and whether lane changing can be carried out or not. The millimeter wave radar with the frequency range of 77GHz is mainly used for being assembled on a front bumper of a vehicle, detecting the distance between the millimeter wave radar and the front vehicle and the speed of the front vehicle, and realizing the functions of active safety fields such as emergency braking, automatic vehicle following and the like. The millimeter wave radar in the 79GHz frequency band has very high resolution, and can reach 5cm. This resolution is very valuable in the field of intelligent driving, because intelligent driving automobiles are intended to distinguish between many delicate objects such as pedestrians. By way of example, the performance parameters of a current millimeter wave radar may be as follows: the frequency is 76-77 GHz, the detection distance is 0.5-250 m, the viewing angle is +/-15 degrees, the resolution is 0.5 m, the relative speed is-75 to +60 m/s, and the speed resolution is 0.6 m/s. The millimeter wave radar according to the present embodiment may be any one of the above-described millimeter wave radars in any frequency band, and the present embodiment is not limited thereto.
The millimeter wave radar sends out directional millimeter waves of corresponding wavebands through the transmitting antenna, when the millimeter waves meet the target point, the millimeter waves are reflected back, and the reflected millimeter waves are received through the receiving antenna. And measuring the position of the target point according to the time difference between millimeter wave receiving and transmitting, and delaying the time td=2R/C, wherein R is the distance between the target point and the vehicle, and C is the light speed. For the direction angle of the target point, the detection principle of the millimeter wave radar is as follows: after millimeter waves are emitted through the transmitting antennas of the millimeter wave radar, the millimeter waves are reflected back by the target points, and the reflected millimeter waves are received by the receiving antennas arranged in parallel of the millimeter wave radar. By receiving the phase difference of the millimeter waves reflected by the same target point, the azimuth angle of the target point relative to the vehicle can be calculated.
In this embodiment, the vehicle positioning device may be specifically integrated in a vehicle controller of a vehicle. And in the running process of the vehicle, the millimeter wave radar arranged on the vehicle acquires cloud data of the target point in real time. In one implementation, the vehicle positioning device may actively acquire target point cloud data acquired by the millimeter wave radar. In another implementation mode, the millimeter wave radar actively transmits the collected cloud data of the target point to the vehicle positioning device.
Alternatively, the target point cloud data in the present embodiment may include the acquired coordinate values of the target point in the vehicle coordinate system. The vehicle coordinate system in the present embodiment is a coordinate system established with a point on the vehicle as the origin. The vehicle coordinate system in the present embodiment may be a three-dimensional coordinate system or a polar coordinate system. More specifically, the origin of the vehicle coordinate system is the center point of the rear axis of the vehicle, the positive X-axis direction of the vehicle coordinate system is the left-hand direction of the vehicle when the vehicle is facing forward, the positive Y-axis direction is the forward direction of the vehicle, and the positive Z-axis direction is the upward direction of the vehicle.
It should be noted that, the target point cloud data may include coordinate values of the collected plurality of target points in a vehicle coordinate system.
In this embodiment, the target point corresponding to the cloud data of the target point acquired by the millimeter wave radar is a scattered point, and is not a determined point on a certain object. According to the implementation mode, on one hand, a millimeter wave radar is not needed to determine whether the target point is a point on a certain object, so that the cost is reduced, and on the other hand, the compatibility is good.
Step 202: and determining the position of a target object corresponding to the target point cloud data in the high-precision map according to the target point cloud data and the high-precision map established in advance.
The high-precision map comprises a feature descriptor determined according to point cloud data acquired by the millimeter wave radar.
Specifically, a high-precision map is pre-established in the embodiment, and specifically, the point cloud data can be collected through millimeter waves Lei Dacai arranged on the collection vehicle. And after the point cloud data are acquired, splicing the point cloud data to form a high-precision map. The specific process is as follows: (1) Clustering the point cloud data to form a plurality of point cloud data sets; (2) Extracting key points from the two point cloud data sets according to the same key point selection standard; (3) Calculating characteristic descriptors of all selected key points respectively; (4) Estimating the corresponding relation of the feature descriptors in the two point cloud data sets by combining the coordinate positions of the feature descriptors in the two point cloud data sets and based on the similarity of the features and the positions between the two point cloud data sets, and primarily estimating corresponding point pairs; (5) Assuming that the data is noisy, removing erroneous corresponding point pairs that have an effect on the splice; (6) And estimating rigid transformation by using the residual correct corresponding relation to finish the splicing of the point cloud data set.
When a high-precision map based on millimeter wave radar is constructed, the running track of the acquisition vehicle is recorded by means of a global positioning system (Global Positioning System, GPS). And the point cloud data are spliced to form an initial high-precision map, and the high-precision map is closed to compare the running track of the vehicle. Because cloud data has angle errors in the splicing process, the accumulation of the construction errors of the high-precision map gradually increases, so that the initial high-precision map needs to be corrected by means of the GPS recorded driving tracks, and the high-precision map is closed. FIG. 3 is a schematic illustration of an unclosed high precision map constructed in an embodiment of a vehicle positioning method. As shown in fig. 3, the dark line is the GPS recorded vehicle track, and the light line is the initial high-precision map that is not closed and is constructed from the point cloud data.
Further, in order to improve the efficiency of the subsequent vehicle positioning, the high-precision map further comprises a feature descriptor of the point cloud data. The high-precision map may also include size information of the object.
Alternatively, the specific implementation procedure of step 202 may be: determining key point cloud data from the target point cloud data; determining feature descriptors of the key point cloud data according to a point cloud data feature descriptor extraction algorithm; and determining the position of the target object corresponding to the target point cloud data in the high-precision map according to the feature descriptors of the key point cloud data and the feature descriptors in the high-precision map.
The keypoint cloud data may be determined from the target point cloud data according to a keypoint extraction criterion. The point cloud data feature descriptor extraction algorithm in the embodiment comprises the following four algorithms: estimating a surface normal of the point cloud, a point feature histogram (Point Feature Histograms, PFH) descriptor, a fast point feature histogram (Fast Point Feature Histograms, FPFH) and a viewpoint feature histogram (Viewpoint Feature Histogram, VFH) descriptor. These four algorithms are described in detail below.
(1) Estimating a surface normal of the point cloud:
surface normals are important attributes of a geometric surface, for a known geometric surface, the normal direction to a point on the surface can be inferred from vectors perpendicular to the point surface. The point cloud data set appears as a set of point samples on the surface of the real object, from which the surface normal can be approximately inferred directly in this embodiment.
The problem of determining a point normal to the surface approximates the problem of estimating a tangent surface normal to the surface, and thus becomes a least squares plane fit estimation problem after conversion. The solution of estimating the surface normal therefore becomes to analyze eigenvectors and eigenvalues of a covariance matrix created from neighboring elements of the query point. The query points herein refer to key point cloud data. More specifically, for each point Pi, the corresponding covariance matrix C is given by:
where k is the number of points adjacent to point Pi,representing the three-dimensional centroid, lambda of nearest neighbor elements j Is the j-th eigenvalue of the covariance matrix,>is the j-th feature vector.
Fig. 4A is a schematic diagram of point cloud data before surface normal estimation in an embodiment of a vehicle positioning method.
Fig. 4B is a schematic diagram of the result of the surface normal estimation in the vehicle positioning method embodiment.
(2) PFH descriptor:
PFH is related to three-dimensional data of coordinate axes and surface normals, and thus can be better used to describe three-dimensional features of a point cloud data set. The PFH calculation mode describes the k neighborhood geometrical attribute of the query point by parameterizing the space difference between the query point and the neighborhood point and forming a multi-dimensional histogram. The Gao Weichao space in which the histogram resides provides a measurable information space for the feature representation that is invariant to the 6-dimensional pose of the point cloud corresponding surface and robust to noise levels at different sampling densities or neighborhoods. The PFH representation is based on the relationship between the query point and its k-neighborhood and their estimated normals, and in short, it considers all interactions between the estimated normals in an attempt to capture the best sample surface variations to describe the geometry of the sample. Thus, the special featureThe synthesis of the symptomatic hyperspace depends on the quality of the surface normal estimate for each point. FIG. 5A is a schematic diagram of an area of influence in determining a point feature histogram description in an embodiment of a vehicle localization method. As shown in fig. 5A, a query point (P q ) Is the PFH calculated area of influence, P q Is positioned in the middle of the sphere and has a radius r and P q All k-neighbor elements of (i.e. with point P q All points whose distance is smaller than the radius r) are all interconnected in one network. The final PFH descriptor is a histogram obtained by computing the relationship between all two points in the neighborhood.
Fig. 5B is a schematic diagram of a coordinate system when determining a point feature histogram description in an embodiment of a vehicle localization method. As shown in fig. 5B, to calculate two points P i And P j And the normal n corresponding to them i And n j The relative deviation between them defines a fixed local coordinate system at one of the points.
u＝n s
u＝n s
w＝u×v
Using the u-v-w coordinate system in FIG. 5B, normal n s And n t The deviation between can be expressed in terms of a set of angles, as follows:
α＝v·n t
θ＝arctan(w·n t ,u·n t )
where d is the Euclidean distance between two points Ps and Pt, d= ||P t –P s || 2 . Four sets of values are calculated for each pair of points within the k-neighborhood.
Creating a final PFH descriptor for the query point, all the quaternions will be put in the histogram in some statistical way, which first fits each range of eigenvaluesDividing into b sub-intervals and counting the number of points falling within each sub-interval, since three-quarters of features are measured as angles between normals in the above, their parameter values can be very easily normalized to the same interval on a triangulated circle. An example of a statistic is: dividing each characteristic interval into equal numbers of equal divisions, for which purpose b is created in a fully associated space 4 Histogram of bins. In this space, the number of statistics of a certain bin in a histogram is increased by 4 eigenvalues corresponding to one point.
In summary, in the process of calculating the PFH descriptor, the actual calculation process of each point P in the point cloud data P is: obtaining nearest neighbor elements of the p point; for each point in the neighborhood, calculating three angle characteristic parameter values of the point; all results are counted into one output histogram.
(3) FPFH descriptor:
knowing that there are n points in the point cloud data P, k is the number of neighborhoods considered when computing the feature vector for each point P in the point cloud P in its theoretical calculation of PFH. In practical applications, the computation of PFH of dense point clouds requires a lot of computation resources. Therefore, after simplifying the PFH calculation mode, FPFH can be obtained to describe the characteristics of the point cloud.
To simplify the feature computation of the histogram, we perform the following procedure:
the first step: for each query point, a tuple between this point and its neighborhood point is computed, called a simplified point feature histogram (Simple Point Feature Histograms, SPFH);
second, the k neighborhood of each point is redetermined, and the final histogram (called FPFH) calculated using the neighboring SPFH values is formulated as follows:
in the above, the weight w k In some given metric space, a query point P is represented q And its adjacent point P k The distance between them can thus be used to assess the distance between a pair of points (P q ,P k ). FIG. 6 is a schematic diagram of an area of influence in determining a fast point feature histogram description in an embodiment of a vehicle localization method.
Thus, for a known query point, the algorithm first uses only P q And its corresponding pair of neighbor points (illustrated by line segment 31 in fig. 6) to estimate its SPFH value, it is apparent that there is less interconnection between neighbor points than standard calculation of PFH. All points in the point cloud dataset need to perform this calculation to obtain the SPFH, which is then used to make its neighbors P k SPFH value and P of (C) q Re-weighting calculation of SPFH value of point to obtain P q Final FPFH value for the point.
(4) VFH descriptor:
the VFH descriptor is mainly applied to the problems of point cloud cluster recognition and six-degree-of-freedom pose estimation. The VFH descriptor is derived from the FPFH descriptor. Features are constructed for application to object recognition problems and pose estimation by two calculations: (1) Expanding the FPFH to enable the FPFH to utilize the whole point cloud object to carry out calculation estimation, wherein point pairs between the center point of the object and all other points on the surface of the object are used as calculation units when the FPFH is calculated; (2) And adding additional statistical information between the viewpoint direction and the estimated normal of each point, and directly integrating the viewpoint direction variable into the calculation of the relative normal angle in the FPFH calculation.
The viewpoint-related feature component is calculated by counting a histogram of angles between the viewpoint direction and each normal line. Not every normal viewing angle, because normal viewing angles have variability in scaling, referred to herein as translating the viewpoint to the viewpoint direction after the query point and the angle between each normal line. The second set of feature components are the three angles taught in PFH, now measured as the angle between the viewpoint direction at the center point and each surface normal. The uniqueness of the VFH is mainly reflected in the following two aspects: a viewpoint-direction-dependent component and a component describing the surface shape containing the extended FPFH.
After the feature descriptors of the key point cloud data are calculated in any one of the four modes, the position of the target object corresponding to the target point cloud data in the high-precision map is determined according to the feature descriptors of the key point cloud data and the feature descriptors in the high-precision map. One point cloud dataset is accurately registered with another by applying an estimated 4*4 rigid body transformation matrix representing translation and rotation. The specific process can be as follows: estimating the corresponding relation of the feature descriptors in the two data sets by combining the coordinate positions of the feature descriptors in the two data sets and based on the similarity of the features and the positions of the feature descriptors, and primarily estimating corresponding point pairs; assuming the data is noisy, removing erroneous corresponding point pairs that have an effect on registration; and estimating the rigid body transformation by using the rest correct corresponding relation to finish registration. The most important of the whole registration process is the extraction of key points and the feature description of the key points. So as to ensure the accuracy and efficiency of the corresponding estimation, and ensure the error-free estimation of the rigid body transformation matrix in the subsequent flow.
It should be noted that, the target object corresponding to the target point cloud data in this embodiment may be a target object corresponding to one or more target points among the target points included in the point cloud data. Illustratively, the target object may be a road edge or a railing on the road edge, or the like.
Alternatively, a coordinate system may be established with a certain point on the high-precision map as an origin, and the position of the target object in the high-precision map may be shown in the form of coordinates.
Step 203: and determining the position of the vehicle in the high-precision map according to the position of the target object in the high-precision map and the target point cloud data.
Specifically, after the position of the target object in the high-precision map is determined, the position of the vehicle in the high-precision map can be determined according to the position of the target object in the high-precision map and the target point cloud data, so that the positioning of the vehicle is realized.
In one implementation, coordinate values of a target object in a vehicle coordinate system can be determined according to target point cloud data; and determining the position of the vehicle in the high-precision map according to the position of the target object in the high-precision map and the coordinate value of the target object in the vehicle coordinate system. In this embodiment, the origin of the vehicle coordinate system is the center point of the rear axis of the vehicle, and when the positive X-axis direction of the vehicle coordinate system is the forward direction of the vehicle, the positive Y-axis direction is the forward direction of the vehicle, and the positive Z-axis direction is the upward direction of the vehicle.
As described in step 202, the target object corresponding to the target point cloud data may be a target object corresponding to one or several target points included in the target point cloud data, that is, some target points included in the target point cloud data do not belong to a point corresponding to the target object. In this embodiment, when determining the coordinate value of the target object in the vehicle coordinate system according to the target point cloud data, the average value of the target point cloud data corresponding to the target point belonging to the target object or the value of the center point may be used as the coordinate value of the target object in the vehicle coordinate system. Then, according to the coordinate value of the target object in the vehicle coordinate system, determining the distance between the target object and the vehicle and the included angle between the target object and the vehicle advancing direction; and determining the position of the vehicle in the high-precision map according to the position of the target object in the high-precision map, the distance between the target object and the vehicle and the included angle between the target object and the vehicle advancing direction.
FIG. 7A is a schematic diagram of one implementation of a vehicle positioning method embodiment for determining the position of a vehicle in a high-precision map. As shown in fig. 7A, the distance between the target object and the vehicle, and the angle between the target object and the vehicle advancing direction, i.e., the Y-axis, can be determined based on the coordinate values of the target object in the vehicle coordinate system. In the present embodiment, the vehicle may be represented by the origin of the vehicle coordinate system. Knowing the position of the target object in the high-precision map, the distance between the target object and the vehicle and the included angle between the target object and the vehicle advancing direction, the position of the vehicle in the high-precision map can be calculated according to the triangle geometry principle.
In an implementation manner, when there are a plurality of target objects corresponding to the target point cloud data, in the above implementation manner, according to a position of the target object in the high-precision map and a coordinate value of the target object in the vehicle coordinate system, a specific process of determining the position of the vehicle in the high-precision map may be: determining a plurality of initial positions of the vehicle in the high-precision map according to the position of each target object in the high-precision map and the coordinate values of the corresponding target object in the vehicle coordinate system; determining a sum of distances between each initial position and other initial positions; the minimum distance and the corresponding initial position are determined as the position of the vehicle in the high-precision map.
FIG. 7B is a schematic diagram of another implementation of a vehicle positioning method embodiment for determining the position of a vehicle in a high-precision map. As shown in fig. 7B, the initial position set (P1, P2, P3 … Pn) of the vehicle may be calculated correspondingly based on the target object 1, the target object 2, the target objects 3, … …, and the target object n. Then, a sum of distances between each initial position and other initial positions is determined, and the minimum distance and the corresponding initial position are determined as the position of the vehicle in the high-precision map. This way, the accuracy of vehicle positioning can be improved.
Further, in order to improve the efficiency of vehicle positioning, the average position of the vehicle in the high-precision map may be determined according to a plurality of initial positions; determining the sum of the distances between each initial position with the distance smaller than a preset threshold value and other initial positions; the minimum distance and the corresponding initial position are determined as the position of the vehicle in the high-precision map. In the implementation mode, when the distance sum is determined, only the distance sum between the initial position, the distance between which and the average position is smaller than the preset threshold value, and other initial positions is determined, so that compared with the calculation of the distance sum in an enumeration mode, the calculation amount is reduced, and the algorithm efficiency is improved.
The vehicle positioning method provided by the embodiment utilizes the millimeter wave radar to collect the cloud data of the target point for vehicle positioning, has the advantages of low cost and high positioning accuracy compared with the modes of laser radar, base station positioning, SLAM and the like, can also identify black objects, can also be used in severe weather such as cloud and fog, and has higher positioning reliability.
The vehicle positioning method provided in the embodiment includes: acquiring target point cloud data acquired through millimeter wave radar arranged on a vehicle in the running process of the vehicle, and determining the position of a target object corresponding to the target point cloud data in a high-precision map according to the target point cloud data and a pre-established high-precision map, wherein the high-precision map comprises a feature descriptor determined according to the point cloud data acquired by the millimeter wave radar, and determining the position of the vehicle in the high-precision map according to the position of the target object in the high-precision map and the target point cloud data. The method and the device have the advantages that the position of the vehicle in the high-precision map is determined according to the cloud data of the target points acquired by the millimeter wave radar, and compared with the prior art, the method and the device are low in cost, high in positioning precision and high in positioning reliability.
Fig. 8 is a schematic structural diagram of an embodiment of a vehicle positioning device provided by the invention. As shown in fig. 8, the vehicle positioning device provided in the present embodiment includes: an acquisition module 81, a first determination module 82 and a second determination module 83.
The acquiring module 81 is configured to acquire target point cloud data acquired by a millimeter wave radar disposed on a vehicle during a driving process of the vehicle.
The first determining module 82 is configured to determine, according to the target point cloud data and a pre-established high-precision map, a position of a target object corresponding to the target point cloud data in the high-precision map.
The high-precision map comprises a feature descriptor determined according to point cloud data acquired by the millimeter wave radar.
Optionally, the first determining module 82 is specifically configured to: determining key point cloud data from the target point cloud data; determining feature descriptors of the key point cloud data according to a point cloud data feature descriptor extraction algorithm; and determining the position of the target object corresponding to the target point cloud data in the high-precision map according to the feature descriptors of the key point cloud data and the feature descriptors in the high-precision map.
The second determining module 83 is configured to determine a position of the vehicle in the high-precision map according to the position of the target object in the high-precision map and the target point cloud data.
Optionally, the second determining module 83 specifically includes: the first determination sub-module and the second determination sub-module. And the first determining submodule is used for determining coordinate values of the target object in a vehicle coordinate system according to the target point cloud data. The vehicle coordinate system is a coordinate system established by taking a point on the vehicle as an origin. And the second determining submodule is used for determining the position of the vehicle in the high-precision map according to the position of the target object in the high-precision map and the coordinate value of the target object in the vehicle coordinate system.
In one implementation, the origin of the vehicle coordinate system is the center point of the rear axle of the vehicle, the positive X-axis direction of the vehicle coordinate system is the left-hand direction of the vehicle when the vehicle is facing forward, the positive Y-axis direction is the forward direction of the vehicle, and the positive Z-axis direction is the upward direction of the vehicle. The second determination submodule is specifically configured to: determining the distance between the target object and the vehicle and the included angle between the target object and the vehicle advancing direction according to the coordinate value of the target object in the vehicle coordinate system; and determining the position of the vehicle in the high-precision map according to the position of the target object in the high-precision map, the distance between the target object and the vehicle and the included angle between the target object and the vehicle advancing direction.
Optionally, when the target object corresponding to the target point cloud data is multiple, the second determining submodule specifically includes: a first determination unit, a second determination unit, and a third determination unit. And the first determining unit is used for determining a plurality of initial positions of the vehicle in the high-precision map according to the position of each target object in the high-precision map and the coordinate value of the corresponding target object in the vehicle coordinate system. And a second determining unit for determining a sum of distances between each initial position and other initial positions. And a third determining unit for determining the minimum distance and the corresponding initial position as the position of the vehicle in the high-precision map. Still further, based on the implementation, the vehicle positioning device may further include: and a fourth determining unit for determining an average position of the vehicle in the high-precision map based on the plurality of initial positions. Accordingly, the second determining unit is specifically configured to: and determining the sum of the distances between each initial position with the distance smaller than the preset threshold value and other initial positions.
The vehicle positioning device provided by the embodiment of the invention can execute the vehicle positioning method provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.
Fig. 9 is a schematic structural view of a vehicle according to the present invention. As shown in fig. 9, the vehicle includes a processor 70, a memory 71, and a millimeter wave radar 72. The number of processors 70 in the vehicle may be one or more, and the number of millimeter wave radars 72 may be one or more, one processor 70 and one millimeter wave radar 72 being exemplified in fig. 9; the processor 70 and memory 71 of the vehicle may be connected by a bus or other means, for example by a bus connection in fig. 9.
The memory 71 is a computer-readable storage medium that can be used to store software programs, computer-executable programs, and modules, such as program instructions and modules corresponding to the vehicle positioning method in the embodiment of the present invention (for example, the acquisition module 81, the first determination module 82, and the second determination module 83 in the vehicle positioning device). The processor 70 executes various functional applications of the vehicle and data processing, namely, implements the above-described vehicle positioning method by running software programs, instructions, and modules stored in the memory 71.
The memory 71 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, at least one application program required for functions; the storage data area may store data created according to the use of the vehicle, etc. In addition, memory 71 may include high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, memory 71 may further include memory remotely located with respect to processor 70, which may be connected to the vehicle via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The present invention also provides a storage medium containing computer executable instructions which, when executed by a computer processor, are for performing a vehicle locating method comprising:
acquiring target point cloud data acquired through millimeter wave radars arranged on a vehicle in the running process of the vehicle;
determining the position of a target object corresponding to the target point cloud data in the high-precision map according to the target point cloud data and a pre-established high-precision map; the high-precision map comprises a feature descriptor determined according to point cloud data acquired by the millimeter wave radar;
and determining the position of the vehicle in the high-precision map according to the position of the target object in the high-precision map and the target point cloud data.
Of course, the storage medium containing the computer executable instructions provided in the embodiments of the present invention is not limited to the method operations described above, and may also perform the related operations in the vehicle positioning method provided in any embodiment of the present invention.
From the above description of embodiments, it will be clear to a person skilled in the art that the present invention may be implemented by means of software and necessary general purpose hardware, but of course also by means of hardware, although in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a FLASH Memory (FLASH), a hard disk or an optical disk of a computer, etc., including several instructions for causing a computer device (which may be a personal computer, a vehicle, a network device, etc.) to execute the method according to the embodiments of the present invention.
It should be noted that, in the embodiment of the vehicle positioning device, each unit and module included are only divided according to the functional logic, but are not limited to the above-mentioned division, so long as the corresponding functions can be implemented; in addition, the specific names of the functional units are also only for distinguishing from each other, and are not used to limit the protection scope of the present invention.
Note that the above is only a preferred embodiment of the present invention and the technical principle applied. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, while the invention has been described in connection with the above embodiments, the invention is not limited to the embodiments, but may be embodied in many other equivalent forms without departing from the spirit or scope of the invention, which is set forth in the following claims.
Claims (6)
1. A vehicle positioning method, characterized by comprising:
acquiring target point cloud data acquired through millimeter wave radars arranged on a vehicle in the running process of the vehicle; wherein, the target point corresponding to the target point cloud data is a scattered point;
determining the position of a target object corresponding to the target point cloud data in the high-precision map according to the target point cloud data and a pre-established high-precision map; the high-precision map comprises a feature descriptor determined according to point cloud data acquired by a millimeter wave radar, wherein a target object corresponding to the point cloud data is a target object corresponding to one or more target points in target points contained in the point cloud data;
according to the target point cloud data, determining coordinate values of the target object in a vehicle coordinate system; the vehicle coordinate system is established by taking a point on the vehicle as an origin;
determining the position of the vehicle in the high-precision map according to the position of the target object in the high-precision map and the coordinate value of the target object in the vehicle coordinate system;
when the target object corresponding to the target point cloud data is a plurality of target objects, determining the position of the vehicle in the high-precision map according to the position of the target object in the high-precision map and the coordinate value of the target object in the vehicle coordinate system includes: determining a plurality of initial positions of the vehicle in the high-precision map according to the position of each target object in the high-precision map and the coordinate values of the corresponding target object in the vehicle coordinate system; determining an average position of the vehicle in the high-precision map according to the initial positions; determining the sum of the distances between each initial position with the distance from the average position being smaller than a preset threshold value and other initial positions; and determining the minimum distance and the corresponding initial position as the position of the vehicle in the high-precision map.
2. The method according to claim 1, wherein an origin of the vehicle coordinate system is a rear axis center point of the vehicle, an X-axis positive direction of the vehicle coordinate system is a left-side direction of the vehicle when the vehicle is facing forward, a Y-axis positive direction is the vehicle forward direction, and a Z-axis positive direction is an upward direction of the vehicle;
the determining the position of the vehicle in the high-precision map according to the position of the target object in the high-precision map and the coordinate value of the target object in the vehicle coordinate system comprises the following steps:
determining the distance between the target object and the vehicle and the included angle between the target object and the vehicle advancing direction according to the coordinate value of the target object in the vehicle coordinate system;
and determining the position of the vehicle in the high-precision map according to the position of the target object in the high-precision map, the distance between the target object and the vehicle and the included angle between the target object and the vehicle advancing direction.
3. The method according to claim 1 or 2, wherein the determining, according to the target point cloud data and a pre-established high-precision map, a position of a target object corresponding to the target point cloud data in the high-precision map includes:
determining key point cloud data from the target point cloud data;
determining feature descriptors of the key point cloud data according to a point cloud data feature descriptor extraction algorithm;
and determining the position of the target object corresponding to the target point cloud data in the high-precision map according to the feature descriptors of the key point cloud data and the feature descriptors in the high-precision map.
4. A vehicle positioning device, characterized by comprising:
the acquisition module is used for acquiring target point cloud data acquired through millimeter wave radars arranged on the vehicle in the running process of the vehicle; wherein, the target point corresponding to the target point cloud data is a scattered point;
the first determining module is used for determining the position of a target object corresponding to the target point cloud data in the high-precision map according to the target point cloud data and the high-precision map established in advance; the high-precision map comprises a feature descriptor determined according to point cloud data acquired by a millimeter wave radar, wherein a target object corresponding to the point cloud data is a target object corresponding to one or more target points in target points contained in the point cloud data;
the second determining module is used for determining the position of the vehicle in the high-precision map according to the position of the target object in the high-precision map and the target point cloud data;
the second determining module specifically includes: the first determining submodule is used for determining coordinate values of the target object in a vehicle coordinate system according to the target point cloud data; the vehicle coordinate system is established by taking a point on the vehicle as an origin; a second determining sub-module for determining a position of the vehicle in the high-precision map according to a position of the target object in the high-precision map and coordinate values of the target object in the vehicle coordinate system;
when the target objects corresponding to the target point cloud data are multiple, the second determining submodule specifically includes: a first determination unit, a second determination unit, a third determination unit, and a fourth determination unit; the first determining unit is used for determining a plurality of initial positions of the vehicle in the high-precision map according to the position of each target object in the high-precision map and the coordinate value of the corresponding target object in the vehicle coordinate system; the fourth determining unit is used for determining the average position of the vehicle in the high-precision map according to the initial positions; the second determining unit is used for determining the sum of the distances between each initial position with the distance from the average position smaller than a preset threshold value and other initial positions; the third determining unit is configured to determine a minimum distance and a corresponding initial position as a position of the vehicle in the high-precision map.
5. A vehicle, characterized in that the vehicle comprises:
one or more processors;
a memory for storing one or more programs;
one or more millimeter wave radars;
the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the vehicle localization method of any one of claims 1-3.
6. A computer-readable storage medium, on which a computer program is stored, characterized in that the program, when being executed by a processor, implements a vehicle positioning method as claimed in any one of claims 1-3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910978766.XA CN110794392B (en) | 2019-10-15 | 2019-10-15 | Vehicle positioning method and device, vehicle and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910978766.XA CN110794392B (en) | 2019-10-15 | 2019-10-15 | Vehicle positioning method and device, vehicle and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110794392A CN110794392A (en) | 2020-02-14 |
| CN110794392B true CN110794392B (en) | 2024-03-19 |
Family
ID=69439175
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910978766.XA Active CN110794392B (en) | 2019-10-15 | 2019-10-15 | Vehicle positioning method and device, vehicle and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110794392B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111505662B (en) * | 2020-04-29 | 2021-03-23 | 北京理工大学 | Unmanned vehicle positioning method and system |
| CN111638528B (en) * | 2020-05-26 | 2023-05-30 | 北京百度网讯科技有限公司 | Positioning method, positioning device, electronic equipment and storage medium |
| CN111812658B (en) * | 2020-07-09 | 2021-11-02 | 北京京东乾石科技有限公司 | Position determination method, device, system and computer readable storage medium |
| CN114252883B (en) * | 2020-09-24 | 2022-08-23 | 北京万集科技股份有限公司 | Target detection method, apparatus, computer device and medium |
| CN112287557B (en) * | 2020-11-09 | 2023-04-07 | 东风汽车集团有限公司 | Radar point cloud data loop playback method and system for assisting driving simulation test |
| CN113419235A (en) * | 2021-05-28 | 2021-09-21 | 同济大学 | Unmanned aerial vehicle positioning method based on millimeter wave radar |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102928816A (en) * | 2012-11-07 | 2013-02-13 | 东南大学 | High-reliably integrated positioning method for vehicles in tunnel environment |
| US9140792B2 (en) * | 2011-06-01 | 2015-09-22 | GM Global Technology Operations LLC | System and method for sensor based environmental model construction |
| WO2017150059A1 (en) * | 2016-03-02 | 2017-09-08 | 三菱電機株式会社 | Autonomous travel assistance device, roadside device, and autonomous travel assistance system |
| CN108732582A (en) * | 2017-04-20 | 2018-11-02 | 百度在线网络技术（北京）有限公司 | Vehicle positioning method and device |
| CN108983248A (en) * | 2018-06-26 | 2018-12-11 | 长安大学 | It is a kind of that vehicle localization method is joined based on the net of 3D laser radar and V2X |
| CN109031269A (en) * | 2018-06-08 | 2018-12-18 | 上海西井信息科技有限公司 | Localization method, system, equipment and storage medium based on millimetre-wave radar |
| CN109425365A (en) * | 2017-08-23 | 2019-03-05 | 腾讯科技（深圳）有限公司 | Method, apparatus, equipment and the storage medium of Laser Scanning Equipment calibration |
| CN109459750A (en) * | 2018-10-19 | 2019-03-12 | 吉林大学 | A kind of more wireless vehicle trackings in front that millimetre-wave radar is merged with deep learning vision |
| WO2019099802A1 (en) * | 2017-11-17 | 2019-05-23 | DeepMap Inc. | Iterative closest point process based on lidar with integrated motion estimation for high definitions maps |
| CN109870689A (en) * | 2019-01-08 | 2019-06-11 | 武汉中海庭数据技术有限公司 | Millimetre-wave radar and the matched lane grade localization method of high-precision map vector and system |
| CN110044371A (en) * | 2018-01-16 | 2019-07-23 | 华为技术有限公司 | A kind of method and vehicle locating device of vehicle location |
| CN110084272A (en) * | 2019-03-26 | 2019-08-02 | 哈尔滨工业大学（深圳） | A kind of cluster map creating method and based on cluster map and the matched method for relocating of location expression |
| CN110148144A (en) * | 2018-08-27 | 2019-08-20 | 腾讯大地通途（北京）科技有限公司 | Dividing method and device, storage medium, the electronic device of point cloud data |
| CN110208739A (en) * | 2019-05-29 | 2019-09-06 | 北京百度网讯科技有限公司 | Assist method, apparatus, equipment and the computer readable storage medium of vehicle location |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10352703B2 (en) * | 2016-04-28 | 2019-07-16 | Rogerson Aircraft Corporation | System and method for effectuating presentation of a terrain around a vehicle on a display in the vehicle |
| CN109840448A (en) * | 2017-11-24 | 2019-06-04 | 百度在线网络技术（北京）有限公司 | Information output method and device for automatic driving vehicle |
| US10935650B2 (en) * | 2017-12-22 | 2021-03-02 | Waymo Llc | Radar based three dimensional point cloud for autonomous vehicles |
-
2019
- 2019-10-15 CN CN201910978766.XA patent/CN110794392B/en active Active
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9140792B2 (en) * | 2011-06-01 | 2015-09-22 | GM Global Technology Operations LLC | System and method for sensor based environmental model construction |
| CN102928816A (en) * | 2012-11-07 | 2013-02-13 | 东南大学 | High-reliably integrated positioning method for vehicles in tunnel environment |
| WO2017150059A1 (en) * | 2016-03-02 | 2017-09-08 | 三菱電機株式会社 | Autonomous travel assistance device, roadside device, and autonomous travel assistance system |
| CN108732582A (en) * | 2017-04-20 | 2018-11-02 | 百度在线网络技术（北京）有限公司 | Vehicle positioning method and device |
| CN109425365A (en) * | 2017-08-23 | 2019-03-05 | 腾讯科技（深圳）有限公司 | Method, apparatus, equipment and the storage medium of Laser Scanning Equipment calibration |
| WO2019099802A1 (en) * | 2017-11-17 | 2019-05-23 | DeepMap Inc. | Iterative closest point process based on lidar with integrated motion estimation for high definitions maps |
| CN110044371A (en) * | 2018-01-16 | 2019-07-23 | 华为技术有限公司 | A kind of method and vehicle locating device of vehicle location |
| CN109031269A (en) * | 2018-06-08 | 2018-12-18 | 上海西井信息科技有限公司 | Localization method, system, equipment and storage medium based on millimetre-wave radar |
| CN108983248A (en) * | 2018-06-26 | 2018-12-11 | 长安大学 | It is a kind of that vehicle localization method is joined based on the net of 3D laser radar and V2X |
| CN110148144A (en) * | 2018-08-27 | 2019-08-20 | 腾讯大地通途（北京）科技有限公司 | Dividing method and device, storage medium, the electronic device of point cloud data |
| CN109459750A (en) * | 2018-10-19 | 2019-03-12 | 吉林大学 | A kind of more wireless vehicle trackings in front that millimetre-wave radar is merged with deep learning vision |
| CN109870689A (en) * | 2019-01-08 | 2019-06-11 | 武汉中海庭数据技术有限公司 | Millimetre-wave radar and the matched lane grade localization method of high-precision map vector and system |
| CN110084272A (en) * | 2019-03-26 | 2019-08-02 | 哈尔滨工业大学（深圳） | A kind of cluster map creating method and based on cluster map and the matched method for relocating of location expression |
| CN110208739A (en) * | 2019-05-29 | 2019-09-06 | 北京百度网讯科技有限公司 | Assist method, apparatus, equipment and the computer readable storage medium of vehicle location |
Non-Patent Citations (3)
| Title |
|---|
| 基于NDT与ICP结合的点云配准算法;王庆闪等;计算机工程与应用;全文 * |
| 张会平 著.《基于可视化技术的知识转化研究》.电子科技大学出版社,2011,第107-108页. * |
| 智能型地图匹配综合算法的研究;王仁礼, 陈天泽, 王冬红;计算机辅助设计与图形学学报(第11期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110794392A (en) | 2020-02-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110794392B (en) | Vehicle positioning method and device, vehicle and storage medium | |
| US11885872B2 (en) | System and method for camera radar fusion | |
| CN108509918B (en) | Target detection and tracking method fusing laser point cloud and image | |
| US10677907B2 (en) | Method to determine the orientation of a target vehicle | |
| US10366310B2 (en) | Enhanced camera object detection for automated vehicles | |
| US9563808B2 (en) | Target grouping techniques for object fusion | |
| WO2021104497A1 (en) | Positioning method and system based on laser radar, and storage medium and processor | |
| CN109521757B (en) | Static obstacle identification method and device | |
| US8260539B2 (en) | Object and vehicle detection and tracking using 3-D laser rangefinder | |
| WO2021012254A1 (en) | Target detection method, system, and mobile platform | |
| Zhou et al. | T-loam: truncated least squares lidar-only odometry and mapping in real time | |
| CN104035071A (en) | Methods And Apparatus Of Fusing Radar/camera Object Data And Lidar Scan Points | |
| Liu et al. | Dynamic vehicle detection with sparse point clouds based on PE-CPD | |
| Li et al. | An adaptive 3D grid-based clustering algorithm for automotive high resolution radar sensor | |
| CN111045022A (en) | Vehicle tracking method and system based on light detection and distance measurement | |
| CN112505671B (en) | Millimeter wave radar target positioning method and device under GNSS signal missing environment | |
| Molineros et al. | Monocular rear-view obstacle detection using residual flow | |
| Ma et al. | Vehicle tracking method in polar coordinate system based on radar and monocular camera | |
| CN114882458A (en) | Target tracking method, system, medium and device | |
| US11668799B2 (en) | Histogram based L-shape detection of target objects | |
| CN116863325A (en) | Method for multiple target detection and related product | |
| US20230204363A1 (en) | Method for improving localization accuracy of a self-driving vehicle | |
| CN114384486A (en) | Data processing method and device | |
| Zhang et al. | Obstacle Detection and Tracking of Unmanned Mine Car Based on 3D Lidar | |
| CN111712731B (en) | Target detection method, target detection system and movable platform |
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 |