WO2020192688A1

WO2020192688A1 - Method and apparatus for eliminating angular velocity measurement error of automobile gear ring

Info

Publication number: WO2020192688A1
Application number: PCT/CN2020/081099
Authority: WO
Inventors: 钟毅
Original assignee: 武汉理工大学
Priority date: 2019-03-26
Filing date: 2020-03-25
Publication date: 2020-10-01
Also published as: CN109991438B; CN109991438A

Abstract

A method for eliminating an angular velocity measurement error of an automobile ring gear, comprising the following steps: 1) assuming that a ring gear comprising N teeth is installed on a rotating shaft, reading signals of the ring gear, obtaining a time t_k at which a tooth k passes, and assuming an error of a tooth i to be δi to obtain an angular velocity for eliminating the error; 2) a time interval between teeth being Δti(k)=t_k‑t_k‑ ₁, and one complete cycle time being Ti(k)=t_k+N _/2–t_k‑N _/2, then Δti/Ti=1/N+Δi(+ek), and the error of the tooth i being δi=2πΔi=2π(Δti/Ti1/N)+ek; and 3) using an adaptive filter to perform filter processing on the error of the tooth i; and 4) using the error of the tooth i to correct the angular velocity of gear teeth according to a correspondence relationship. The present invention also relates to an apparatus for eliminating an angular velocity measurement error of an automobile gear ring. Learning speed is adaptively updated according to the statistical characteristics of input signals to achieve a faster convergence speed, and effectively avoid overshoot, which provides more accurate data for subsequent tire pressure monitoring.

Description

一种汽车齿圈角速度测量误差消除方法与装置Method and device for eliminating measurement error of automobile ring gear angular velocity

技术领域Technical field

本发明涉及汽车电子技术，尤其涉及一种汽车齿圈角速度测量误差消除方法与装置。The invention relates to automobile electronic technology, in particular to a method and device for eliminating the angular velocity measurement error of an automobile ring gear.

背景技术Background technique

统计表明，交通事故中爆胎原因所占的比例高达60％，如果车速超过160公里/小时，前轮爆胎的生存几率几乎为0。75％的爆胎都是由胎压不足造成的，当胎压不足时，轮胎侧面因受挤压而弯曲，进而造成轮胎温度升高引发爆胎。近年来，随着我国最新胎压监测强制性法规GB26149-2017的发布，胎压监测***(TPMS)受到前所未有的重视。Statistics show that the proportion of tire blowouts in traffic accidents is as high as 60%. If the speed exceeds 160 km/h, the survival probability of front tire blowouts is almost 0.75%. The tire blowouts are caused by insufficient tire pressure. When the tire pressure is insufficient, the side of the tire bends due to squeezing, causing the temperature of the tire to rise and cause a puncture. In recent years, with the release of my country's latest mandatory tire pressure monitoring regulations GB26149-2017, tire pressure monitoring systems (TPMS) have received unprecedented attention.

按照工作原理的不同，胎压监测***可以分为两种：直接式胎压监测***和间接式胎压监测***。其中间接式胎压监测***通过轮胎的转速差和轮胎的振动频率判断轮胎是否漏气。提取轮胎振动频率需要对轮速信号进行频谱分析，但由于汽车齿圈在生产制造时存在不可避免的误差，在汽车行驶速度恒定时，此生产误差会导致轮速信号频谱中出现幅值较大的谐波分量，影响下一步的频谱分析，无法准确提取轮胎的共振频率。因此进行频谱分析前需要消除由汽车齿圈生产误差导致的轮速信号频谱中的谐波分量。现有间接式胎压监测***中消除齿圈生产误差的常规方法是采用最小均方差(LMS)滤波算法来实现原始轮速信号误差的自学习，采用梯度下降法逼近真实误差信号。此方法缺点在于，实施过程中需要设定合理的收敛速度，收敛速度过快则容易造成在真实值附近的超调与震荡，收敛速度过慢则导致学习时间过长，有效数据减少。According to different working principles, tire pressure monitoring systems can be divided into two types: direct tire pressure monitoring systems and indirect tire pressure monitoring systems. Among them, the indirect tire pressure monitoring system judges whether the tire is leaking through the difference in tire rotation speed and the vibration frequency of the tire. Extracting the tire vibration frequency requires spectrum analysis of the wheel speed signal. However, due to the inevitable error in the production of the car ring gear, when the car is running at a constant speed, this production error will cause a large amplitude in the wheel speed signal spectrum The harmonic components of, affect the next frequency spectrum analysis, and it is impossible to accurately extract the resonance frequency of the tire. Therefore, it is necessary to eliminate the harmonic components in the wheel speed signal spectrum caused by the production error of the automobile ring gear before performing the spectrum analysis. In the existing indirect tire pressure monitoring system, the conventional method to eliminate the production error of the ring gear is to use the least mean square error (LMS) filtering algorithm to realize the self-learning of the original wheel speed signal error, and the gradient descent method to approximate the real error signal. The disadvantage of this method is that a reasonable convergence speed needs to be set during the implementation process. If the convergence speed is too fast, it will easily cause overshoot and oscillation near the true value. If the convergence speed is too slow, the learning time will be too long and the effective data will be reduced.

发明内容Summary of the invention

本发明要解决的技术问题在于针对现有技术中的缺陷，提供一种汽车齿圈测量误差消除方法。The technical problem to be solved by the present invention is to provide a method for eliminating the measurement error of the automobile gear ring in view of the defects in the prior art.

本发明解决其技术问题所采用的技术方案是：一种汽车齿圈角速度测量误差消除方法，包括以下步骤：The technical solution adopted by the present invention to solve the technical problem is: a method for eliminating the measurement error of the angular velocity of the automobile ring gear, including the following steps:

1)设旋转轴上安装有含有N个齿的齿圈，读取齿圈信号，获得齿k经过的时刻t _k，设齿i的误差为δi，则消除误差的角速度ω _k’＝ω _k-δi,i＝k mod N； 1) Suppose a ring gear containing N teeth is installed on the rotating shaft, read the signal of the ring gear, and obtain the time t _k when the tooth k passes, and set the error of the tooth i as δi, then the angular velocity ω _k' = ω _{k to} eliminate the error -δi,i=k mod N;

2)齿与齿之间的时间间隔Δti(k)＝t _k-t _k-1，一个完整周期时间为Ti(k)＝t _k+N/2-t _k-N/2，则Δti/Ti＝1/N+Δi(+ek)，齿i的误差δi＝2πΔi＝2π(Δti/Ti-1/N)+ek； 2) The time interval between teeth Δti(k)=t _k -t _k-1 , a complete cycle time is Ti(k)=t _k+N/2 -t _kN/2 , then Δti/Ti= 1/N+Δi(+ek), the error of tooth i δi=2πΔi=2π(Δti/Ti-1/N)+ek;

3)使用自适应滤波器对齿i的误差进行滤波处理；3) Use an adaptive filter to filter the error of tooth i;

4)使用齿i的误差对轮齿按对应关系对角速度进行修正。4) Use the error of tooth i to correct the angular velocity of the tooth according to the corresponding relationship.

按上述方案，所述步骤3)中自适应滤波器是长度为M的FIR滤波器。According to the above solution, the adaptive filter in step 3) is an FIR filter with a length of M.

按上述方案，所述步骤3)中在每个时间步长中，滤波器系数都使用误差进行更新en＝dn-yn，其中dn是所需的响应，yn是输入误差序列通过滤波器产生的序列。According to the above solution, in the step 3) in each time step, the filter coefficients are updated with error en=dn-yn, where dn is the desired response, and yn is the input error sequence generated by the filter sequence.

按上述方案，所述步骤3)中损失函数：MSE ei(k)＝(2π*(Δti(k)/Ti(k)-1/N)+δi(k+1))。According to the above scheme, the loss function in the step 3): MSE ei(k)=(2π*(Δti(k)/Ti(k)-1/N)+δi(k+1)).

一种汽车齿圈角速度测量误差消除装置，包括：An error elimination device for measuring the angular velocity of the automobile ring gear includes:

齿圈信号读取和储存模块，用于对旋转轴上安装有含有N个齿的齿圈，读取齿圈信号，获得齿k经过的时刻t _k，并存储数据； The ring gear signal reading and storage module is used to read the ring gear signal for a ring gear with N teeth installed on the rotating shaft, obtain the time t _k when the tooth k passes, and store the data;

齿圈误差估计模块，用于根据齿圈信号获得齿i的误差δi＝2πΔi＝2π(Δti/Ti-1/N)+ek；Ring gear error estimation module, used to obtain tooth i error δi=2πΔi=2π(Δti/Ti-1/N)+ek according to the ring gear signal;

并使用自适应滤波器对齿i的误差进行滤波处理；And use an adaptive filter to filter the error of tooth i;

误差修正模块，使用齿i的误差对轮齿按对应关系对角速度进行修正，消除误差的角速度ω _k’＝ω _k-δi,i＝k mod N。 The error correction module uses the error of the tooth i to correct the angular velocity of the tooth according to the corresponding relationship, and eliminates the angular velocity of the error ω _k' = ω _k- δi, i = k mod N.

按上述方案，所述齿圈误差估计模块中自适应滤波器是长度为M的FIR滤波器。According to the above solution, the adaptive filter in the ring gear error estimation module is an FIR filter with a length of M.

按上述方案，所述齿圈误差估计模块中在每个时间步长中，滤波器系数都使用误差进行更新en＝dn-yn，其中dn是所需的响应，yn是输入误差序列通过滤波器产生的序列。According to the above solution, in the ring gear error estimation module, in each time step, the filter coefficients are updated with error en=dn-yn, where dn is the required response, and yn is the input error sequence through the filter The resulting sequence.

按上述方案，所述齿圈误差估计模块中代价函数：MSE ei(k)＝(2π*(Δti(k)/Ti(k)-1/Ncog)+δi(k+1))。According to the above solution, the cost function in the ring gear error estimation module: MSE ei(k)=(2π*(Δti(k)/Ti(k)-1/Ncog)+δi(k+1)).

本发明产生的有益效果是：The beneficial effects produced by the present invention are:

本发明依据输入信号的统计特性自适应更新学习速度，实现更快的收敛速度，并且有效避免超调。The invention updates the learning speed adaptively according to the statistical characteristics of the input signal, realizes a faster convergence speed, and effectively avoids overshoot.

本发明方法只需要根据当前测量值对状态估计值进行更新，因此具有较低的计算复杂度和存储空间要求，能够较好地满足车辆控制的实时性要求。The method of the present invention only needs to update the state estimation value according to the current measurement value, so it has lower calculation complexity and storage space requirements, and can better meet the real-time requirements of vehicle control.

附图说明Description of the drawings

下面将结合附图及实施例对本发明作进一步说明，附图中：The present invention will be further described below in conjunction with the accompanying drawings and embodiments. In the accompanying drawings:

图1是本发明实施例的结构示意图；Figure 1 is a schematic structural diagram of an embodiment of the present invention;

图2是本发明实施例的修正前轮速信号频谱；Figure 2 is a modified front wheel speed signal spectrum of an embodiment of the present invention;

图3是本发明实施例的修正后轮速信号频谱；Figure 3 is a modified rear wheel speed signal spectrum of an embodiment of the present invention;

图4是本发明实施例的误差拟合的变化趋势。Fig. 4 shows the variation trend of error fitting in the embodiment of the present invention.

具体实施方式detailed description

为了使本发明的目的、技术方案及优点更加清楚明白，以下结合实施例，对本发明进行进一步详细说明。应当理解，此处所描述的具体实施例仅用以解释本发明，并不用于限定本发明。In order to make the objectives, technical solutions, and advantages of the present invention clearer and more comprehensible, the present invention will be further described in detail below in conjunction with embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, but not to limit the present invention.

如图1所示，一种汽车齿圈角速度测量误差消除方法，包括以下步骤：As shown in Figure 1, a method for eliminating the measurement error of the angular velocity of an automobile ring gear includes the following steps:

2)齿与齿之间的时间间隔Δti(k)＝t _k-t _k-1，一个完整周期时间为Ti(k)＝t _k+N/2–t _k-N/2，则Δti/Ti＝1/N+Δi(+ek)，齿i的误差δi＝2πΔi＝2π(Δti/Ti-1/N)+ek； 2) The time interval between teeth Δti(k)=t _k -t _k-1 , a complete cycle time is Ti(k)=t _k+N/2 -t _kN/2 , then Δti/Ti= 1/N+Δi(+ek), the error of tooth i δi=2πΔi=2π(Δti/Ti-1/N)+ek;

本发明获取轮速信号的原理如下：含有Ncog个齿的齿圈安装在旋转轴上，假设齿圈是理想的，那么每个齿之间的角度是α＝2π/Ncog，传感器给出具有变化幅度和频率的正弦信号，在比较器中被转换为具有恒定幅度的方波。该信号中的每个边缘对应于每个齿的边缘。齿k经过的时间表示为t _k，也可表示相应的角度。但是在真实环境下，齿圈都具有生产误差δi。由此我们可以获得真实的角速度ω _k＝ω _k-δi,i＝k mod Ncog。由于每个齿的误差是不同且未知的，为了进行精确的角速度计算，我们需要对误差进行预测。 The principle of obtaining wheel speed signals in the present invention is as follows: a ring gear containing Ncog teeth is installed on the rotating shaft, assuming that the ring gear is ideal, then the angle between each tooth is α=2π/Ncog, and the sensor gives a change The sine signal of amplitude and frequency is converted into a square wave with constant amplitude in the comparator. Each edge in the signal corresponds to the edge of each tooth. The elapsed time of tooth k is denoted as t _k , and the corresponding angle can also be expressed. But in a real environment, the gear ring has a production error δi. From this we can obtain the true angular velocity ω _k =ω _k -δi, i=k mod Ncog. Since the error of each tooth is different and unknown, in order to perform accurate angular velocity calculation, we need to predict the error.

为了估计齿圈误差，我们需要计算齿与齿之间要时间间隔Δti(k)＝t _k-t _k-1，一个完整周期时间为Ti(k)＝t(k+Ncog/2)–t(k-Ncog/2)。先假设短时间内车速为匀速，即Δti/Ti＝1/Ncog，所以在有误差的情况下Δti/Ti＝1/Ncog+Δi(+ek)，由此可以计算出误差δi＝2πΔi＝2π(Δti/Ti-1/Ncog)+ek，此误差的评估运用自适应滤波器。 In order to estimate the ring gear error, we need to calculate the time interval between teeth Δti(k)=t _k -t _k-1 , and a complete cycle time is Ti(k)=t(k+Ncog/2)–t (k-Ncog/2). First assume that the vehicle speed is uniform in a short time, that is, Δti/Ti=1/Ncog, so if there is an error, Δti/Ti=1/Ncog+Δi(+ek), from which the error δi=2πΔi=2π can be calculated (Δti/Ti-1/Ncog)+ek, this error is evaluated using an adaptive filter.

自适应滤波器是长度为M的FIR，系数为bk,k＝0,1,2...M-1。输入流{fn}通过滤波器产生序列{yn}。在每个时间步长中，滤波器系数都使用误差进行更新en＝dn-yn，其中dn是所需的响应。定义代价函数：MSE ei(k)＝(2π*(Δti(k)/Ti(k)-1/Ncog)+δi(k+1))，该梯度用来更新误差。The adaptive filter is an FIR of length M, and the coefficients are bk, k=0,1,2...M-1. The input stream {fn} is passed through the filter to generate a sequence {yn}. In each time step, the filter coefficients are updated with error en=dn-yn, where dn is the desired response. Define the cost function: MSE ei(k)=(2π*(Δti(k)/Ti(k)-1/Ncog)+δi(k+1)), and the gradient is used to update the error.

误差函数的获得：在误差函数中需要对学习率和学习预设值进行预测，以达到最好效果。因此需要建立梯度下降模型，通过仿真软件进行效果验证。步长决定了在梯度下降迭代的过程中，每一步沿梯度负方向前进的长度。特征(feature)：指的是样本中输入部分，比如样本(x(0)，y(0))就是对应齿的信号，第一个样本特征为x(0)，第一个样本输出为y(0)。假设函数(hypothesis function)：在监督学习中，为了拟合输入样本使用的函数，记为hθ(x)，在此采用拟合函数如：δ _i(0)。损失函数(loss function)：为了评估模型拟合的好坏，通常用损失函数来度量拟合的程度。损失函数极小化，意味着拟合程度最好，对应的模型参数即为最优参数。在线性回归中，损失函数通常为样本输出和假设函数的差取平方。比如对于m个样本(xi，yi)(i＝1，2，...m)，采用线性回归，损失函数为：MSE ei(k)＝(2π*(Δti(k)/Ti(k)-1/Ncog)-δi(k-1))*2。递归步长取决于数据样本，可以试验不同的值对比迭代效果，如果损失函数在变小，说明取值有效，否则要增大步长。 Obtaining the error function: The learning rate and the learning preset value need to be predicted in the error function to achieve the best effect. Therefore, it is necessary to establish a gradient descent model and verify the effect through simulation software. The step size determines the length of each step in the negative direction of the gradient during the gradient descent iteration. Feature: refers to the input part of the sample, for example, the sample (x(0), y(0)) is the signal corresponding to the tooth, the first sample feature is x(0), and the first sample output is y (0). Hypothesis function: In supervised learning, the function used to fit the input sample is denoted as hθ(x), and the fitting function is used here, such as δ _i (0). Loss function (loss function): In order to evaluate how well the model fits, the loss function is usually used to measure the degree of fit. Minimization of the loss function means that the degree of fit is the best, and the corresponding model parameters are the optimal parameters. In linear regression, the loss function is usually the square of the difference between the sample output and the hypothesis function. For example, for m samples (xi, yi) (i=1, 2,...m), using linear regression, the loss function is: MSE ei(k)=(2π*(Δti(k)/Ti(k) -1/Ncog)-δi(k-1))*2. The recursive step size depends on the data sample. You can test different values to compare the iteration effect. If the loss function is getting smaller, the value is valid, otherwise the step size should be increased.

图2至图4是经过本发明方法修正后的轮速信号频谱以及误差拟合的变化趋势。Figures 2 to 4 are the wheel speed signal frequency spectrum and the changing trend of the error fitting after being corrected by the method of the present invention.

根据本发明的方法可以得到对应的装置，具体如下：According to the method of the present invention, the corresponding device can be obtained, which is specifically as follows:

并使用自适应滤波器对齿i的误差进行滤波处理；齿圈误差估计模块中自适应滤波器是长度为M的FIR滤波器；在每个时间步长中，滤波器系数都使用误差进行更新en＝dn-yn，其中dn是所需的响应，yn是输入误差序列通过滤波器产生的序列；And use an adaptive filter to filter the error of tooth i; the adaptive filter in the ring gear error estimation module is an FIR filter of length M; in each time step, the filter coefficient is updated with the error en=dn-yn, where dn is the desired response, and yn is the sequence generated by the input error sequence through the filter;

应当理解的是，对本领域普通技术人员来说，可以根据上述说明加以改进或变换，而所有这些改进和变换都应属于本发明所附权利要求的保护范围。It should be understood that those of ordinary skill in the art can make improvements or transformations based on the above description, and all these improvements and transformations should fall within the protection scope of the appended claims of the present invention.

Claims

一种汽车齿圈角速度测量误差消除方法，其特征在于，包括以下步骤：A method for eliminating the measurement error of the angular velocity of an automobile ring gear is characterized in that it comprises the following steps:

1)设旋转轴上安装有含有N个齿的齿圈，读取齿圈信号，获得齿k经过的时刻t _k，设齿i的误差为δi，则消除误差的角速度ω _k’＝ω _k-δi,i＝k mod N； 1) Suppose a ring gear containing N teeth is installed on the rotating shaft, read the signal of the ring gear, and obtain the time t _k when the tooth k passes, and set the error of the tooth i as δi, then the angular velocity ω _k' = ω _{k to} eliminate the error -δi,i=k mod N;

2)齿与齿之间的时间间隔Δti(k)＝t _k-t _k-1，一个完整周期时间为Ti(k)＝t _k+N/2–t _k-N/2，则Δti/Ti＝1/N+Δi(+ek)，齿i的误差δi＝2πΔi＝2π(Δti/Ti-1/N)+ek； 2) The time interval between teeth Δti(k)=t _k -t _k-1 , a complete cycle time is Ti(k)=t _k+N/2 -t _kN/2 , then Δti/Ti= 1/N+Δi(+ek), the error of tooth i δi=2πΔi=2π(Δti/Ti-1/N)+ek;

3)使用自适应滤波器对齿i的误差进行滤波处理；3) Use an adaptive filter to filter the error of tooth i;

4)使用齿i的误差对轮齿按对应关系对角速度进行修正。4) Use the error of tooth i to correct the angular velocity of the tooth according to the corresponding relationship.
根据权利要求1所述的汽车齿圈角速度测量误差消除方法，其特征在于，所述步骤3)中自适应滤波器是长度为M的FIR滤波器。The method for eliminating the measurement error of the angular velocity of the automobile ring gear according to claim 1, wherein the adaptive filter in the step 3) is an FIR filter with a length of M.
根据权利要求2所述的汽车齿圈角速度测量误差消除方法，其特征在于，所述步骤3)中在每个时间步长中，滤波器系数都使用误差进行更新en＝dn-yn，其中dn是所需的响应，yn是输入误差序列通过滤波器产生的序列。The method for eliminating the measurement error of the automobile ring gear angular velocity according to claim 2, wherein, in the step 3), in each time step, the filter coefficients are updated with the error en=dn-yn, where dn Is the desired response, and yn is the sequence generated by the input error sequence through the filter.
根据权利要求2所述的汽车齿圈角速度测量误差消除方法，其特征在于，所述步骤3)中损失函数：MSE ei(k)＝(2π*(Δti(k)/Ti(k)-1/N)+δi(k+1))。The method for eliminating the measurement error of the automobile ring gear angular velocity according to claim 2, wherein the loss function in step 3): MSE ei(k)=(2π*(Δti(k)/Ti(k)-1 /N)+δi(k+1)).
一种汽车齿圈角速度测量误差消除装置，其特征在于，包括：An error elimination device for measuring angular velocity of automobile ring gear, which is characterized in that it comprises:

齿圈信号读取和储存模块，用于对旋转轴上安装有含有N个齿的齿圈，读取齿圈信号，获得齿k经过的时刻t _k，并存储数据； The ring gear signal reading and storage module is used to read the ring gear signal for a ring gear with N teeth installed on the rotating shaft, obtain the time t _k when the tooth k passes, and store the data;

齿圈误差估计模块，用于根据齿圈信号获得齿i的误差δi＝2πΔi＝2π(Δti/Ti-1/N)+ek；Ring gear error estimation module, used to obtain tooth i error δi=2πΔi=2π(Δti/Ti-1/N)+ek according to the ring gear signal;

并使用自适应滤波器对齿i的误差进行滤波处理；And use an adaptive filter to filter the error of tooth i;

误差修正模块，使用齿i的误差对轮齿按对应关系对角速度进行修正，消除误差的角速度ω _k’＝ω _k-δi,i＝k mod N。 The error correction module uses the error of the tooth i to correct the angular velocity of the tooth according to the corresponding relationship, and eliminates the angular velocity of the error ω _k' = ω _k- δi, i = k mod N.
根据权利要求5所述的汽车齿圈角速度测量误差消除装置，其特征在于，所述齿圈误差估计模块中自适应滤波器是长度为M的FIR滤波器。5. The automobile ring gear angular velocity measurement error elimination device according to claim 5, wherein the adaptive filter in the ring gear error estimation module is an FIR filter with a length of M.
根据权利要求6所述的汽车齿圈角速度测量误差消除装置，其特征在于，所述齿圈误差估计模块中在每个时间步长中，滤波器系数都使用误差进行更新en＝dn-yn，其中dn是所需的响应，yn是输入误差序列通过滤波器产生的序列。The automobile ring gear angular velocity measurement error elimination device according to claim 6, characterized in that, in each time step in the ring gear error estimation module, the filter coefficients are updated with error en=dn-yn, Where dn is the desired response, and yn is the sequence generated by the input error sequence through the filter.