WO2021057894A1 - 一种基于车辆零速检测的惯性导航误差修正方法 - Google Patents

一种基于车辆零速检测的惯性导航误差修正方法 Download PDF

Info

Publication number
WO2021057894A1
WO2021057894A1 PCT/CN2020/117703 CN2020117703W WO2021057894A1 WO 2021057894 A1 WO2021057894 A1 WO 2021057894A1 CN 2020117703 W CN2020117703 W CN 2020117703W WO 2021057894 A1 WO2021057894 A1 WO 2021057894A1
Authority
WO
WIPO (PCT)
Prior art keywords
zero
speed
vehicle
data
gyroscope
Prior art date
Application number
PCT/CN2020/117703
Other languages
English (en)
French (fr)
Inventor
熊璐
魏琰超
夏新
陆逸适
高乐天
宋舜辉
Original Assignee
同济大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 同济大学 filed Critical 同济大学
Publication of WO2021057894A1 publication Critical patent/WO2021057894A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01CMEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/10Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration
    • G01C21/12Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning
    • G01C21/16Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation
    • G01C21/165Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation combined with non-inertial navigation instruments
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01CMEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C25/00Manufacturing, calibrating, cleaning, or repairing instruments or devices referred to in the other groups of this subclass
    • G01C25/005Manufacturing, calibrating, cleaning, or repairing instruments or devices referred to in the other groups of this subclass initial alignment, calibration or starting-up of inertial devices

Definitions

  • the invention relates to the field of vehicle navigation and positioning, in particular to an inertial navigation error correction method based on vehicle zero-speed detection.
  • Inertial navigation system Inertial Navigation System, INS
  • INS Inertial Navigation System
  • inertial navigation systems have the advantages of independent work, high accuracy, and resistance to electronic interference, but the most obvious shortcoming is that their navigation errors will accumulate over time, so they will produce different degrees of accumulated errors after long hours of work.
  • the zero-speed correction technology is the most practical constraint method for overcoming the divergence of accumulated errors in inertial navigation positioning and attitude determination.
  • This technology includes two parts: zero-speed zone detection and zero-speed update. Among them, zero-speed detection is the basis for zero-speed update.
  • the zero-speed detection information can be used as a reference for the INS relative to the geostationary state, and the speed, attitude, and IMU (Inertial Measurement Unit) errors of the INS can be calibrated through quasi-static alignment.
  • the traditional INS zero-speed detection technology can be used to correct the inertial navigation error.
  • the traditional INS zero-speed detection technology uses the data collected by the vehicle accelerometer and gyroscope as the detection quantity. However, the detection accuracy of this method is not high enough, and The drift of navigation error cannot be suppressed.
  • the purpose of the present invention is to overcome the above-mentioned defects in the prior art and provide an inertial navigation error correction method based on vehicle zero-speed detection with high correction accuracy.
  • An inertial navigation error correction method based on vehicle zero-speed detection including a zero-speed detection step and a zero-speed correction step:
  • Three-axis acceleration data and three-axis gyroscope are used for data collection.
  • the vehicle INS data includes accelerometer data and gyroscope data.
  • the zero speed detection method is used to detect the zero speed of the vehicle in real time. If the zero speed is detected, the zero speed correction step is entered.
  • H 0 means that the carrier is in a moving state
  • H 1 means that the carrier is in a static state
  • H 1 ) and the false alarm probability P FA p(H 1
  • H 0 ) decides, according to Neyman-Pearson theory, for a given P FA ⁇ , the detection probability P D is maximized when the state is judged as H 1 , which needs to satisfy:
  • L(q n ) is the likelihood ratio of the observation sequence q n , where the threshold ⁇ is determined by the following formula:
  • the observation window period is from n to n+N-1, n is the data at the start time of the window period, N is the number of data in the window period, l is any data in the window period, and g is the acceleration of gravity.
  • n is the data at the start time of the window period
  • N is the number of data in the window period
  • l is any data in the window period
  • g is the acceleration of gravity.
  • ⁇ ' -2(ln ⁇ )/N, and ⁇ is the threshold; among them Is the observation vector of the accelerometer at time k, Is the observation vector of the gyroscope at time k, Is the observation vector of the wheel speed sensor at time k; They are the noise variance values of the accelerometer, gyroscope, and wheel speed sensor.
  • a x , a y , a z are the output of the three-axis accelerometer
  • the observation window period is from n to n+N-1, where n is the data at the start time of the window period, N is the number of data in the window period, and l is any data in the window period.
  • ⁇ x , ⁇ y , ⁇ z are the output angular velocities of the original three-axis gyroscope.
  • the present invention has the following advantages:
  • the present invention is oriented to vehicle integrated navigation, and proposes a zero-speed detection method and an inertial navigation error correction method. Compared with the traditional INS zero-speed detection, the method adds a wheel speed sensor as a new detection quantity. Speed sensor and INS data, using Riemann Pearson hypothesis theory to construct the maximum likelihood ratio of the multi-sensor joint probability density, to detect the zero-speed state of the vehicle, after the zero-speed of the vehicle is detected, the speed and attitude are measured by the zero-speed information. Update and estimate the zero offset of the gyroscope and the roll angle and pitch angle of the vehicle. The method of the present invention improves the accuracy of detection;
  • the method of the present invention takes into account the characteristics of the vehicle that needs to start and stop, and can be effectively applied to the process of vehicle driving, using zero-speed detection and zero-speed correction to suppress the drift of navigation error, and effectively correct the accumulated navigation error .
  • Figure 1 is a schematic flow diagram of the method of the present invention.
  • the invention relates to an inertial navigation error correction method based on vehicle zero-speed detection.
  • the invention is divided into two modules, namely a zero-speed detection module and a zero-speed correction module.
  • the zero-speed detection module is the input of the zero-speed correction module, which provides the judgment of whether the vehicle is at zero speed, hereinafter referred to as the zero-speed Flag signal. If the vehicle is at zero speed, the zero speed correction module is run; if the vehicle is not at zero speed, the zero speed correction module is not allowed.
  • the zero-speed detection module includes the following steps:
  • Step 1 After the vehicle is started, collect the wheel speed data collected by the accelerometer, gyroscope and wheel speed sensor.
  • the acceleration variance detector Through the acceleration variance detector, acceleration amplitude detector, and angular velocity energy detector, the acceleration variance, acceleration amplitude, and angular velocity energy are calculated.
  • three-axis acceleration data and a three-axis gyroscope are used for data collection. That is, the data collected during the driving of the vehicle includes inertial measurement data, that is, three-axis acceleration data, three-axis gyroscope data, and wheel speed data collected by a wheel speed sensor.
  • Step 2 Calculate the generalized likelihood ratio of the joint probability of all sensors by constructing the maximum likelihood estimation of the distribution parameters.
  • ⁇ 3 represents a three-dimensional unit matrix
  • 0 1 ⁇ 3 (0 3 ⁇ 1 ) represents a zero matrix with a size of 1 ⁇ 3 (3 ⁇ 1)
  • Zero-speed detection can be studied as a binary hypothesis testing problem, and the hypothesis is defined as:
  • H 1 The carrier is in a static state
  • H 1 )) and the false alarm probability (P FA p(H 1
  • P D p(H 0
  • P FA p(H 1
  • L(q n ) is the likelihood ratio of the signal q n
  • this formula is the Likelihood Ratio Test.
  • the threshold ⁇ is determined by the following formula:
  • the output of the three-axis accelerometer should be the local acceleration of gravity
  • the output of the three-axis gyroscope should be 0,
  • the output of the wheel speed sensor should also be Is 0.
  • Step 3 Pass the generalized likelihood ratio test, compare the generalized likelihood ratio with the set test threshold, and output the result of zero-speed detection. That is, if the zero-speed detection does not meet the requirements of the above formula, and the zero-speed Flag output by the zero-speed detection is Fasle, the zero-speed correction module is not operated and the INS outputs normally. If the zero-speed detection meets the requirements of the above formula and the zero-speed Flag output by the zero-speed detection is True, then the zero-speed correction module is run.
  • the zero-speed correction module first detects the zero-speed Flag signal. If the zero-speed Flag signal is True, the zero-speed correction module is run; if the zero-speed Flag signal is False, the zero-speed correction module stops.
  • the zero-speed correction module includes the following three steps:
  • Error estimation Use the output of the sensor when the speed is zero to re-estimate the zero bias of the gyroscope.

Landscapes

  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Manufacturing & Machinery (AREA)
  • Automation & Control Theory (AREA)
  • Navigation (AREA)
  • Gyroscopes (AREA)

Abstract

一种基于车辆零速检测的惯性导航误差修正方法,在车辆行驶时,采集轮速传感数据及INS数据,在车辆行驶过程中零速检测算法通过轮速、加速度计、陀螺仪进行零速检测;零速检测的原理是对轮速、加速度计、陀螺仪信号进行基于黎曼皮尔逊准则的联合概率,零速修正的原理是:1、在车辆检测出零速时,对速度和角速度进行置零处理,防止速度误差和角速度误差累积,导致位置和姿态出现偏差;2、在检测出零速后通过加速度计和角速度计的输出,分别对姿态、陀螺仪零偏进行再估计;3、在车辆再次运动退出零速状态后,利用再估计出的姿态和陀螺仪零偏对INS进行修正。与现有技术相比,该方法具有抑制导航误差的漂移,提高检测的准确度等优点。

Description

一种基于车辆零速检测的惯性导航误差修正方法 技术领域
本发明涉及车载导航与定位领域,尤其是涉及一种基于车辆零速检测的惯性导航误差修正方法。
背景技术
惯性导航系统(Inertial Navigation System,INS)能根据惯性传感器(陀螺仪、加速度计)提供的载体相对于惯性空间的线速度和角速度信息,成为车载组合定位系统关键部件。惯性导航系统具有独立工作、精度高、不易被电子干扰的优点,但最明显的缺点是其导航误差会随时间积累,因而长时间工作后会产生不同程度的积累误差。
零速修正技术是克服惯导定位定姿累计误差发散最为实用的一种约束方法,该技术包含零速区间探测和零速更新两个部分,其中,零速检测是零速更新的基础。此外,零速检测信息可以作为INS相对于地球静止的参考,通过准静态对准来标定INS的速度、姿态和IMU(Inertial measurement unit,惯性测量单元)误差。利用传统的INS零速检测技术可对惯性导航误差进行修正,传统的INS零速检测技术将车辆加速度计和陀螺仪采集的数据作为检测量,然而,该方法的检测的准确度不够高,且无法抑制导航误差的漂移。
发明内容
本发明的目的就是为了克服上述现有技术存在的缺陷而提供一种修正准确率高的基于车辆零速检测的惯性导航误差修正方法。
本发明的目的可以通过以下技术方案来实现:
一种基于车辆零速检测的惯性导航误差修正方法,包括零速检测步骤和零速修正步骤:
零速检测步骤:
采用三轴加速度数据、三轴陀螺仪进行数据采集。
1.1、车辆启动后,采集车辆INS数据及轮速传感器的轮速数据,车辆INS数据包括加速度计数据和陀螺仪数据。
假设某一时刻k的所有传感器观测值为y k
Figure PCTCN2020117703-appb-000001
式中,
Figure PCTCN2020117703-appb-000002
为k时刻三轴加速度计的观测向量,
Figure PCTCN2020117703-appb-000003
为k时刻三轴陀螺仪的观测向量,
Figure PCTCN2020117703-appb-000004
为k时刻轮速传感器的观测向量。选择观测窗口期为n至n+N-1,n为窗口期开始时间点的数据,N为窗口期数据数量,得到一组连续观测序列
Figure PCTCN2020117703-appb-000005
1.2、基于采集的数据,在车辆静止时,利用零速检测方法对车辆零速进行实时检测,若检测出零速,则进入零速修正步骤。
121)基于采集的数据,构建分布参数的最大似然估计,计算所有传感器联合概率的广义似然比。具体地:
a)定义H 0表示载体为运动状态,H 1表示载体为静止状态,零速检测器的性能由检测概率P D=p(H 0|H 1)和虚警概率P FA=p(H 1|H 0)决定,根据Neyman-Pearson理论,对于给定的P FA=α,使判断为H 1状态时检测概率P D最大,需满足:
Figure PCTCN2020117703-appb-000006
式中,L(q n)为观测序列q n的似然比,其中,阈值γ由下式决定:
Figure PCTCN2020117703-appb-000007
b)考虑到假设H 0下信号较难描述,但H 1情况下车辆静止且比力只有重力,三轴加速度计输出应为当地重力加速度,三轴陀螺仪输出应为0,轮速传感器输出也应为0,则设定信号约束条件为:
Figure PCTCN2020117703-appb-000008
or
Figure PCTCN2020117703-appb-000009
or
Figure PCTCN2020117703-appb-000010
Figure PCTCN2020117703-appb-000011
then
Figure PCTCN2020117703-appb-000012
and
Figure PCTCN2020117703-appb-000013
and
Figure PCTCN2020117703-appb-000014
式中,
Figure PCTCN2020117703-appb-000015
为方向为重力方向的单位矢量,
Figure PCTCN2020117703-appb-000016
观测窗口期为n至n+N-1,n为窗口期开始时间点的数据,N为窗口期数据数量,l为窗口期中任意的一个数据,g为重力加速度,
Figure PCTCN2020117703-appb-000017
分别为加速度计、陀螺仪及轮速传感器真实值;
Figure PCTCN2020117703-appb-000018
分别为加速度计、陀螺仪及轮速传感器的测量噪声。
c)结合步骤a)、b)获取所有传感器联合概率的广义似然比的表达式。
所有传感器联合概率的广义似然比的表达式为:
Figure PCTCN2020117703-appb-000019
式中,γ'=-2(lnγ)/N,γ为阈值;
Figure PCTCN2020117703-appb-000020
其中
Figure PCTCN2020117703-appb-000021
为k时刻加速度计的观测向量,
Figure PCTCN2020117703-appb-000022
为k时刻陀螺仪的观测向量,
Figure PCTCN2020117703-appb-000023
为k时刻轮速传感器的观测向量;
Figure PCTCN2020117703-appb-000024
分别为加速度计、陀螺仪及轮速传感器的噪声方差值。
d)计算IMU及轮速传感器输出数据,若满足所有传感器联合概率的广义似然比的表达式,则认为载体处于静止状态。
122)计算IMU及轮速传感器输出数据,若满足所有传感器联合概率的广义似然比,则认为载体处于静止状态,即检测出零速。
零速修正步骤:
2.1、利用零速状态下的加速度计数据,对车辆姿态角中的俯仰角θ和翻滚角Φ的估计,其表达式为:
Figure PCTCN2020117703-appb-000025
Figure PCTCN2020117703-appb-000026
式中,a x、a y、a z为三轴加速度计输出,
Figure PCTCN2020117703-appb-000027
观测窗口期为n至n+N-1,n为窗口期开始时间点的数据,N为窗口期数据数量,l为窗口期中任意的一个数据。
2.2、利用零速时各传感器的输出,对陀螺仪的零偏进行再估计。
假设陀螺仪零偏为[b x b y b z],利用零速时各传感器的输出,对陀螺仪的零偏进行再估计的表达式为:
Figure PCTCN2020117703-appb-000028
式中,ω xyz分别为原始的三轴陀螺仪的输出角速度。
2.3、当车辆启动后,利用估计的车辆姿态及陀螺仪的零偏对INS误差进行修正。修正方法的表达式为:
Figure PCTCN2020117703-appb-000029
式中,
Figure PCTCN2020117703-appb-000030
分别为修正后的三轴陀螺仪的输出角速度。
与现有技术相比,本发明具有以下优点:
一、本发明面向车载组合导航,提出了一种零速检测方法和惯导误差修正方法,该方法相比于传统的INS零速检测,加入了轮速传感器作为新的检测量,通过采集轮速传感器和INS数据,利用黎曼皮尔逊假设理论构造多传感器联合概率密度的最大似然比,对车辆零速状态进行检测,在检测出车辆零速后,通过零速信息对速度、姿态进行更新,并对陀螺仪零偏以及车辆这翻滚角、俯仰角进行估计,本发明方法提高了检测的准确度;
二、本发明方法考虑到车辆需要启停的特性,能够有效地应用在车辆行驶的过程中,利用零速检测和零速修正对导航误差的漂移进行抑制,并对导航累计误差进行有效的修正。
附图说明
图1为本发明方法的流程示意图。
具体实施方式
下面结合附图和具体实施例对本发明进行详细说明。显然,所描述的实施例是本发明的一部分实施例,而不是全部实施例。基于本发明中的实施例,本领域普通技术人员在没有做出创造性劳动的前提下所获得的所有其他实施例,都应属于本发明保护的范围。
本发明涉及一种基于车辆零速检测的惯性导航误差修正方法,本发明分为两个模块,分别是零速检测模块和零速修正模块。零速检测模块为零速修正模块的输入,提供车辆是否处于零速的判断,以下称零速Flag信号。若车辆处于零速,则运行零速修正模块;若车辆不处于零速状态,则不允许零速修正模块。
零速检测模块包括以下步骤:
步骤1、车辆启动后,采集加速度计、陀螺仪数据和轮速传感器采集的轮速数 据。
通过加速度方差检测器、加速度幅值检测器、角速度能量检测器,计算得到加速度方差、加速度幅值、角速度能量。优选地,采用三轴加速度数据、三轴陀螺仪进行数据采集。即车辆行驶过程中采集的数据包括惯性测量数据,即三轴加速度数据、三轴陀螺仪数据以及轮速传感器采集的轮速数据。
假设某一时刻k的所有传感器观测值为y k
Figure PCTCN2020117703-appb-000031
式中,
Figure PCTCN2020117703-appb-000032
为k时刻三轴加速度计的观测向量,
Figure PCTCN2020117703-appb-000033
为k时刻三轴陀螺仪的观测向量,
Figure PCTCN2020117703-appb-000034
为k时刻轮速传感器的观测向量。选择观测窗口期为n至n+N-1,n为窗口期开始时间点的数据,N为窗口期数据数量,得到一组连续观测序列
Figure PCTCN2020117703-appb-000035
步骤2、通过构建分布参数的最大似然估计,计算所有传感器联合概率的广义似然比。
假设传感器测量值可以用下式进行描述:
y k=s k+v k
其中:
Figure PCTCN2020117703-appb-000036
Figure PCTCN2020117703-appb-000037
式中,
Figure PCTCN2020117703-appb-000038
分别表示三轴加速度计、三轴陀螺仪及轮速传感器真实值;
Figure PCTCN2020117703-appb-000039
分别表示三轴加速度计、三轴陀螺仪及轮速传感器的测量噪声。假设示三轴加速度计、三轴陀螺仪及轮速传感器的噪声项均为独立分布的高斯白噪声,即协方差矩阵为:
Figure PCTCN2020117703-appb-000040
式中,Ι 3表示三维单位矩阵,0 1×3(0 3×1)表示大小为1×3(3×1)的零矩阵;
Figure PCTCN2020117703-appb-000041
Figure PCTCN2020117703-appb-000042
分别表示三轴加速度计、三轴陀螺仪及轮速传感器的噪声方差值。
零速检测可作为二元假设检验问题进行研究,并定义假设为:
H 0:载体为运动状态
H 1:载体为静止状态
零速检测器的性能由检测概率(P D=p(H 0|H 1))和虚警概率(P FA=p(H 1|H 0))决定。根据Neyman-Pearson理论,对于给定的P FA=α,要使得判断为H 1状态时检测概率P D最大,有:
Figure PCTCN2020117703-appb-000043
式中,L(q n)为信号q n的似然比,该式即为似然比检验(Likelihood Ratio Test)。式中,阈值γ由下式决定:
Figure PCTCN2020117703-appb-000044
考虑到假设H 0下信号较难描述,但H 1情况下车辆静止且比力只有重力,三轴加速度计输出应为当地重力加速度,三轴陀螺仪输出应为0,轮速传感器输出也应为0。两种假设条件下的信号约束应为:
Figure PCTCN2020117703-appb-000045
or
Figure PCTCN2020117703-appb-000046
or
Figure PCTCN2020117703-appb-000047
Figure PCTCN2020117703-appb-000048
then
Figure PCTCN2020117703-appb-000049
and
Figure PCTCN2020117703-appb-000050
and
Figure PCTCN2020117703-appb-000051
式中,
Figure PCTCN2020117703-appb-000052
为方向为重力方向的单位矢量,
Figure PCTCN2020117703-appb-000053
l表示窗口期中任意的一个数据,g为重力加速度。推导得到融合轮速的零速检测器似然比检验式如下:
Figure PCTCN2020117703-appb-000054
式中,γ'=-2(lnγ)/N。
计算IMU及轮速传感器输出数据,若满足上式,则认为载体处于静止状态。
步骤3、通过广义似然比检验,比较广义似然比与设定的检验阈值,并输出零速检测的结果。即若零速检测不符合上式要求,零速检测输出的零速Flag为Fasle,则不运行零速修正模块,INS正常输出。若零速检测符合上式要求,零速检测输出的零速Flag为True,则运行零速修正模块。
若运行零速修正模块,则零速修正模块首先检测零速Flag信号,若零速Flag信号为True,则运行零速修正模块;若零速Flag信号为False,则停止运行零速修 正模块。
零速修正模块包括以下三个步骤:
1)速度姿态位置更新:
利用零速信息,对速度和位置进行置零更新,即v=0,[ω x ω y ω z]=[0 0 0],式中ω x、ω y、ω z分别为原始的三轴陀螺仪的角速度输出。此时由于速度和角速度输入均为0,则载体的位置和姿态会维持停止前状态,不会进行更新。此时陀螺仪噪声及加速度计噪声无法对车辆位姿造成影响。
利用零速状态下的加速度计,对车辆姿态进行重新估计。具体地:
假设加速度输出
Figure PCTCN2020117703-appb-000055
则姿态角中的俯仰角θ和翻滚角Φ可以通过加速度进行计算:
Figure PCTCN2020117703-appb-000056
Figure PCTCN2020117703-appb-000057
2)误差估计:利用零速时候传感器的输出,对陀螺仪的零偏进行再估计。
假设陀螺仪零偏为[b x b y b z],利用零速状态下陀螺仪的输出对陀螺仪零偏重新进行估计,具体如下式:
Figure PCTCN2020117703-appb-000058
3)零偏、姿态修正:当车辆启动后,利用步骤1)、步骤2)估计的车辆姿态和陀螺零偏对INS误差进行修正。修正方法如下式所示:
Figure PCTCN2020117703-appb-000059
式中,
Figure PCTCN2020117703-appb-000060
分别表示修正后的三轴陀螺仪输出角速度。
以上所述,仅为本发明的具体实施方式,但本发明的保护范围并不局限于此,任何熟悉本技术领域的工作人员在本发明揭露的技术范围内,可轻易想到各种等效的修改或替换,这些修改或替换都应涵盖在本发明的保护范围之内。因此,本发明 的保护范围应以权利要求的保护范围为准。

Claims (8)

  1. 一种基于车辆零速检测的惯性导航误差修正方法,其特征在于,包括:
    零速检测步骤:
    11)车辆启动后,采集车辆INS数据及轮速传感器的轮速数据,车辆INS数据包括加速度计数据和陀螺仪数据;
    12)基于采集的数据,在车辆静止时,利用零速检测方法对车辆零速进行实时检测,若检测出零速,则进入零速修正步骤;
    零速修正步骤:
    21)利用零速状态下的加速度计,对车辆姿态进行重新估计;
    22)利用零速时各传感器的输出,对陀螺仪的零偏进行再估计;
    23)当车辆启动后,利用估计的车辆姿态及陀螺仪的零偏对INS误差进行修正。
  2. 根据权利要求1所述的一种基于车辆零速检测的惯性导航误差修正方法,其特征在于,所述的零速检测步骤中,步骤12)的具体内容为:
    121)基于采集的数据,构建分布参数的最大似然估计,计算所有传感器联合概率的广义似然比;
    122)计算IMU及轮速传感器输出数据,若满足所有传感器联合概率的广义似然比,则认为载体处于静止状态,即检测出零速。
  3. 根据权利要求2所述的一种基于车辆零速检测的惯性导航误差修正方法,其特征在于,步骤121)具体包括以下内容:
    a)定义H 0表示载体为运动状态,H 1表示载体为静止状态,零速检测器的性能由检测概率P D=p(H 0|H 1)和虚警概率P FA=p(H 1|H 0)决定,根据Neyman-Pearson理论,对于给定的P FA=α,使判断为H 1状态时检测概率P D最大,需满足:
    Figure PCTCN2020117703-appb-100001
    式中,L(q n)为观测序列q n的似然比,其中,阈值γ由下式决定:
    Figure PCTCN2020117703-appb-100002
    b)设定信号约束条件,即:
    Figure PCTCN2020117703-appb-100003
    Figure PCTCN2020117703-appb-100004
    式中,
    Figure PCTCN2020117703-appb-100005
    为方向为重力方向的单位矢量,
    Figure PCTCN2020117703-appb-100006
    观测窗口期为n至n+N-1,n为窗口期开始时间点的数据,N为窗口期数据数量,l为窗口期中任意的一个数据,g为重力加速度,
    Figure PCTCN2020117703-appb-100007
    分别为加速度计、陀螺仪及轮速传感器真实值;
    Figure PCTCN2020117703-appb-100008
    分别为加速度计、陀螺仪及轮速传感器的测量噪声;
    c)结合步骤a)、b)获取所有传感器联合概率的广义似然比的表达式;
    d)计算IMU及轮速传感器输出数据,若满足所有传感器联合概率的广义似然比的表达式,则认为载体处于静止状态。
  4. 根据权利要求3所述的一种基于车辆零速检测的惯性导航误差修正方法,其特征在于,所有传感器联合概率的广义似然比的表达式为:
    Figure PCTCN2020117703-appb-100009
    式中,γ'=-2(lnγ)/N,γ为阈值;
    Figure PCTCN2020117703-appb-100010
    其中
    Figure PCTCN2020117703-appb-100011
    为k时刻加速度计的观测向量,
    Figure PCTCN2020117703-appb-100012
    为k时刻陀螺仪的观测向量,
    Figure PCTCN2020117703-appb-100013
    为k时刻轮速传感器的观测向量;
    Figure PCTCN2020117703-appb-100014
    分别为加速度计、陀螺仪及轮速传感器的噪声方差值。
  5. 根据权利要求1所述的一种基于车辆零速检测的惯性导航误差修正方法,其特征在于,采用三轴加速度数据、三轴陀螺仪进行数据采集。
  6. 根据权利要求5所述的一种基于车辆零速检测的惯性导航误差修正方法,其特征在于,步骤21)的具体内容为:
    利用零速状态下的加速度计数据,对车辆姿态角中的俯仰角θ和翻滚角Φ的估计,其表达式为:
    Figure PCTCN2020117703-appb-100015
    Figure PCTCN2020117703-appb-100016
    式中,a x、a y、a z为三轴加速度计输出,
    Figure PCTCN2020117703-appb-100017
    观测窗口期为n至n+N-1,n为窗口期开始时间点的数据,N为窗口期数据数量,l为窗口期中任意的一个数据。
  7. 根据权利要求5所述的一种基于车辆零速检测的惯性导航误差修正方法,其特征在于,步骤22)中,假设陀螺仪零偏为[b x b y b z],则利用零速时各传感器的输出,对陀螺仪的零偏进行再估计的表达式为:
    Figure PCTCN2020117703-appb-100018
    式中,w x,w y,w z分别为原始的三轴陀螺仪的输出角速度。
  8. 根据权利要求7所述的一种基于车辆零速检测的惯性导航误差修正方法,其特征在于,步骤23)中,修正方法的表达式为:
    Figure PCTCN2020117703-appb-100019
    式中,
    Figure PCTCN2020117703-appb-100020
    分别为修正后的三轴陀螺仪的输出角速度。
PCT/CN2020/117703 2019-09-27 2020-09-25 一种基于车辆零速检测的惯性导航误差修正方法 WO2021057894A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910924281.2 2019-09-27
CN201910924281.2A CN110702104B (zh) 2019-09-27 2019-09-27 一种基于车辆零速检测的惯性导航误差修正方法

Publications (1)

Publication Number Publication Date
WO2021057894A1 true WO2021057894A1 (zh) 2021-04-01

Family

ID=69196718

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/117703 WO2021057894A1 (zh) 2019-09-27 2020-09-25 一种基于车辆零速检测的惯性导航误差修正方法

Country Status (2)

Country Link
CN (1) CN110702104B (zh)
WO (1) WO2021057894A1 (zh)

Families Citing this family (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110702104B (zh) * 2019-09-27 2023-09-26 同济大学 一种基于车辆零速检测的惯性导航误差修正方法
CN111693723B (zh) * 2020-05-29 2022-04-22 阿波罗智联(北京)科技有限公司 速度预测方法、装置和电子设备
CN111750897B (zh) * 2020-07-03 2022-01-28 南京晓庄学院 一种基于龙贝格观测器的横摆角速度陀螺仪偏差估计方法
CN112284378A (zh) * 2020-09-17 2021-01-29 坤泰车辆系统(常州)有限公司 惯性测量单元零漂移的自动驾驶控制方法
CN112815980A (zh) * 2020-12-31 2021-05-18 天通盛邦通信科技(苏州)有限公司 一种动中收传感器自动校准方法
CN112577527B (zh) * 2021-02-25 2021-09-17 北京主线科技有限公司 车载imu误差标定方法及装置
CN113092819B (zh) * 2021-04-14 2022-11-18 东方红卫星移动通信有限公司 足部加速度计动态零速校准方法及系统
CN115246417B (zh) * 2021-07-29 2023-08-25 上海仙途智能科技有限公司 作业执行方法、装置、设备及计算机可读存储介质
CN114120252B (zh) * 2021-10-21 2023-09-01 阿波罗智能技术(北京)有限公司 自动驾驶车辆状态的识别方法、装置、电子设备及车辆
CN114370885B (zh) * 2021-10-29 2023-10-13 北京自动化控制设备研究所 一种惯性导航系统误差补偿方法及系统
CN114019182B (zh) * 2021-11-04 2024-02-02 苏州挚途科技有限公司 零速状态检测方法、装置及电子设备
CN114440867A (zh) * 2021-12-17 2022-05-06 际络科技(上海)有限公司 重卡的零速检测方法及装置

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090254276A1 (en) * 2008-04-08 2009-10-08 Ensco, Inc. Method and computer-readable storage medium with instructions for processing data in an internal navigation system
CN104296750A (zh) * 2014-06-27 2015-01-21 大连理工大学 一种零速检测方法和装置以及行人导航方法和系统
CN104897155A (zh) * 2015-06-05 2015-09-09 北京信息科技大学 一种个人携行式多源定位信息辅助修正方法
CN105043385A (zh) * 2015-06-05 2015-11-11 北京信息科技大学 一种行人自主导航定位的自适应卡尔曼滤波方法
CN105783923A (zh) * 2016-01-05 2016-07-20 山东科技大学 基于rfid和mems惯性技术的人员定位方法
CN106017461A (zh) * 2016-05-19 2016-10-12 北京理工大学 基于人体/环境约束的行人导航系统三维空间定位方法
CN106153069A (zh) * 2015-03-31 2016-11-23 日本电气株式会社 自主导航系统中的姿态修正装置和方法
CN107869998A (zh) * 2016-09-26 2018-04-03 恩智浦美国有限公司 使用胎压监测系统信号来校准惯性导航数据
CN110057381A (zh) * 2019-03-20 2019-07-26 深圳市元征科技股份有限公司 一种导航系统的零速修正方法及系统
CN110702104A (zh) * 2019-09-27 2020-01-17 同济大学 一种基于车辆零速检测的惯性导航误差修正方法

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101447617B1 (ko) * 2013-04-01 2014-10-07 (주)나노포인트 교통정지 표지판 인식 후 가속센서 바이어스 값을 이용한 정차 및 무정차 검출 장치 및 방법
CN103499354B (zh) * 2013-09-24 2017-01-18 哈尔滨工程大学 一种基于内曼‑皮尔逊准则的零速检测方法

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090254276A1 (en) * 2008-04-08 2009-10-08 Ensco, Inc. Method and computer-readable storage medium with instructions for processing data in an internal navigation system
CN104296750A (zh) * 2014-06-27 2015-01-21 大连理工大学 一种零速检测方法和装置以及行人导航方法和系统
CN106153069A (zh) * 2015-03-31 2016-11-23 日本电气株式会社 自主导航系统中的姿态修正装置和方法
CN104897155A (zh) * 2015-06-05 2015-09-09 北京信息科技大学 一种个人携行式多源定位信息辅助修正方法
CN105043385A (zh) * 2015-06-05 2015-11-11 北京信息科技大学 一种行人自主导航定位的自适应卡尔曼滤波方法
CN105783923A (zh) * 2016-01-05 2016-07-20 山东科技大学 基于rfid和mems惯性技术的人员定位方法
CN106017461A (zh) * 2016-05-19 2016-10-12 北京理工大学 基于人体/环境约束的行人导航系统三维空间定位方法
CN107869998A (zh) * 2016-09-26 2018-04-03 恩智浦美国有限公司 使用胎压监测系统信号来校准惯性导航数据
CN110057381A (zh) * 2019-03-20 2019-07-26 深圳市元征科技股份有限公司 一种导航系统的零速修正方法及系统
CN110702104A (zh) * 2019-09-27 2020-01-17 同济大学 一种基于车辆零速检测的惯性导航误差修正方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
LI, LIUSONG ET AL.: "Transducer and Microsystem Technologies", PEDESTRIAN ZERO VELOCITY DETECTION ALGORITHM BASED ON INERTIAL NAVIGATION, vol. 38, no. 3, 31 March 2019 (2019-03-31), XP009526923, ISSN: 1000-9787 *

Also Published As

Publication number Publication date
CN110702104B (zh) 2023-09-26
CN110702104A (zh) 2020-01-17

Similar Documents

Publication Publication Date Title
WO2021057894A1 (zh) 一种基于车辆零速检测的惯性导航误差修正方法
CN105300379B (zh) 一种基于加速度的卡尔曼滤波姿态估计方法及系统
JP5328252B2 (ja) ナビゲーションシステムの位置検出装置および位置検出方法
CN107289930B (zh) 基于mems惯性测量单元的纯惯性车辆导航方法
CN102692225B (zh) 一种用于低成本小型无人机的姿态航向参考系统
CN107560613B (zh) 基于九轴惯性传感器的机器人室内轨迹跟踪系统及方法
CN112505737B (zh) 一种gnss/ins组合导航方法
Youn et al. Combined quaternion-based error state Kalman filtering and smooth variable structure filtering for robust attitude estimation
CN106053879A (zh) 通过数据融合的失效操作的车辆速度估计
CN113029139B (zh) 基于运动检测的机场飞行区车辆差分北斗/sins组合导航方法
JP2009019992A (ja) 位置検出装置及び位置検出方法
CN106153069B (zh) 自主导航系统中的姿态修正装置和方法
CN114485641B (zh) 一种基于惯导卫导方位融合的姿态解算方法及装置
CN112650281B (zh) 多传感器三余度系统、控制方法、无人机、介质及终端
CN112362057B (zh) 基于零速修正与姿态自观测的惯性行人导航算法
JP5164645B2 (ja) カルマンフィルタ処理における繰り返し演算制御方法及び装置
CN112683269A (zh) 一种附有运动加速度补偿的marg姿态计算方法
JP2014240266A (ja) センサドリフト量推定装置及びプログラム
CN113008229B (zh) 一种基于低成本车载传感器的分布式自主组合导航方法
CN109916399B (zh) 一种阴影下的载体姿态估计方法
CN108828644B (zh) Gnss/mems紧组合导航系统中动态突变识别方法
CN115540854A (zh) 一种基于uwb辅助的主动定位方法、设备和介质
CN111553933B (zh) 一种应用于不动产测量基于优化的视觉惯性组合测量方法
CN110132267B (zh) 空天地一体化飞行器的光纤惯导系统及光纤惯导在轨对准方法
CN113985466A (zh) 一种基于模式识别的组合导航方法及系统

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20868031

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20868031

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 20868031

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: OTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 23/01/2023)

122 Ep: pct application non-entry in european phase

Ref document number: 20868031

Country of ref document: EP

Kind code of ref document: A1