Attitude angle estimation method and anti-shake method of electronic rearview mirror, electronic rearview mirror and vehicle

Abstract

The invention provides an attitude angle estimation method and an anti-shake method of an electronic rearview mirror, the electronic rearview mirror and a vehicle, and relates to the technical field of electronic rearview mirrors. The motion data of the electronic rearview mirror is acquired, multiple rounds of filtering processing are performed on the motion data, the weight is calculated according to the residual error between the current round of filtering value and the last round of estimation value in each processing round, and weighted fusion is performed by combining the filtering values of multiple rounds and the corresponding weight, so that the motion data of the electronic rearview mirror can be obtained. According to the method, the final estimation value of the motion data is obtained, so that the final estimation value after multiple rounds of processing is more accurate, the attitude angle of the electronic rearview mirror is determined based on the final estimation value, the influence of abnormal data on the estimation result can be effectively inhibited, and the precision and stability of attitude angle estimation are improved.

Description

Technical Field

[0001] The present application relates to the technical field of electronic rearview mirrors, and in particular to a posture angle estimation method and an anti-shake method of an electronic rearview mirror, an electronic rearview mirror, and a vehicle. Background Art

[0002] In recent years, with the increasing intelligence of vehicles, electronic rearview mirrors have become widely used in vehicle safety assistance systems as an alternative to traditional optical rearview mirrors. Electronic rearview mirrors use cameras to capture real-time images of the vehicle's rear environment and display them on the vehicle's display. Compared to traditional optical rearview mirrors, they provide a clearer and more stable field of view in complex weather and lighting conditions.

[0003] However, during vehicle driving, especially in dynamic driving scenarios such as bumpy roads, steering and acceleration, the electronic rearview mirror body is prone to vibration or displacement, causing the captured image to shake, which in turn affects the driver's observation accuracy of the rear environment and poses a potential safety hazard.

[0004] Existing technologies typically process the motion data of electronic rearview mirrors through single filtering or simple integration to determine attitude angles. However, these methods suffer from issues such as insufficient filtering accuracy, large estimation errors, and sensitivity to abnormal data, making them difficult to meet the demands of applications in complex working conditions.

[0005] Therefore, this application proposes a new attitude angle estimation method. Summary of the Invention

[0006] In view of the above problems, embodiments of the present application provide an attitude angle estimation method, an anti-shake method, an electronic rearview mirror, and a vehicle for an electronic rearview mirror, so as to overcome the above problems or at least partially solve the above problems.

[0007] In a first aspect of an embodiment of the present application, a method for estimating an attitude angle of an electronic rearview mirror is provided, the method comprising: Obtain motion data of electronic rearview mirror; Performing multiple rounds of filtering on the motion data to obtain a filtered value of the motion data in each round; For each of the multiple rounds, calculating a residual of the motion data in the round based on a filtered value of the motion data in the round and an estimated value of the motion data in the previous round; Determining weights corresponding to filtered values ​​of the motion data in the multiple rounds according to residuals of the motion data in the multiple rounds; Obtaining a final estimated value of the motion data according to filtered values ​​of the motion data in multiple rounds and corresponding weights; The attitude angle of the electronic rearview mirror is obtained according to the final estimated value of the motion data.

[0008] Optionally, the number of the multiple rounds is K, where K is an integer greater than 1; for the kth round of the K rounds, the estimated value of the motion data in the kth round is determined according to the following steps: Calculating a residual of the motion data at the kth round based on a filtered value of the motion data at the kth round and an estimated value of the motion data at the k−1th round; Determining a weight corresponding to a filtered value of the motion data in the kth round based on a residual of the motion data in the kth round; Based on the filtered values ​​of the motion data in the 1st to kth rounds, and the respective weights of the filtered values ​​of the motion data in the 1st to kth rounds, an estimated value of the motion data in the kth round is obtained to determine the residual of the motion data in the k+1th round; wherein k is an integer between 2 and K.

[0009] Optionally, the number of the multiple rounds is K, where K is an integer greater than 1; and determining, based on the residuals of the motion data in the multiple rounds, weights corresponding to the filtered values ​​of the motion data in the multiple rounds includes: Determine a magnitude relationship between a residual of the motion data in the kth round and a robust threshold value among the multiple rounds, where k is an integer between 2 and K; When the residual of the motion data in the kth round among the multiple rounds is less than or equal to the robust threshold, assigning a weight corresponding to the filtered value of the motion data in the kth round to 1; When the residual of the motion data in the kth round among multiple rounds is greater than the robust threshold, the weight corresponding to the filtered value of the motion data in the kth round is determined as the ratio of the robust threshold to the residual of the kth round.

[0010] Optionally, the robust threshold is determined according to the following steps, including: determining noise characteristics of the motion data based on the motion data of the electronic rearview mirror; When the noise characteristic belongs to the first type of noise characteristic, the robust threshold is determined to be the first threshold based on a first mapping relationship between the noise characteristic and the robust threshold; when the noise characteristic belongs to the second type of noise characteristic, the robust threshold is determined to be the second threshold based on a second mapping relationship between the noise characteristic and the robust threshold; wherein the first threshold is greater than the second threshold, and the noise amplitude of the first type of noise characteristic is greater than the noise amplitude of the second type of noise characteristic.

[0011] Optionally, the method further includes: Determine the difference between the estimated values ​​of the motion data obtained in each two adjacent rounds; Obtaining a final estimated value of the motion data according to the filtered values ​​of the motion data in multiple rounds and corresponding weights includes: When the difference between the estimated value in the kth round and the estimated value in the k-1th round is less than the target threshold, or when the number of rounds reaches K, the final estimated value of the motion data is obtained based on the filtered values ​​of the motion data in K rounds and the corresponding weights; wherein K is an integer greater than 1, and k is an integer between 2 and K.

[0012] Optionally, the motion data includes angular velocity and linear acceleration, and the attitude angle includes angular velocity attitude angle and linear acceleration attitude angle. After obtaining the attitude angle of the electronic rearview mirror, the method further includes: determining a current driving scene of the electronic rearview mirror based on the motion data; Determining a target filtering weight based on a mapping relationship between different driving scenarios and filtering weights according to the current driving scenario of the electronic rearview mirror; Assigning the target filter weight to the angular velocity attitude angle, and assigning a weight obtained by subtracting the target filter weight from 1 to the linear acceleration attitude angle; The angular velocity attitude angle and the linear acceleration attitude angle are summed in a weighted manner to obtain a fusion attitude angle.

[0013] Optionally, determining the target filtering weight according to the current driving scene of the electronic rearview mirror and based on a mapping relationship between different driving scenes and filtering weights includes: When the driving scene in which the electronic rearview mirror is located is the first type of driving scene, determining the target filtering weight to be the first target filtering weight based on a third mapping relationship between the driving scene and the filtering weight; When the driving scene in which the electronic rearview mirror is located is a second type of driving scene, based on the fourth mapping relationship between the driving scene and the filtering weight, the target filtering weight is determined to be the second target filtering weight, the first target filtering weight is greater than the second target filtering weight, and the dynamic change speed of the first type of driving scene is faster than the dynamic change speed of the second type of driving scene.

[0014] In a second aspect of the present application, a method for anti-shake of an electronic rearview mirror is provided, the method comprising: Determine the attitude angle of the electronic rearview mirror according to the method described in the first aspect of the present application based on the motion data of the electronic rearview mirror; determining a jitter parameter value of the electronic rearview mirror according to the attitude angle of the electronic rearview mirror; A jitter compensation value is determined according to the jitter parameter value, so as to control the vibration unit of the electronic rearview mirror to perform jitter compensation based on the jitter compensation value.

[0015] In a third aspect of the present application, an electronic rearview mirror is provided, which is used to execute the steps of the electronic rearview mirror attitude angle estimation method as described in the first aspect of the present application, or the electronic rearview mirror is used to execute the steps of the electronic rearview mirror anti-shake method as described in the second aspect of the present application.

[0016] In a fourth aspect of the present application, a vehicle is provided, comprising the electronic rearview mirror as described in the third aspect of the present application.

[0017] Beneficial effects of this application: The present application proposes a method for estimating the attitude angle of an electronic rearview mirror, the method comprising: acquiring motion data of the electronic rearview mirror; performing multiple rounds of filtering processing on the motion data to obtain a filtered value of the motion data in each round; for each of the multiple rounds, calculating the residual of the motion data in the round based on the filtered value of the motion data in the round and the estimated value of the motion data in the previous round; determining the weights corresponding to the filtered values ​​of the motion data in the multiple rounds based on the residuals of the motion data in the multiple rounds; obtaining a final estimated value of the motion data based on the filtered values ​​of the motion data in the multiple rounds and the corresponding weights; and obtaining a final estimated value of the motion data based on the final estimated value of the motion data.

[0018] This application obtains the motion data of the electronic rearview mirror, performs multiple rounds of filtering processing on the motion data, and in each processing round, calculates the weight according to the residual between the current round filter value and the previous round estimation value, combines the filter values ​​of multiple rounds and the corresponding weights for weighted fusion to obtain the final estimated value of the motion data, so that the final estimated value after multiple rounds of processing is more accurate, and determines the attitude angle of the electronic rearview mirror based on the final estimated value, thereby effectively suppressing the influence of abnormal data on the estimation result, and improving the accuracy and stability of attitude angle estimation. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0020] Figure 1 This is a schematic flow chart of the steps of a method for estimating the attitude angle of an electronic rearview mirror provided in an embodiment of the present application; Figure 2This is a schematic flow chart of the steps of a method for determining a fusion posture angle provided in an embodiment of the present application; Figure 3 This is a schematic flow chart of the steps of an anti-shake method for an electronic rearview mirror provided in an embodiment of the present application; Figure 4 This is a structural block diagram of an electronic rearview mirror provided in an embodiment of the present application. DETAILED DESCRIPTION

[0021] The exemplary embodiments of the present application will be described in more detail below in conjunction with the accompanying drawings in the embodiments of the present application. Although the accompanying drawings show exemplary embodiments of the present application, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present application and to fully convey the scope of the present application to those skilled in the art.

[0022] In the first aspect of the present application, a method for estimating the attitude angle of an electronic rearview mirror is provided, such as Figure 1 As shown, the method includes: Step S101, obtaining motion data of the electronic rearview mirror.

[0023] In this step, motion data is collected by a motion detection module installed on the electronic rearview mirror. The motion detection module can be a high-precision gyroscope and an accelerometer. The high-precision gyroscope outputs the angular velocity of the electronic rearview mirror, and the accelerometer synchronously outputs the linear acceleration of the electronic rearview mirror. The angular velocity of the electronic rearview mirror includes three-axis angular velocity, such as: x-axis angular velocity , y-axis angular velocity , and the z-axis angular velocity , the linear acceleration of the electronic rearview mirror also includes three-axis linear acceleration, such as: x-axis linear acceleration , y-axis linear acceleration and z-axis linear acceleration The motion data is used to characterize the motion state of the electronic rearview mirror, specifically, the motion state of the vehicle on which the electronic rearview mirror is installed. It should be noted that for motion data, such as: x-axis angular velocity , y-axis angular velocity , and the z-axis angular velocity , and, x-axis linear acceleration , y-axis linear acceleration and z-axis linear acceleration Therefore, the processing of motion data mentioned below in this application refers to the processing of all motion data, rather than specifically the processing of a certain type of motion data.

[0024] Step S102 : performing multiple rounds of filtering on the motion data to obtain a filtering value of the motion data in each round.

[0025] In this step, the acquired motion data is subjected to multiple rounds of filtering, and each round may adopt the same or different filtering methods. In this application, the filtering method is not limited. In some cases, it may be mean filtering, median filtering, Kalman filtering, adaptive filtering, etc., so as to obtain the filtered values ​​of the motion data under different filtering rounds to reduce the impact of noise on the motion data. This application adopts first-order low-pass filtering for angular velocity and first-order high-pass filtering for linear acceleration.

[0026] Step S103 : For each of the multiple rounds, the residual of the motion data in the round is calculated based on the filtered value of the motion data in the round and the estimated value of the motion data in the previous round.

[0027] In this step, for each round of filtering results, the residual of the round is calculated based on the difference between its filtered value and the estimated value of the previous round.

[0028] Step S104 : determining weights corresponding to the filter values ​​of the motion data in multiple rounds according to the residuals of the motion data in multiple rounds.

[0029] In this step, the weight corresponding to each filter value of each round is calculated based on the residual of each round calculated in step S103. Generally, the smaller the residual, the larger the weight corresponding to the filter value, and the larger the residual, the smaller the weight corresponding to the filter value, so as to enhance the reliable filtering results and suppress abnormal data.

[0030] Step S105 : obtaining a final estimated value of the motion data according to the filtered values ​​of the motion data in multiple rounds and the corresponding weights.

[0031] In this step, based on the filtering values ​​of each round and their weights, weighted fusion processing is performed to integrate the effective information of the filtering results of each round to generate the final estimated value of the motion data, thereby improving the accuracy and robustness of the motion data estimation.

[0032] Step S106 , obtaining the attitude angle of the electronic rearview mirror according to the final estimated value of the motion data.

[0033] In this step, based on the final estimated value obtained in step S105, the attitude angle information of the electronic rearview mirror is calculated to provide a data basis for the subsequent implementation of anti-shake of the electronic rearview mirror.

[0034] This application obtains the motion data of the electronic rearview mirror, performs multiple rounds of filtering processing on the motion data, and in each processing round, calculates the weight according to the residual between the current round filter value and the previous round estimation value, combines the filter values ​​of multiple rounds and the corresponding weights for weighted fusion to obtain the final estimated value of the motion data, so that the final estimated value after multiple rounds of processing is more accurate, and determines the attitude angle of the electronic rearview mirror based on the final estimated value, thereby effectively suppressing the influence of abnormal data on the estimation result and improving the accuracy and stability of attitude angle estimation.

[0035] In one embodiment, the number of the multiple rounds is K, where K is an integer greater than 1; for the kth round of the K rounds, the estimated value of the motion data in the kth round is determined according to the following steps: Calculating a residual of the motion data at the kth round based on a filtered value of the motion data at the kth round and an estimated value of the motion data at the k−1th round; Determining a weight corresponding to a filtered value of the motion data in the kth round based on a residual of the motion data in the kth round; Based on the filtered values ​​of the motion data in the 1st to kth rounds, and the respective weights of the filtered values ​​of the motion data in the 1st to kth rounds, an estimated value of the motion data in the kth round is obtained to determine the residual of the motion data in the k+1th round; wherein k is an integer between 2 and K.

[0036] In this embodiment, the number of the multiple rounds is K, where K is an integer greater than 1. For the kth round of the K rounds, the estimated value of the motion data in the kth round is determined according to the following steps, where k is an integer between 2 and K: First, the residual of the motion data at round k is calculated based on the filtered value of the motion data at round k and the estimated value of the motion data at round k−1. Specifically, the residual is used to represent the degree of deviation of the filtered value of the current round relative to the estimated value of the previous round, which is used for subsequent weighting processing.

[0037] Secondly, based on the residual of the motion data at round k, the weight corresponding to the filtered value of the motion data at round k is determined. Generally, filter values ​​with smaller residuals are given higher weights to enhance the influence of reliable data; filter values ​​with larger residuals are given lower weights to suppress the interference of abnormal data.

[0038] Then, a weighted fusion process is performed based on the filtered values ​​of the motion data from rounds 1 to k, and their corresponding weights, to obtain the estimated value of the motion data at round k. Specifically, the filtered values ​​from rounds 1 to k are weighted averaged according to their corresponding weights to generate the estimated value of round k, which serves as the basis for the subsequent calculation of the residual of round k+1.

[0039] In this embodiment, historical filtering information and current filtering results can be dynamically fused in each round to improve the stability and accuracy of motion data estimation, thereby further improving the reliability of electronic rearview mirror attitude angle estimation.

[0040] In this embodiment, the residual of the motion data at the kth round satisfies the following formula: (1) in, represents the residual of the motion data at the kth round; represents the filtered value of motion data at the kth round; Represents the estimated value of the motion data at the k-1th round.

[0041] In this embodiment, the estimated value of the motion data in the kth round satisfies the following formula: (2) in, represents the estimated value of the motion data at the kth round; Represents the weight corresponding to the filtered value of the motion data in the i-th round; represents the filtered value of motion data in round i, where .

[0042] In one embodiment, for the case where k=2, since multiple rounds of filtering and fusion have not yet been performed, the processing method of the second round is different, specifically including: First, the filtered value of the motion data in the second round is compared with the estimated value in the first round, and the residual between the two is calculated. In this embodiment, for the estimated value in the first round, the motion data can be preliminarily filtered, and the filtered value obtained by the preliminary filtering is determined as the initial estimated value in the first round; then, the weight of the filtered value in the second round is determined based on the residual between the filtered value in the second round and the estimated value in the first round. In some cases, the weight of the filtered value in the first round can be preset to a fixed value or dynamically determined according to the initial residual; then, the estimated value of the motion data in the second round is calculated by weighted averaging based on the filtered values ​​in the first and second rounds and their corresponding weights, wherein the estimated value in the second round serves as a reference benchmark for the subsequent residual calculation in the third round.

[0043] In one embodiment, the number of the multiple rounds is K, where K is an integer greater than 1; and determining, based on the residuals of the motion data in the multiple rounds, the weights corresponding to the filtered values ​​of the motion data in the multiple rounds include: Determine a magnitude relationship between a residual of the motion data in the kth round and a robust threshold value among the multiple rounds, where k is an integer between 2 and K; When the residual of the motion data in the kth round among the multiple rounds is less than or equal to the robust threshold, assigning a weight corresponding to the filtered value of the motion data in the kth round to 1; When the residual of the motion data in the kth round among multiple rounds is greater than the robust threshold, the weight corresponding to the filtered value of the motion data in the kth round is determined as the ratio of the robust threshold to the residual of the kth round.

[0044] In this embodiment, the process of determining the weights corresponding to the filtered values ​​of the motion data in multiple rounds based on the residuals of the motion data in multiple rounds specifically includes the following steps: First, a pre-set robust threshold is obtained. The robust threshold can be set according to the noise characteristics of the motion data or the expected error level, and is used to suppress the impact of abnormal residuals on the estimation results during weight allocation.

[0045] Then, for the kth round in the multiple rounds, where k is an integer between 2 and K, the following judgment and processing are performed respectively: Determine the size relationship between the residual of the motion data in the kth round and the robust threshold; wherein, when the residual of the motion data in the kth round is less than or equal to the robust threshold, directly assign the weight corresponding to the filtered value of the motion data in the kth round to 1, that is, it is considered that the reliability of the filtered value of this round is higher; when the residual of the motion data in the kth round is greater than the robust threshold, the weight corresponding to the filtered value of the motion data in the kth round is determined according to the ratio of the robust threshold to the residual of the kth round. The specific weight value is the robust threshold divided by the residual of the kth round, which reflects that a lower weight is given to the filtered value of the abnormally large residual, thereby reducing the interference of abnormal data on the final estimation result.

[0046] The weight corresponding to the filtered value of the motion data in the kth round satisfies the following formula: (3) in, Indicates the weight corresponding to the filtered value of the motion data in the kth round; represents the robust threshold.

[0047] Through the above processing method, robust weighting of multiple rounds of filtering results can be effectively achieved, and the stability and reliability of the attitude angle estimation results in the presence of noise or abnormal motion can be enhanced.

[0048] In one embodiment, the robust threshold is determined according to the following steps, including: determining noise characteristics of the motion data based on the motion data of the electronic rearview mirror; When the noise characteristic belongs to the first type of noise characteristic, the robust threshold is determined to be the first threshold based on a first mapping relationship between the noise characteristic and the robust threshold; when the noise characteristic belongs to the second type of noise characteristic, the robust threshold is determined to be the second threshold based on a second mapping relationship between the noise characteristic and the robust threshold; wherein the first threshold is greater than the second threshold, and the noise amplitude of the first type of noise characteristic is greater than the noise amplitude of the second type of noise characteristic.

[0049] In this embodiment, a method for determining a robust threshold is provided, which specifically includes: First, based on the motion data of the electronic rearview mirror, the noise characteristics of the motion data are analyzed and determined. The noise characteristics may include indicators such as noise amplitude and noise distribution. In this embodiment, the noise characteristics of the motion data are analyzed by noise amplitude. Then, the noise characteristics are determined to which category the noise characteristics belong. If the noise characteristics belong to the first category of noise characteristics, the robust threshold is determined to be a first threshold based on a first mapping relationship between the noise characteristics and the robust threshold. If the noise characteristics belong to the second category of noise characteristics, the robust threshold is determined to be a second threshold based on a second mapping relationship between the noise characteristics and the robust threshold. The first threshold is greater than the second threshold, and the noise amplitude of the first category of noise characteristics is greater than the noise amplitude of the second category of noise characteristics.

[0050] Through the above method, the robust threshold can be adaptively adjusted according to different noise characteristics, thereby improving the tolerance to outliers when the noise amplitude is large and enhancing the sensitivity of the filter when the noise amplitude is small, thereby further improving the accuracy and robustness of the estimated value of the motion data.

[0051] In one embodiment, a difference between estimated values ​​of the motion data obtained in two adjacent rounds is determined; Obtaining a final estimated value of the motion data according to the filtered values ​​of the motion data in multiple rounds and corresponding weights includes: When the difference between the estimated value in the kth round and the estimated value in the k-1th round is less than the target threshold, or when the number of rounds reaches K, the final estimated value of the motion data is obtained based on the filtered values ​​of the motion data in K rounds and the corresponding weights; wherein K is an integer greater than 1, and k is an integer between 2 and K.

[0052] In this embodiment, the difference between the estimated values ​​of the motion data obtained in every two adjacent rounds is first determined.

[0053] Specifically, according to the filtered values ​​of the motion data in multiple rounds and the corresponding weights, a final estimated value of the motion data is obtained, which specifically includes the following steps: In this embodiment, after each filtering round is completed, the difference between the estimated value of the current k-th round and the estimated value of the previous round (i.e., the k-1th round) is calculated; If the difference is less than the preset target threshold, or K rounds of filtering have been completed (i.e., the maximum number of rounds K has been reached), the execution of subsequent filtering rounds is stopped; After stopping the filtering round, a weighted average process is performed based on the filtering values ​​of each of the K completed rounds and their corresponding weights to obtain the final estimated value of the motion data.

[0054] In this way, when the estimated value of the motion data converges to a certain degree, the filtering iteration can be terminated in advance, thereby avoiding unnecessary computational costs while ensuring the stability and accuracy of the estimation results.

[0055] In one embodiment, the motion data collected from the electronic rearview mirror is processed using a robust estimation method to improve the accuracy and robustness of attitude angle estimation. The specific steps are as follows: First, an initial estimation value is set, which can be obtained by performing a preliminary filtering on the original motion data.

[0056] Subsequently, multiple rounds of filtering and estimation are performed. In each round, the difference between the filtered value of the motion data in the current round and the estimated value in the previous round, i.e., the residual, is calculated. Then, the corresponding weight is determined based on the size of the residual.

[0057] Next, based on the filtered values ​​of the motion data in the 1st to the kth rounds, and the respective weights of the filtered values ​​of the motion data in the 1st to the kth rounds, the estimated value of the motion data in the kth round is calculated by weighted averaging, and the estimated value of the kth round is used to update the estimated value of the k-1th round, so that the accuracy of the estimation result gradually increases during the iteration process of each round.

[0058] The above iterative process continues until a preset convergence condition is met. For example, when the difference in the estimated value between two consecutive rounds is less than a minimum threshold, or when the number of rounds reaches a set maximum number (such as 10), the iteration stops. The final estimated value is the robust estimation result of the motion data.

[0059] Since the motion data includes the angular velocity and linear acceleration of the x-axis, y-axis, and z-axis, the present application uses the same processing as described above to calculate the estimated values ​​of the angular velocity and linear acceleration of the x-axis, y-axis, and z-axis, respectively. The specific processing process is not further described in this embodiment.

[0060] In one embodiment, the motion data includes: angular velocity and linear acceleration, and the attitude angle includes: angular velocity attitude angle and linear acceleration attitude angle; after obtaining the attitude angle of the electronic rearview mirror, the method further includes a method for determining the fusion attitude angle, such as Figure 2 As shown: Step S201, determining the current driving scene of the electronic rearview mirror based on the motion data; Step S202, determining a target filtering weight according to the current driving scene of the electronic rearview mirror and based on a mapping relationship between different driving scenes and filtering weights; Step S203, assigning the target filter weight to the angular velocity attitude angle, and assigning a weight obtained by subtracting the target filter weight from 1 to the linear acceleration attitude angle; Step S204 : summing the angular velocity attitude angle and the linear acceleration attitude angle in a weighted manner to obtain a fusion attitude angle.

[0061] In this embodiment, the motion data includes angular motion data and linear motion data, and the attitude angle includes an angular velocity attitude angle obtained based on angular velocity and a linear acceleration attitude angle obtained based on linear acceleration. After obtaining the attitude angle of the electronic rearview mirror, the method further includes: In step S201, the driving scene that the electronic rearview mirror is currently in is determined based on the motion data. In some cases, the driving scenes can be divided into different categories such as constant speed driving, acceleration, deceleration, turning, and driving on bumpy roads by analyzing the size, change trend, or stability characteristics of the motion data.

[0062] In step S202, a target filtering weight is determined based on the preset mapping relationship between different driving scenarios and filtering weights, depending on the current driving scenario of the electronic rearview mirror. Specifically, different driving scenarios have different reliance on angular velocity attitude angles and linear acceleration attitude angles. For example, in bumpy driving, the angular motion data may be more reliable, so a higher weight should be assigned to the angular velocity attitude angle. In contrast, in smooth driving, the linear motion data may be more stable, so a higher weight should be assigned to the linear acceleration attitude angle. The preset mapping relationship can be obtained through experimental calibration.

[0063] In step S203, the determined target filter weight is assigned to the angular velocity attitude angle, and a value obtained by subtracting the target filter weight from 1 is assigned to the linear acceleration attitude angle as the weight of the linear acceleration attitude angle.

[0064] In step S204, the angular velocity attitude angle and the linear acceleration attitude angle are summed in a weighted manner to obtain a fused attitude angle, where the fused attitude angle = target filter weight × angular velocity attitude angle + (1 − target filter weight) × linear acceleration attitude angle.

[0065] Through the above method, the fusion ratio can be dynamically adjusted according to the reliability of each sensor data in different driving scenarios, thereby improving the accuracy and robustness of attitude angle estimation.

[0066] In some embodiments, the angular velocity estimate after Huber robust estimation is 、 、 By integrating, we can get the x-axis angular velocity attitude angle, the y-axis angular velocity attitude angle, and the z-axis angular velocity attitude angle:

[0067] (4)

[0068] in, Indicates the x-axis angular velocity attitude angle; Indicates the y-axis angular velocity attitude angle; Indicates the z-axis angular velocity attitude angle; In some embodiments, the acceleration estimate after Huber robust estimation is 、 、 , we can get the x-axis linear acceleration attitude angle, y-axis linear acceleration attitude angle and z-axis linear acceleration attitude angle:

[0069] (5)

[0070] In this application, based on the calculated x-axis angular velocity attitude angle, x-axis linear acceleration attitude angle; y-axis angular velocity attitude angle, y-axis linear acceleration attitude angle; and z-axis angular velocity attitude angle, z-axis linear acceleration attitude angle, the calculation formula of the fusion attitude angle is: fusion attitude angle = target filter weight × angular velocity attitude angle + (1-target filter weight) × linear acceleration attitude angle, the x-axis fusion attitude angle is calculated respectively. , y-axis fusion attitude angle And the z-axis fusion attitude angle .

[0071] In one embodiment, determining the target filtering weight according to the current driving scene of the electronic rearview mirror and based on a mapping relationship between different driving scenes and filtering weights includes: When the driving scene in which the electronic rearview mirror is located is the first type of driving scene, determining the target filtering weight to be the first target filtering weight based on a third mapping relationship between the driving scene and the filtering weight; When the driving scene in which the electronic rearview mirror is located is a second type of driving scene, based on the fourth mapping relationship between the driving scene and the filtering weight, the target filtering weight is determined to be the second target filtering weight, the first target filtering weight is greater than the second target filtering weight, and the dynamic change speed of the first type of driving scene is faster than the dynamic change speed of the second type of driving scene.

[0072] In this embodiment, first, the current driving scene that the electronic rearview mirror is in is obtained. The driving scene can be identified by analyzing the motion data (such as acceleration and angular velocity) collected by the electronic rearview mirror and the driving state of the vehicle (such as vehicle speed and acceleration rate).

[0073] Subsequently, the corresponding target filter weight is determined based on the mapping relationship between different driving scenarios and filter weights. Specifically, when the driving scenario in which the electronic rearview mirror is located is a first type of driving scenario (for example, scenarios with rapid dynamic changes, such as high-speed driving or driving on rough roads), the target filter weight is determined to be a first target filter weight based on a pre-set third mapping relationship. This first target filter weight is relatively large to improve responsiveness and adaptability to rapidly changing posture data.

[0074] When the driving scenario in which the electronic rearview mirror is located is a second type of driving scenario (e.g., a scenario with a relatively slow dynamic change rate, such as constant speed driving or parking), the target filtering weight is determined to be a second target filtering weight based on a pre-set fourth mapping relationship. The second target filtering weight is relatively small, thereby improving filtering smoothness and suppressing errors caused by minor noise.

[0075] Among them, the first target filter weight is greater than the second target filter weight, which reflects the adaptive adjustment of different dynamic change characteristics: In this embodiment, in the first type of driving scenario, due to the rapid dynamic change rate, it is necessary to give a higher weight to the angular velocity attitude angle to respond to the attitude change more quickly; while in the second type of driving scenario, due to the slow dynamic change rate, the weight of the angular velocity attitude angle is appropriately reduced, and the influence of the linear acceleration attitude angle is enhanced, thereby improving the stability and smoothness of the attitude estimation.

[0076] In this embodiment, the speed of dynamic change refers to the degree to which the motion data (such as angular velocity and linear acceleration) detected by the electronic rearview mirror changes over time. Specifically, the speed of the dynamic change depends on the magnitude of the change in the motion data value per unit time. A fast dynamic change refers to a dramatic change in the motion data within a short period of time. For example, when a vehicle quickly changes lanes, turns, or encounters a pothole, causing severe body vibration, the angle and acceleration of the electronic rearview mirror can change significantly and rapidly. This manifests in the data as a significant difference in angular velocity and linear acceleration between two consecutive moments. A slow dynamic change refers to a smooth and subtle change in the motion data per unit time. For example, when a vehicle is traveling at a constant speed in a straight line, accelerating slowly, or making a slight lane change, the angle and linear acceleration of the electronic rearview mirror change very little. This manifests in the data as nearly identical values ​​between two consecutive moments, with no noticeable change.

[0077] In one embodiment, the attitude angle of an electronic rearview mirror is accurately estimated under dynamic driving conditions, and its jitter state is identified and compensated, thereby improving image stability and driving safety. In the attitude angle acquisition process, the gyroscope and accelerometer installed inside the electronic rearview mirror first collect three-axis angular velocity and linear acceleration data in real time to form motion data. This motion data is then filtered multiple times. In each filtering round, the reliability is evaluated based on the residual between the current filtered value and the previous estimated value, and a weight is dynamically assigned accordingly. The weighted average of the filtered values ​​from multiple rounds and the corresponding weights is then taken to obtain the final estimated value of the motion data. Subsequently, the attitude angle of the electronic rearview mirror is calculated based on the estimated value, including the angular velocity attitude angle and the linear acceleration attitude angle. In terms of attitude angle fusion, the characteristics of the driving scene in which the vehicle is currently located are further combined. By analyzing the dynamic change speed of the motion state, the fusion weight is adaptively determined based on the preset scene and filter weight mapping relationship. A higher weight is assigned to the angular velocity attitude angle to improve the responsiveness in violent dynamic scenes, while in stable scenes, the influence of the linear acceleration attitude angle is enhanced to improve the stability of the estimation. Finally, based on the change amplitude and direction of the fused attitude angle at adjacent moments, the jitter parameter value of the electronic rearview mirror is calculated, and the corresponding jitter compensation value is determined to drive the Zhendong unit to make real-time corrections to the electronic rearview mirror. Through the above scheme, the interference of noise and abnormal data on the attitude angle estimation can be effectively filtered out, and the accurate suppression of the electronic rearview mirror jitter in multiple types of dynamic scenes can be achieved, which significantly improves the rear view imaging quality during vehicle driving.

[0078] Through the above method, the filter weight distribution can be flexibly adjusted according to different actual driving scenarios to improve the accuracy and robustness of the electronic rearview mirror posture estimation.

[0079] Based on the same inventive concept, the second aspect of the present application provides an anti-shake method for an electronic rearview mirror, the method being as follows: Figure 3 As shown, including: Determine the attitude angle of the electronic rearview mirror according to the method described in the first aspect of the present application based on the motion data of the electronic rearview mirror; Step S301, determining a jitter parameter value of the electronic rearview mirror according to the attitude angle of the electronic rearview mirror; Step S302 : determining a jitter compensation value according to the jitter parameter value, so as to control the vibration unit of the electronic rearview mirror to perform jitter compensation based on the jitter compensation value.

[0080] In this embodiment, a method for stabilizing an electronic rearview mirror is provided. The method is based on the attitude angle of the electronic rearview mirror obtained by the attitude angle estimation method of the electronic rearview mirror described in the first aspect of the present application, and further determines the jitter parameters of the electronic rearview mirror. The specific method is as follows: Acquire motion data from the electronic rearview mirror. The motion data may include, but is not limited to, parameters such as angular velocity and linear acceleration. This motion data can be collected by an inertial motion detection module or other motion sensor within the electronic rearview mirror for subsequent attitude angle estimation and jitter analysis. The inertial motion detection module may be a gyroscope or accelerometer, for example. The motion data is filtered in multiple rounds to obtain filtered values ​​for each round. Specifically, different filtering algorithms and parameters can be employed in each round, or multi-scale filtering can be employed to gradually extract stable components from the motion data and suppress high-frequency noise or abnormal fluctuations. For each of the multiple rounds, calculate the residual of the motion data in that round based on the filtered value of the motion data in that round and the estimated value of the motion data in the previous round. This residual calculation reflects the consistency between the current filtered value and the historical estimated value, further providing a basis for subsequent weight adjustment. Based on the residuals of the motion data in the multiple rounds, determine the weights corresponding to the filtered values ​​of the motion data in the multiple rounds. Specifically, a robust threshold can be preset. When the residual of a certain round is less than or equal to the threshold, a higher weight (such as a weight of 1) is given to the filter value of this round; when the residual is greater than the robust threshold, the weight is reduced. The weight value can be obtained by the ratio of the robust threshold to the residual, so as to suppress abnormal data. Based on the filter values ​​of the motion data in multiple rounds and the corresponding weights, the final estimated value of the motion data is obtained. Specifically, by taking a weighted average of the filter values ​​of each round and their corresponding weights, and combining the filtering effects of different rounds, a more stable and accurate motion data estimation result is obtained. Based on the final estimated value of the motion data, the attitude angle of the electronic rearview mirror is obtained. The attitude angle can include angular velocity attitude angle and linear acceleration attitude angle, reflecting the attitude change of the electronic rearview mirror in space.

[0081] Step S301: Determine the jitter parameter value of the electronic rearview mirror according to the attitude angle of the electronic rearview mirror. Specifically, based on the attitude angle change rate, amplitude and other characteristics, relevant parameters for characterizing the degree of jitter, such as jitter direction, jitter amplitude, etc., can be extracted.

[0082] In step S302, a jitter compensation value is determined based on the jitter parameter value, thereby controlling the vibration unit of the electronic rearview mirror to perform jitter compensation based on the jitter compensation value. Specifically, a reverse adjustment signal is applied to the vibration unit to offset jitter caused by external disturbances, thereby stabilizing the imaging effect of the electronic rearview mirror and improving the driver's visual experience and safety while driving.

[0083] In this application, a method for determining the jitter parameter value is provided for the posture change of the electronic rearview mirror, specifically including: Get the fusion attitude angle of the electronic rearview mirror at the current time t and the previous time t-1. Among them, the fusion attitude angle at the current time t includes: x-axis fusion attitude angle , y-axis fusion attitude angle And the z-axis fusion attitude angle The fusion posture angles at the previous moment t−1 include: x-axis fusion posture angle , y-axis fusion attitude angle And the z-axis fusion attitude angle .

[0084] Then, based on the fusion attitude angle difference at adjacent moments, the jitter amplitude of the electronic rearview mirror is calculated. Specifically, the jitter amplitude can be calculated according to the following formula: (6) in, Indicates the vibration amplitude of the electronic rearview mirror.

[0085] Based on the change in the fusion attitude angle of each axis, the shaking direction vector of the electronic rearview mirror is determined. Specifically, the shaking direction vector can be calculated according to the following formula: (7) in, Indicates the shaking direction vector of the electronic rearview mirror.

[0086] This application detects the vibration of the vehicle in real time, and the controller controls the vibration unit to perform jitter compensation based on the detected signal, effectively eliminating the electronic rearview mirror jitter caused by vehicle vibration, improving the driver's observation of the road conditions behind, and enhancing driving safety; the structure is simple, easy to install and maintain, and suitable for various types of vehicles; under different road conditions, the working parameters of the vibration unit in the electronic rearview mirror can be adjusted according to actual conditions to achieve the best anti-shake effect.

[0087] Based on the same inventive concept, the third aspect of the present application provides an electronic rearview mirror, which is used to execute the steps of the attitude angle estimation method of the electronic rearview mirror as described in the first aspect of the present application, or the electronic rearview mirror is used to execute the steps of the anti-shake method of the electronic rearview mirror as described in the second aspect of the present application.

[0088] In one embodiment, the structural block diagram of the electronic rearview mirror is as follows: Figure 4 As shown: It includes motion detection module, image sensor, main control chip, display screen and vibration unit.

[0089] Among them, the motion detection module includes a gyroscope and an accelerometer, which are used to measure the angular velocity and linear acceleration of the electronic rearview mirror respectively; the image sensor is used to collect image data of the environment behind the vehicle; the main control chip is used to receive data transmitted by the motion detection module and the image sensor, and process the motion data and image data to generate a control signal for anti-shake control and an image signal for display; the display screen is used to display the image signal for the driver to observe the road conditions behind, and the vibration unit is used to perform vibration compensation according to the control signal of the anti-shake control; among them, in some cases, the electronic rearview mirror is also provided with a shell for accommodating the above-mentioned components and providing waterproof, dustproof and shockproof protection; and a mounting bracket for fixing the shell to a specified position of the vehicle and maintaining the stability of the electronic rearview mirror during driving.

[0090] Specifically, the motion detection module is connected to the data input of the main control chip via a signal line to transmit angular motion data and linear motion data to the main control chip in real time. The image sensor is connected to the image input of the main control chip via a high-speed data bus to transmit image data. The image output of the main control chip is connected to the display screen via a display interface to output image signals after anti-shake processing. The motion signal output of the main control chip is connected to the vibration unit via a signal interface to output control signals for anti-shake control. Through this design, the various components can achieve efficient collaboration, ensuring the real-time transmission and processing of motion data, image data, and control signals, thereby improving the stability and display quality of the rearview image.

[0091] Based on the same inventive concept, the fourth aspect of the present application provides a vehicle, which includes the electronic rearview mirror as described in the third aspect of the present application.

[0092] Each embodiment in this specification focuses on the differences from other embodiments, and the same or similar parts between the embodiments can be referenced to each other.

[0093] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, devices, or computer program products. Therefore, the embodiments of the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the embodiments of the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0094] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0095] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0096] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0097] Although preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they become aware of the basic inventive concepts. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present invention.

[0098] Finally, it should be noted that, in this document, relational terms such as first and second, etc., are used only 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 terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or terminal device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or terminal device that includes the element.

[0099] The above is a detailed introduction to the provided electronic rearview mirror attitude angle estimation method, anti-shake method, electronic rearview mirror and vehicle. This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method of this application and its core idea; at the same time, for general technical personnel in this field, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on this application.

Claims

1. A method for estimating the attitude angle of an electronic rearview mirror, characterized in that: The method comprises: Obtain motion data of electronic rearview mirror; Performing multiple rounds of filtering on the motion data to obtain a filtered value of the motion data in each round; For each of the multiple rounds, calculating a residual of the motion data in the round based on a filtered value of the motion data in the round and an estimated value of the motion data in the previous round; Determining weights corresponding to filtered values ​​of the motion data in the multiple rounds according to residuals of the motion data in the multiple rounds; Obtaining a final estimated value of the motion data according to filtered values ​​of the motion data in multiple rounds and corresponding weights; The attitude angle of the electronic rearview mirror is obtained according to the final estimated value of the motion data.

2. The method for estimating the attitude angle of an electronic rearview mirror according to claim 1, wherein: The number of the multiple rounds is K, where K is an integer greater than 1; for the kth round of the K rounds, the estimated value of the motion data in the kth round is determined according to the following steps: Calculating a residual of the motion data at the kth round based on a filtered value of the motion data at the kth round and an estimated value of the motion data at the k−1th round; Determining a weight corresponding to a filtered value of the motion data in the kth round based on a residual of the motion data in the kth round; Based on the filtered values ​​of the motion data in the 1st to kth rounds, and the respective weights of the filtered values ​​of the motion data in the 1st to kth rounds, an estimated value of the motion data in the kth round is obtained to determine the residual of the motion data in the k+1th round; wherein k is an integer between 2 and K.

3. The method for estimating the attitude angle of an electronic rearview mirror according to claim 1, wherein: The number of the multiple rounds is K, where K is an integer greater than 1; and determining, based on the residuals of the motion data in the multiple rounds, weights corresponding to the filtered values ​​of the motion data in the multiple rounds include: Determine a magnitude relationship between a residual of the motion data in the kth round and a robust threshold value among the multiple rounds, where k is an integer between 2 and K; When the residual of the motion data in the kth round among the multiple rounds is less than or equal to the robust threshold, assigning a weight corresponding to the filtered value of the motion data in the kth round to 1; When the residual of the motion data in the kth round among multiple rounds is greater than the robust threshold, the weight corresponding to the filtered value of the motion data in the kth round is determined as the ratio of the robust threshold to the residual of the kth round.

4. The method for estimating the attitude angle of an electronic rearview mirror according to claim 3, wherein: The robust threshold is determined according to the following steps, including: determining noise characteristics of the motion data based on the motion data of the electronic rearview mirror; When the noise characteristic belongs to the first type of noise characteristic, the robust threshold is determined to be the first threshold based on a first mapping relationship between the noise characteristic and the robust threshold; when the noise characteristic belongs to the second type of noise characteristic, the robust threshold is determined to be the second threshold based on a second mapping relationship between the noise characteristic and the robust threshold; wherein the first threshold is greater than the second threshold, and the noise amplitude of the first type of noise characteristic is greater than the noise amplitude of the second type of noise characteristic.

5. The method for estimating the attitude angle of an electronic rearview mirror according to claim 1, wherein: The method further comprises: Determine the difference between the estimated values ​​of the motion data obtained in each two adjacent rounds; Obtaining a final estimated value of the motion data according to the filtered values ​​of the motion data in multiple rounds and corresponding weights includes: When the difference between the estimated value in the kth round and the estimated value in the k-1th round is less than the target threshold, or when the number of rounds reaches K, the final estimated value of the motion data is obtained based on the filtered values ​​of the motion data in K rounds and the corresponding weights; wherein K is an integer greater than 1, and k is an integer between 2 and K.

6. The method for estimating the attitude angle of an electronic rearview mirror according to claim 1, wherein: The motion data includes: angular velocity and linear acceleration, and the attitude angle includes: angular velocity attitude angle and linear acceleration attitude angle; after obtaining the attitude angle of the electronic rearview mirror, the method further includes: determining a current driving scene of the electronic rearview mirror based on the motion data; Determining a target filtering weight based on a mapping relationship between different driving scenarios and filtering weights according to the current driving scenario of the electronic rearview mirror; Assigning the target filter weight to the angular velocity attitude angle, and assigning a weight obtained by subtracting the target filter weight from 1 to the linear acceleration attitude angle; The angular velocity attitude angle and the linear acceleration attitude angle are summed in a weighted manner to obtain a fusion attitude angle.

7. The method for estimating the attitude angle of an electronic rearview mirror according to claim 6, wherein: The determining of the target filtering weight according to the current driving scene of the electronic rearview mirror and based on the mapping relationship between different driving scenes and filtering weights includes: When the driving scene in which the electronic rearview mirror is located is the first type of driving scene, determining the target filtering weight to be the first target filtering weight based on a third mapping relationship between the driving scene and the filtering weight; When the driving scene in which the electronic rearview mirror is located is a second type of driving scene, based on the fourth mapping relationship between the driving scene and the filtering weight, the target filtering weight is determined to be the second target filtering weight, the first target filtering weight is greater than the second target filtering weight, and the dynamic change speed of the first type of driving scene is faster than the dynamic change speed of the second type of driving scene.

8. An anti-shake method for an electronic rearview mirror, characterized in that: The method comprises: Determining the attitude angle of the electronic rearview mirror according to the method of any one of claims 1 to 7 based on the motion data of the electronic rearview mirror; determining a jitter parameter value of the electronic rearview mirror according to the attitude angle of the electronic rearview mirror; A jitter compensation value is determined according to the jitter parameter value, so as to control the vibration unit of the electronic rearview mirror to perform jitter compensation based on the jitter compensation value.

9. An electronic rearview mirror, characterized in that: The electronic rearview mirror is used to execute the steps of the attitude angle estimation method of the electronic rearview mirror as described in any one of claims 1 to 7, or the electronic rearview mirror is used to execute the steps of the anti-shake method of the electronic rearview mirror as described in claim 8.

10. A vehicle, characterized in that: The vehicle includes the electronic rearview mirror according to claim 9.

Images