CN110632607B - Object boundary determining method and system based on ultrasonic data and vehicle - Google Patents

Object boundary determining method and system based on ultrasonic data and vehicle Download PDF

Info

Publication number
CN110632607B
CN110632607B CN201910869730.8A CN201910869730A CN110632607B CN 110632607 B CN110632607 B CN 110632607B CN 201910869730 A CN201910869730 A CN 201910869730A CN 110632607 B CN110632607 B CN 110632607B
Authority
CN
China
Prior art keywords
data point
existing
current data
point
data points
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
Application number
CN201910869730.8A
Other languages
Chinese (zh)
Other versions
CN110632607A (en
Inventor
欧阳湛
邓志权
蒋少峰
张博
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangzhou Xiaopeng Motors Technology Co Ltd
Original Assignee
Guangzhou Xiaopeng Autopilot Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Guangzhou Xiaopeng Autopilot Technology Co Ltd filed Critical Guangzhou Xiaopeng Autopilot Technology Co Ltd
Priority to CN201910869730.8A priority Critical patent/CN110632607B/en
Publication of CN110632607A publication Critical patent/CN110632607A/en
Application granted granted Critical
Publication of CN110632607B publication Critical patent/CN110632607B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S15/00—Systems using the reflection or reradiation of acoustic waves, e.g. sonar systems
    • G01S15/88—Sonar systems specially adapted for specific applications
    • G01S15/93—Sonar systems specially adapted for specific applications for anti-collision purposes
    • G01S15/931—Sonar systems specially adapted for specific applications for anti-collision purposes of land vehicles
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines

Abstract

An object boundary determining method, system and vehicle based on ultrasonic data are disclosed, wherein the method comprises the following steps: after receiving a current data point acquired by an ultrasonic sensor, judging whether line segment segmentation is needed; if necessary, performing line fitting by using the data points collected after the current data point to obtain a new object boundary; if not, performing line fitting by using the current data point and the existing data point to update the boundary of the existing object; wherein the new object boundary is different from the existing object boundary, the existing object boundary being fitted by the existing data points, the existing data points being data points acquired before the current data point. By implementing the embodiment of the invention, the calculation amount required for determining the boundary of the object can be reduced, so that the calculation efficiency is improved.

Description

Object boundary determining method and system based on ultrasonic data and vehicle
Technical Field
The invention relates to the technical field of ultrasonic data processing, in particular to an object boundary determining method and system based on ultrasonic data and a vehicle.
Background
In the solution of autonomous driving, the environment around the vehicle can be detected by means of ultrasound to identify obstacles and drivable areas. Generally, the position coordinates of the data points of the reflected ultrasonic waves can be calculated according to the distance measured by the ultrasonic sensor and the position coordinates of the vehicle when the distance measured is measured. If the number of data points obtained is sufficiently large, it is possible to fit the boundaries of the obstacle using a plurality of data points, thereby determining the position of the obstacle and the area in which the vehicle can travel.
However, in practice, it is found that the computing device on the vehicle is generally an embedded device, and the computing power of the embedded device is limited, and if the number of data points is too large, the computing amount is increased, so that the computing efficiency is reduced, and the requirement of real-time detection is difficult to adapt.
Disclosure of Invention
The embodiment of the invention discloses an object boundary determining method and system based on ultrasonic data and a vehicle, which can reduce the calculation amount required for determining the object boundary, thereby improving the calculation efficiency.
The embodiment of the invention discloses a method for determining an object boundary based on ultrasonic data in a first aspect, which comprises the following steps:
after receiving a current data point acquired by an ultrasonic sensor, judging whether line segment segmentation is needed;
if necessary, performing line fitting by using the data points collected after the current data point to obtain a new object boundary;
if not, performing line fitting by using the current data point and the existing data point to update the boundary of the existing object;
wherein the new object boundary is different from the existing object boundary, the existing object boundary being fitted by the existing data points, the existing data points being data points acquired before the current data point.
As an optional implementation manner, in the first aspect of the embodiment of the present invention, the determining whether segment division is required includes:
judging whether the variation between the ranging distance corresponding to the current data point and the ranging distance corresponding to the adjacent data point is larger than a preset first distance threshold value or not; if yes, judging that segment segmentation is needed; the adjacent data points are data points of the existing data points, the acquisition time of which is adjacent to the current data point;
or, judging whether the number of the existing data points exceeds a preset number threshold; if yes, judging that the line segment division is needed.
As an optional implementation manner, in the first aspect of the embodiment of the present invention, after determining that segment segmentation is not required, and before performing segment fitting using the current data point and the existing data point to update the existing object boundary, the method further includes:
judging whether the deviation degree of the current data point from the set of the existing data points is lower than a preset first error threshold value or not; if yes, executing the step of performing line fitting by using the current data point and the existing data point to update the boundary of the existing object;
or judging whether the current data point is the Nth data point collected after the first data point; if yes, executing the step of performing line fitting by using the current data point and the existing data point to update the boundary of the existing object; wherein the first data point is the data point with the earliest acquisition time in the existing data points fitted with the existing object boundary; n is an integral multiple of a preset interval value;
or, judging whether the distance between the spatial position of the current data point and the spatial position of the adjacent data point exceeds a preset second distance threshold value; if yes, executing the step of performing line fitting by using the current data point and the existing data point to update the boundary of the existing object; wherein the adjacent data point is a data point of the existing data points, the acquisition time of which is adjacent to the current data point;
or, judging whether the current data point is an angular point; if not, executing the step of performing line segment fitting by using the current data point and the existing data point to update the boundary of the existing object.
As an optional implementation manner, in the first aspect of the embodiment of the present invention, if it is determined that the distance between the spatial position of the current data point and the spatial position of the adjacent data point does not exceed the preset second distance threshold, the current data point and the adjacent data point are merged to obtain a merged data point;
and performing line fitting by using the merged data point and the rest data points except the adjacent data points in the existing data points so as to update the existing object boundary.
As an optional implementation manner, in the first aspect of the embodiment of the present invention, the determining whether the current data point is a corner point includes:
acquiring other data points with the same distance measurement distance as the current data point in the existing data points;
judging whether the difference value of the echo energy of the current data point lower than the echo energy of the other data points exceeds a preset difference threshold value or not;
if yes, determining the current data point as a corner point;
or, the determining whether the current data point is an angular point includes:
inputting the current data point into a classification model, and determining whether the current data point is an angular point according to an output result of the classification model;
the classification model is obtained by utilizing pre-marked angular point data and non-angular point data for training, and is a support vector machine.
As an optional implementation manner, in the first aspect of the embodiment of the present invention, the determining whether segment division is required includes:
judging whether the variation between the ranging distance corresponding to the current data point and the ranging distance corresponding to the adjacent data point is larger than a preset distance threshold value or not; if yes, judging that segment segmentation is needed; the adjacent data point is the last data point collected before the current data point in the existing data points;
and after judging that the line segment division is needed, the method further comprises the following steps:
judging whether the deviation degree of the current data point from the set of the existing data points is lower than a preset error threshold value or not;
if yes, updating the existing object boundary by using the current data point, and executing the step of performing line segment fitting by using the data point acquired after the current data point to obtain a new object boundary when receiving the data point acquired after the current data point;
if not, when a data point collected after the current data point is received, executing the step of performing line segment fitting by using the data point collected after the current data point to obtain a new object boundary.
As an optional implementation manner, in the first aspect of the embodiment of the present invention, the performing a line segment fitting by using the current data point and the existing data point to update the existing object boundary includes:
identifying segmentation points for segmenting the sub-line segments from a data point set consisting of the current data point and the existing data points;
respectively fitting the data points in the data point set according to a positive direction and a negative direction to obtain a first fitting result fitted according to the positive direction and a second fitting result fitted according to the negative direction;
selecting a fitting result with a smaller fitting error from the first fitting result and the second fitting result as the updated existing object boundary;
wherein the sub-line segment to which the division point belongs in the first fitting result is different from the sub-line segment to which the division point belongs in the second fitting result; the forward direction is a direction in which the arrangement direction of the data points is consistent with the traveling direction of the vehicle, and the reverse direction is a direction in which the arrangement direction of the data points is inconsistent with the traveling direction of the vehicle.
The second aspect of the embodiments of the present invention discloses an object boundary determining system based on ultrasonic data, including:
the judging unit is used for judging whether line segment segmentation is needed or not after receiving the current data point acquired by the ultrasonic sensor;
the segmentation unit is used for performing line segment fitting by using data points acquired after the current data point when the judgment unit judges that line segment segmentation is required to be performed so as to obtain a new object boundary;
the updating unit is used for performing line segment fitting by using the current data point and the existing data point when the judging unit judges that the line segment segmentation is not needed so as to update the boundary of the existing object;
wherein the new object boundary is different from the existing object boundary, the existing object boundary being fitted by the existing data points, the existing data points being data points acquired before the current data point.
The third aspect of the embodiment of the invention discloses a vehicle, which comprises the second aspect of the embodiment of the invention and an object boundary determining system based on ultrasonic data.
A fourth aspect of the present invention discloses a computer-readable storage medium storing a computer program, wherein the computer program causes a computer to execute any one of the methods disclosed in the first aspect of the embodiments of the present invention.
A fifth aspect of the embodiments of the present invention discloses a computer program product, which, when running on a computer, causes the computer to execute any one of the methods disclosed in the first aspect of the embodiments of the present invention.
Compared with the prior art, the embodiment of the invention has the following beneficial effects:
the ultrasonic sensor continuously collects data points of surrounding objects, after the current data point is collected, if the fact that line segment segmentation is needed is judged, the current data point is taken as a critical point, the data point collected before the current data point does not participate in generation calculation of a new object boundary, and therefore calculation amount during line segment fitting is reduced; meanwhile, if the judgment shows that the line segment segmentation is not needed, the current data point is used for updating the boundary of the existing object, namely, the current data point and the data points collected among the current data points are used for line segment fitting. It can be seen that, in the embodiment of the present invention, the number of data points participating in the line segment fitting may be controlled within a certain range, and not all the collected data points are used for performing global line segment fitting, so that the amount of calculation in performing line segment fitting may be reduced, and the efficiency of calculating the object boundary by the embedded devices such as the car machine and the like may be improved.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a schematic flow chart of a method for determining object boundaries based on ultrasonic data according to an embodiment of the present invention;
FIG. 2 is an exemplary graph of a path distance as disclosed in embodiments of the present invention;
fig. 3 is an exemplary diagram of line segment fitting performed in the forward direction and the reverse direction, respectively, according to an embodiment of the present invention;
FIG. 4 is a schematic flow chart illustrating another method for determining object boundaries based on ultrasonic data according to an embodiment of the present disclosure;
FIG. 5 is a diagram illustrating a distribution of current data points and existing data points according to an embodiment of the present invention;
FIG. 6 is a schematic structural diagram of an object boundary determining system based on ultrasonic data according to an embodiment of the present invention;
fig. 7 is a schematic structural diagram of another object boundary determining system based on ultrasonic data according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It is to be noted that the terms "comprises" and "comprising" and any variations thereof in the embodiments and drawings of the present invention are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements listed, but may alternatively include other steps or elements not listed, or inherent to such process, method, article, or apparatus.
The embodiment of the invention discloses an object boundary determining method and system based on ultrasonic data and a vehicle, which can reduce the calculation amount required for determining the object boundary and improve the calculation efficiency. The following are detailed below.
Example one
Referring to fig. 1, fig. 1 is a schematic flowchart illustrating a method for determining an object boundary based on ultrasonic data according to an embodiment of the present invention. As shown in fig. 1, the object boundary determining method based on ultrasonic data may include the steps of:
101. after receiving a current data point acquired by an ultrasonic sensor, judging whether line segment segmentation is needed; if yes, executing step 102; if not, step 103 is performed.
In an embodiment of the present invention, one ultrasound transceiving operation of the ultrasound sensor acquires one data point, and the attribute of each data point may include, but is not limited to, at least one of the following information: corresponding position coordinates, corresponding distance measurement distance and corresponding ultrasonic echo energy. The corresponding ranging distance may include a ranging distance measured by a primary echo and/or a ranging distance measured by a secondary echo, and the corresponding ultrasonic echo energy may include ultrasonic energy of the primary echo and/or ultrasonic energy of the secondary echo.
The ultrasonic sensor performs a transmission/reception operation of ultrasonic waves at a certain period, so that a plurality of data points can be collected. In an embodiment of the present invention, the current data point is a data point collected by the ultrasonic sensor at the current time. The data points collected before the current data point are called existing data points, and when the number of the existing data points is at least two, the existing data points can be used for line segment fitting to obtain the boundary of the existing object. Optionally, existing data points may be stored in a cache for later recall when updating existing boundaries.
As an optional implementation manner, step 101 may specifically include:
judging whether the variation dt between the ranging distance L1 corresponding to the current data point and the ranging distance L2 corresponding to the adjacent data point is larger than a preset first distance threshold value or not;
if yes, judging that the line segment division is needed.
And the adjacent data points are data points of the existing data points, wherein the acquisition time of the existing data points is adjacent to the current data point. In practice, it is found that if two adjacent data points actually belong to the same object boundary, the distance measurement distance corresponding to the two data points has small change; otherwise, the ranging distance corresponding to the two data points has a large change. Thus, if dt is determined to be greater than the first threshold, the existing data point may actually belong to two different object boundaries than the data point collected after the current data point.
As another optional implementation, step 101 may further include:
judging whether the number of the existing data points exceeds a preset number threshold value or not; if yes, judging that the line segment division is needed.
By implementing the above-described embodiment, when the data points used for line segment fitting are accumulated to a certain number, line segment segmentation may be forced to reduce the number of data points used for line segment fitting each time, thereby reducing the amount of calculation. In addition, the number threshold may be set according to the moving speed of the vehicle and/or the period of the ultrasonic wave transmitted and received by the ultrasonic sensor, so that the distance moved by the vehicle is long enough when the number of the existing data points is accumulated to the number threshold. For example, the number threshold may be set to 100, and when the number of existing data points is accumulated to 100, the moving distance of the vehicle may already exceed the boundary length of a general object, so that segment segmentation is necessary.
That is to say, in the embodiment of the present invention, the line segment may be segmented when it is determined that two different object boundaries may be detected, or the line segment may be segmented when the existing data points are accumulated to a certain number, so as to reduce the number of data points used for line segment fitting, thereby improving the calculation efficiency. In addition, the variation of the distance measurement distance and/or the number of data points are used as the judgment standard of segment segmentation, and the line segmentation accords with the distribution rule of the actual object boundary to a certain extent, so that the object boundary with higher accuracy can be fitted by using relatively fewer data points.
102. And performing line segment fitting by using data points acquired after the current data point to obtain a new object boundary.
103. And performing line segment fitting by using the current data point and the existing data point so as to update the boundary of the existing object.
In the embodiment of the present invention, the line fitting may adopt a data fitting algorithm such as a least square method, and the embodiment of the present invention is not limited. When the data points used for fitting the line segment are changed, the line segment fitted is also changed correspondingly. And step 103 is executed, and the object boundary fitted after the current data point is added has a certain change compared with the object boundary fitted before the current data point is added, so that the existing object boundary is updated.
It will be appreciated that in step 102, the data points used in fitting the new object boundary are data points collected after the current data point, excluding the current data point as well as existing data points. Thus, the new object boundary generated is different from the existing object boundary. Furthermore, generating new object boundaries is a process that may actually translate into updating new existing boundaries. Specifically, after it is determined that segment segmentation is required, if at least two new data points are received and collected after the current data point, a new object boundary can be fitted by using the two new data points; when a third and subsequent new data point is received, if it is determined that segment segmentation is not required, the new object boundary may continue to be updated.
Further optionally, in the embodiment of the present invention, in order to improve the accuracy of the line segment fitting, a specific implementation manner of the step 103 may include the following steps:
1031. and identifying a segmentation point for segmenting the sub-line segment from the data point set consisting of the current data point and the existing data point.
Firstly, preliminarily fitting a fitting line segment by using data points in the data point set, identifying the projection distance from each data point in the data point set to the fitting line segment, and identifying the data point with the farthest projection distance to the fitting line segment as a segmentation point;
alternatively, a path distance between two adjacent data points in the data point set on the fitting line segment (i.e., a distance between projected points of the two data points on the fitting line segment) may be obtained, and a data point with the path distance exceeding a preset path distance threshold may be identified, and a data point exceeding the path distance threshold may be identified as a segmentation point. Referring to fig. 2, fig. 2 is a diagram illustrating an exemplary path distance according to an embodiment of the disclosure.
1032. And respectively fitting the data points in the data point set according to the forward direction, the reverse direction and the reverse direction to obtain a first fitting result fitted according to the forward direction and a second fitting result fitted according to the reverse direction.
The forward direction is a direction in which the arrangement direction of the data points is consistent with the driving direction of the vehicle, and the reverse direction is a direction in which the arrangement direction of the data points is inconsistent with the driving direction of the vehicle. Referring to fig. 3, fig. 3 is an exemplary diagram of performing line fitting in a forward direction and a reverse direction respectively according to an embodiment of the present invention.
1033. And selecting the fitting result with smaller fitting error from the first fitting result and the second fitting result as the updated existing object boundary.
Wherein the fitting error may comprise at least one of the following errors: standard deviation, mean square deviation, partial variance, polar difference, path distance between two adjacent data points in the same sub-line segment, and projection distance from any data point to the sub-line segment to which the data point belongs.
Referring to fig. 3, when the line segment fitting is performed in the forward and reverse directions, the sub-line segment of the division point in the first fitting result may be different from the sub-line segment of the division point in the second fitting result. In the embodiment of the invention, the fitting result with smaller fitting error is selected as the updated boundary of the existing object, so that the fitting accuracy can be improved. Taking fig. 3 as an example, in the sub-line segment fitted in the positive direction, the path distance between two adjacent data points is relatively average, and there is no situation of relatively large path distance, so that the first fitting result fitted in the positive direction can be selected as the updated existing boundary.
In summary, in the method described in fig. 1, if it is determined that segment segmentation is required, the data point acquired before the current data point does not participate in the generation calculation of the new object boundary; if the fact that segment segmentation is not needed is judged, the current data point is used for updating the boundary of the existing object, so that the number of data points participating in segment fitting each time can be controlled within a certain range, instead of performing global segment fitting by using all collected data points, the calculation amount during segment fitting can be reduced, the calculation efficiency is improved, and the method described in the figure 1 can be suitable for embedded equipment with low calculation capacity. Further, in the method described in fig. 1, specifically, the variation of the ranging distance and/or the number of data points may be used as a criterion for segment segmentation, and an object boundary with a higher accuracy may be fitted by using relatively fewer data points. Furthermore, when the existing object boundary is updated, the data points in the data point set can be further segmented to fit a plurality of sub-line segments; and, through the two fitting at least times of positive and negative direction, can improve the degree of accuracy of fitting.
Example two
Referring to fig. 4, fig. 4 is a schematic flowchart illustrating another method for determining an object boundary based on ultrasonic data according to an embodiment of the present invention. As shown in fig. 4, the object boundary determining method based on ultrasonic data may include the steps of:
201. after receiving a current data point collected by an ultrasonic sensor, judging whether the variation dt between the ranging distance L1 corresponding to the current data point and the ranging distance L2 corresponding to an adjacent data point is larger than a preset first distance threshold value; judging whether the number NUM of the existing data points exceeds a preset number threshold value or not; if dt is less than or equal to the first distance threshold and NUM is less than or equal to the number threshold, executing step 202 to step 205; if dt > the first distance threshold or NUM > the quantity threshold, step 208 is performed.
In the embodiment of the invention, if dt is less than or equal to a first distance threshold and NUM is less than or equal to a quantity threshold, it is judged that segment segmentation is not needed; and if dt is larger than the first distance threshold or NUM is larger than the quantity threshold, judging that the line segment is required to be segmented.
202. Judging whether the deviation degree of the current data point from the set of the existing data points is lower than a preset first error threshold value or not; if yes, go to step 203; if not, return to execute step 201.
In the embodiment of the present invention, the deviation degree may be specifically calculated by statistical errors such as standard deviation, variance, range, and the like. That is, step 202 may specifically be to calculate whether at least one error of standard deviation, variance or range of the set is lower than a preset first error threshold after adding the current data point to the set of existing data points.
Step 202 is performed to determine whether the current data point is significantly deviated from the set of existing data points (i.e., the existing object boundary), and if so, the current data point is not suitable for updating the existing object boundary.
203. Judging whether the current data point is the Nth data point collected after the first data point; if yes, go to step 204; if not, return to execute step 202.
In an embodiment of the present invention, the first data point is a data point with the earliest acquisition time among the existing data points that are fitted to the boundary of the existing object; n is an integral multiple of the preset interval value. Wherein the preset interval value may be set with reference to a quantity threshold. For example, if the quantity threshold is set to 100, the interval value may be set to 5. That is, every 5 data points in the data points collected by the ultrasonic sensor are used for updating the boundary of the existing object.
204. Judging whether the distance between the spatial position of the current data point and the spatial position of the adjacent data point exceeds a preset second distance threshold value or not; if yes, go to step 205; if not, step 206 is performed.
In the embodiment of the present invention, the second distance threshold may be set with reference to the moving speed of the vehicle and the cycle of the ultrasonic wave sensor transmitting and receiving the ultrasonic wave. If the distance between the spatial location (i.e., the location coordinate) of the current data point and the spatial location of the adjacent data point does not exceed the second distance threshold, it can be assumed that the vehicle has not moved or has moved a lesser distance in the time period from the detection of the adjacent data point to the detection of the current data point. Therefore, step 206 is performed to merge several data points with small spatial position variation into one merged data point to further reduce the number of data points for line segment fitting.
205. Judging whether the current data point is an angular point; if yes, returning to execute the step 201; if not, step 207 is performed.
In the embodiment of the present invention, the corner point is a redundant data point generated when the beam angle of the ultrasonic wave is large. Since the ultrasonic sensor cannot accurately identify at which position in the range covered by the beam angle the echo is reflected, when the beam angle of the ultrasonic wave is large, the position of the object located at the boundary of the beam angle may be calculated as the position of the object located at the center of the beam angle, thereby generating redundant data points. The redundant data points have low accuracy and are not suitable for updating the boundary of the existing object.
As an optional implementation manner, the implementation manner of step 205 may specifically include the following steps:
acquiring other data points with the same distance measurement distance as the current data point in the existing data points;
judging whether the difference value of the echo energy of the current data point lower than the echo energy of other data points exceeds a preset difference threshold value or not;
if so, the current data point is determined to be a corner point.
Since the ultrasonic energy at the beam angle boundary is generally small, by implementing the above-described embodiment, it can be determined whether the current data point is a corner point according to the echo energy.
As another optional implementation, the implementation of step 205 may specifically include the following steps:
inputting the current data point into a classification model, and determining whether the current data point is an angular point according to an output result of the classification model;
the classification model is obtained by training with pre-marked angular point data and non-angular point data, and is a Support Vector Machine (SVM).
By implementing the above embodiment, whether the current data point is a corner point can be identified through a pre-trained SVM classification model.
In addition, in the embodiment of the present invention, the above-mentioned steps 202 to 205 are executed to determine whether the current data point is suitable for updating the boundary of the existing object; and after the current data point is judged to be suitable for updating the existing object boundary, executing step 206 or step 207 to update the existing object boundary.
The foregoing steps 202 to 205 do not have a logically inevitable sequence, and only a part or all of the steps 202 to 20 may be executed, and the execution sequence disclosed in the embodiment of the present invention should not be limited.
206. And merging the current data point and the adjacent data points to obtain merged data points, and performing line segment fitting by using the merged data points and the rest data points except the adjacent data points in the existing data points to update the boundary of the existing object.
In the embodiment of the present invention, the line segment fitting is performed by using the merged data points and the remaining data points, and a method for updating the existing object boundary may specifically be:
the merged data point and the remaining data points are used to form a data point set, and step 1032 to step 1033 in the first embodiment are performed on the formed data point set, which is not described in detail below.
207. And performing line segment fitting by using the current data point and the existing data point so as to update the boundary of the existing object.
In this embodiment of the present invention, a specific implementation manner of step 207 may be as shown in step 1031 to step 1033 in embodiment one, and details are not described below.
208. Judging whether the deviation degree of the current data point from the set of the existing data points is lower than a preset first error threshold value or not; if so, step 209; if not, step 210 is performed.
In the embodiment of the present invention, please refer to fig. 5, in which fig. 5 is a diagram illustrating a distribution of current data points and existing data points according to the embodiment of the present invention. As shown in fig. 5, dt > the first error threshold, but in practice the degree of deviation of the current data point from the entire set of existing data points is low, and in this case, it is considered that the determination that segment division is necessary is actually a false determination, and the current data point is actually suitable for updating the existing object boundary. Step 209, described below, is therefore performed, and after the last update to the existing boundary, step 210 is performed to generate a new object boundary.
209. The existing object boundary is updated with the current data point.
In an embodiment of the present invention, step 209 is used to perform the last update on the existing object boundary.
210. And when receiving the data points collected after the current data point, performing line segment fitting by using the data points collected after the current data point to obtain a new object boundary.
It can be seen that in the method described in fig. 4, the number of data points used for line segment fitting can be reduced, the amount of calculation required for determining the object boundary can be reduced, and the calculation efficiency can be improved. Whether the current data point is suitable for updating the existing object boundary may also be determined by at least one of: judging the deviation degree of the current data point, judging whether the current data point is the Nth data point, judging whether the moving distance of the vehicle is enough, and judging whether the current data point is not an angular point. After the judgment is suitable for updating, the current data point is used for updating the existing object boundary, so that the accuracy of the fitted object boundary can be further improved. Furthermore, after the judgment that the segment segmentation is not needed, whether the current data point is suitable for updating the boundary of the existing object is also judged, so that the influence caused by the misjudgment of the segment segmentation can be reduced.
EXAMPLE III
Referring to fig. 6, fig. 6 is a schematic structural diagram of an object boundary determining system based on ultrasonic data according to an embodiment of the present invention. As shown in fig. 6, the object boundary determining system based on ultrasonic data may include:
the judging unit 601 is configured to judge whether segment segmentation is required after receiving a current data point acquired by the ultrasonic sensor;
a segmentation unit 602, configured to perform line segment fitting by using data points acquired after a current data point when the determination unit 601 determines that line segment segmentation is required, so as to obtain a new object boundary;
an updating unit 603, configured to perform line segment fitting by using the current data point and the existing data point when the determining unit 601 determines that line segment segmentation is not needed, so as to update the existing object boundary;
wherein the new object boundary is different from an existing object boundary, the existing object boundary being fitted from existing data points, the existing data points being data points acquired before the current data point.
Optionally, the determining unit 601 may be specifically configured to determine whether a variation dt between the ranging distance L1 corresponding to the current data point and the ranging distance L2 corresponding to the adjacent data point is greater than a preset first distance threshold; if yes, judging that the line segment division is needed. And the adjacent data points are data points of the existing data points, wherein the acquisition time of the existing data points is adjacent to the current data point.
Alternatively, the determining unit 601 may be specifically configured to determine whether the number of the existing data points exceeds a preset number threshold; if yes, judging that the line segment division is needed.
Further optionally, the updating unit 603 may be specifically configured to identify, from a data point set formed by the current data point and the existing data point, a segmentation point for performing sub-line segment segmentation; and the number of the first and second groups,
respectively fitting the data points in the data point set according to the forward direction, the reverse direction and the reverse direction to obtain a first fitting result fitted according to the forward direction and a second fitting result fitted according to the reverse direction;
and selecting the fitting result with smaller fitting error from the first fitting result and the second fitting result as the updated existing object boundary.
The forward direction is a direction in which the arrangement direction of the data points is consistent with the driving direction of the vehicle, and the reverse direction is a direction in which the arrangement direction of the data points is inconsistent with the driving direction of the vehicle. The sub-line segment to which the segmentation point belongs in the first fitting result is different from the sub-line segment to which the segmentation point belongs in the second fitting result.
It can be seen that, by implementing the object boundary determining system based on ultrasonic data shown in fig. 6, when it is determined that segment segmentation is required, new object boundary generation calculation can be performed without using data points acquired before the current data point; when the line segment segmentation is judged not to be needed, the current data point is used for updating the boundary of the existing object, so that the number of data points participating in line segment fitting each time can be controlled within a certain range, instead of performing global line segment fitting by using all collected data points, the calculation amount during line segment fitting is reduced, and the calculation efficiency is improved. Furthermore, the variation of the distance measurement distance and/or the number of data points can be used as a judgment standard for segment segmentation, and the object boundary with higher accuracy can be fitted by adopting relatively fewer data points. Furthermore, when the existing object boundary is updated, the data points in the data point set can be further segmented to fit a plurality of sub-line segments; and, through the two fitting at least times of positive and negative direction, can improve the degree of accuracy of fitting.
Example four
Referring to fig. 7, fig. 7 is a schematic structural diagram of another object boundary determining system based on ultrasonic data according to an embodiment of the present disclosure. The object boundary determining system based on ultrasonic data shown in fig. 7 is optimized by the object boundary determining system based on ultrasonic data shown in fig. 6. As shown in fig. 7, the object boundary determining system based on ultrasonic data may further include:
a screening unit 604, configured to determine whether a deviation degree of the current data point from the existing data point set is lower than a preset first error threshold after the determining unit 601 determines that the line segment segmentation is required, and before the updating unit 603 performs line segment fitting by using the current data point and the existing data point to update the existing object boundary; alternatively, the first and second electrodes may be,
judging whether the current data point is the Nth data point collected after the first data point; the first data point is the data point with the earliest acquisition time in the existing data points which are fitted with the boundary of the existing object; n is an integral multiple of a preset interval value; alternatively, the first and second electrodes may be,
judging whether the distance between the spatial position of the current data point and the spatial position of the adjacent data point exceeds a preset second distance threshold value or not; alternatively, the first and second electrodes may be,
judging whether the current data point is an angular point;
the screening unit 604 is further configured to determine that a deviation degree of the current data point from the existing set of data points is lower than a preset first error threshold; or when the current data point is judged to be the Nth data point collected after the first data point; or judging that the distance between the spatial position of the current data point and the spatial position of the adjacent data point exceeds a preset second distance threshold; or, when the current data point is determined to be an angular point, the updating unit 603 is triggered to perform a line segment fitting operation using the current data point and the existing data point to update the existing object boundary.
Further optionally, the manner that the screening unit 604 is used to determine whether the current data point is a corner point may specifically be:
the screening unit 604 is configured to obtain other data points, which have the same distance as the current data point, in the existing data points; judging whether the difference value of the echo energy of the current data point lower than the echo energy of other data points exceeds a preset difference threshold value or not; if yes, determining the current data point as a corner point;
or, the data processing device is used for inputting the current data point into the classification model and determining whether the current data point is an angular point according to the output result of the classification model; the classification model is obtained by utilizing pre-marked angular point data and non-angular point data for training, and is a support vector machine.
Further optionally, the updating unit 603 may be further configured to, after the screening unit 604 determines that the distance between the spatial position of the current data point and the spatial position of the adjacent data point does not exceed the preset second distance threshold, merge the current data point and the adjacent data point to obtain a merged data point; and performing line fitting by using the merged data point and the rest data points except for the adjacent data points in the existing data points so as to update the boundary of the existing object.
Still further optionally, the screening unit 604 may be further configured to, when the determining unit 601 determines that the variation between the ranging distance corresponding to the current data point and the ranging distance corresponding to the adjacent data point is greater than the preset distance threshold, perform an operation of determining whether the deviation degree of the current data point from the existing set of data points is lower than a preset error threshold, and when the deviation degree of the current data point from the existing set of data points is determined to be lower than the preset error threshold, trigger the updating unit 603 to update the boundary of the existing object by using the current data point; and when receiving the data point collected after the current data point, the trigger segmentation unit 602 performs a step of performing line segment fitting using the data point collected after the current data point to obtain a new object boundary.
It can be seen that implementing the object boundary determining system based on ultrasonic data as shown in fig. 7 can reduce the amount of calculation required to determine the boundary of an object, thereby improving the calculation efficiency. Whether the current data point is suitable for updating the existing object boundary may also be determined by at least one of: judging the deviation degree of the current data point, judging whether the current data point is the Nth data point, judging whether the moving distance of the vehicle is enough, and judging whether the current data point is not an angular point. After the judgment is suitable for updating, the current data point is used for updating the existing object boundary, so that the accuracy of the fitted object boundary can be further improved. Furthermore, after the judgment that the segment segmentation is not needed, whether the current data point is suitable for updating the boundary of the existing object is also judged, so that the influence caused by the misjudgment of the segment segmentation can be reduced.
An embodiment of the present invention discloses a computer-readable storage medium storing a computer program, wherein the computer program causes a computer to execute any one of the object boundary determining methods based on ultrasonic data shown in fig. 1 or 4.
An embodiment of the present invention discloses a computer program product comprising a non-transitory computer readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute any one of the object boundary determining methods based on ultrasonic data shown in fig. 1 or fig. 4.
It should be appreciated that reference throughout this specification to "one embodiment" or "an embodiment" means that a particular feature, structure or characteristic described in connection with the embodiment is included in at least one embodiment of the present invention. Thus, the appearances of the phrases "in one embodiment" or "in an embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. Those skilled in the art should also appreciate that the embodiments described in this specification are exemplary and alternative embodiments, and that the acts and modules illustrated are not required in order to practice the invention.
In various embodiments of the present invention, it should be understood that the sequence numbers of the above-mentioned processes do not imply an inevitable order of execution, and the execution order of the processes should be determined by their functions and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated units, if implemented as software functional units and sold or used as a stand-alone product, may be stored in a computer accessible memory. Based on such understanding, the technical solution of the present invention, which is a part of or contributes to the prior art in essence, or all or part of the technical solution, can be embodied in the form of a software product, which is stored in a memory and includes several requests for causing a computer device (which may be a personal computer, a server, a network device, or the like, and may specifically be a processor in the computer device) to execute part or all of the steps of the above-described method of each embodiment of the present invention.
It will be understood by those skilled in the art that all or part of the steps in the methods of the embodiments described above may be implemented by hardware instructions of a program, and the program may be stored in a computer-readable storage medium, where the storage medium includes Read-Only Memory (ROM), Random Access Memory (RAM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), One-time Programmable Read-Only Memory (OTPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Compact Disc Read-Only Memory (CD-ROM), or other Memory, such as a magnetic disk, or a combination thereof, A tape memory, or any other medium readable by a computer that can be used to carry or store data.
The object boundary determining method, system and vehicle based on ultrasonic data disclosed in the embodiments of the present invention are described in detail above, and the principles and embodiments of the present invention are explained herein by applying specific examples, and the above description of the embodiments is only used to help understanding the method and its core ideas of the present invention. Meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.

Claims (9)

1. A method for object boundary determination based on ultrasound data, the method comprising:
after receiving a current data point acquired by an ultrasonic sensor, judging whether line segment segmentation is needed according to an existing data point, wherein the existing data point is acquired by the ultrasonic sensor periodically executing the receiving and sending operation of ultrasonic waves;
if necessary, performing line fitting by using the data points collected after the current data point to obtain a new object boundary;
if not, performing line fitting by using the current data point and the existing data point to update the boundary of the existing object;
wherein the new object boundary is different from the existing object boundary, the existing object boundary being fitted by the existing data points, the existing data points being data points acquired before the current data point.
2. The method of claim 1, wherein the determining whether segment segmentation is required comprises:
judging whether the variation between the ranging distance corresponding to the current data point and the ranging distance corresponding to the adjacent data point is larger than a preset first distance threshold, and judging whether the number of the current data points exceeds a preset number threshold, wherein the adjacent data points are data points adjacent to the current data point in the acquisition time; if the variation is larger than the first distance threshold and the quantity is larger than the quantity threshold, judging that segment segmentation is needed; if the variation is smaller than or equal to the first distance threshold or the quantity is smaller than or equal to the quantity threshold, judging that segment segmentation is not needed;
after determining that segment segmentation is not required, and before performing segment fitting using the current data point and the existing data point to update the existing object boundary, the method further includes:
judging whether the deviation degree of the current data point from the set of the existing data points is lower than a preset first error threshold value or not; if so, judging whether the current data point is the Nth data point collected after the first data point; if not, returning to the step of judging whether the line segment division is needed or not;
if the current data point is the Nth data point collected after the first data point, judging whether the distance between the spatial position of the current data point and the spatial position of the adjacent data point exceeds a preset second distance threshold value; if not, returning to the step of judging whether the line segment division is needed or not; wherein the first data point is the data point with the earliest acquisition time in the existing data points fitted with the existing object boundary; n is an integral multiple of a preset interval value;
if the distance between the spatial position of the current data point and the spatial position of the adjacent data point exceeds a preset second distance threshold, judging whether the current data point is an angular point; if not, merging the current data point and the adjacent data point to obtain a merged data point; wherein the adjacent data point is a data point of the existing data points, the acquisition time of which is adjacent to the current data point;
if the current data point is not an angular point, executing the step of performing line fitting by using the current data point and the existing data point to update the boundary of the existing object; if yes, returning to the step of judging whether the line segment division is needed or not.
3. The method of claim 2, further comprising, after said obtaining merged data points:
and performing line fitting by using the merged data point and the rest data points except the adjacent data points in the existing data points so as to update the existing object boundary.
4. The method of claim 2, wherein said determining whether the current data point is a corner point comprises:
acquiring other data points with the same distance measurement distance as the current data point in the existing data points;
judging whether the difference value of the echo energy of the current data point lower than the echo energy of the other data points exceeds a preset difference threshold value or not;
if yes, determining the current data point as a corner point;
or, the determining whether the current data point is an angular point includes:
inputting the current data point into a classification model, and determining whether the current data point is an angular point according to an output result of the classification model;
the classification model is obtained by utilizing pre-marked angular point data and non-angular point data for training, and is a support vector machine.
5. The method of claim 1, wherein the determining whether segment segmentation is required comprises:
judging whether the variation between the ranging distance corresponding to the current data point and the ranging distance corresponding to the adjacent data point is larger than a preset first distance threshold value or not; if yes, judging that segment segmentation is needed; the adjacent data point is the last data point collected before the current data point in the existing data points;
and after judging that the line segment division is needed, the method further comprises the following steps:
judging whether the deviation degree of the current data point from the set of the existing data points is lower than a preset first error threshold value or not;
if yes, updating the existing object boundary by using the current data point, and executing the step of performing line segment fitting by using the data point acquired after the current data point to obtain a new object boundary when receiving the data point acquired after the current data point;
if not, when a data point collected after the current data point is received, executing the step of performing line segment fitting by using the data point collected after the current data point to obtain a new object boundary.
6. The method of claim 1, wherein said line segment fitting using said current data points and existing data points to update existing object boundaries comprises:
identifying segmentation points for segmenting the sub-line segments from a data point set consisting of the current data point and the existing data points;
respectively fitting the data points in the data point set according to a positive direction and a negative direction to obtain a first fitting result fitted according to the positive direction and a second fitting result fitted according to the negative direction;
selecting a fitting result with a smaller fitting error from the first fitting result and the second fitting result as the updated existing object boundary;
wherein the sub-line segment to which the division point belongs in the first fitting result is different from the sub-line segment to which the division point belongs in the second fitting result; the forward direction is a direction in which the arrangement direction of the data points is consistent with the traveling direction of the vehicle, and the reverse direction is a direction in which the arrangement direction of the data points is inconsistent with the traveling direction of the vehicle.
7. An object boundary determination system based on ultrasonic data, comprising:
the device comprises a judging unit, a judging unit and a judging unit, wherein the judging unit is used for judging whether line segment segmentation is needed according to the existing data point after receiving the current data point acquired by an ultrasonic sensor, and the existing data point is acquired by the ultrasonic sensor periodically executing the receiving and sending operation of ultrasonic waves;
the segmentation unit is used for performing line segment fitting by using data points acquired after the current data point when the judgment unit judges that line segment segmentation is required to be performed so as to obtain a new object boundary;
the updating unit is used for performing line segment fitting by using the current data point and the existing data point when the judging unit judges that the line segment segmentation is not needed so as to update the boundary of the existing object;
wherein the new object boundary is different from the existing object boundary, the existing object boundary being fitted by the existing data points, the existing data points being data points acquired before the current data point.
8. A vehicle comprising the ultrasonic data-based object boundary determination system of claim 7.
9. A computer-readable storage medium characterized by storing a computer program, wherein the computer program causes a computer to execute the object boundary determining method based on ultrasonic data according to any one of claims 1 to 6.
CN201910869730.8A 2019-09-16 2019-09-16 Object boundary determining method and system based on ultrasonic data and vehicle Active CN110632607B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910869730.8A CN110632607B (en) 2019-09-16 2019-09-16 Object boundary determining method and system based on ultrasonic data and vehicle

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910869730.8A CN110632607B (en) 2019-09-16 2019-09-16 Object boundary determining method and system based on ultrasonic data and vehicle

Publications (2)

Publication Number Publication Date
CN110632607A CN110632607A (en) 2019-12-31
CN110632607B true CN110632607B (en) 2022-02-01

Family

ID=68972622

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910869730.8A Active CN110632607B (en) 2019-09-16 2019-09-16 Object boundary determining method and system based on ultrasonic data and vehicle

Country Status (1)

Country Link
CN (1) CN110632607B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113311437B (en) * 2021-06-08 2022-04-19 安徽域驰智能科技有限公司 Method for improving angular point position accuracy of vehicle-mounted radar positioning side parking space

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP5271202B2 (en) * 2009-08-31 2013-08-21 パナソニック株式会社 Ultrasonic sensor
JP2011196943A (en) * 2010-03-23 2011-10-06 Denso Corp Recognition device
CN103760569B (en) * 2013-12-31 2016-03-30 西安交通大学 A kind of drivable region detection method based on laser radar
CN105137412B (en) * 2015-08-19 2017-10-20 重庆大学 A kind of 2D laser radars range image middle conductor feature Accurate Curve-fitting method
CN105930833B (en) * 2016-05-19 2019-01-22 重庆邮电大学 A kind of vehicle tracking and dividing method based on video monitoring
CN106840176B (en) * 2016-12-28 2020-01-31 济宁中科先进技术研究院有限公司 GPS time-space data increment road network real-time updating and track matching system
CN109035363B (en) * 2018-07-05 2023-11-10 凌云光技术股份有限公司 Line-circle optimal fitting method for rapid iteration

Also Published As

Publication number Publication date
CN110632607A (en) 2019-12-31

Similar Documents

Publication Publication Date Title
EP3384360B1 (en) Simultaneous mapping and planning by a robot
JP3708130B2 (en) How to create an area map for the cell structure of an autonomous movement unit
CN109635816B (en) Lane line generation method, apparatus, device, and storage medium
JP7042905B2 (en) Methods and devices for generating inverse sensor models, as well as methods for detecting obstacles
CN110674705B (en) Small-sized obstacle detection method and device based on multi-line laser radar
JPWO2018221453A1 (en) Output device, control method, program, and storage medium
US20190391583A1 (en) Method and device for detecting road boundary
US20200116509A1 (en) Assistance control system
JP7152355B2 (en) Obstacle detection device and obstacle detection method
US8744752B2 (en) Apparatus and method for detecting locations of vehicle and obstacle
CN111886597A (en) Obstacle detection method and device for movable platform and movable platform
CN110632607B (en) Object boundary determining method and system based on ultrasonic data and vehicle
CN114384491B (en) Point cloud processing method and device for laser radar and storage medium
CN114384492B (en) Point cloud processing method and device for laser radar and storage medium
CN113536850A (en) Target object size testing method and device based on 77G millimeter wave radar
CN111951552A (en) Method and related device for risk management in automatic driving
CN116699596A (en) Method and device for correcting speed of millimeter wave radar of vehicle
JP2004309166A (en) Target tracking apparatus
US11508147B2 (en) Streaming object detection within sensor data
CN115327529A (en) 3D target detection and tracking method fusing millimeter wave radar and laser radar
CN116863325A (en) Method for multiple target detection and related product
JP5644673B2 (en) Perimeter monitoring device, perimeter monitoring method, and driving support device
JP2022090373A (en) Information processor, control method, program and storage medium
CN113552574B (en) Region detection method and device, storage medium and electronic equipment
CN117250610B (en) Laser radar-based intruder early warning method and system

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
TA01 Transfer of patent application right
TA01 Transfer of patent application right

Effective date of registration: 20210107

Address after: Room 46, room 406, No.1, Yichuang street, Zhongxin knowledge city, Huangpu District, Guangzhou City, Guangdong Province 510000

Applicant after: Guangzhou Xiaopeng Automatic Driving Technology Co.,Ltd.

Address before: Room 245, No. 333, jiufo Jianshe Road, Zhongxin Guangzhou Knowledge City, Guangzhou City, Guangdong Province

Applicant before: GUANGZHOU XIAOPENG MOTORS TECHNOLOGY Co.,Ltd.

GR01 Patent grant
GR01 Patent grant
TR01 Transfer of patent right
TR01 Transfer of patent right

Effective date of registration: 20240227

Address after: 510000 No.8 Songgang street, Cencun, Tianhe District, Guangzhou City, Guangdong Province

Patentee after: GUANGZHOU XIAOPENG MOTORS TECHNOLOGY Co.,Ltd.

Country or region after: China

Address before: Room 46, room 406, No.1, Yichuang street, Zhongxin knowledge city, Huangpu District, Guangzhou City, Guangdong Province 510000

Patentee before: Guangzhou Xiaopeng Automatic Driving Technology Co.,Ltd.

Country or region before: China