A dynamic calibration method for installation angle of vehicle-mounted millimeter-wave radar

Abstract

The present invention discloses a method for dynamically calibrating the installation angle of an in-vehicle millimeter-wave radar. During the calibration process, the radar does not need to obtain vehicle driving information, nor does it need a specific reference target, and it also does not need the assistance of other sensors. It can completely calculate whether the vehicle driving state meets the calibration conditions through its own data processing. After meeting the calibration conditions, it automatically performs calibration and informs the driver of the calibration result through specific sound and light signals. This method has simple steps, does not rely on a specific calibration reference target, nor on other sensor data, has accurate calibration angles, small calibration errors, is easy to implement in engineering, and is not prone to errors.

Description

Technical Field

[0001] The present invention relates to the field of automotive electronics technology, and particularly to a method for dynamically calibrating the installation angle of an in-vehicle millimeter-wave radar. Background Art

[0002] Safe driving has increasingly become the focus of people's attention. One of the major factors affecting safe driving of automobiles is the blind spot in the driver's field of vision. During the driving process of an automobile, there are large blind spots in the left rear and right rear, resulting in the driver being unable to observe the vehicles in the adjacent lanes through the rearview mirror, thus causing a collision with the vehicles in the adjacent lanes during the lane-changing process, triggering traffic accidents and endangering personal safety. At the same time, it is difficult for the driver to judge the vehicle speed of the adjacent lane vehicles during high-speed driving and unable to correctly predict the collision time, thus triggering traffic accidents during the lane-changing process.

[0003] The millimeter-wave radar with a frequency-modulated continuous-wave system has the characteristics of small size, light weight, high reliability, and all-weather operation. The millimeter-wave radar emits and receives millimeter waves of a certain frequency, and then performs a series of signal processing, so as to obtain the speed, angle, and distance information of the targets within the detection range. The radar can predict in advance whether there is a collision risk between the target vehicle and the vehicle itself based on this information, and warn the driver in advance to avoid traffic accidents.

[0004] The side-rear millimeter-wave angle radar is usually installed in the bumpers at the left and right rear corners of the vehicle. Due to installation errors, the installation errors will cause a large error in the radar's estimation of the target angle. Therefore, calibration is required after the radar is installed. The calibrated radar can compensate for the installation errors, thereby improving the accuracy of angle measurement and driving safety.

[0005] In related technologies, most of them need to add a reference device to achieve the calibration function. For example, the patent with the application number 202010146529.X proposes a method and device for calibrating the azimuth angle installation deviation of an in-vehicle millimeter-wave radar. This method needs to preset corner reflectors at multiple fixed positions and then perform calibration; this method requires a specific calibration site and needs to accurately preset the corner reflectors as a reference, with high requirements for the site; for example, the patent with the application number 202010026381.6 proposes an online calibration method for the external parameters of a millimeter-wave radar. Although this method does not require a specific reference or a specific site, it requires the cooperation of other sensors, with relatively large limitations and difficult to use; for example, the patent with the application number 201710720609.X proposes a calibration method for an automobile and an in-vehicle radar. This method needs to obtain a large amount of vehicle body information, such as vehicle speed, steering wheel angle, and vehicle acceleration, and requires the vehicle to be in a uniform linear driving state, and still needs to place fixed target reference objects for calibration. This method has many limiting conditions and is not suitable for engineering applications.

[0006] Most of the existing angle calibration methods require special sites and equipment support. The site construction is expensive, covers a large area, and is not easy to use. Or simple corner reflector calibration is used, which has low accuracy and large human errors. Summary of the Invention

[0007] The purpose of the present invention is to provide a method for dynamically calibrating the installation angle of a vehicle-mounted millimeter-wave radar to solve the problems raised in the above-mentioned background technology.

[0008] To solve the above technical problems, the present invention provides the following technical solution: A method for dynamically calibrating the installation angle of a vehicle-mounted millimeter-wave radar, the method includes the following steps:

[0009] Step S100: Drive the vehicle straight on the road, and start the millimeter-wave radar calibration by performing vehicle body operations on the vehicle; the starting installation angle automatically set by the millimeter-wave radar is 0 degrees;

[0010] Step S200: The millimeter-wave radar starts to obtain relevant data of the detection target, and the millimeter-wave radar screens and calculates useful data from the relevant data;

[0011] Step S300: The millimeter-wave radar judges the driving state of the vehicle based on the useful data;

[0012] Step S400: Calibrate the driving state of the vehicle; when the driving state meets the calibration requirements, check the quantity of the screened data; when the quantity threshold is met, perform Step S500; when the quantity of the screened data that meets the conditions does not meet the quantity threshold, return to Step S200 until the data quantity meets the quantity threshold. If the time exceeds the limit, end the calibration; when the driving state does not meet the calibration requirements, return to Step S200 and prompt the driver to continue waiting for the conditions that meet the calibration requirements;

[0013] Step S500: Judge the range of the calculated installation angle; when the installation angle is within the preset range, return that the calibration is successful; when the installation angle is not within the preset range, return that the calibration fails;

[0014] This method can realize that the radar performs calculations completely by the program during the dynamic calibration process, without the need for manual participation to set up corresponding reference objects, avoiding the generation of human errors; and this method does not require a specific calibration site or placing specific target benchmarks or target reference objects, nor does it require the cooperation of other sensors, and also does not require the input of vehicle body signals. It completely calibrates the angle through its own data; the advantages of this method are that the requirements for the operator's level, calibration site and equipment are extremely low, it can effectively avoid human errors, and it is easy to implement in engineering.

[0015] Further, the activation of the millimeter-wave radar calibration by performing a vehicle body operation in step S100 includes:

[0016] Step S101: Continuously turn on and off the vehicle's hazard lights and repeat the operation. The number of repetitions meets the number threshold.

[0017] Step S102: Continuously turn on the left turn signal. When the number of times meets the number threshold, activate the left millimeter-wave radar calibration.

[0018] Step S103: Continuously turn on the right turn signal. When the number of times meets the number threshold, activate the right millimeter-wave radar calibration.

[0019] Through the above operation settings, it is possible to activate the radar calibration without expensive equipment such as a diagnostic instrument, and completely through vehicle body operations.

[0020] Further, step S200 includes:

[0021] Step S201: Perform the first screening on the millimeter-wave radar data to remove targets with a speed greater than zero.

[0022] Step S202: Perform coordinate conversion on the remaining millimeter-wave radar data, converting it from polar coordinates to rectangular coordinates.

[0023] Step S203: Calculate the lateral distance disx, longitudinal distance disy, and longitudinal speed vely of the target based on the data obtained in step S202. The formulas are as follows:

[0024] disx = dis * sin(angle * π / 180)

[0025] disy = dis * cos(angle * π / 180)

[0026] vely = vel / sin(angle * π / 180)

[0027] Where dis is the distance to the target detected by the millimeter-wave radar; vel is the radial speed of the target detected by the millimeter-wave radar; angle is the angle of the target detected by the millimeter-wave radar.

[0028] Step S204: Calculate the mutual correlation degree of the vely data in this frame, and retain the m vely data with the highest mutual correlation degree, deleting the remaining data.

[0029] Step S205: Calculate the mutual correlation degree of the disx data in this frame, and retain the n disx data with the highest mutual correlation degree, deleting the remaining data.

[0030] The target data detected by the radar usually contains many false targets. Instead of using the threshold and track method, the data correlation method is used for screening, and the data required for angle calibration can be screened out. The screening method based on data correlation pays more attention to the characteristics of the target than the threshold limit method, which is more conducive to screening out useful information.

[0031] Furthermore, the calculation of the mutual correlation degree of the vely data of this frame and the mutual correlation degree of the disx data of this frame includes:

[0032] Step S211: Obtain the measurement set of the millimeter-wave radar for the target at the t-th frame where i is the serial number of the target in this frame of data, and t is the serial number of this frame;

[0033] Step S212: For all the vely data and disx data in the measurement set sort them respectively according to the sorting method of vely (i,t) <vely (i+1,t) ; disx (i,t) <disx (i+1,t) ; where i is the serial number of the target in this frame of data, and t is the serial number of this frame;

[0034] Step S213: Calculate the deviation values of the sorted vely data and disx data respectively, and the formula is as follows:

[0035] diff1 (i+1,t) = vely (i+1,t) - vely (i,t)

[0036] diff2 (i+1,t) = disx (i+1,t) - disx (i,t)

[0037] where diff1 (i+1,t) represents the deviation value of the vely data of the target with serial number i + 1 and the vely data of the target with serial number i at the t-th frame; diff2 (i+1,t) represents the deviation value of the disx data of the target with serial number i + 1 and the disx data of the target with serial number i at the t-th frame;

[0038] Step 214: Calculate the error thresholds diff vth and diff dth of the vely data and disx data respectively, and the formula is as follows:

[0039]

[0040] Based on the error threshold diffvth Filter the vely data to form a set Based on the error threshold diff dth Filter the disx data to form a set Where i is the serial number of the target in this frame of data, and t is the serial number of this frame; the filtering needs to meet the filtering formula, and the data that does not meet the filtering formula will be deleted. The filtering formulas are: diff1 (i+1,t) <diff vth ; diff2 (i+1,t) <diff dth 。

[0041] Furthermore, step S300 includes:

[0042] Step S301: Estimate the vehicle speed V of this vehicle based on the historical vely data of the detected target; the vehicle speed V needs to meet the vehicle speed calibration formula: |v * 3.6 - 30| < V th ; where V th represents a fixed threshold value; if the vehicle speed V does not meet the vehicle speed calibration formula, delete all target data of this frame;

[0043] Step S302: Estimate the driving slope difference S of this vehicle based on the historical disx data of the detected target; the driving slope difference S needs to meet the driving slope difference calibration formula: S < S th ; where S th represents the driving slope difference threshold; if the driving slope difference S does not meet the driving slope difference calibration formula, delete all target data of this frame;

[0044] The above method can enable the radar to estimate the driving speed and driving direction of the vehicle only based on the target data obtained by itself, without the need for external signals and information from external sensors, greatly reducing the dependence on other components.

[0045] Furthermore, estimating the vehicle speed V in step S301 includes:

[0046] Step S311: Sort the data in the set according to the sorting method of vely (i,t) <vely (i+1,t) and remove the minimum and maximum values in the set;

[0047] Step S312: Calculate the mean square error U1 of the set The formula is as follows:

[0048]

[0049] Step S313: The estimation formula of the vehicle speed V is as follows:

[0050]

[0051] Further, estimating the driving slope difference S of the vehicle in step S302 includes:

[0052] Step 321: Sort the data in the set according to the sorting method of disx (i,t) <disx (i+1,t) and remove the minimum and maximum values in the data;

[0053] Step 322: Calculate the mean square error U2 of the set The formula is as follows:

[0054]

[0055] Step 323: Calculate the most credible distance disx of the vehicle's disx based on the mean square error U2 t , and the formula is as follows:

[0056]

[0057] Step 324: Calculate the slope a of this frame of data t , and the formula is as follows:

[0058]

[0059] where disy t-e represents the longitudinal distance of the (t - e)-th frame of data in the historical data, and disy t represents the longitudinal distance of the data at the t-th frame;

[0060] Step S325: Calculate the relationship between the slope of this frame and the slopes of the previous e frames:

[0061]

[0062] In the formula, is the mean value of the previous e frames of data, and S is the slope coefficient;

[0063] Step S326: Determine whether the vehicle is changing lanes according to the S value. When S ≥ S th , the vehicle is changing lanes. When S < S th , the vehicle is driving straight. Here, S th represents a fixed threshold value; if the vehicle is driving straight, calculate the installation angle, otherwise delete the data;

[0064] In the above method, the radar estimates the vehicle speed and driving state of the vehicle through the data measured by itself. Calibrating around a fixed vehicle speed is beneficial to improving the accuracy of calibration. And by judging whether the vehicle is in a straight driving state and only calibrating when driving straight, the calibration accuracy and calibration success rate are greatly improved.

[0065] Further, the calculation of the installation angle of the vehicle-mounted millimeter-wave radar in step S400 includes:

[0066] Step S401: Calculate the angle value that minimizes the root mean square of the vely error of this frame of data within the physical installation angle range. The calculation method is as follows:

[0067]

[0068] where N r represents the number of data points filtered out that meet the conditions; vel is the radial velocity of the target detected by the millimeter-wave radar; A j ∈[a, b] represents a frame of data that meets the conditions within the angle range [a, b];

[0069] Step S402: Find the minimum value R j in R min . If R min < R th retain the frame of data A j ; if R min > R th delete the frame of data A j ;

[0070] Step S403: When the number of the frame of data A j is greater than the quantity threshold, calculate the mean square error R je of all passing angles; if R je is less than the mean square error threshold, proceed to the next step; if R je is greater than the mean square error threshold, return calibration failure; the mean square error R je formula is as follows:

[0071]

[0072] where W represents the quantity value of the frame of data A j ; represents the average value of the frame of data A j ;

[0073] Step S404: Calculate the installation angle. The formula is as follows:

[0074]

[0075] Step S405: If the Angle meets the preset range, the calibration is returned to success; if the Angle does not meet the preset range, the calibration is returned to failure;

[0076] Using multiple frames of data to calculate the installation angle, we traverse all angles within the conditions to find the most suitable angle value, improve the calibration accuracy, and make the data more consistent with the actual installation angle. After the calculation is completed, two more checks are performed to confirm that the calculated installation angle is correct and in line with expectations, which greatly improves the accuracy and stability of the calibrated angle.

[0077] Furthermore, in step S400, the driver is prompted by sound and light signals; if the radar determines that the vehicle is speeding too fast, too slow, or not traveling in a straight line, different frequencies of beeps are set to prompt the driver to operate in a standardized manner; when the driving conditions are met, the buzzer does not work and the indicator light flashes.

[0078] Further, when the return calibration is successful in step S500, the prescribed indicator light flashing frequency and flashing time are set; when the return calibration fails, the indicator light is set to be always on and the buzzer is set to be off after the buzzer sounds for a full time threshold;

[0079] The above-mentioned prompting method can realize that the operator does not need to hold a diagnostic instrument to observe the calibration status, thus avoiding the use of expensive equipment such as a diagnostic instrument. The driver is directly prompted by sound and light signals, and no observer is required, thus reducing the need for personnel and greatly facilitating the entire calibration method.

[0080] Compared with the prior art, the beneficial effects achieved by the present invention are: the method of the present invention has simple steps, does not need to rely on vehicle body signals, does not need to rely on specific calibration reference targets, and also does not rely on other sensor data, has a small calibration error, and is easy to implement in engineering. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0082] Figure 1 It is a schematic diagram of a module combination of a method for calibrating the installation angle of a vehicle-mounted side and rear angle radar according to the present invention;

[0083] Figure 2 It is a schematic diagram of the installation position of the millimeter wave radar in a method for calibrating the installation angle of a vehicle-mounted side rear angle radar of the present invention, wherein angle a is the installation angle;

[0084] Figure 3 It is a flow chart of a calibration method of an embodiment of a vehicle-mounted side rear angle radar installation angle calibration method of the present invention;

[0085] Figure 4 It is the flowchart of the method for screening useful data in a method for calibrating the installation angle of a vehicle-mounted side-rear corner radar according to the present invention;

[0086] Figure 5 It is the flowchart of the method for the radar to judge the driving state of the vehicle according to the data in a method for calibrating the installation angle of a vehicle-mounted side-rear corner radar according to the present invention;

[0087] Figure 6 It is the flowchart of the method for calculating the installation angle in a method for calibrating the installation angle of a vehicle-mounted side-rear corner radar according to the present invention;

[0088] Figure 7 It is the schematic diagram for judging the calibration process state and calibration result of the embodiment in a method for calibrating the installation angle of a vehicle-mounted side-rear corner radar according to the present invention. Detailed implementation manners

[0089] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0090] Please refer to Figure 1-7 , the present invention provides a technical solution: a method for dynamically calibrating the installation angle of a vehicle-mounted millimeter-wave radar, and the method includes the following steps:

[0091] Step S100: Drive the vehicle to drive straight on the road, and start the millimeter-wave radar calibration by performing vehicle body operations; the starting installation angle automatically set by the millimeter-wave radar is 0 degrees; among them, starting the millimeter-wave radar calibration by performing vehicle body operations includes:

[0092] Step S101: Continuously turn on and off the vehicle's hazard lights, and repeat the operation three times;

[0093] Step S102: Continuously turn on the left turn signal three times to start the calibration of the left-side millimeter-wave radar;

[0094] Step S103: Continuously turn on the right turn signal three times to start the calibration of the right-side millimeter-wave radar;

[0095] Step S200: The millimeter-wave radar starts to obtain relevant data of the detection target, and the millimeter-wave radar screens and calculates useful data from the relevant data; among them, step S200 includes:

[0096] Step S201: Perform the first screening on the millimeter-wave radar data to remove targets with a speed greater than zero;

[0097] Step S202: Perform coordinate transformation on the remaining millimeter-wave radar data, converting it from polar coordinates to rectangular coordinates;

[0098] Step S203: Calculate the lateral distance disx, longitudinal distance disy, and longitudinal velocity vely of the target based on the data obtained in Step S202. The formulas are as follows:

[0099] disx = dis * sin(angle * π / 180)

[0100] disy = dis * cos(angle * π / 180)

[0101] vely = vel / sin(angle * π / 180)

[0102] where dis is the distance to the target detected by the millimeter-wave radar; vel is the radial velocity of the target detected by the millimeter-wave radar; angle is the target angle detected by the millimeter-wave radar;

[0103] Step S204: Calculate the mutual correlation degree of the vely data in this frame, and retain the m vely data with the highest mutual correlation degree, deleting the remaining data;

[0104] Step S205: Calculate the mutual correlation degree of the disx data in this frame, and retain the n disx data with the highest mutual correlation degree, deleting the remaining data;

[0105] The calculation of the mutual correlation degree of the vely data in this frame and the mutual correlation degree of the disx data in this frame includes:

[0106] Step S211: Obtain the measurement set of the millimeter-wave radar for the target at the t-th frame where i is the serial number of the target in the data of this frame, and t is the serial number of this frame;

[0107] Step S212: For all the vely data and disx data in the measurement set sort them respectively according to the sorting methods of vely (i,t) <vely (i+1,t) ; disx (i,t) <disx (i+1,t) where i is the serial number of the target in the data of this frame, and t is the serial number of this frame;

[0108] Step S213: Calculate the deviation values of the sorted vely data and disx data respectively. The formulas are as follows:

[0109] diff1 (i+1,t) = vely (i+1,t) - vely(i,t)

[0110] diff2 (i+1,t) = disx (i+1,t) -disx (i,t)

[0111] Among them, diff1 (i+1,t) represents the deviation value of the vely data of the target with target serial number i + 1 and the vely data of the target with target serial number i at frame t; diff2 (i+1,t) represents the deviation value of the disx data of the target with target serial number i + 1 and the disx data of the target with target serial number i at frame t;

[0112] Step 214: Calculate the error thresholds diff vth and diff dth of the vely data and the disx data respectively, and the formula is as follows:

[0113]

[0114] Based on the error threshold diff vth screen the vely data to form a set Based on the error threshold diff dth screen the disx data to form a set Among them, i is the serial number of the target in the current frame data, and t is the serial number of the current frame; the screening needs to meet the screening formula, and the data that does not meet the screening formula is deleted. The screening formulas are respectively: diff1 (i+1,t) < diff vth ; diff2 (i+1,t) < diff dth ;

[0115] Step S300: The millimeter-wave radar judges the driving state of the vehicle based on the useful data; among them, step S300 includes:

[0116] Step S301: Estimate the vehicle speed V of the vehicle based on the historical vely data of the detected target; among them, estimating the vehicle speed V of the vehicle includes:

[0117] Step S311: Sort the data in the set according to the sorting method of vely (i,t) < vely (i+1,t) and remove the minimum value and the maximum value in the set;

[0118] Step S312: Calculate the mean square error U1 of the set , and the formula is as follows:

[0119]

[0120] Step S313: The estimation formula for vehicle speed V is as follows:

[0121]

[0122] The vehicle speed V needs to satisfy the vehicle speed calibration formula: |v * 3.6 - 30| < V th ; where V th represents a fixed threshold value, and V th = 5; if the vehicle speed V does not satisfy the vehicle speed calibration formula, delete all target data in this frame;

[0123] Step S302: Estimate the driving slope difference S of the vehicle based on the historical disx data of the detected target; among them, estimating the driving slope difference S of the vehicle includes:

[0124] Step 321: Sort the data in the said set in ascending order according to disx (i,t) <disx (i+1,t) and remove the minimum and maximum values from the data;

[0125] Step 322: Calculate the mean square error U2 of the said set using the following formula:

[0126]

[0127] Step 323: Calculate the most credible distance disx of the vehicle's disx based on the mean square error U2 t , using the following formula:

[0128]

[0129] Step 324: Calculate the slope a of this frame of data t , using the following formula:

[0130]

[0131] where disy t-e represents the longitudinal distance of the data in the (t - e)th frame in the historical data, and disy t represents the longitudinal distance of the data at time t;

[0132] Step S325: Calculate the relationship between the slope of this frame and the slopes of the previous 10 frames:

[0133]

[0134] In the formula, n = 10, is the mean value of the previous 10 frames of data, and S is the slope coefficient;

[0135] Step S326: Determine whether the vehicle is changing lanes based on the S value, S≥S th When the vehicle is changing lanes, S th When the vehicle is in a straight line, S th Represents a fixed threshold value, S th =0.6; if the vehicle is in a straight line, calculate the installation angle, otherwise delete the data. The driving slope difference S must satisfy the driving slope difference calibration formula: S th ; Among them, S th Indicates the driving slope difference threshold, S th =0.6; if the driving slope difference S does not satisfy the driving slope difference calibration formula, delete all target data of this frame;

[0136] Step S400: Calibrate the driving state of the car; when the driving state meets the calibration requirements, the number of filtered data N r Check the situation; meet N r >3, proceed to step S500; when the number of data filtered out that meets the conditions does not meet N r >3, return to step S200 until the number of data meets N r >3, if the time is over, the calibration is terminated; when the driving state does not meet the calibration requirements, the process returns to step S200 and prompts the driver, and continues to wait for the calibration requirements to be met; the prompting method for the driver is sound and light signals; if the radar determines that the vehicle is driving too fast, the buzzer will beep at a frequency of 50HZ; if the radar determines that the vehicle is driving too slowly, the buzzer will beep at a frequency of 10HZ; if the radar determines that the vehicle is not driving in a straight line, the buzzer will beep alternately for a long and short time to prompt the driver to operate in a standardized manner. When the driving conditions are met, the buzzer will not work and the indicator light will flash at a frequency of 50HZ;

[0137] Among them, the calculation of the installation angle of the vehicle-mounted millimeter-wave radar includes:

[0138] Step S401: Calculate the angle value that minimizes the root mean square error of the current frame data within the physical installation angle range. The calculation method is as follows:

[0139]

[0140] Among them, N r Indicates the number of data that meet the conditions and are filtered out; vel is the radial velocity of the target detected by the millimeter-wave radar; A j ∈[20,50] represents the frame data that meets the conditions in the angle interval [20, 50];

[0141] Step S402: Find R j The minimum value R min , if R min <R​​th Reserve the data A of this frame j ; If R min >

[0142] R th Delete the data A of this frame j ; Where R th is 100;

[0143] Step S403: When the quantity of the data A of this frame j satisfies 300, calculate the mean square error R of all passing angles je ; If R je is less than 50, proceed to the next step; if R je is greater than 50, return calibration failure; The mean square error R je The formula is as follows:

[0144]

[0145] Where W represents the quantity value of the data A of this frame j ; represents the average value of the data A of this frame j ;

[0146] Step S404: Calculate the installation angle, and the formula is as follows:

[0147]

[0148] Step S405: If Angle satisfies |Angle a -35| < 5, return calibration success; if Angle does not satisfy |Angle a -35| < 5, return calibration failure;

[0149] Step S500: Judge the range of the calculated installation angle; When the installation angle is within the preset range, return calibration success; When the installation angle is not within the preset range, return calibration failure; When returning calibration success, the indicator light flashes at a frequency of 10HZ for 5 seconds and then goes out; When returning calibration failure, the indicator light is on constantly and the buzzer sounds for 5 seconds and then goes out.

[0150] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "including", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device.

[0151] Finally, it should be noted that the above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for dynamically calibrating the installation angle of a vehicle-mounted millimeter-wave radar, characterized in that, The method includes the following steps: Step S100: Drive the vehicle straight on the road and start millimeter-wave radar calibration by performing vehicle body operations on the vehicle; the starting installation angle automatically set by the millimeter-wave radar is 0 degrees; Step S200: The millimeter-wave radar starts to obtain relevant data of the detection target, and the millimeter-wave radar screens and calculates useful data from the relevant data; Step S300: The millimeter-wave radar judges the driving state of the vehicle based on the useful data; Step S400: Calibrate the driving state of the vehicle; when the driving state meets the calibration requirements, check the number of screened data; when the number meets the quantity threshold, proceed to step S500; when the number of screened data that meets the conditions does not meet the quantity threshold, return to step S200 until the data quantity meets the quantity threshold, and if the time expires, end the calibration; when the driving state does not meet the calibration requirements, return to step S200 and prompt the driver to continue waiting for the conditions that meet the calibration requirements; Step S500: Judge the range of the calculated installation angle; when the installation angle is within the preset range, return that the calibration is successful; when the installation angle is not within the preset range, return that the calibration fails; The said step S300 includes: Step S301: Set the longitudinal speed of the target detected by the millimeter-wave radar as vely, and estimate the vehicle speed V based on the historical vely data of the detected target; the vehicle speed V needs to satisfy the vehicle speed calibration formula: |V * 3.6 - 30| < V th ; where V th represents a fixed threshold value; if the vehicle speed V does not satisfy the vehicle speed calibration formula, delete all target data of this frame; Step S302: Estimate the vehicle driving slope difference S based on the historical disx data of the detection target; the driving slope difference S needs to satisfy the driving slope difference calibration formula: S < S th ; where S th represents the driving slope difference threshold; if the driving slope difference S does not satisfy the driving slope difference calibration formula, delete all target data of this frame; In the said step S301, estimating the vehicle speed V of the vehicle includes: Step S311: Calculate the error thresholds diff of the vely data and the disx data respectively vth and diff dth , and based on the error threshold diff vth screen the vely data to form a set Based on the error threshold diff dth screen the disx data to form a set Sort the data in the set according to the sorting method of vely (i,t) <vely (i+1,t) , and remove the minimum and maximum values in the set; Step S312: For the set calculate the mean square error U1, and the formula is as follows: Step S313: The estimation formula of the vehicle speed V is as follows: In the said step S302, estimating the driving slope difference S of the vehicle includes: Step 321: Take the set and sort the data in it according to the sorting method of disx (i,t) <disx (i+1,t) , and remove the minimum and maximum values from the data; Step 322: For the said set Calculate the mean square error U2, and the formula is as follows: Step 323: Obtain the most credible distance disx of the vehicle based on the mean square error U2 t , and the formula is as follows: Step 324: Calculate the slope a of the data of this frame t , and the formula is as follows: Among them, disy t-e represents the vertical distance of the (t - e)-th frame data in the historical data, and disy t represents the vertical distance of the data at the t-th frame; Step S325: Calculate the relationship between the slope of this frame and the slope of the previous e frames: Wherein, is the mean value of the first e-frame data, and S is the slope coefficient; Step S326: Determine whether the vehicle is changing lanes according to the S value. When S≥S, the vehicle is changing lanes; when S<S, the vehicle is driving straight, where S represents a fixed threshold value. If the vehicle is driving straight, calculate the installation angle; otherwise, delete the data. th When the vehicle is changing lanes, when S<S th the vehicle is driving straight, where S th represents a fixed threshold value; if the vehicle is driving straight, calculate the installation angle, otherwise delete the data.

2. The method for dynamically calibrating the installation angle of a vehicle-mounted millimeter-wave radar according to claim 1, characterized in that, In the said step S100, starting the millimeter-wave radar calibration by performing vehicle body operations on the vehicle includes: Step S101: Continuously turn on and off the vehicle's hazard lights, and repeat the operation. The number of repetitions meets the number threshold; Step S102: Continuously turn on the left turn signal for a number of times that meets the threshold to turn on the left millimeter-wave radar calibration; Step S103: Continuously turn on the right turn signal for a number of times that meets the threshold to turn on the right millimeter-wave radar calibration.

3. The method for dynamically calibrating the installation angle of a vehicle-mounted millimeter-wave radar according to claim 1, characterized in that, The said step S200 includes: Step S201: Perform the first screening on the millimeter-wave radar data to remove targets with a speed greater than zero; Step S202: Perform coordinate transformation on the remaining millimeter-wave radar data, and convert it from the polar coordinate system to the rectangular coordinate system; Step S203: Calculate the lateral distance disx, longitudinal distance disy, and longitudinal speed vely of the target according to the data obtained in step S202. The formulas are as follows: disx = dis * sin(angle * π / 180) disy = dis * cos(angle * π / 180) vely = vel / sin(angle * π / 180) where dis is the distance to the target detected by the millimeter-wave radar; vel is the radial speed of the target detected by the millimeter-wave radar; angle is the target angle detected by the millimeter-wave radar; Step S204: Calculate the correlation degree of the vely data of this frame, and retain the m vely data with the highest correlation degree, and delete the remaining data; Step S205: Calculate the mutual correlation degree of the disx data of this frame, retain the n disx data with the highest mutual correlation degree, and delete the remaining data.

4. A method for dynamically calibrating the installation angle of an in-vehicle millimeter-wave radar according to claim 3, characterized in that, The calculation of the mutual correlation degree of the vely data of this frame and the mutual correlation degree of the disx data of this frame includes: Step S211: Obtain the measurement set of the millimeter-wave radar for the target at the t-th frame where i is the serial number of the target in the data of this frame, and t is the serial number of this frame; Step S212: For the measurement set all the vely data and disx data therein are sorted respectively according to the sorting method of vely (i,t) <vely (i+1,t) ; disx (i,t) <disx (i+1,t) ; where i is the serial number of the target in this frame of data, and t is the serial number of this frame; Step S213: Calculate the deviation values of the sorted vely data and disx data respectively. The formula is as follows: diff1 (i+1,t) = positively (i+1,t) = negatively (i,t) diff2 (i+1,t) = disx (i+1,t) -disx (i,t) Among them, diff1 (i+1,t) represents the deviation value between the vely data of the target with target sequence number i + 1 and the vely data of the target with target sequence number i at time t; diff2 (i+1,t) represents the deviation value between the disx data of the target with target sequence number i + 1 and the disx data of the target with target sequence number i at time t; Step 214: Calculate the error thresholds diff of vely data and disx data respectively vth and diff dth , as shown in the following formula: Based on the error threshold diff vth Filter the vely data to form a set Based on the error threshold diff dth Filter the disx data to form a set where i is the serial number of the target in this frame of data, and t is the serial number of this frame; the filtering needs to satisfy the filtering formula, and the data that does not satisfy the filtering formula is deleted. The filtering formulas are respectively: diff1 (i+1,t) <diff vth ; diff2 (i+1,t) <diff dth .

5. A method for dynamically calibrating the installation angle of an in-vehicle millimeter-wave radar according to claim 1, characterized in that, The calculation of the installation angle of the vehicle-mounted millimeter-wave radar in the step S400 includes: Step S401: Calculate the angle value that minimizes the root mean square of the vely error of this frame of data within the physical installation angle range. The calculation method is as follows: Among them, N r represents the number of data filtered out that meet the conditions; vel is the radial velocity of the target detected by the millimeter-wave radar; A j ∈[a, b] means a piece of data in this frame that meets the conditions within the angular range [a, b]; Step S402: Find R j the minimum value R in min , if R min < R th retain the data of this frame A j ; if R min > R th delete the data of this frame A j ; Step S403: When the quantity of the current frame of data A j is greater than the quantity threshold, calculate the mean square error R of all passing angles je ; if R je is less than the mean square error threshold, proceed to the next step; if R je is greater than the mean square error threshold, return calibration failure; the formula for the mean square error R je is as follows: Among them, W represents the quantity value of the data A of this frame j ; represents the average value of the data A of this frame j ; Step S404: Calculate the installation angle. The formula is as follows: Step S405: If Angle meets the preset range, return calibration success; if Angle does not meet the preset range, return calibration failure.

6. A method for dynamically calibrating the installation angle of an in-vehicle millimeter-wave radar according to claim 1, characterized in that, The method of prompting the driver in the step S400 is an acoustic and optical signal; different frequencies of beeping are set for the radar to judge that the vehicle speed is too fast, the vehicle speed is too slow, and the vehicle is not driving straight to prompt the driver to operate in a standardized manner; when the driving conditions are met, the buzzer does not work and the indicator light flashes.

7. A method for dynamically calibrating the installation angle of an in-vehicle millimeter-wave radar according to claim 1, characterized in that, When calibration success is returned in the step S500, set the specified flashing frequency and flashing time of the indicator light; when calibration failure is returned, set the indicator light to be always on and the buzzer to sound continuously for the full time threshold and then go out.

Images