CN115235477A - Vehicle positioning inspection method and device, storage medium and equipment - Google Patents
Vehicle positioning inspection method and device, storage medium and equipment Download PDFInfo
- Publication number
- CN115235477A CN115235477A CN202111444170.5A CN202111444170A CN115235477A CN 115235477 A CN115235477 A CN 115235477A CN 202111444170 A CN202111444170 A CN 202111444170A CN 115235477 A CN115235477 A CN 115235477A
- Authority
- CN
- China
- Prior art keywords
- target
- vehicle
- positioning
- local
- positioning information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000007689 inspection Methods 0.000 title claims abstract description 14
- 238000004422 calculation algorithm Methods 0.000 claims description 28
- 230000008859 change Effects 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 6
- 230000000875 corresponding effect Effects 0.000 description 15
- 230000004807 localization Effects 0.000 description 15
- 238000010586 diagram Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 8
- 238000012545 processing Methods 0.000 description 8
- 238000001514 detection method Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 238000013519 translation Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000002596 correlated effect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 241001417527 Pempheridae Species 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
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
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/86—Combinations of lidar systems with systems other than lidar, radar or sonar, e.g. with direction finders
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Computer Networks & Wireless Communication (AREA)
- Electromagnetism (AREA)
- Navigation (AREA)
Abstract
The present specification provides a vehicle positioning inspection method, apparatus, storage medium, and device, in which local positioning information is determined based on laser point cloud data and odometer data, global positioning is performed on a high-precision map using the local positioning information and the laser point cloud data to obtain global positioning information, and accuracy of vehicle positioning is determined according to a trajectory error between a plurality of local positioning information and a plurality of global positioning information corresponding to a plurality of target times. Therefore, the accuracy of vehicle positioning is checked under the condition of no GPS information by comparing the global positioning information with low frequency and the local positioning information with high frequency for many times.
Description
Technical Field
The present disclosure relates to the field of automatic driving technologies, and in particular, to a vehicle positioning inspection method, device, storage medium, and apparatus.
Background
In recent years, automatic driving has become a popular development direction for artificial intelligence. In the related technology, a positioning module on the automatic driving vehicle generally acquires pose variation quantity of a previous moment and a next moment by using a wheel speed meter and other sensors under the condition of no GPS information, and calculates an initial value of point cloud matching of the current moment by combining the vehicle pose of the previous moment, so as to estimate the vehicle pose of the current moment. However, due to the road environment, the point cloud matching at a certain time may be erroneous, which will result in the subsequent vehicle positioning also being inaccurate at all times.
Disclosure of Invention
According to a first aspect of embodiments herein, there is provided a vehicle positioning inspection method including: determining local positioning information of the target vehicle at each moment in a plurality of moments based on the laser point cloud data and the odometer data collected in the target time period; global positioning is carried out on a high-precision map by using local positioning information and laser point cloud data of each target moment in a plurality of target moments to obtain global positioning information of a target vehicle at each target moment in the plurality of target moments; the target moments are determined from the moments based on a preset frequency; and judging whether the positioning of the target vehicle in the target time period is accurate or not according to the error between the track formed by the local positioning information corresponding to the target times and the track formed by the global positioning information corresponding to the target times.
In some examples, the global positioning information of the target vehicle at a target time is obtained by performing global positioning on the local high-precision map through a global registration algorithm based on the laser point cloud data acquired at the target time; the local high-precision map is extracted from the high-precision map based on local positioning information of the target vehicle at the target time.
In some examples, the odometer data is data collected by a wheel speed meter indicating an amount of change in the pose of the target vehicle between two times.
In some examples, the error includes an absolute trajectory error and/or a relative pose error.
In some examples, the errors include absolute trajectory errors and relative pose errors; and if the absolute track error is smaller than a first threshold value and the relative pose error is smaller than a second threshold value, determining that the positioning of the target vehicle in the target time period is accurate.
In some examples, the method further comprises: and if the target vehicle is determined to be inaccurately positioned in the target time period, sending alarm information to a management terminal, so that the management terminal determines whether the target vehicle is accurately positioned in the target time period again according to the alarm information.
In some examples, the method further comprises: and if receiving feedback information which is sent by the management terminal and used for indicating accurate positioning, clearing the obtained global positioning information.
According to a second aspect of embodiments herein, there is provided a vehicle positioning inspection apparatus comprising: the local positioning module is used for determining local positioning information of the target vehicle at each moment in a plurality of moments based on the laser point cloud data and the odometer data which are collected in the target time period; the global positioning module is used for carrying out global positioning on a high-precision map by using local positioning information and laser point cloud data of each target moment in a plurality of target moments to obtain global positioning information of the target vehicle at each target moment in the plurality of target moments; the target moments are determined from the moments based on a preset frequency; and the positioning judgment module is used for judging whether the positioning of the target vehicle in the target time period is accurate or not according to the error between the track formed by utilizing the local positioning information corresponding to the target times and the track formed by utilizing the global positioning information corresponding to the target times.
According to a third aspect of embodiments of the present specification, there is provided a computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs any one of the methods of the embodiments of the specification.
According to a fourth aspect of embodiments herein, there is provided a computer apparatus comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements any of the methods in the embodiments herein when executing the program.
The technical scheme provided by the embodiment of the specification can have the following beneficial effects:
in the method, local positioning information is determined based on laser point cloud data and odometer data, global positioning is performed on a high-precision map by using the local positioning information and the laser point cloud data to obtain global positioning information, and then the accuracy of vehicle positioning is judged according to track errors between a plurality of pieces of local positioning information and a plurality of pieces of global positioning information corresponding to a plurality of target moments. Therefore, the accuracy of vehicle positioning is checked under the condition of no GPS information by comparing the global positioning information with low frequency and the local positioning information with high frequency for many times.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the specification.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present specification and together with the description, serve to explain the principles of the specification.
FIG. 1 is a flow chart diagram illustrating a vehicle location checking method according to one exemplary embodiment;
FIG. 2 is a schematic diagram of a position detection system for vehicle position checking according to an embodiment of the present disclosure;
fig. 3 is a schematic diagram illustrating a flow of performing a positioning check by a positioning detection system according to an embodiment of the present disclosure;
fig. 4 is a hardware configuration diagram of a computer device in which a vehicle positioning inspection apparatus according to an embodiment of the present disclosure is located;
FIG. 5 is a block diagram of a vehicle alignment checking device shown in the present specification according to an exemplary embodiment.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present specification. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the specification, as detailed in the appended claims.
The terminology used in the description herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the description. As used in this specification and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.
It should be understood that although the terms first, second, third, etc. may be used herein to describe various information, these information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, the first information may also be referred to as second information, and similarly, the second information may also be referred to as first information, without departing from the scope of the present specification. The word "if," as used herein, may be interpreted as "at … …" or "at … …" or "in response to a determination," depending on the context.
In recent years, automatic driving has become a popular development direction for artificial intelligence. In the related technology, a positioning module on the automatic driving vehicle generally acquires pose variation quantity of a previous moment and a next moment by using a wheel speed meter and other sensors under the condition of no GPS information, and calculates an initial value of point cloud matching of the current moment by combining the vehicle pose of the previous moment, so as to estimate the vehicle pose of the current moment. However, due to the road environment, the point cloud matching at a certain time may be erroneous, which will result in the subsequent vehicle positioning also being inaccurate at all times.
Based on this, the embodiments of the present specification provide a vehicle positioning detection scheme to solve the above problems.
The following provides a detailed description of examples of the present specification.
As shown in fig. 1, fig. 1 is a flow chart illustrating a vehicle localization check method according to an exemplary embodiment of the present description, the method comprising:
the method of the present embodiment may be applied to a target vehicle, which may be any type of vehicle, and may be any one of a car, a truck, a bus, an electric vehicle, and the like, and alternatively, the target vehicle may be an unmanned vehicle, i.e., a vehicle with automatic driving capability, such as an unmanned sweeper, an unmanned patrol car, and the like. Of course, in other embodiments, this method may also be applied to a computing device that is communicatively connected to the target vehicle, that is, the target vehicle may send its own collected data to the computing device, which checks the accuracy of the current location of the target vehicle. For the sake of brevity, the following description will be made with a device as an execution subject of the present embodiment.
The positioning result of the vehicle indicates the pose of the vehicle in a specific coordinate system, the specific coordinate system generally adopts a three-dimensional coordinate system, in the three-dimensional coordinate system, the pose of the vehicle can be represented by three-dimensional coordinates, a pitch angle, a course angle and a roll angle, and the pose of the vehicle can also be represented by only four dimensions of the three-dimensional coordinates and the course angle under most conditions because the vehicle generally runs on the flat ground. Of course, in the case where the specific coordinate system is a two-dimensional coordinate system, the pose of the vehicle may also be represented by three dimensions, i.e., a two-dimensional coordinate and a heading angle. In this embodiment, the positioning result of the target vehicle is the local positioning information, and is obtained by the device based on the laser point cloud data and the mileage data.
The laser point cloud data mentioned in this step may be acquired by a laser scanning device mounted on the target vehicle. The laser scanning equipment can be a laser radar, such as a single-line laser radar and a multi-line laser radar, and can also be a binocular stereo camera capable of generating point cloud data. The environmental information of the whole body of the target vehicle can be acquired through the laser scanning device, in the embodiment, the device can acquire laser point cloud data acquired by the laser scanning device in real time, the pose of the target vehicle is obtained through a point cloud registration mode, and positioning is completed.
The odometer data mentioned in this step may be acquired by the device through an odometer mounted on the target vehicle. The odometer can be a sensor for measuring the stroke, optionally, the odometer can be a wheel speed meter, and the working principle of the wheel speed meter is that the distance and radian of the movement of the wheel within a certain time are detected according to photoelectric encoders arranged on motors of a left driving wheel and a right driving wheel, so that the change of the relative pose of the vehicle is calculated. Thus, the odometer data may include data indicating an amount of change in the pose of the target vehicle between two times. Therefore, when the device knows the pose of the target vehicle at the previous moment, the pose estimation value of the target vehicle at the current moment can be obtained by combining the mileage data and the pose of the target vehicle at the previous moment, and then higher-precision adjustment can be performed on the basis of the pose estimation value by utilizing the laser point cloud data, and the positioning information obtained after adjustment is the pose of the target vehicle at the current moment, namely the local positioning information. Correspondingly, this positioning process may be referred to as local positioning.
The above local positioning method can be realized by using a point cloud registration algorithm. The Point cloud registration algorithm used may include any one of ICP (Iterative Closest Point) algorithm, GICP (Generalized Iterative Closest Point) algorithm, NDT (Normal Distribution Transform) algorithm, and the like. Taking the ICP algorithm as an example, the process of locally positioning the device may include: determining laser point cloud data of a target vehicle at the current moment as target point cloud P, and acquiring source point cloud Q from an existing high-precision map based on a pose estimation value of the vehicle at the current moment; selecting a point set P from a target point cloud P i Selecting a point set Q from a source point cloud Q i Minimizing the Euclidean distance between the two point sets; the optimal rotation R and the optimal translation t are calculated so that the value of the error function shown in the following formula (1)Minimum; obtaining a new point set p based on the optimal rotation R and the optimal translation t i’ (ii) a Computing a new set of points p i’ Arrival set q i The average distance d of; if d is smaller than a preset threshold value or exceeds the iteration times, stopping, otherwise, reselecting the point set q i And executing iterative processing until a convergence condition is met.
The high-precision map is an electronic map for automatic driving, and is generally made of point cloud data acquired by a map acquisition vehicle through sensors such as a laser radar and a camera mounted on the map acquisition vehicle. Compared with the traditional map, the high-precision map has higher precision, generally reaching the centimeter level, and has more data dimensions, and besides the road level data (such as road shape, gradient, curvature, direction and the like) recorded by the traditional map, the high-precision map also comprises lane level data, such as lane line type, lane width and the like. The positioning result of the vehicle can be presented in the form of coordinates in a high-precision map. During local positioning, the device can adjust the pose estimation value of the target vehicle at the current moment through the optimal rotation and the optimal translation when the convergence condition is met, and then according to the adjusted value, the corresponding coordinate is searched in the high-precision map, namely the local positioning information of the target vehicle at the current moment. It should be noted that, in some examples, before local positioning is performed by using a high-precision map, a local high-precision map may be extracted from the high-precision map by using an estimated value of vehicle positioning, and then when point cloud matching is performed by using an algorithm, the processing amount in searching on the map may be reduced, thereby improving the processing efficiency. Wherein the estimated value may be a position where the target vehicle is estimated by the apparatus based on the historical trip data of the target vehicle.
In an alternative embodiment, the device may perform local location of the first frequency during the target time period, resulting in local location information for the target vehicle at each of a plurality of time instants. The first frequency is the number of times that the device performs local positioning in a unit time, for example, if the unit time is 1 minute and the first frequency is 20 times/minute, the device performs local positioning 60 times in a target time period of 3 minutes before the current time to obtain 60 pieces of local positioning information, and each piece of local positioning information corresponds to one time. The first frequency may be determined according to the scanning frequency of the laser scanning device, or may be set according to the requirements of a specific scene. Of course, it should be noted that, in the target time period, a time interval between times corresponding to two adjacent local positioning may be a fixed value, or may not be fixed, and this specification does not limit this.
102, carrying out global positioning on a high-precision map by using local positioning information and laser point cloud data of each target moment in a plurality of target moments to obtain global positioning information of a target vehicle at each target moment in the plurality of target moments; the target moments are determined from the moments based on a preset frequency;
in the related art, in a case where GPS (Global Positioning System) information is blocked or disabled, the vehicle usually acquires its own pose by using the local Positioning information mentioned in step 101. However, due to the influence of the road environment, the laser point cloud data acquired by the vehicle at a certain time may contain fewer features, which may cause the positioning of the device at the certain time to be incorrect, and further may cause the positioning of the device to be incorrect at subsequent times. Based on this, the present embodiment checks the accuracy of the positioning of the target vehicle by the low-frequency global positioning result.
The method comprises the following steps of utilizing local positioning information and laser point cloud data of a target vehicle at a target moment to carry out global positioning aiming at a high-precision map. In some examples, this step may refer to: extracting a local high-precision map from the high-precision map based on the local positioning information of the target vehicle at the target moment; and carrying out global positioning on the local high-precision map through a global registration algorithm based on the laser point cloud data acquired at the target moment. The local high-precision map can be extracted from the high-precision map according to the local positioning information of the target vehicle at the target moment, if global positioning is carried out on the whole high-precision map, the searching range in the point cloud matching process is too large, huge data processing capacity is a large burden for equipment, the local positioning information is a positioning result to be checked, and the local high-precision map extracted according to the local positioning information has high probability of containing the surrounding environment where the target vehicle is actually located, so that the method of the embodiment can effectively reduce the data quantity which needs to be processed during global positioning.
The global registration algorithm does not need to rely on an initial pose, so that the global positioning result at a certain moment is not influenced by the global positioning result at the previous moment. Optionally, the global registration algorithm may be any one of a BnB (Branch and Bound) algorithm, a GO-ICP (global Optimal-Iterative Closest Point based on a global Optimal solution) algorithm, and the like. Taking the GO-ICP algorithm as an example, the device can register laser point cloud data of the target vehicle at the target time with point cloud data in the local high-precision map under L2 error measurement defined by ICP through the GO-ICP algorithm, so as to obtain a final pose as global positioning information of the target vehicle at the target time. The specific calculation process of the global registration algorithm may refer to descriptions in the related art, which are not described in detail herein.
The plurality of target moments mentioned in this step are screened from the plurality of moments mentioned in step 101 based on a preset frequency. The preset frequency value may be recorded as 1/n, that is, a target time is determined every n times, for example, the multiple times in the target time period include nine times T1, T2, T3, T4, T5, T6, T7, T8, and T9, and if the preset frequency is 1/3, the target time may include T1, T4, and T7, may also include T2, T5, and T8, or may also include T3, T6, and T9. The larger the value of the preset frequency is, the more the number of times of global positioning is executed, the more processing resources the device needs to consume, and in practical application, the value of the preset frequency can be set according to the requirements of a specific scene.
In some examples, the device may perform local localization and global localization based on two different threads, respectively, for the thread performing local localization, the number of times is generally higher than that of the thread performing global localization, but the time consumption for performing local localization once is generally lower than that of performing global localization once. Therefore, the cooperation processing of local positioning and global positioning can be better realized.
And 103, judging whether the positioning of the target vehicle in the target time period is accurate or not according to the error between the track formed by the local positioning information corresponding to the target times and the track formed by the global positioning information corresponding to the target times.
In the embodiment, whether the vehicle positioning is accurate is judged by using the error between the track formed by the plurality of local positioning information and the track formed by the plurality of global positioning information, and it can also be considered that the high-frequency local positioning result is verified by using the low-frequency global positioning result. Under the condition that the error is small, the result of the local positioning is basically consistent with the result of the global positioning, and the positioning of the vehicle in the target time period can be judged to be accurate; in the case where the error is large, it is described that the result of the local localization is not consistent with the result of the global localization, and in this case, it is likely that the local localization at a certain time is erroneous, and therefore it can be determined that the localization of the vehicle within the target time period is erroneous. For simplicity, a track formed by using a plurality of local positioning information may be denoted as a local track, and a track formed by using a plurality of global positioning information may be denoted as a global track. The number of target times in the local trajectory and the global trajectory is positively correlated with the size of the calculated amount, and is also positively correlated with the accuracy of the determination result, in order to balance the calculated amount and the accuracy, in some examples, the number may be 5, of course, in other embodiments, the number may also be set according to the requirements of a specific scene, and this embodiment does not limit this.
In an alternative embodiment, the error may comprise an absolute trajectory error and/or a relative pose error. Absolute Track Error (ATE) And Relative Pose Error (RPE) are typically indicators for evaluating SLAM (Simultaneous Localization And Mapping) system performance, where ATE is a direct difference between an estimated Pose And a real Pose, which can intuitively reflect algorithm accuracy And track global consistency, and RPE is a difference between Pose variations on the same two timestamps, which can be considered as a real-time comparison of the real Pose And the estimated Pose. In the embodiment, the error between the local trajectory and the global trajectory can be calculated by using at least one of the two indexes, so as to determine the accuracy of vehicle positioning.
Respectively recording the n-frame positions in the local track as P 1 、P 2 、P 3 、……、P n Respectively recording the n-frame attitude in the global track as Q 1 、Q 2 、Q 3 、……、Q n Where the subscript indicates the frame order, in this embodiment, the frames (i.e., target time instants) are time aligned and the total number of frames is the same. Thus, the ATE between the local trace and the global trace can be calculated based on the following equation:
in the above formula, F i ATE, RMSE (F) of the ith frame 1:n And delta) is ATE for both local and global traces. S is a transformation matrix from the pose in the local track to the pose in the global track, and can be obtained through least square calculation; Δ is the time interval between two frames, optionally, Δ =1; m is the difference between n and Δ; trans (F) i ) Is F i The translation component. As can be seen from the above formula, when calculating the ATE of the local track and the global track, the ATE of the ith frame can be calculated first, and then the calculation is performed through the mean square error RMSEMaking statistics to obtain an overall value RMSE (F) 1:n And delta) ATE as two traces. It should be noted that, in other embodiments, the ATE for the local trace and the global trace may be calculated in other manners, for example, after the ATE for the ith frame is calculated, the ATE for the local trace and the global trace may also be calculated by a mean value or a median value.
The smaller the ATE of the local track and the global track, the closer the shape of the local track is to the shape of the global track, and thus, in some instances, where the ATE is less than a first threshold, the target vehicle's position within the target time period may be determined to be accurate.
And the RPE between the local track and the global track may be calculated based on the following formula:
above formula, E i RPE, RMSE (E) of the ith frame 1:n Δ) are the RPE of the local trace and the global trace, and the remaining parameters are defined consistent with the definitions in the ATE's equations. Correspondingly, in other embodiments, the RPEs of the local track and the global track may also be calculated in other manners, for example, after the RPE of the ith frame is calculated, the RPEs of the local track and the global track may also be calculated by a mean value or a median, which is not limited in this specification.
The smaller the RPE of the local track and the global track is, the smaller the relative transformation amount between the local track and the global track is, that is, the smaller the error existing between the coordinates is, and therefore, in some examples, in the case where the RPE is smaller than the second threshold value, it may be determined that the positioning of the target vehicle within the target time period is accurate.
In other examples, to further improve the accuracy of the determination result, the error between the two trajectories may include both ATE and RPE, and if ATE is smaller than the first threshold and RPE is smaller than the second threshold, it is determined that the target vehicle is accurately located in the target time period. That is, the ATE and the RPE between the two tracks can be calculated, and the positioning of the vehicle is determined to be accurate only when the conditions that both the ATE and the RPE are small are satisfied, and the positioning of the vehicle is determined to be wrong if either one of the ATE and the RPE is large or both of the ATE and the RPE are large. The first threshold and the second threshold may be set according to the requirements of a specific scenario.
In some examples, if ATE is less than the first threshold and RPE is greater than or equal to the second threshold, indicating that the shapes of the two tracks are close but the coordinates have a large error, which may be caused by a positioning check error, such as a global positioning error at a certain target time, an alarm message may be sent to the management end, so that the management end may re-determine whether the target vehicle is accurately positioned in the target time period according to the alarm message. The management terminal can be a service program arranged in a remote operation and maintenance center of the target vehicle, the management terminal can have a more precise high-precision map and more efficient computing capability, and the alarm information can carry laser point cloud data acquired by the target vehicle at the current moment so that the management terminal can check whether the laser point cloud data is matched with the high-precision map, namely whether the positioning is accurate. When the management terminal checks that the positioning is accurate, the fact that the global positioning information obtained by the equipment is possible to be wrong is shown, and based on the fact that the equipment can clear the obtained global positioning information and continue to operate when the equipment receives feedback information which is sent by the management terminal and used for indicating the accurate positioning; and when the management terminal checks that the positioning error is obtained, the checking result of the equipment is correct, and based on the checking result, the equipment can restart the positioning module of the target vehicle and restart the task of vehicle positioning when receiving the feedback information which is sent by the management terminal and used for indicating the positioning error.
According to the method, the local positioning information is determined based on the laser point cloud data and the odometer data, global positioning is carried out on a high-precision map by utilizing the local positioning information and the laser point cloud data to obtain global positioning information, and then the accuracy of vehicle positioning is judged according to track errors between a plurality of local positioning information and a plurality of global positioning information corresponding to a plurality of target moments. That is to say, the accuracy of vehicle positioning without GPS information is checked by comparing the low-frequency global positioning information with the high-frequency local positioning information for many times.
To describe aspects of embodiments of the present disclosure in more detail, a specific embodiment is described below.
As shown in fig. 2, fig. 2 is a schematic diagram of a positioning detection system for vehicle positioning inspection, which includes a computing device 21, a laser scanning device 22, and a wheel speed meter 23, and is installed on a smart driving vehicle 24, where the laser scanning device 22 is used to collect point cloud data of the environment around the smart driving vehicle 24 in real time, and the wheel speed meter 23 is used to collect odometer data indicating the amount of change in the pose of the smart driving vehicle 24 at two moments in front and at two moments in back in real time; the computing device 21 is configured to acquire the point cloud data acquired by the laser scanning device 22 and the odometer data acquired by the wheel speed meter 23, and perform a positioning inspection based on these data to output a positioning inspection result.
The flow of the computing device 21 performing the positioning check is shown in fig. 3, and includes:
s301, acquiring a local point cloud map around the vehicle according to an existing high-precision map and a current positioning estimation value of the intelligent driving vehicle;
s302, local positioning at the current moment is carried out according to the point cloud data, the mileage data and the pose data of the intelligent driving vehicle at the previous moment, so that the pose data of the intelligent driving vehicle at the current moment, namely the local positioning information of the intelligent driving vehicle at the current moment, is obtained;
specifically, the local positioning information of the intelligent driving vehicle at the current moment is obtained by calculating the pose estimation value of the intelligent driving vehicle at the current moment through the pose data of the intelligent driving vehicle at the previous moment and combining mileage data, and then adjusting the point cloud data of the intelligent driving vehicle at the current moment in a local point cloud map by adopting a point cloud registration algorithm; the adopted point cloud registration algorithm comprises but is not limited to an ICP algorithm, a GICP algorithm and an NDT algorithm; in this embodiment, the smart-driven vehicle is in a state where the GPS information is disabled, in which case the smart-driven vehicle actually utilizes the local positioning information as a result of the current positioning of the vehicle;
in addition, the computing device performs local positioning at a first frequency, which is 600 times/hour, that is, the computing device performs local positioning every 6s on average;
s303, carrying out global positioning aiming at the high-precision map according to the local positioning information, the point cloud data and the existing high-precision map of the intelligent driving vehicle at the current moment to obtain the global positioning information of the intelligent driving vehicle at the current moment;
specifically, the global positioning information of the intelligent driving vehicle at the current moment is obtained by extracting a local high-precision map from an existing high-precision map by using the local positioning information of the intelligent driving vehicle at the current moment and then performing point cloud registration on the local high-precision map by using a global positioning algorithm through point cloud data of the intelligent driving vehicle at the current moment; the adopted global positioning algorithm comprises but is not limited to BnB search and GO-ICP algorithm;
in addition, the computing device is opened another thread to perform the more time-consuming global localization with low frequency, the computing device performs the global localization with the second frequency, which is 30 times/hour, that is, the computing device performs the global localization once every 120s on average;
s304, calculating absolute track errors ATE and relative pose errors RPE among local positioning information, global positioning information and global positioning information of a plurality of target moments;
specifically, the results of local positioning at the five target times T1, T2, T3, T4, T5, and the results of global positioning at the five target times may be selected for calculating ATE and RPE; because the target time is the same, the time of each frame is aligned, and the total frame number is the same;
s305, judging whether the ATE is smaller than a first threshold value and whether the RPE is smaller than a second threshold value, if so, executing S302, otherwise, executing S306;
when the ATE is smaller than the first threshold and the RPE is smaller than the second threshold, it indicates that the shape of the trajectory obtained by the global positioning algorithm is close to that of the trajectory obtained by the local positioning algorithm, and the error between the coordinates is small, in which case the result of the local positioning is considered to be correct, so the process returns to S302 again to continue the task;
s306, sending a positioning alarm signal to remind a management end to check whether the point cloud data at the current moment is matched with a high-precision map, namely whether positioning is accurate;
the management end is arranged in a remote operation and maintenance center connected with the intelligent driving vehicle, and a user of the management end comprises a safety supervisor which can carry out more precise manual inspection on a positioning result when receiving a positioning alarm signal;
s307, if feedback information indicating correct positioning by the management terminal is received, executing S308, and if feedback information indicating wrong positioning by the management terminal is received, executing S309;
s308, eliminating alarm and clearing the calculated global positioning information;
and S309, if feedback information indicating positioning errors by the management terminal is received, controlling the whole positioning detection system to restart so as to restart the task.
The scheme of the embodiment provides an automatic driving positioning checking method based on a high-precision map, whether positioning is correct is checked by comparing a low-frequency laser global positioning result and a high-frequency laser local positioning result for many times, and the technical problem that vehicle positioning is easy to make mistakes without GPS information is solved.
Corresponding to the embodiment of the method, the specification also provides an embodiment of the vehicle positioning checking device and the terminal applied to the vehicle positioning checking device.
The embodiment of the vehicle positioning checking device in the specification can be applied to computer equipment, such as a server or terminal equipment. The device embodiments may be implemented by software, or by hardware, or by a combination of hardware and software. The software implementation is taken as an example, and as a logical device, the device is formed by reading corresponding computer program instructions in the nonvolatile memory into the memory for operation through the processor in which the file processing is located. From a hardware aspect, as shown in fig. 4, which is a hardware structure diagram of a computer device in which a vehicle positioning and checking apparatus is located in an embodiment of this specification, except for the processor 410, the memory 430, the network interface 420, and the nonvolatile memory 440 shown in fig. 4, a server or an electronic device in which an apparatus 431 is located in an embodiment may also include other hardware according to an actual function of the computer device, which is not described again.
Accordingly, the embodiments of the present specification also provide a computer storage medium, in which a program is stored, and the program, when executed by a processor, implements the method in any of the above embodiments.
Embodiments of the present description may take the form of a computer program product embodied on one or more storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having program code embodied therein. Computer-usable storage media include permanent and non-permanent, removable and non-removable media, and may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of the storage medium of the computer include, but are not limited to: phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technologies, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape storage or other magnetic storage devices, or any other non-transmission medium, may be used to store information that may be accessed by a computing device.
As shown in fig. 5, fig. 5 is a block diagram of a vehicle positioning inspection apparatus shown in the present specification according to an exemplary embodiment, the apparatus including:
the local positioning module is used for determining local positioning information of the target vehicle at each moment in a plurality of moments based on the laser point cloud data and the odometer data which are collected in the target time period;
the global positioning module is used for carrying out global positioning on a high-precision map by using local positioning information and laser point cloud data of each target moment in a plurality of target moments to obtain global positioning information of the target vehicle at each target moment in the plurality of target moments; the target moments are determined from the moments based on a preset frequency;
and the positioning judgment module is used for judging whether the positioning of the target vehicle in the target time period is accurate or not according to the error between the track formed by utilizing the local positioning information corresponding to the target times and the track formed by utilizing the global positioning information corresponding to the target times.
The implementation process of the functions and actions of each module in the above device is specifically described in the implementation process of the corresponding step in the above method, and is not described herein again.
For the device embodiments, since they substantially correspond to the method embodiments, reference may be made to the partial description of the method embodiments for relevant points. The above-described embodiments of the apparatus are merely illustrative, wherein the modules described as separate parts may or may not be physically separate, and the parts displayed as modules may or may not be physical modules, may be located in one place, or may be distributed on a plurality of network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution in the present specification. One of ordinary skill in the art can understand and implement it without inventive effort.
The foregoing description has been directed to specific embodiments of this disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
Other embodiments of the present description will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This specification is intended to cover any variations, uses, or adaptations of the specification following, in general, the principles of the specification and including such departures from the present disclosure as come within known or customary practice within the art to which the specification pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the specification being indicated by the following claims.
It will be understood that the present description is not limited to the precise arrangements that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present description is limited only by the appended claims.
The above description is only a preferred embodiment of the present disclosure, and should not be taken as limiting the present disclosure, and any modifications, equivalents, improvements and the like made within the spirit and principle of the present disclosure should be included in the protection scope of the present disclosure.
Claims (10)
1. A vehicle positioning inspection method, characterized by comprising:
determining local positioning information of the target vehicle at each moment in a plurality of moments based on the laser point cloud data and the odometer data collected in the target time period;
carrying out global positioning on a high-precision map by using local positioning information and laser point cloud data of each target time in a plurality of target times to obtain global positioning information of a target vehicle at each target time in the plurality of target times; the target moments are determined from the moments based on a preset frequency;
and judging whether the positioning of the target vehicle in the target time period is accurate or not according to the error between the track formed by the local positioning information corresponding to the target times and the track formed by the global positioning information corresponding to the target times.
2. The method according to claim 1, wherein the global positioning information of the target vehicle at a target moment is obtained by global positioning on the local high-precision map through a global registration algorithm based on the laser point cloud data collected at the target moment; the local high-precision map is extracted from the high-precision map based on local positioning information of the target vehicle at the target time.
3. The method of claim 1, wherein the odometer data is data acquired by a wheel speed meter indicating an amount of pose change of the target vehicle between two moments in time.
4. The method of claim 1, wherein the errors comprise absolute trajectory errors and/or relative pose errors.
5. The method of claim 1, wherein the errors include absolute trajectory errors and relative pose errors; and if the absolute track error is smaller than a first threshold value and the relative pose error is smaller than a second threshold value, determining that the target vehicle is accurately positioned in the target time period.
6. The method of claim 5, further comprising:
and if the target vehicle is determined to be inaccurately positioned in the target time period, sending alarm information to a management terminal, so that the management terminal determines whether the target vehicle is accurately positioned in the target time period again according to the alarm information.
7. The method of claim 6, further comprising:
and if receiving feedback information which is sent by the management terminal and used for indicating accurate positioning, clearing the obtained global positioning information.
8. A vehicle positioning inspection device, characterized by comprising:
the local positioning module is used for determining local positioning information of the target vehicle at each moment in a plurality of moments based on the laser point cloud data and the odometer data which are collected in the target time period;
the global positioning module is used for carrying out global positioning on a high-precision map by using local positioning information and laser point cloud data of each target moment in a plurality of target moments to obtain global positioning information of the target vehicle at each target moment in the plurality of target moments; the target moments are determined from the moments based on a preset frequency;
and the positioning judgment module is used for judging whether the positioning of the target vehicle in the target time period is accurate or not according to the error between the track formed by utilizing the local positioning information corresponding to the target times and the track formed by utilizing the global positioning information corresponding to the target times.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program implements the method of any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that a computer program is stored thereon which, when being executed by a processor, carries out the method of any one of claims 1 to 7.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111444170.5A CN115235477A (en) | 2021-11-30 | 2021-11-30 | Vehicle positioning inspection method and device, storage medium and equipment |
| PCT/CN2022/071303 WO2023097873A1 (en) | 2021-11-30 | 2022-01-11 | Method and apparatus for checking vehicle positioning, and storage medium and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111444170.5A CN115235477A (en) | 2021-11-30 | 2021-11-30 | Vehicle positioning inspection method and device, storage medium and equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115235477A true CN115235477A (en) | 2022-10-25 |
Family
ID=83665907
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111444170.5A Pending CN115235477A (en) | 2021-11-30 | 2021-11-30 | Vehicle positioning inspection method and device, storage medium and equipment |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN115235477A (en) |
| WO (1) | WO2023097873A1 (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107144292A (en) * | 2017-06-08 | 2017-09-08 | 杭州南江机器人股份有限公司 | The odometer method and mileage counter device of a kind of sports equipment |
| CN109887028A (en) * | 2019-01-09 | 2019-06-14 | 天津大学 | A kind of unmanned vehicle assisted location method based on cloud data registration |
| WO2019215987A1 (en) * | 2018-05-09 | 2019-11-14 | ソニー株式会社 | Information processing device, information processing method, and program |
| CN111536964A (en) * | 2020-07-09 | 2020-08-14 | 浙江大华技术股份有限公司 | Robot positioning method and device, and storage medium |
| CN111947671A (en) * | 2020-03-02 | 2020-11-17 | 北京百度网讯科技有限公司 | Method, apparatus, computing device and computer-readable storage medium for positioning |
| CN112462372A (en) * | 2021-01-29 | 2021-03-09 | 北京主线科技有限公司 | Vehicle positioning method and device |
| CN113376650A (en) * | 2021-08-09 | 2021-09-10 | 浙江华睿科技股份有限公司 | Mobile robot positioning method and device, electronic equipment and storage medium |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107782320A (en) * | 2016-08-26 | 2018-03-09 | 法乐第(北京)网络科技有限公司 | A kind of vehicle positioning method |
| CN109870157B (en) * | 2019-02-20 | 2021-11-02 | 苏州风图智能科技有限公司 | Method and device for determining pose of vehicle body and mapping method |
| CN111161353B (en) * | 2019-12-31 | 2023-10-31 | 深圳一清创新科技有限公司 | Vehicle positioning method, device, readable storage medium and computer equipment |
| CN111272165B (en) * | 2020-02-27 | 2020-10-30 | 清华大学 | Intelligent vehicle positioning method based on characteristic point calibration |
-
2021
- 2021-11-30 CN CN202111444170.5A patent/CN115235477A/en active Pending
-
2022
- 2022-01-11 WO PCT/CN2022/071303 patent/WO2023097873A1/en unknown
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107144292A (en) * | 2017-06-08 | 2017-09-08 | 杭州南江机器人股份有限公司 | The odometer method and mileage counter device of a kind of sports equipment |
| WO2019215987A1 (en) * | 2018-05-09 | 2019-11-14 | ソニー株式会社 | Information processing device, information processing method, and program |
| CN109887028A (en) * | 2019-01-09 | 2019-06-14 | 天津大学 | A kind of unmanned vehicle assisted location method based on cloud data registration |
| CN111947671A (en) * | 2020-03-02 | 2020-11-17 | 北京百度网讯科技有限公司 | Method, apparatus, computing device and computer-readable storage medium for positioning |
| CN111536964A (en) * | 2020-07-09 | 2020-08-14 | 浙江大华技术股份有限公司 | Robot positioning method and device, and storage medium |
| CN112462372A (en) * | 2021-01-29 | 2021-03-09 | 北京主线科技有限公司 | Vehicle positioning method and device |
| CN113376650A (en) * | 2021-08-09 | 2021-09-10 | 浙江华睿科技股份有限公司 | Mobile robot positioning method and device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2023097873A1 (en) | 2023-06-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114526745B (en) | Drawing construction method and system for tightly coupled laser radar and inertial odometer | |
| CN109444932B (en) | Vehicle positioning method and device, electronic equipment and storage medium | |
| CN104677361B (en) | A kind of method of comprehensive location | |
| CN114034307A (en) | Lane line-based vehicle pose calibration method and device and electronic equipment | |
| CN112455502B (en) | Train positioning method and device based on laser radar | |
| US11983890B2 (en) | Method and apparatus with motion information estimation | |
| CN114758504A (en) | Online vehicle overspeed early warning method and system based on filtering correction | |
| CN115494533A (en) | Vehicle positioning method, device, storage medium and positioning system | |
| CN114754778B (en) | Vehicle positioning method and device, electronic equipment and storage medium | |
| CN111612818A (en) | Novel binocular vision multi-target tracking method and system | |
| CN114280583B (en) | Laser radar positioning accuracy verification method and system without GPS signal | |
| CN115546705A (en) | Target identification method, terminal device and storage medium | |
| CN114137562B (en) | Multi-target tracking method based on improved global nearest neighbor | |
| CN110345924A (en) | A kind of method and apparatus that distance obtains | |
| CN117367419A (en) | Robot positioning method, apparatus and computer readable storage medium | |
| CN117197182B (en) | Lei Shibiao method, apparatus and storage medium | |
| CN117387966A (en) | Evaluation method and system of intelligent driving system | |
| CN112965076A (en) | Multi-radar positioning system and method for robot | |
| CN115235477A (en) | Vehicle positioning inspection method and device, storage medium and equipment | |
| CN114358038B (en) | Two-dimensional code coordinate calibration method and device based on vehicle high-precision positioning | |
| CN112987010B (en) | System and method for multi-radar mapping of robot | |
| CN113034687B (en) | Real-time grid map generation method for large scene | |
| CN116228820B (en) | Obstacle detection method and device, electronic equipment and storage medium | |
| JP7540559B1 (en) | Self-position estimation device and self-position estimation method | |
| CN118405152A (en) | Target positioning method and device, electronic equipment and storage medium |
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: 20221025 |