WWB radar-based sentry mode implementation method and storage medium

Abstract

The invention relates to the technical field of intelligent automobiles, and provides a UWB radar-based sentry mode implementation method and a storage medium, after a vehicle sentry mode is entered, original sampling data is obtained based on a radar function of a UWB chip of a vehicle and based on channel pulse response data of an echo signal, and moving target detection is performed through static clutter filtering, so that a vehicle sentry mode is obtained. Real-time data including a moving target is acquired, a preset threshold value is calculated based on the real-time data, feature extraction is performed to obtain target features, the relative distance between the moving target and a vehicle is determined by combining the preset threshold value after further feature extraction, and then whether the vehicle is awakened to perform remote alarm is judged based on the relative distance, so that the safety of the vehicle is guaranteed. Meanwhile, based on the product characteristics of the UWB radar, the production cost and the operation power consumption of the product are greatly reduced.

Description

Technical Field

[0001] The present invention relates to the field of intelligent automobile technology, and in particular to a sentinel mode implementation method based on UWB radar and a storage medium. Background Art

[0002] Sentry Mode is a safety feature built for smart connected cars based on the vehicle's existing sensors, cameras and other hardware. It helps the driver to obtain real-time information about the vehicle's safety when leaving the vehicle by monitoring the vehicle's surroundings, automatically recording videos, and providing remote alarms. It also automatically collects and records images around the vehicle and sends alarms to the driver's mobile phone. Car owners can use their mobile phones to control the vehicle's horn, flashing lights, and remote voice functions, thus avoiding many losses caused by accidents while the vehicle is parked.

[0003] The current vehicle sentry mode mainly relies on camera technology. Although this method can provide visual monitoring, it also has some significant deficiencies and disadvantages. First, the camera consumes a lot of power when running all the time, which poses a challenge to vehicle battery life and energy management. Second, the camera has a low recognition rate when there is insufficient light or obstructions, which may cause the monitoring function to fail or cause false alarms. In addition, the camera needs to process a large amount of image data in real time, which has high requirements for processor and storage resources, increasing system cost and complexity. Summary of the invention

[0004] The present invention provides a sentinel mode implementation method based on UWB radar and a storage medium, which solves the technical problems that the existing vehicle sentinel mode relies on camera detection, has high cost, low recognition accuracy and high power consumption.

[0005] In order to solve the above technical problems, the present invention provides a method for implementing a sentinel mode based on a UWB radar, comprising:

[0006] Read the channel impulse response data of the ultra-wideband radar echo signal to obtain the original sampling data;

[0007] Calculate a preset threshold based on the original sampled data;

[0008] Performing static clutter filtering on the original sampled data to detect moving targets;

[0009] Acquire real-time data containing the mobile target, and perform feature extraction to obtain target features;

[0010] Calculating the relative distance between the moving target and the vehicle according to the preset threshold and target features;

[0011] Based on the relative distance, it is determined whether to wake up the vehicle for remote alarm.

[0012] After entering the vehicle sentry mode, this basic solution obtains the original sampling data based on the radar function of the vehicle's UWB chip and the channel pulse response data of the echo signal, detects the moving target through static clutter filtering, and further extracts features and determines the relative distance between the moving target and the vehicle based on the preset threshold. Based on the relative distance, it determines whether to wake up the vehicle for remote alarm to ensure vehicle safety. At the same time, based on the product characteristics of UWB radar, it greatly reduces product production costs and operating power consumption.

[0013] In a further embodiment, the channel impulse response data of the ultra-wideband radar echo signal is read to obtain the original sampling data, including:

[0014] Read the channel impulse response data of the ultra-wideband radar echo signal;

[0015] The real part I and the imaginary part Q of each frame of channel impulse response data are obtained, the corresponding data amplitude is calculated, and the corresponding distance resolution is obtained as the original sampling data.

[0016] This solution uses ultra-wideband radar for environmental detection, taking advantage of UWB's high spatial resolution to give it a strong target recognition capability, thereby ensuring the accuracy of target detection.

[0017] In a further embodiment, static clutter filtering is performed on the raw sampled data to detect moving targets, including:

[0018] Obtain two adjacent frames of sampling data from the original sampling data, compare the two frames of sampling data to perform difference calculation, perform static clutter filtering, and detect the moving target from the static clutter background:

[0019] The difference is calculated as follows:

[0020] Diff_data_all(k,n)=abs(data_all(k,n)-data_all(k-1,n))

[0021] In the formula, Diff represents the difference between adjacent elements, data_all is the data amplitude, n is the distance dimension of each frame, data_all(k,n) represents the data amplitude of the kth frame sampling data, and abs() represents the absolute value.

[0022] This scheme subtracts the data of the previous frame from the data of the next frame of the original sampling data. The main purpose is to detect moving targets in a static clutter background. The processed difference data can keenly show the fluctuation of the signal, and use the fluctuation to simply and intuitively determine whether there is a moving target.

[0023] In a further embodiment, the calculating of the preset threshold based on the original sampled data comprises:

[0024] Obtaining N range gate data from each frame of the original sampled data;

[0025] The corresponding tap coordinates are obtained from each range gate data, and the corresponding preset threshold is calculated according to the tap coordinates. The calculation formula of the preset threshold is as follows:

[0026]

[0027] Where i is the tap coordinate, a represents the threshold amplitude, b represents the threshold corner curvature, and c represents the detection setting parameters.

[0028] This solution specifically brings the tap coordinates of each range gate data into the set calculation formula, thereby giving a preset threshold for each range gate data, so as to fit the actual collected data and optimize the detection effect.

[0029] In a further embodiment, feature extraction is performed to obtain target features, including:

[0030] Acquire N range gate data from the real-time data;

[0031] Calculate the preset number of range gate data, obtain the peak and valley values ​​as the first feature, and obtain the standard deviation as the second feature;

[0032] The first feature and the second feature are fused to obtain a target feature.

[0033] This scheme combines standard deviation and peak-to-valley value to measure target characteristics, which can effectively quantify the distribution characteristics and variation range of data, thereby helping to analyze the stability and discreteness of data and improve the recognition accuracy of moving targets.

[0034] In a further embodiment, the relative distance between the moving target and the vehicle is calculated according to the preset threshold and the target feature, including:

[0035] Compare the target feature extracted at the current moment with the corresponding preset threshold, obtain the tap coordinates of all feature values ​​greater than the preset threshold, and obtain a coordinate set;

[0036] Obtaining the tap coordinate with the smallest value from the target coordinate set as the target coordinate of the moving target;

[0037] The relative distance between the moving target and the vehicle is calculated according to the target coordinates.

[0038] In a further embodiment, calculating the relative distance between the moving target and the vehicle according to the target coordinates includes:

[0039] Acquire a corresponding distance resolution from the original sampling data according to the target coordinates;

[0040] The relative distance between the moving target and the vehicle is calculated according to the target coordinates and the distance resolution, and the calculation formula is as follows:

[0041] Distance=SitTap*Rang_reso

[0042] Wherein, Distance is the relative distance, SitTap is the tap coordinate of the moving target, and Rang_reso is the distance resolution.

[0043] This solution compares the characteristic value of the target feature with the preset threshold, and screens out the tap coordinates that are greater than the pre-screening threshold, thereby screening out the tap coordinates with an obvious approaching trend, and further obtains the tap coordinate with the smallest value from the target coordinate set as the coordinate position of the current moving target. Finally, combined with the distance resolution, the actual distance between the moving target and the vehicle can be determined, which has the effect of accurate positioning and low latency.

[0044] In a further embodiment, it also includes:

[0045] Detect whether the vehicle owner has locked the vehicle. If so, control the image acquisition module to enter a shutdown mode, control the vehicle to enter a sentinel mode, and start an ultra-wideband radar to obtain an echo signal.

[0046] This solution starts the ultra-wideband radar to obtain echo signals after entering the sentinel mode, which can achieve ultra-low power consumption environmental detection and high vehicle safety.

[0047] In a further embodiment, determining whether to wake up the vehicle for remote alarm based on the relative distance includes:

[0048] Determine whether the relative distance is greater than the safety distance. If not, determine that there is an abnormal environment, wake up the image acquisition module to collect images to obtain the moving trajectory of the moving target, synchronize the video to the user end for on-site display, and execute remote alarm.

[0049] This solution uses UWB radar to perform environmental detection in the vehicle sentry mode to achieve low-power operation; after determining that there is an abnormal environment, the image acquisition module is started to collect images, which can effectively reduce the standby power consumption of the entire vehicle.

[0050] The present invention provides a storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned method for implementing a sentinel mode based on a UWB radar are implemented. The storage medium may be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 is a system flow chart of a method for implementing a sentinel mode based on a UWB radar provided in an embodiment of the present invention;

[0052] Figure 2 It is a workflow diagram of a method for implementing a sentinel mode based on a UWB radar provided in an embodiment of the present invention;

[0053] Figure 3 It is a curve diagram of the threshold function provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0054] The following specifically illustrates the implementation mode of the present invention in conjunction with the accompanying drawings. The embodiments are provided for illustrative purposes only and cannot be understood as limiting the present invention. The accompanying drawings are provided for reference and illustration only and do not constitute a limitation on the scope of patent protection of the present invention, because many changes can be made to the present invention without departing from the spirit and scope of the present invention.

[0055] Example 1

[0056] The embodiment of the present invention provides a method for implementing a sentinel mode based on a UWB radar, such as Figure 1 , Figure 2 As shown, in this embodiment, it includes:

[0057] S0, detecting whether the vehicle owner has locked the vehicle, and if so, controlling the image acquisition module to enter shutdown mode, controlling the vehicle to enter sentinel mode, and starting the ultra-wideband radar to obtain echo signals.

[0058] In this embodiment, after entering the sentinel mode, the ultra-wideband radar is started to obtain the echo signal, which can achieve ultra-low power consumption environmental detection and high vehicle safety.

[0059] S1. Read the channel impulse response data of the ultra-wideband radar echo signal to obtain the original sampling data, including:

[0060] Read the channel impulse response data of the ultra-wideband radar echo signal;

[0061] The real part I and the imaginary part Q of each frame of channel impulse response data are obtained, the corresponding data amplitude is calculated, and the corresponding distance resolution is obtained as the original sampling data.

[0062] Specifically, the data amplitude data_all is a matrix of size m×n, where m is the number of data frames sampled in the slow time domain, n is the distance dimension of each frame, and the distance resolution is Rang_reso=15 cm.

[0063] This embodiment uses ultra-wideband radar for environmental detection, and uses the high spatial resolution of UWB to enable it to have a strong target recognition capability, thereby ensuring the accuracy of target detection.

[0064] S2. Calculating a preset threshold based on the original sampled data, including:

[0065] Obtaining N range gate data from each frame of the original sampled data;

[0066] The corresponding tap coordinates are obtained from each range gate data, and the corresponding preset threshold is calculated according to the tap coordinates. The formula of the threshold function for calculating the preset threshold is as follows:

[0067]

[0068] Where i is the tap coordinate, a represents the threshold amplitude, b represents the threshold corner curvature, and c represents the detection setting parameters.

[0069] See also Figure 3 , specifically:

[0070] Parameter a affects the overall amplitude of the threshold function, making the threshold function trajectory rise or fall. Appropriate parameter a can improve the accuracy of close-range target detection.

[0071] Parameter b mainly controls the curvature of the threshold function corner, affects the steepness of the threshold function, and controls its trend. The determination of this parameter is related to the UWB data signal-to-noise ratio.

[0072] The parameter c mainly affects the overall up and down translation of the entire threshold function. This parameter is mainly used to detect long-distance moving targets.

[0073] This embodiment specifically brings the tap coordinates of each range gate data into a set calculation formula, thereby providing a preset threshold for each range gate data, thereby being able to fit the actual collected data and optimize the detection effect.

[0074] S3, performing static clutter filtering on the original sampled data to detect moving targets, including:

[0075] Obtain two adjacent frames of sampling data from the original sampling data, compare the two frames of sampling data to perform difference calculation, perform static clutter filtering, and detect the moving target from the static clutter background:

[0076] The difference is calculated as follows:

[0077] Diff_data_all(k,n)=abs(data_all(k,n)-data_all(k-1,n))

[0078] In the formula, Diff represents the difference between adjacent elements, data_all is the data amplitude, n is the distance dimension of each frame, data_all(k,n) represents the data amplitude of the kth frame sampling data, and abs() represents the absolute value.

[0079] This embodiment subtracts the data of the previous frame from the data of the next frame of the original sampled data, and its main purpose is to detect moving targets in a static clutter background. The processed difference data can keenly show the fluctuation of the signal, and use the fluctuation to simply and intuitively determine whether there is a moving target.

[0080] S4, acquiring real-time data containing the mobile target, and performing feature extraction to obtain target features;

[0081] In this embodiment, the real-time data including the moving target is obtained, specifically, the real-time data Diff_data_cur is accumulated with a length of win_curtime_size=25.

[0082] In this embodiment, feature extraction is performed based on the real-time data to obtain target features, including:

[0083] Obtain N range gate data from the real-time data (ie, N=win_curtime_size=25);

[0084] Calculate the preset number of range gate data, obtain the peak and valley values ​​as the first feature, and obtain the standard deviation as the second feature;

[0085] Get the peak and valley values ​​of real-time data as the first feature:

[0086] fea1=max(Diff_data_cur)-min(Diff_data_cur)

[0087] Among them, max() means taking the maximum value, and min() means taking the minimum value.

[0088] Get the standard deviation of the real-time data as the second feature:

[0089] fea2 = std (Diff_data_cur)

[0090] Among them, the standard deviation can be used to express the fluctuation of the moving target; the std() function calculates the standard deviation of function data.

[0091] The target feature is obtained by fusing the first feature and the second feature, as follows:

[0092] Fea_cur=sqrt(fea1*fea2)

[0093] Where sqrt() is used to find the square root of a given value, Fea_cur represents the target feature, fea1 represents the first feature, and fea2 represents the second feature.

[0094] The target feature Fea_cur is used for sentinel trajectory detection.

[0095] This embodiment combines the standard deviation and the peak-to-valley value to measure the target characteristics, which can effectively quantify the distribution characteristics and variation range of the data, thereby helping to analyze the stability and discreteness of the data and improving the recognition accuracy of moving targets.

[0096] S5. Calculating the relative distance between the moving target and the vehicle according to the preset threshold and the target feature, including:

[0097] S51, comparing the target feature extracted at the current moment with the corresponding preset threshold, obtaining the tap coordinates of all feature values ​​greater than the preset threshold, and obtaining a coordinate set, the calculation formula is as follows;

[0098] sit_all = find(Fea_cur>Thr)

[0099] sit_all is the search result, find() function is a function used to find specific elements or texts, Fea_cur is the target feature, and Thr is the preset threshold.

[0100] S52, obtaining the tap coordinate with the smallest value from the target coordinate set as the target coordinate of the moving target, as follows;

[0101] SitTap = min(sit_all)

[0102] Since the UWB radar mode will produce a multipath effect, but the first path is close, find the minimum value of all feature values ​​greater than the threshold coordinate set sit_all, which is the Tap where the current moving target is located.

[0103] S53, calculating the relative distance between the moving target and the vehicle according to the target coordinates, including:

[0104] Acquire a corresponding distance resolution from the original sampling data according to the target coordinates;

[0105] The relative distance between the moving target and the vehicle is calculated according to the target coordinates and the distance resolution, and the calculation formula is as follows:

[0106] Distance=SitTap*Rang_reso

[0107] Wherein, Distance is the relative distance, SitTap is the tap coordinate of the moving target, and Rang_reso is the distance resolution.

[0108] This embodiment compares the characteristic value of the target feature with a preset threshold, screens out the tap coordinates that are greater than the pre-screening threshold, and thus screens out the tap coordinates with an obvious approaching trend, and further obtains the tap coordinate with the smallest value from the target coordinate set as the coordinate position of the current moving target. Finally, combined with the distance resolution, the actual distance between the moving target and the vehicle can be determined, which has the effects of accurate positioning and low latency.

[0109] S6, judging whether to wake up the vehicle for remote alarm based on the relative distance, including:

[0110] Determine whether the relative distance is greater than the safety distance. If not, determine that there is an abnormal environment, wake up the image acquisition module to collect images to obtain the moving trajectory of the moving target, synchronize the video to the user end for on-site display, and execute remote alarm.

[0111] In this embodiment, the UWB radar performs environmental detection in the vehicle sentry mode to achieve low power consumption operation; after determining that there is an abnormal environment, the image acquisition module is started to collect images, which can effectively reduce the standby power consumption of the entire vehicle.

[0112] After entering the vehicle sentry mode, the embodiment of the present invention obtains the original sampling data based on the radar function of the UWB chip of the vehicle itself and the channel impulse response data of the echo signal, performs mobile target detection through static clutter filtering, further performs feature extraction and combines the preset threshold to determine the relative distance between the mobile target and the vehicle, and then determines whether to wake up the vehicle for remote alarm based on the relative distance to ensure vehicle safety. At present, the price of a UWB chip is 4-4.5 US dollars (equivalent to about 30 yuan in RMB), and usually a car needs to be equipped with about 5 UWB chips. The cost price of the UWB solution is about 150 yuan. Based on the product characteristics of the UWB radar, compared with the ADAS vehicle camera module and the in-cabin millimeter wave radar, this embodiment greatly reduces the product production cost and operating power consumption.

[0113] Example 2

[0114] An embodiment of the present invention provides a storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the steps of a method for implementing a sentinel mode based on a UWB radar as described in the above embodiment 1 are implemented. The storage medium may be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0115] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

Claims

1. A method for implementing sentinel mode based on UWB radar, characterized in that: include: Read the channel impulse response data of the ultra-wideband radar echo signal to obtain the original sampling data; Calculate a preset threshold based on the original sampled data; Performing static clutter filtering on the original sampled data to detect moving targets; Acquire real-time data containing the mobile target, and perform feature extraction to obtain target features; Calculating the relative distance between the moving target and the vehicle according to the preset threshold and target features; Based on the relative distance, it is determined whether to wake up the vehicle for remote alarm.

2. A method for implementing sentinel mode based on UWB radar as claimed in claim 1, characterized in that: Read the channel impulse response data of the ultra-wideband radar echo signal to obtain the original sampling data, including: Read the channel impulse response data of the ultra-wideband radar echo signal; The real part I and the imaginary part Q of each frame of channel impulse response data are obtained, the corresponding data amplitude is calculated, and the corresponding distance resolution is obtained as the original sampling data.

3. A method for implementing sentinel mode based on UWB radar as claimed in claim 2, characterized in that: Performing static clutter filtering on the original sampled data to detect moving targets includes: Obtain two adjacent frames of sampling data from the original sampling data, compare the two frames of sampling data to perform difference calculation, perform static clutter filtering, and detect the moving target from the static clutter background: The difference is calculated as follows: Diff_data_all(k,n)=abs(data_all(k,n)-data_all(k-1,n)) In the formula, Diff represents the difference between adjacent elements, data_all is the data amplitude, n is the distance dimension of each frame, data_all(k,n) represents the data amplitude of the kth frame sampling data, and abs() represents the absolute value.

4. A method for implementing sentinel mode based on UWB radar as claimed in claim 1, characterized in that: The calculating a preset threshold based on the original sampled data comprises: Obtaining N range gate data from each frame of the original sampled data; The corresponding tap coordinates are obtained from each range gate data, and the corresponding preset threshold is calculated according to the tap coordinates. The calculation formula of the preset threshold is as follows: Where i is the tap coordinate, a represents the threshold amplitude, b represents the threshold corner curvature, and c represents the detection setting parameters.

5. The method for implementing sentinel mode based on UWB radar as claimed in claim 1, characterized in that: Feature extraction is performed to obtain target features, including: Acquire N range gate data from the real-time data; Calculate the preset number of range gate data, obtain the peak and valley values ​​as the first feature, and obtain the standard deviation as the second feature; The first feature and the second feature are fused to obtain a target feature.

6. A method for implementing sentinel mode based on UWB radar as claimed in claim 1, characterized in that: Calculating the relative distance between the moving target and the vehicle according to the preset threshold and the target feature includes: Compare the target feature extracted at the current moment with the corresponding preset threshold, obtain the tap coordinates of all feature values ​​greater than the preset threshold, and obtain a coordinate set; Obtaining the tap coordinate with the smallest value from the target coordinate set as the target coordinate of the moving target; The relative distance between the moving target and the vehicle is calculated according to the target coordinates.

7. A method for implementing sentinel mode based on UWB radar as claimed in claim 6, characterized in that: Calculating the relative distance between the moving target and the vehicle according to the target coordinates includes: Acquire a corresponding distance resolution from the original sampling data according to the target coordinates; The relative distance between the moving target and the vehicle is calculated according to the target coordinates and the distance resolution, and the calculation formula is as follows: Distance=SitTap*Rang_reso Wherein, Distance is the relative distance, SitTap is the tap coordinate of the moving target, and Rang_reso is the distance resolution.

8. A method for implementing sentinel mode based on UWB radar as claimed in claim 1, characterized in that: Also includes: Detect whether the vehicle owner has locked the vehicle. If so, control the image acquisition module to enter a shutdown mode, control the vehicle to enter a sentinel mode, and start an ultra-wideband radar to obtain an echo signal.

9. A method for implementing sentinel mode based on UWB radar as claimed in claim 1, characterized in that: Determining whether to wake up the vehicle for remote alarm based on the relative distance includes: Determine whether the relative distance is greater than the safety distance. If not, determine that there is an abnormal environment, wake up the image acquisition module to collect images to obtain the moving trajectory of the moving target, synchronize the video to the user end for on-site display, and execute remote alarm.

10. A storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the steps of a method for implementing a sentry mode based on a UWB radar as described in any one of claims 1 to 9 are implemented.

Images