CN113108808A

Movatterモバイル変換

Info

Publication number: CN113108808A
Application number: CN202110280143.2A
Authority: CN
Inventors: 费庆; 邵勇强; 刘其志; 郑子伟
Original assignee: Beijing Institute of Technology BIT
Current assignee: Beijing Institute of Technology BIT
Priority date: 2021-03-16
Filing date: 2021-03-16
Publication date: 2021-07-13
Anticipated expiration: 2041-03-16
Also published as: CN113108808B

Abstract

The invention discloses an online verification system and a verification method for a vehicle odometer, and belongs to the technical field of vehicle detection. The invention adopts the combined navigation system (GNSS + INS) to automatically calibrate the vehicle odometer, and can automatically select the road section meeting the calibration standard and the driving state meeting the calibration standard to calibrate the odometer when the GPS positioning effect is good. When the vehicle is considered to have the change of the driving condition, the verified odometer K value is automatically put into use, and the calibration data is automatically updated, so that the stability of the K value is ensured, and the automatic adjustment is carried out when the vehicle condition is changed for a long time. Meanwhile, the method saves verification key data, can perform historical backtracking and is convenient for analyzing verification results. The method can be widely applied to the field of vehicle travel mileage verification such as taximeters, freight vehicle odometers and the like.

Description

Vehicle odometer online verification system and method

Technical Field

The invention relates to a vehicle odometer and a verification system and method, belonging to the technical field of vehicle detection.

Background

In the course of the certification of a vehicle Odometer, a vehicle Odometer (Odometer) is generally used. The vehicle odometer is a wheel revolution pulse detection device which is arranged on a wheel spindle or a speed reducer and is used for counting the running mileage of a vehicle.

For example, a taximeter is a typical odometer application, and the mileage of the taximeter is calculated by the following method:

the odometer was first calibrated. The vehicle runs for a certain distance L under a standard verification scene_cal(e.g., 1 km, or a number of revolutions of the wheel equivalent to 1 km on the verification device), recording the number of pulses n detected by the odometer during the trip_calThen calculating the K value of the odometer, K being n_cal/L_calThe unit of K is: number of pulses per kilometer. In normal driving, the relationship between the driving mileage of the vehicle and the number of pulses detected by the odometer is: and L is n/K, so that the driving mileage of the vehicle is obtained.

The traditional vehicle odometer verification method comprises a roller-type verification method and is generally applied to verification work of a taxi odometer. The method comprises the steps of placing a front wheel of a vehicle on a rotary roller device, enabling the vehicle to run (the front wheel rotates) for a period of time, and calculating the K value of a speedometer according to the running mileage calculated by detection equipment and the number of pulses displayed by the speedometer of the taxi.

However, this assay method requires dedicated assay equipment, field, and assay personnel, and is prone to assay errors due to equipment or human factors. In addition, the verification method only simulates the actual running state of the vehicle, and although some compensation can be performed on the temperature, the tire pressure and the like in the process, the method still cannot completely meet the actual operation condition.

Another more commonly used odometer calibration scheme utilizes a vehicle calibration site with a standard straight road built therein, the length of which can meet the calibration requirements. And during detection, the detected vehicle is enabled to run linearly from the starting point to the end point of the road, and then the K value is calculated according to the actual running distance and the odometer pulse number. However, the method needs larger site conditions and standard lanes, has high site requirements and low use efficiency, is only suitable for specific occasions requiring high-precision vehicle mileage calibration, and is not suitable for large-scale application.

Currently, there are several new vehicle odometer verification methods disclosed. For example, chinese patent application "a taximeter calibrating apparatus based on high precision positioning technology" (CN201420363714.4) discloses a method for calibrating using a high precision navigation system (RTK differential GPS), which is to install a high precision GPS device on a vehicle, then obtain the precise position and the driving distance in real time during the normal driving of the vehicle, and calculate the K value of the odometer together with the actual number of pulses of the odometer. The method adopts high-precision differential GPS positioning, and the positioning precision reaches centimeter level. However, since this method requires a communication link to be established with the RTK reference station, there are various problems such as complicated equipment, narrow use range, and payment required in use, and it cannot be used when the satellite fails.

In addition, in documents such as the chinese patent application "taxi pricing and monitoring system based on GPS-GPRS" (CN201710170212.8), and the chinese master paper "research on GPS-GPRS taxi pricing monitoring system" (china marine university, 2011), methods for measuring and calibrating mileage based on an electronic map using GPS and an electronic map, wireless communication are proposed. The method uses an electronic map for auxiliary positioning, improves positioning precision and mileage calculation precision, and is commonly used in navigation positioning systems of mobile phones, such as Goods navigation and Baidu navigation. Although the method for detecting the odometer can improve the positioning accuracy of the vehicle through an accurate map, the method has the problems of high use cost, limited application range and the like because the method needs to carry out wireless communication with an electronic map server or a vehicle-mounted electronic map needs to be updated in time.

In documents such as 'design and development of BJ-a type multifunctional taximeter verification instrument system' (modern measurement and laboratory management, 2016), a method for implementing taximeter verification by using a combined navigation system is proposed. This method uses integrated navigation to calibrate the odometer in a process similar to that described above using differential GPS. However, the positioning accuracy of the integrated navigation system is generally in the meter level and is subjected to fixed-point test and experiment, so that the problems of manually specified calibration road sections, complex operation process and the like exist.

In summary, the existing vehicle odometer verification method mainly has the following problems and disadvantages:

1. the drum-type odometer identification method needs special equipment, places and personnel, the identification result is influenced by tire pressure, temperature, load and the like, and the identification result can be maliciously adjusted by drivers (such as replacing small-size tires, reducing tire pressure and the like), so that the measurement error is increased during actual use.

2. The method for carrying out the odometer calibration on the standard road section needs a larger field, has long detection time and is not suitable for large-scale popularization.

3. The method of differential GPS positioning needs to establish communication with a reference station, requires good communication signals without interference, and has high equipment cost and limited use.

4. By adopting the method of the common GPS and the electronic map, the communication connection with the map server is required to be established, the map is required to be ensured to be accurate, the use cost is increased, and the use scene is limited.

5. The existing method for calibrating the odometer by adopting combined navigation needs human participation, specifies a detection road section and length, and cannot reduce malicious adjustment of users by real-time updating.

Disclosure of Invention

The invention aims to overcome the problems and the defects of the conventional vehicle odometer calibrating method, provides a novel vehicle odometer online calibrating system and a novel vehicle odometer online calibrating method, and can be widely applied to the field of vehicle travel mileage calibration such as taximeters, freight car odometers and the like.

The innovation points of the invention are as follows: when the pulse type wheel revolution detecting device is used for calculating the running distance of a vehicle (namely, a vehicle odometer), a combined navigation system (GNSS + INS) is used for automatically calibrating the vehicle odometer, and when the GPS positioning effect is good, a road section meeting the calibration standard and a running state meeting the calibration standard are automatically selected for calibrating the odometer. When the vehicle is considered to have the change of the driving condition, the verified K value is automatically put into use, and the calibration data is automatically updated, so that the stability of the K value is ensured, and the automatic adjustment is carried out when the vehicle condition changes for a long time. Meanwhile, the method saves verification key data, can perform historical backtracking and is convenient for analyzing verification results.

The technical scheme adopted by the invention is as follows:

an on-line verification system for a vehicle odometer comprises a GNSS antenna, a pulse wheel revolution detecting device interface, a GNSS module, an inertial navigation device, a processor, a storage unit and an odometer display screen. The GNSS module and the inertial navigation device form a basic navigation unit of the integrated navigation system, the processor is used for finishing integrated navigation calculation and odometer verification calculation, the storage unit is used for storing verification results and intermediate data elements, and the odometer display screen is used for displaying the vehicle mileage and verification result information.

The GNSS module, the inertial navigation device, the processor, the storage unit and the odometer display screen are arranged in the odometer of the detected vehicle. The GNSS antenna is connected with the GNSS module, the pulse type wheel revolution detecting device is connected with the processor, the inertial navigation device is connected with the GNSS module, and the processor is respectively connected with the GNSS module, the storage unit and the odometer display. The pulse type wheel revolution detecting device interface is connected with a pulse type wheel revolution detecting device of the detected vehicle. The GNSS antenna can be placed under the front windshield of the detected vehicle, and good satellite signal receiving is guaranteed.

Based on the system, the invention provides an online verification method for the vehicle odometer.

When the vehicle runs, the processor calculates the running mileage according to the number of wheel revolutions and the initial K value. The vehicle mileage calculation formula is as follows: and the driving mileage is equal to the value of the wheel revolution K. The method comprises the following steps:

step 1: and judging whether the GNSS signal is effective or not, and judging that the vehicle speed is greater than a set threshold value T. And (3) if the signal is effective and the vehicle speed is greater than the set threshold value T, executing the step 2, otherwise, continuously detecting and judging until the requirements are met, and then entering the next step.

Specifically, when the number of satellites in the navigation system is not less than 4, the GNSS signal is determined to be an effective signal, and the vehicle speed threshold T is required to be not less than 20 km/h.

Step 2: and processing and storing the vehicle running data at fixed time, setting a sliding time window for judging straight line running, and judging whether the vehicle is in a straight line running state according to the running data in the time window. If the vehicle is in the straight-line driving state, continuously storing the vehicle driving data until the vehicle does not meet the straight-line driving state, judging whether the straight-line driving distance of the vehicle is more than 1 kilometer, if so, executing the step 3, if not, clearing the vehicle driving data, restarting to store the data and judging the straight-line driving state.

When the step is executed, if any one of the situations that the GNSS signal is invalid or the running speed is lower than the threshold value T occurs, the vehicle running data is cleared, and the step 1 is returned again.

Wherein the vehicle driving data processed at regular time includes: and the attitude angle of the vehicle obtained by the combined navigation system comprises a pitch angle, a roll angle and a course angle after filtering. The timed interval may be set to 1 second and the sliding time window size may be determined based on the speed of travel, or may be fixed at 10 seconds or other values. The window is scrolled as it moves once per second.

The straight-line running of the vehicle is judged according to the data in the sliding window, and the standards comprise:

a. whether the vehicle is in horizontal travel. That is, whether the pitch angle and the roll angle of the vehicle are smaller than given threshold values, if smaller, it indicates a horizontal running state, and if not, it indicates no horizontal running state.

b. Whether the vehicle is traveling in one direction. That is, whether the heading angle of the vehicle is always consistent or not is within a set threshold range.

The stored running data at least comprises the accumulated running time (seconds) t from the time of entering the straight running state, the pulse number n of the odometer, the average running speed, the average attitude angle and the speed of the vehicleMinimum and maximum values of degree, minimum and maximum values of attitude angle, minimum and maximum values of altitude, time for starting straight-line driving of vehicle, longitude and latitude coordinates, and driving mileage value L calculated by combined navigation_startAnd the K value currently used by the odometer.

And step 3: and calculating the K value and judging whether the K value is effective or not. If the data is valid, executing the step 4, otherwise, clearing the vehicle running data and returning to the step 1.

Wherein, the K value is the driving mileage value L recorded according to the straight driving start of the vehicle_start(unit: centimeter) combined navigation driving range L when finishing straight line driving_end(unit: cm) and the accumulated pulse number n (unit: pulse) of the odometer are calculated, and the specific formula is as follows:

K＝n*100000/(L_end-L_start) [ pulse/km] (1)

Judging whether the K value is effective or not, namely judging whether the currently calculated K value is effective or not and whether the theoretical value is effective or not_iniFor comparison, the formula is:

Δ＝Abs(K-K_ini)/K_ini (2)

wherein Abs represents an absolute value.

If the calculated change Δ is greater than a given threshold, for example 10%, the newly calculated value of K is considered invalid.

And 4, step 4: and calculating the reliability correlation quantity of the K value, and storing the correlation data in a K value table on the day.

The reliability related quantity of the K value at least comprises the vehicle running distance, the altitude range, the heading angle range and the speed range for calculating the K value.

The data stored in the K value table of the current day at least comprises: the vehicle driving starting time, the longitude and latitude coordinates of the starting point, the longitude and latitude coordinates of the driving end point, the calculated K value, the K value reliability, the vehicle driving average speed, the vehicle driving average attitude angle, the vehicle driving speed minimum value and maximum value, and the vehicle driving attitude angle minimum value and maximum value.

And 5: and judging whether the vehicle running time reaches 24 hours or not, namely whether the current day is finished or not. And if not, clearing the driving data, returning to the step 1, and continuously calculating and storing a new K value. If so, step 6 is performed.

Step 6: if the records in the K value table of the current day exceed the set upper limit (for example, 10 records), deleting the records in the table sorted from high to low according to the reliability, and then arranging the records which are sorted at the end and exceed the set record upper limit in the record table, so that the number of the records in the K value table is the upper limit, and then executing the step 7; if the records are smaller than the set lower limit value (such as 5), clearing the running data, returning to the step 1, and continuing to calculate and store a new K value. If the record is between the upper and lower limits, step 7 is performed directly.

The reliability is in direct proportion to the running distance of the vehicle and in inverse proportion to the non-linear uniform speed running. The confidence level is calculated according to the following formula:

reliability coefficient driving distance/(height variation range linear driving range speed variation range)

And 7: and saving the data in the current-day K value table into a total K value table. And judging whether the records of the total K value table exceed a set upper limit (such as 20) of the records, if so, deleting the K value records which are sorted from near to far according to time and have the earliest time and exceed the upper limit of the record number, and keeping the record number at the upper limit. And then calculating whether the standard deviation of the K values in the total K value table is within an allowable range, if so, such as: the standard deviation of the K value/mean of the K value < 5%, step 8 is performed. Otherwise, the data is not processed, the driving data is cleared, the step 1 is returned, and the data of the next 24 hours is collected.

And 8: judging whether the difference between the K value average value in the total K value table and the currently used K value is larger than a given threshold value, such as: k average value-K current value)/K current value > 2%, if greater than, executing step 9, otherwise, not updating the K value, clearing the driving data, and returning to step 1. Therefore, the updating rate of the K value can be reduced, and the stability of the K value is kept.

And step 9: the currently used value of K is updated. Then, the running data is cleared, and the procedure returns to step 1.

The updating of the K value can be directly replaced by the average value of the K value in the record table. Alternatively, partial correction may be performed by the following method:

updated K value α current K value + (1- α) K average value (3)

Wherein alpha is a forgetting factor and takes a value between [0 and 1 ].

Thus, the online verification of the vehicle odometer is completed.

Advantageous effects

Compared with the prior art, the method has the advantages of low cost and convenient use, and can automatically finish the K value verification work of the odometer in the normal running process of the vehicle and ensure the long-term accuracy of the metering of the odometer.

Drawings

FIG. 1 is a schematic diagram of the system of the present invention.

FIG. 2 is a flow chart of the method of the present invention.

Detailed Description

The method of the present invention is further described in detail below with reference to the drawings and examples.

Examples

As shown in fig. 1, an on-line verification system for a vehicle odometer.

The system (with the built-in GNSS antenna) is fixed below a front windshield, a 12V power supply interface is connected to a vehicle-mounted power supply, and a wheel revolution input interface is connected to the output of a wheel revolution detection device of a vehicle. The vehicle driving mileage of the system is calculated according to the wheel rotating speed and the K value, and the calculation formula is as follows: and the driving mileage is equal to the value of the wheel revolution K. The mileage value is displayed on the display screen.

As shown in fig. 2, a method for online verification of a vehicle odometer.

step 1: and according to the satellite number received by the GNSS, whether the GNSS signal is effective or not is judged. When the number of satellites is more than 4, the GNSS positioning is considered to be effective, and whether the vehicle running speed is more than a speed threshold value T (30 km/h) or not is judged according to the running speed output by the GNSS. When the GNSS is active and the travel speed is greater than the threshold T, the next step may be entered. Otherwise, continuously detecting until the conditions are met, and then entering the next step. Meanwhile, in the subsequent step, once the GNSS signal is invalid or the driving speed is lower than the threshold value T, the step 1 is returned again.

Step 2: the pitch angle and roll angle of the vehicle are sampled periodically at 0.2 second intervals, and the data are filtered (e.g., moving average filtered) at 1 second intervals. The driving data of the vehicle is saved every second. And judging whether the vehicle is in a straight-line driving state according to the sliding time window of 10 seconds. If the vehicle is in straight line driving and the driving distance is more than 1 kilometer, continuously storing the vehicle driving data until the vehicle is in a state of no straight line driving, and recording the driving mileage value L at the moment_endExecuting the step 3; if not, the vehicle driving data is cleared, and the data is restarted to be stored and the straight line driving is judged.

Whether the vehicle is in straight line driving is judged by data in the sliding window, and the judging method can be as follows:

a. judging horizontal driving: and the standard deviation of the pitch angle and the roll angle in the sliding window is less than 5 degrees. If yes, the vehicle is in the horizontal driving state, and if not, the vehicle is not in the horizontal driving state.

b. Judging straight line driving: the standard deviation of the heading angle within the sliding window is less than 5 degrees. If yes, the straight driving state is indicated, and if not, the straight driving state is not indicated.

The stored travel data per second includes the accumulated travel time (seconds) and the number of pulses n of the odometer from the start of entering the straight-line travel state, the average vehicle travel speed, the average attitude angle, the minimum value and the maximum value of the speed, the minimum value and the maximum value of the attitude angle, the minimum value and the maximum value of the altitude, the time for starting the straight-line travel of the vehicle, the longitude and latitude coordinates, and the travel mileage value L calculated by the integrated navigation_startAnd the K value currently used by the odometer.

And step 3: and calculating the K value and judging whether the K value is effective or not. If yes, executing step 4, otherwise, returning to step 2.

Wherein, the K value is the driving mileage value L recorded according to the straight driving start of the vehicle_start(unit: centimeter) combined navigation driving range L when finishing straight line driving_end(unit: cm) and the number of accumulated pulses n (unit: pulse) of the odometer. The formula is as follows:

K＝n*100000/(L_end-L_start) [ pulse/km] (1)

Δ＝Abs(K-K_ini)/K_ini (2)

wherein Abs represents an absolute value.

If the calculated change Δ is greater than a given threshold, e.g., 10%, the newly calculated value of K is considered invalid.

Here, the mileage value at the start of straight-line running is set to 15.2 km, i.e., L_startThe mileage for ending straight line travel is 18.3 km, L1520000 cm_endWhen the accumulated pulse number n is 1820 and 1830000 cm, K is calculated as: k1820 × 100000/(1830000-. If the current K is_iniWith a value of 600, the variance is: Δ ═ Abs (587 + 600)/600 ═ 2.16%.

And 4, step 4: and calculating the reliability correlation quantity of the K value, and storing the correlation data in a K value table of the current day.

And acquiring the travel mileage, the altitude change range (the maximum altitude minus the minimum altitude), the straight line travel range (the maximum difference of course angles) and the speed change range when the K value is calculated according to the data when the K value is calculated. These data serve as K-value reliability-related quantities.

The data stored in the K-value table for the current day includes: the vehicle driving starting time, the longitude and latitude coordinates of the starting point, the longitude and latitude coordinates of the driving ending point, the calculated K value, the reliability related quantity of the K value, the average vehicle driving speed, the average vehicle driving attitude angle, the minimum value and the maximum value of the vehicle driving speed and the minimum value and the maximum value of the vehicle driving attitude angle.

And 5: and judging whether the vehicle running time reaches 24 hours or not, namely ending the current day. If not, returning to the step 1, and continuing to calculate and save the new K value. If so, step 6 is performed.

Step 6: if the number of records in the K value table exceeds 10, the records with lower reliability of the table are deleted, and then step 7 is executed. If the number of records is less than 5, returning to the step 1, and continuing to calculate and save the new K value. If between the upper and lower limits, step 7 is performed directly.

The confidence level is calculated according to the following formula:

And 7: and saving the data in the current-day K value table into a total K value table. And judging whether the records of the total K value table exceed the upper limit of the records by 20, and if so, deleting the early K value records. And then calculating whether the standard deviation of the K values in the total K value table is within an allowable range. If it is within the allowable range (e.g., standard deviation of K value/mean of K value < 5%), then step 8 is performed. Otherwise, the data is not processed and the data of the next day is collected in the step 1.

And 8: whether the average value of the K values in the total K value table is within the requirement of the threshold value (2%) is calculated according to the following formula:

(K mean-K current)/K current value (3)

If the threshold value is exceeded, step 9 is executed, otherwise, the K value is not updated, and the step 1 is directly returned. Therefore, the updating rate of the K value can be reduced, and the stability of the K value is kept.

And step 9: the currently used K value is updated as follows. Then, the procedure returns to step 1.

Updated K value α current K value + (1- α) K average value (4)

Wherein, alpha is a forgetting factor, the value is between [0,1], and can be set to 0.5.

Thus, the online verification of the vehicle odometer is completed.

In the method, the driving data which can be used for verifying the K value is selected on line (namely, the driving data of a longer straight line driving road section is selected as verification calculation data by acquiring the driving information in real time).

The K value table is divided into a current K value table and a total K value table. The day K value table can acquire the latest K value verification data, and the total K value table can ensure the stability of the K value.

And finally updating the K value by adopting an updating formula in the total K value table, so that the stability of the K value is ensured, and the K value can be changed to the latest driving state.

Claims

Translated fromChinese

1.一种车辆里程计在线检定系统，其特征在于，包括GNSS天线、脉冲式车轮转数检测装置接口、GNSS模块、惯性导航器件、处理器、存储单元和里程计显示屏；1. a vehicle odometer online verification system, is characterized in that, comprises GNSS antenna, pulse type wheel revolution detection device interface, GNSS module, inertial navigation device, processor, storage unit and odometer display screen;

GNSS模块和惯性导航器件组成组合导航系统基本导航单元，处理器用于完成组合导航计算、里程计检定计算工作，存储单元用于存储检定结果和中间数据元，里程计显示屏用于显示车辆行驶里程和检定结果信息；The GNSS module and inertial navigation device constitute the basic navigation unit of the integrated navigation system. The processor is used to complete the integrated navigation calculation and odometer verification calculation, the storage unit is used to store the verification results and intermediate data elements, and the odometer display screen is used to display the vehicle mileage. and test result information;

其中，GNSS模块、惯性导航器件、处理器、存储单元、里程计显示屏安装在被检定车辆的里程计内；GNSS天线与GNSS模块相连，脉冲式车轮转数检测装置与处理器相连，惯性导航器件与GNSS模块相连，处理器与GNSS模块、存储单元、里程计显示器分别相连；脉冲式车轮转数检测装置接口与被检定车辆的脉冲式车轮转数检测装置相连接。Among them, the GNSS module, inertial navigation device, processor, storage unit, and odometer display screen are installed in the odometer of the tested vehicle; the GNSS antenna is connected to the GNSS module, and the pulse-type wheel revolution detection device is connected to the processor. The device is connected with the GNSS module, the processor is connected with the GNSS module, the storage unit and the odometer display respectively; the interface of the pulse-type wheel revolution detection device is connected with the pulse-type wheel revolution detection device of the vehicle to be verified.

2.一种如权利要求1所述系统的车辆里程计在线检定方法，其特征在于，当车辆行驶时，处理器根据车轮转数和初始K值计算行驶里程，车辆行驶里程计算公式为：行驶里程＝车轮转数*K值；2. a vehicle odometer online verification method of the system as claimed in claim 1 is characterized in that, when the vehicle is running, the processor calculates the mileage according to the number of wheel revolutions and the initial K value, and the vehicle mileage calculation formula is: Mileage = wheel revolutions * K value;

包括以下步骤：Include the following steps:

步骤1：判断GNSS信号是否有效，并且车速大于设定阈值T；如果信号有效且车速大于设定阈值T，则执行步骤2，否则持续进行检测和判断，直到满足上述要求后，再进入下一步骤；Step 1: Determine whether the GNSS signal is valid and the vehicle speed is greater than the set threshold T; if the signal is valid and the vehicle speed is greater than the set threshold T, go to step 2, otherwise continue to detect and judge until the above requirements are met, and then enter the next step. step;

步骤2：定时处理和保存车辆行驶数据，并设置用于直线行驶判断的滑动时间窗，根据时间窗内的行驶数据判断车辆是否处于直线行驶状态；Step 2: regularly process and save the vehicle driving data, and set a sliding time window for straight-line driving judgment, and judge whether the vehicle is in a straight-line driving state according to the driving data in the time window;

如果车辆处于直线行驶状态，则持续保存车辆行驶数据，直到车辆不满足直线行驶状态时，判断车辆直线行驶距离是否大于1公里，如果是，则执行步骤3，如果不是，则清除车辆行驶数据，重新开始保存数据并进行直线行驶状态判断；If the vehicle is in a straight-line driving state, keep saving the vehicle driving data until the vehicle does not meet the straight-line driving state, determine whether the vehicle's straight-line driving distance is greater than 1 km, if so, go to step 3, if not, clear the vehicle driving data, Restart to save the data and judge the straight driving state;

在执行本步骤时，如果出现GNSS信号失效或者行驶速度低于阈值T中的任意一种情况，则清除车辆行驶数据，重新返回步骤1；When performing this step, if any one of the GNSS signal failure or the driving speed is lower than the threshold value T occurs, clear the vehicle driving data, and return to step 1;

其中，定时处理的车辆行驶数据包括：由组合导航系统得到的车辆的姿态角，包括经过滤波后的俯仰角、侧滚角、航向角；Wherein, the vehicle driving data processed periodically includes: the attitude angle of the vehicle obtained by the integrated navigation system, including the filtered pitch angle, roll angle, and heading angle;

车辆直线行驶根据在滑动窗内的数据进行判断，标准包括：The straight-line driving of the vehicle is judged according to the data in the sliding window, and the criteria include:

a.车辆是否处于水平行驶；即，车辆的俯仰角和侧滚角是否小于给定的阈值，如果小于，则说明处于水平行驶状态，如果不小于，则说明未处于水平行驶状态；a. Whether the vehicle is driving horizontally; that is, whether the pitch angle and roll angle of the vehicle are less than the given thresholds, if it is less than that, it means it is in a horizontal driving state, if not, it means that it is not in a horizontal driving state;

b.车辆是否沿一个方向行驶，即，车辆的航向角否始终保持一致，均处在一个设定的阈值范围内；b. Whether the vehicle is driving in one direction, that is, whether the heading angle of the vehicle is always consistent, all within a set threshold range;

步骤3：计算K值并判断是否有效。如果有效，则执行步骤4，否则，清除车辆行驶数据，返回步骤1；Step 3: Calculate the K value and judge whether it is valid. If it is valid, go to step 4, otherwise, clear the vehicle driving data and return to step 1;

其中，K值根据车辆直线行驶开始记录的行驶里程值L_start结束直线行驶时的组合导航行驶里程L_end，以及里程计的累计脉冲数n进行计算，具体公式为：Among them, the K value is calculated according to the mileage value L_start recorded at the beginning of the straight-line driving of the vehicle and the combined navigation mileage L_end when the vehicle ends the straight-line driving, and the accumulated pulse number n of the odometer. The specific formula is:

K＝n*100000/(L_end-L_start)[脉冲/公里] (1)K=n*100000/(L_end -L_start )[pulse/km] (1)

判断K值是否有效，是将当前计算的K值和理论值K_ini进行比较，公式为：To judge whether the K value is valid, it is to compare the currently calculated K value with the theoretical value K_ini . The formula is:

Δ＝Abs(K-K_ini)/K_ini (2)Δ=Abs(KK_ini )/K_ini (2)

其中，Abs表示绝对值；Among them, Abs represents the absolute value;

若计算的变化量Δ大于给定阈值，则认为新计算的K值无效；If the calculated change Δ is greater than the given threshold, the newly calculated K value is considered invalid;

当日K值表中保存的数据包括车辆行驶起始时间，起始点经纬度坐标，行驶结束点经纬度坐标，计算的K值，K值可信度，车辆行驶平均速度，车辆行驶平均姿态角，车辆行驶速度最小值和最大值，车辆行驶姿态角的最小值和最大值；The data stored in the K value table of the day includes the starting time of the vehicle, the longitude and latitude coordinates of the starting point, the longitude and latitude coordinates of the driving end point, the calculated K value, the reliability of the K value, the average speed of the vehicle, the average attitude angle of the vehicle, the The minimum and maximum speed, the minimum and maximum value of the vehicle attitude angle;

步骤5：判断车辆行驶时间是否达到24小时，即当日是否结束；如果没有达到，则清除行驶数据，返回步骤1，继续计算并保存新的K值；如果达到，则执行步骤6；Step 5: Determine whether the vehicle travel time reaches 24 hours, that is, whether the day ends; if not, clear the driving data, return to step 1, continue to calculate and save the new K value; if it does, go to step 6;

步骤6：若当日K值表中记录超过设置的上限值，则删除表中按可信度从高到低排序后，排在最后且超出记录表中设定记录上限的记录，使K值表中记录数为上限值，然后执行步骤7；如果记录小于设定的下限值，则清除行驶数据，返回步骤1，继续计算并保存新的K值；如果记录处于上下限之间，则直接执行步骤7；Step 6: If the records in the K value table of the day exceed the set upper limit, delete the records in the table sorted by reliability from high to low, and the records that are ranked last and exceed the set record upper limit in the record table, make the K value The number of records in the table is the upper limit, then go to step 7; if the record is less than the set lower limit, clear the driving data, return to step 1, continue to calculate and save the new K value; if the record is between the upper and lower limits, Then directly execute step 7;

其中，可信度和车辆行驶距离成正比，和非直线匀速行驶成反比；Among them, the reliability is proportional to the distance traveled by the vehicle, and inversely proportional to the non-linear uniform speed;

可信度的高低按照如下公式计算：The level of reliability is calculated according to the following formula:

可信度＝系数*行驶距离/(高度变化极差*直线行驶极差*速度变化极差)Reliability = coefficient * distance traveled / (height change range * straight line driving range * speed change range)

步骤7：将当日K值表中数据保存到总K值表中；Step 7: Save the data in the K value table of the day to the total K value table;

判断总K值表的记录是否超过设定的记录上限，如果超出上限，则删除按时间由近到远排序后时间最早且超出记录数上限的K值记录，使记录数维持在上限值；然后计算总K值表中的K值标准差是否在允许范围内，如果在设定的允许范围内，如：K值的标准差/K值平均值<5％，则执行步骤8；否则，不进行处理，清除行驶数据，返回步骤1，采集下一个24小时的数据；Determine whether the records of the total K value table exceed the set record upper limit. If it exceeds the upper limit, delete the K value record with the earliest time and exceed the upper limit of the number of records after sorting by time from near to far, so that the number of records is maintained at the upper limit; Then calculate whether the standard deviation of the K value in the total K value table is within the allowable range. If it is within the set allowable range, such as: the standard deviation of the K value/the average value of the K value is less than 5%, then go to step 8; otherwise, No processing, clear driving data, return to step 1, and collect data for the next 24 hours;

步骤8：判断总K值表中的K值平均值与当前使用的K值的差是否大于给定阈值，如果大于，则执行步骤9，否则，不对K值进行更新，清除行驶数据，返回步骤1；Step 8: Determine whether the difference between the average K value in the total K value table and the currently used K value is greater than the given threshold. If it is greater than the given threshold, go to Step 9. Otherwise, do not update the K value, clear the driving data, and return to the step 1;

步骤9：更新当前使用的K值，然后，清除行驶数据，返回步骤1。Step 9: Update the currently used K value, then clear the driving data and return to Step 1.

3.一种如权利要求2所述的车辆里程计在线检定方法，其特征在于，步骤1中，当使用的导航系统中卫星数不少于4颗时，则判定GNSS信号为有效信号，车速阈值T要求不小于20公里/小时。3. a vehicle odometer online verification method as claimed in claim 2 is characterized in that, in step 1, when the number of satellites in the navigation system used is no less than 4, then it is determined that the GNSS signal is an effective signal, and the vehicle speed The threshold T is required to be no less than 20 km/h.

4.一种如权利要求2所述的车辆里程计在线检定方法，其特征在于，步骤2中，定时时间间隔设置为1秒，滑动时间窗窗口尺寸根据行驶速度来确定，该窗口按照每秒移动1次进行滚动。4. A vehicle odometer on-line verification method as claimed in claim 2, is characterized in that, in step 2, time interval is set to 1 second, sliding time window window size is determined according to traveling speed, and this window is according to per second. Move 1 time to scroll.

5.一种如权利要求2所述的车辆里程计在线检定方法，其特征在于，步骤2中，滑动时间窗窗口尺寸固定为10秒。5. A method for on-line verification of vehicle odometer according to claim 2, wherein in step 2, the size of the sliding time window is fixed to 10 seconds.

6.一种如权利要求2所述的车辆里程计在线检定方法，其特征在于，步骤2中保存的行驶数据，包括从进入直线行驶状态开始累计行驶时间t和里程计的脉冲数n、车辆行驶平均速度、平均姿态角、速度的最小值和最大值、姿态角的最小值和最大值、海拔高度的最小值和最大值，以及车辆开始直线行驶的时间、经纬度坐标、组合导航计算的行驶里程值L_start和里程计当前使用的K值。6. A vehicle odometer on-line verification method as claimed in claim 2, is characterized in that, the driving data saved in step 2, comprises the pulse number n of accumulative travel time t and odometer starting from entering straight travel state, vehicle Driving average speed, average attitude angle, minimum and maximum speed, minimum and maximum attitude angle, minimum and maximum altitude, and the time when the vehicle starts to travel straight, latitude and longitude coordinates, and travel calculated by integrated navigation The mileage value L_start and the K value currently used by the odometer.

7.一种如权利要求2所述的车辆里程计在线检定方法，其特征在于，步骤9中K值的更新方法为：直接更换为记录表中K值平均值。7 . The method for on-line verification of vehicle odometer according to claim 2 , wherein the updating method of the K value in step 9 is: directly replacing the average value of the K value in the record table. 8 .

8.一种如权利要求2所述的车辆里程计在线检定方法，其特征在于，步骤9中K值的更新方法为，对其进行部分修正，方法如下：8. a vehicle odometer online verification method as claimed in claim 2 is characterized in that, the updating method of K value in step 9 is, it is partially corrected, and the method is as follows:

更新后K值＝α*当前K值+(1-α)*K平均值 (3)Updated K value=α*current K value+(1-α)*K average value (3)

其中，α为遗忘因子，取值在[0,1]之间。Among them, α is the forgetting factor, and the value is between [0, 1].