Method for removing road sound insulation cover and removing device therefor
By finding target points on vehicle radar, fitting and screening straight lines, and deleting the target points of sound insulation covers, the problem of misjudgment of sound insulation covers is solved, and the retention and identification of normal target tracks is achieved.
Patent Information
- Application Number
- CN202110880501.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-08-02
- Publication Date
- 2025-07-25
- Estimated Expiration
- 2041-08-02
AI Technical Summary
When the vehicle radar is on an elevated or special section of the road, the echo of the sound insulation cover is displayed directly in front of the radar, causing the radar to mistakenly think that there are obstacles in front and cannot distinguish the normal target tracks in front.
The random sampling consistency algorithm is used to find the target point in front of the vehicle, fit the straight line, filter the straight line that meets the set conditions, select the optimal straight line and delete the target point, reduce randomness through the state machine, and retain the normal target track in front of the vehicle.
Effectively distinguish and remove the road sound insulation cover, retain the normal target track in front of the vehicle, reduce misjudgment obstacles, and improve the accuracy of radar target recognition.
Smart Images

Figure CN115701548B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of vehicle radar signal processing, and more specifically, to a method and a device for removing a road sound insulation cover applicable to vehicle-mounted millimeter-wave radar. Background Art
[0002] When a vehicle is driving on an elevated road or a special section, the sound insulation cover on the side section is exactly within the height detection range of the vehicle radar. The echo of the sound insulation cover is displayed directly in front of the radar. Since the radar has no height information, it causes the radar to mistakenly think that there is an obstacle in front. Moreover, the track of the sound insulation cover is very long, and the tracks of normal targets in front will be submerged in the track of the sound insulation cover, making it impossible to distinguish normal targets in front. Summary of the Invention
[0003] The technical problem to be solved by the present invention is to provide a method and a device for removing a road sound insulation cover from a vehicle radar, which can retain the tracks of normal targets in front of the vehicle.
[0004] To solve the above technical problem, the present invention provides a method for removing a road sound insulation cover, applicable to a vehicle radar, including:
[0005] Step S1, finding target points, and using the random sample consensus algorithm to find the target points within a first set width range in front of the vehicle;
[0006] Step S2, fitting a straight line, traversing every two of the target points and fitting them into a straight line;
[0007] Step S3, performing a first screening on the fitted straight line, screening the fitted straight line within a second set distance range in front of the vehicle, and the slope of the fitted straight line relative to the forward direction of the vehicle is within a third set ratio range;
[0008] Step S4, performing a second screening on the fitted straight line, traversing the first screening result, and selecting the fitted straight line that can cover the most target points as the optimal straight line;
[0009] Step S5, selecting target points, finding the speed mode of the target points covered by the optimal straight line, and selecting the target points corresponding to the sound insulation cover according to the speed mode;
[0010] Step S6, deleting the selected target points.
[0011] According to an embodiment of the present invention, in step S1, the first set width is on the perpendicular bisector of the vehicle's central axis, and the first set width is 1m to 2m.
[0012] According to an embodiment of the present invention, in step S3, the second set distance is 10m to 30m.
[0013] According to an embodiment of the present invention, in step S3, the third set ratio is less than 8 and greater than -8.
[0014] According to an embodiment of the present invention, in step S4, if the distance from the target point to the fitted straight line is less than a set threshold, then the target point can be covered by the fitted straight line.
[0015] According to an embodiment of the present invention, before performing step S5, a state machine is used to reduce the randomness of straight line fitting.
[0016] According to an embodiment of the present invention, the state machine uses a counter, and a threshold value is set for the counter. The execution steps of the state machine include:
[0017] Step T1, the state machine searches for the optimal straight line among the fitted straight lines. If the optimal straight line is found, the counter is incremented by 1;
[0018] Step T2, determine whether the counter has reached the set threshold value. If not, return to step S1; otherwise, proceed to the next step;
[0019] Step T3, determine the validity of the target point.
[0020] According to an embodiment of the present invention, in step S5, the speed mode is obtained by searching for the target points with the same speed or by the vehicle speed.
[0021] According to an embodiment of the present invention, in step S5, during the process of searching for the speed mode of the target points, if the number of target points that meet the speed mode reaches the set quantity, the obtained speed mode is valid.
[0022] The present invention also provides a device for removing a road sound insulation cover, which is applicable to a vehicle radar and includes:
[0023] A search module that uses the random sample consensus algorithm to search for target points within a first set width range in front of the vehicle radar;
[0024] A straight line fitting module that receives the target points found by the search module and fits a straight line by traversing every two of the target points;
[0025] A first screening module that receives the straight lines fitted by the straight line fitting module, screens and retains the straight lines within a second set distance range in front of the vehicle, and the slope of the fitted straight line relative to the forward direction of the vehicle is within a third set ratio range;
[0026] The second screening module receives the straight lines screened by the first screening module and selects the fitted straight line that can cover the most of the target points as the optimal straight line;
[0027] The selected target point module finds the mode of the speeds of the target points covered by the optimal straight line and selects the target points corresponding to the sound insulation cover according to the mode of the speeds;
[0028] The deletion module deletes the target points selected by the selected target point module.
[0029] A method and a device for removing a road sound insulation cover provided by the present invention can distinguish and remove the road sound insulation cover, so that the normal target track in front of the vehicle can be retained. Description of the Drawings
[0030] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the following detailed description of the specific embodiments of the present invention is provided in conjunction with the drawings, where:
[0031] Figure 1 is a flowchart of a method for removing a road sound insulation cover according to an embodiment of the present invention;
[0032] Figure 2 is a schematic diagram of an embodiment of the present invention.
[0033] Figure 3 is a schematic structural diagram of a device for removing a road sound insulation cover according to an embodiment of the present invention. Detailed Embodiments
[0034] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the following detailed description of the specific embodiments of the present invention is provided in conjunction with the drawings.
[0035] In the following description, many specific details are set forth in order to fully understand the present invention, but the present invention may also be implemented in other ways different from those described herein. Therefore, the present invention is not limited by the specific embodiments disclosed below.
[0036] As shown in the present application and the claims, unless the context clearly indicates an exception, the words "a", "an", "one", and / or "the" are not specifically singular and may also include plural. Generally speaking, the terms "including" and "comprising" only indicate the inclusion of the clearly identified steps and elements, and these steps and elements do not constitute an exclusive list, and the method or device may also include other steps or elements.
[0037] When describing the embodiments of the present application in detail, for the convenience of explanation, the cross-sectional views showing the device structure will be enlarged locally in a non-general proportion, and the schematic diagrams are only examples and should not limit the scope of protection of the present application herein. In addition, in actual production, three-dimensional spatial dimensions including length, width, and depth should be included.
[0038] For the convenience of description, spatial relationship terms such as "beneath", "below", "lower", "under", "above", "on", etc. may be used herein to describe the relationship between one element or feature shown in the drawings and other elements or features. It will be understood that these spatial relationship terms are intended to encompass other directions of the device in use or operation in addition to the directions depicted in the drawings. For example, if the device in the drawings is flipped, the direction of the element described as "beneath" or "below" or "under" other elements or features will be changed to "above" the other elements or features. Thus, the exemplary terms "below" and "under" can encompass both upward and downward directions. The device may also have other orientations (rotated 90 degrees or in other directions), so the spatial relationship descriptive terms used herein should be interpreted accordingly. In addition, it will also be understood that when a layer is referred to as being "between" two layers, it can be the only layer between the two layers, or there can be one or more intervening layers.
[0039] In the context of the present application, the structure in which the first feature is "on" the second feature may include embodiments where the first and second features are formed in direct contact, and may also include embodiments where additional features are formed between the first and second features, such that the first and second features may not be in direct contact.
[0040] Since the sound insulation cover is usually arranged directly above the road and is mostly in a semi-cylindrical tubular shape, it cannot be distinguished from ordinary targets based on the external shape shown by the vehicle radar. The sound insulation cover is not like a fence that can be seen as a neat straight line from the appearance. Therefore, only the echo data characteristics of the sound insulation cover can be used to find the features of the sound insulation cover. Due to the pitch angle of the radar and the beam width of the radar, the radar wave has echoes only at the topmost points of the semi-cylindrical tubular sound insulation cover, and these points form a slender straight line. Based on this straight line, the sound insulation cover data can be filtered. Define the points that make up this slender straight line as target points.
[0041] Figure 1 It is a flowchart of a method for removing a road sound insulation cover according to an embodiment of the present invention. Figure 2 It is a schematic diagram of an embodiment of the present invention. As shown, a method for removing a road sound insulation cover, applicable to a vehicle radar, includes:
[0042] Step S1: Locate the target points. Use the Random Sample Consensus (RANSAC) algorithm to find target points within a first preset width range in front of the vehicle. The target points are the echo signals of the obstacles in front received by the vehicle's radar. Refer to Figure 2 , where the X-axis direction is the horizontal direction of the vehicle's front, and the Y-axis direction is the central axis of the vehicle. Locate the target points 201 within the first preset width L1.
[0043] Step S2: Fit a straight line. Traverse every two target points and fit them into a straight line. Specifically, it is necessary to traverse every combination of two target points to fit a straight line, and then find the global optimal solution, that is, the most suitable straight line.
[0044] Step S3: First screening of the fitted straight lines. Make the fitted straight lines within a second preset distance range in front of the vehicle, and the slope formed by the fitted straight lines and the forward direction of the vehicle is within a third preset ratio range. Here, the second preset distance and the third preset ratio are used as limiting conditions to select the fitted straight lines that meet the requirements, which is equivalent to removing those straight lines not fitted by the sound insulation cover. Refer to Figure 2 , find the fitted straight lines that meet the requirements within the second preset distance L2.
[0045] Step S4: Second screening of the fitted straight lines. Traverse the results of the first screening and select the fitted straight line that can cover the most target points as the optimal straight line.
[0046] Step S5: Select the target points. Find the mode of the speeds of the target points 201 covered by the optimal straight line, and select the target points 201 corresponding to the sound insulation cover according to the mode of the speeds.
[0047] Step S6: Delete the selected target points 201, which is equivalent to deleting the trajectory of the sound insulation cover on the vehicle's radar.
[0048] Preferably, in step S1, the first preset width is 1m to 2m. Since there is only data directly in front of the vehicle's radar for the sound insulation cover, when fitting a straight line, only the target points 201 directly in front need to be found, that is, in front of the vehicle's radar, which is equivalent to extracting a straight line within a range of 0.5 to 1 meter on each side of the vehicle's head position along the central axis. In this way, other obstacles such as fences will not be misfitted. It is easy to understand that, refer to Figure 2 , the central axis of the vehicle is the Y-axis, and the vehicle travels along the Y-axis. Locate the target points 201 within a width range of 1m on each side of the Y-axis. The direction of the first preset width L1 is perpendicular to the Y-axis and is consistent with the width direction of the vehicle body. In addition, this method can effectively reduce the computational amount of the Random Sample Consensus (RANSAC) algorithm and improve the removal efficiency.
[0049] Preferably, in step S3, the second set distance L2 is 10 m to 30 m. Since the fences on the roadside close to the vehicle radar also satisfy the fitted straight line and can also meet other conditions, the difference between the fences and the sound insulation covers is used to ensure that the target points 201 corresponding to the fences are not mistakenly deleted. Through actual observation, it is found that the sound insulation covers start to form a straight line at a position more than ten meters directly in front of the radar, while the fences start to form a straight line from the horizontal position directly to the left or right of the vehicle radar. Therefore, controlling the starting point of the fitted straight line of the sound insulation cover to be in the range of 10 meters to 30 meters can effectively reduce the situation of mistakenly deleting the fences. Controlling the area of 10 meters to 30 meters is because the vehicle radar has a pitch angle. For example, when the pitch angle of the radar is between -15 degrees and 15 degrees, for a 3-meter-high sound insulation cover, the detection distance is approximately 12 meters. Therefore, only the fitted straight line that appears beyond 10 meters may be the sound insulation cover, and the fitted straight line that appears within 10 meters is very likely to be the fence within the range directly in front.
[0050] Preferably, in step S3, the third set ratio is less than 8 and greater than -8. When fitting a straight line, since the vehicle can only drive forward along the elevated sound insulation cover and there will be no turning or U-turn phenomena, the slope of the fitted straight line with respect to the vehicle's forward direction is controlled within a range with a small inclination, where the slope range is less than 8 and greater than -8, so that the straight line fitted by the target points 201 corresponding to the sound insulation cover can be correctly found.
[0051] Preferably, in step S4, if the distance from the target point 201 to the fitted straight line is less than a set threshold, then the target point 201 can be covered by the fitted straight line. Or it can be said that the target point 201 belongs to the fitted straight line. The optimal straight line is to find a fitted straight line that has the most target points 201 belonging to it.
[0052] Preferably, before executing step S5, a state machine is used to reduce the randomness of straight line fitting. Since the effect of straight line fitting will fluctuate due to the echoes of the vehicle radar, a state machine is used to reduce the randomness of straight line fitting and increase the credibility of straight line fitting.
[0053] Preferably, the state machine uses a counter, and this counter sets a threshold value. The execution steps of the state machine include:
[0054] Step T1, the state machine searches for the optimal straight line among the fitted straight lines. If the optimal straight line is found, the counter is incremented by 1;
[0055] Step T2, determine whether the counter has reached the set threshold value. If not, return to step S1; otherwise, proceed to the next step;
[0056] Step T3, determine the effectiveness of the target point 201.
[0057] Since the effect of linear fitting fluctuates due to the echoes of vehicle radars, a state machine method is adopted to reduce the randomness of linear fitting and increase the credibility of linear fitting. The state machine uses a counting method. If the optimal line can be found at the current time, the counter is incremented by one. After several time periods, when the counter reaches the threshold number of times, it is determined that the vehicle radar is currently in a soundproof enclosure environment, and then the target points 201 that make up the fitted line are valid, and subsequent operations to filter the soundproof enclosure trajectory can be performed. In addition, after filtering the soundproof enclosure trajectory stably and continuously many times, if the state machine fails to find the optimal line occasionally several times, according to the principle of the state machine, the line fitted based on historical information is used to eliminate the soundproof enclosure trajectory. Using the state machine can ensure that the soundproof enclosure trajectory can be stably filtered even in the case of missing data, and can also ensure that it is not filtered in the case of occasional false optimals.
[0058] Preferably, in step S5, the speed mode is obtained by finding target points 201 with the same speed, or by the vehicle speed. If the vehicle speed is known, the speed mode can be set to an absolute speed of zero.
[0059] Preferably, in step S5, during the process of finding the speed mode of target points 201, when the number of target points 201 that meet the speed mode reaches the set quantity, the speed mode is valid. More preferably, the set quantity is 4. It is easy to understand that when obtaining the speed mode, only when the number of target points 201 that meet the speed mode reaches a certain amount, the speed mode is used to select the target points 201 corresponding to the soundproof enclosure echoes, that is, the target points 201 whose speeds are considered to be close to the speed mode are considered to be generated by the soundproof enclosure, and then these target points 201 are deleted to avoid deleting the targets moving normally in front.
[0060] The present invention also provides a device for removing road soundproof enclosures applicable to vehicle radars. Figure 3 It is a schematic structural diagram of a device for removing road soundproof enclosures according to an embodiment of the present invention. As shown in the figure, the removal device 300 mainly includes a search module 301, a linear fitting module 302, a first screening module 303, a second screening module 304, a selected target point module 305, and a deletion module 306.
[0061] Among them, the search module 301 uses the random sample consensus algorithm to find target points 201 within a first set width L1 in front of the vehicle radar.
[0062] The linear fitting module 302 is used to receive the target points 201 found by the search module 301, and traverse every two target points 201 to fit a straight line.
[0063] The first screening module 303 receives the straight line fitted by the straight line fitting module 302, and screens and retains the straight line within the range of a second set distance L2 in front of the vehicle, and the slope of the fitted straight line relative to the forward direction of the vehicle is within a third set ratio range.
[0064] The second screening module 304 receives the straight line screened by the first screening module 303, and selects the fitted straight line that can cover the most target points 201 as the optimal straight line.
[0065] The selected target point module 305 is used to find the speed mode of the target points 201 covered by the optimal straight line, and select the target points 201 corresponding to the sound insulation cover according to the speed mode.
[0066] The deletion module 306 is used to delete the target points 201 selected by the selected target point module 305 on the radar.
[0067] Although the present invention has been described with reference to the current specific embodiments, those of ordinary skill in the art should recognize that the above embodiments are only used to illustrate the present invention, and various equivalent changes or substitutions can be made without departing from the spirit of the present invention. Therefore, as long as the changes and modifications of the above embodiments are within the scope of the spirit of the present invention, they will fall within the scope of the claims of this application.
Claims
1. A method for removing a road sound insulation cover, applicable to vehicle radar, comprising: Step S1, finding a target point, and using the random sample consensus algorithm to find the target point within a first set width range in front of the vehicle; Step S2, fitting a straight line, traversing every two of the target points and fitting them into a straight line; Step S3, performing the first screening on the fitted straight line, screening the fitted straight line within a second set distance range in front of the vehicle, and the slope of the fitted straight line relative to the forward direction of the vehicle being within a third set ratio range; Step S4, performing the second screening on the fitted straight line, traversing the results of the first screening, and selecting the fitted straight line that can cover the most target points as the optimal straight line; Step S5, selecting the target points, finding the speed mode of the target points covered by the optimal straight line, and selecting the target points corresponding to the sound insulation cover according to the speed mode; Step S6, deleting the selected target points.
2. The removal method according to claim 1, characterized in that, In step S1, the first set width is on the central axis perpendicular to the vehicle, and the first set width is 1 m to 2 m.
3. The removal method according to claim 1, characterized in that, In step S3, the second set distance is 10 m to 30 m.
4. The removal method according to claim 1, characterized in that, In step S3, the third set ratio is less than 8 and greater than -8.
5. The removal method according to claim 1, characterized in that, In step S4, if the distance from the target point to the fitted straight line is less than a set threshold, then the target point can be covered by the fitted straight line.
6. The removal method according to claim 1, characterized in that, Before performing step S5, a state machine is used to reduce the randomness of straight line fitting.
7. The removal method according to claim 6, wherein, The state machine uses a counter, and the counter sets a threshold value. The execution steps of the state machine include: Step T1, the state machine searches for the optimal straight line among the fitted straight lines. If the optimal straight line is found, the counter is incremented by 1; Step T2, determining whether the counter has reached the set threshold value. If not, return to step S1; otherwise, proceed to the next step; Step T3, determining the validity of the target point.
8. The removal method according to claim 1, characterized in that, In step S5, the speed mode is obtained by finding the target points with the same speed, or by the vehicle speed.
9. The removal method according to claim 7, characterized in that, In step S5, during the process of finding the speed mode of the target points, if the number of target points meeting the speed mode reaches the set quantity, the obtained speed mode is valid.
10. A device for removing a road sound insulation cover, applicable to vehicle radar, comprising: A search module, using the random sample consensus algorithm to find target points within a first set width range in front of the vehicle radar; A straight line fitting module, receiving the target points found by the search module, traversing every two of the target points and fitting them into a straight line; A first screening module, receiving the straight lines fitted by the straight line fitting module, screening and retaining the straight lines within a second set distance range in front of the vehicle, and the slope of the fitted straight lines relative to the forward direction of the vehicle being within a third set ratio range; A second screening module, receiving the straight lines screened by the first screening module, and selecting the fitted straight line that can cover the most target points as the optimal straight line; Selected target point module, find the speed mode of the target points covered by the optimal straight line, and select the target point corresponding to the sound insulation cover according to the speed mode; Deletion module, delete the target points selected by the selected target point module.
Citation Information
Patent Citations
Urban road element extraction method
CN111062958A
Target detection method and device for vehicle and vehicle-mounted radar
CN112526498A