CN111998828B - A Road Slope Estimation Method Based on Portable GPS - Google Patents

Abstract

The invention discloses a road gradient estimation method based on portable GPS. The method includes the following steps: placing a plurality of GPS receivers on a vehicle to collect data; preprocessing the collected data; testing road division, considering the difference of data points. Quantity, determine the length of the road segment and divide the test road into multiple sub-sections according to the determined length; align the elevation data of the sub-sections, align the acquired elevation data; The slope calculation of the sub-segments is performed, using the aligned elevation data of each sub-segment to perform a linear regression to estimate the road slope and infer the accuracy of the measured road slope based on the standard error of the slope estimate of the linear regression. The method of the invention can be conveniently and accurately obtained through GPS data analysis, is not affected by the tested road type and vehicle driving state, and can obtain more accurate gradient results by correcting the accumulated distance and elevation.

Description

A Road Slope Estimation Method Based on Portable GPS

technical field

The invention belongs to the technical field of traffic environmental protection, and in particular relates to a road gradient estimation method based on portable GPS.

Background technique

The information provided by the high-precision map is conducive to the system control of the car while driving, and is conducive to reducing the occurrence of traffic accidents, and the mature development of automatic driving technology is also inseparable from the support of high-precision map data, and the road slope is also a high-precision map. Therefore, it is of great significance to measure the road slope accurately and efficiently.

At present, in the road slope measurement at home and abroad, obtaining high-precision road slope data requires very expensive professional surveying and mapping instruments and involves data fusion and calculation, which is relatively complicated. It is very difficult for non-surveying and mapping professionals to obtain road gradient data. It is difficult to obtain high-precision road gradient data with simple GPS receivers or level meters, and cannot meet the needs of dynamic and fast road gradient measurement.

Patent CN101838958A proposes a road gradient detection method, which is processed and calculated based on the contour map and road planning map of the surveying and mapping department. It is mainly aimed at obtaining the gradient in road design and planning. Line drawings and road plans. The gradient obtained by this method is static data, and there is often some deviation between the road planning and the actual road gradient after construction, which cannot accurately reflect the gradient information of the road where the motor vehicle is actually driving. Moreover, it is difficult for ordinary users to obtain relevant road plans from the surveying and mapping department and then conduct professional analysis and calculation.

Patent CN102313535A proposes a gradient detection method. The method obtains the gradient value according to the corrected correction value, acceleration signal, vehicle acceleration value and earth gravitational acceleration through an acceleration sensor and a speed signal module. The invention relies on the estimation of the sensor signal installed on the frame, and needs to be installed and debugged separately when different vehicles are used. This method is more to reflect the undulating state changes of the vehicle, but cannot obtain the real road gradient value and is easily affected by the sensor installation and the accuracy of the sensor. The gradient obtained by this method is difficult to meet the high-precision and portable measurement requirements for vehicle exhaust emission measurement and estimation.

SUMMARY OF THE INVENTION

The main purpose of the present invention is to overcome the shortcomings and deficiencies of the prior art, and to propose a road gradient estimation method based on portable GPS. The improvement of high-precision map slope data and the estimation of dynamic vehicle emissions have important practical value.

In order to achieve the above object, the present invention adopts the following technical solutions:

A kind of road gradient estimation method based on portable GPS, is characterized in that, comprises the following steps:

Utilize one or more portable GPS receivers to collect data, the data collected by each GPS receiver includes road latitude, longitude and elevation data;

Preprocess the collected data, calculate the cumulative distance of each route track according to the latitude and longitude data collected by each GPS, and correct it with reference to the actual road length;

The test road is divided into road sections, considering the number of data points in the road section, determining the length of the road section division and dividing the test road into a plurality of sub-road sections according to the determined length, and the data points are the data recorded by the GPS receiver in the test;

Align the elevation data of the sub-sections, combine multiple sets of GPS data involved in each sub-section into a data set, and align the elevation data of each group to solve the problem of the average elevation data error;

Calculate the road gradient of the test road and quantify the accuracy, perform the gradient calculation of the sub-section for each sub-section, use the aligned elevation data of each sub-section to perform a linear regression to estimate the road gradient and estimate the road gradient according to the standard error of the slope of the linear regression. Infer the accuracy of measuring road slope.

Further, the cumulative distance of the calculated road section is specifically:

Assuming that the number of GPS receivers is m, and the vehicle travels through the test road repeatedly n times, then according to the latitude and longitude data of the GPS receiver, m*n route trajectories are obtained, and the cumulative driving distance of the route trajectory is calculated based on the second-by-second speed of each GPS trajectory, The average cumulative distance of m*n route trajectories is the cumulative distance of route trajectories, and the formula is as follows:

Among them, d is the cumulative distance of the route track, i is the i-th GPS, 1≤i≤n, j is the j-th test, 1≤j≤n, l _i,j is the accumulation of the i-th GPS and the j-th test distance.

Further, the correction of the reference actual road length is specifically by inserting a consistent starting point and end point into the cumulative distance of each route track, and calibrating it to the real distance, and the correction factor of the cumulative distance of each route track is expressed as follows:

θ _i,j =d/l _i,j

Among them, θ _i,j is the cumulative distance correction factor of the ith GPS jth test.

Further, the determining the segment length of the road segment is specifically:

The number of data points affects the accuracy of the test road slope, so the length of road segment segmentation needs to ensure the number of data points for each road segment after segmentation, and the number of data points depends on the frequency of data recording, vehicle speed and the length of the road segment;

The longer the length of the road segment, the more data points it obtains, the higher the grade accuracy is, but if the length of the road segment is too long, the actual change within the road segment will be averaged, resulting in an underestimation of the true gradient change; if the length of the road segment is too short, then Too few data points within a split can result in inaccurate slope estimates.

Further, the calculation formula of the number of data points is as follows:

Among them, number is the number of data points in the sub-section, Δl _i,j is the length of the road section of the i-th GPS jth test; Δv _i,j is the average speed of the road section of the i-th GPS jth test, and H is the GPS data record frequency.

Further, the number of data points in each segmented subsection is greater than 20.

Further, the alignment processing of the road section elevation data is as follows:

The multiple sets of GPS data obtained from each sub-road segment are fused and combined into a data set. The average elevation is calculated for all the elevation data on the test road as the reference absolute elevation, and then the reference absolute elevation is subtracted from the elevation of each data point. The relative elevation difference between the data points does not change.

Further, the road gradient of the calculated road section is specifically:

Calculate the road slope for the sub-sections, and use the aligned elevation data points of each sub-section to perform a univariate linear regression fitting to estimate the road gradient. The slope of the fitted line is multiplied by 100 to obtain the test road gradient percentage. The slope connection is the slope information of the test road.

Further, the accuracy of the road gradient is specifically the standard error of the slope of the fitted straight line.

Compared with the prior art, the present invention has the following advantages and beneficial effects:

1. The method of the present invention improves the accuracy of gradient calculation by segmenting the test road and calculating the gradient, and aligning the elevation data of different road sections; The measurement is not affected by the type of road tested and the driving state of the vehicle, which solves the problem that ordinary portable GPS cannot obtain high-precision slope data.

2. The method of the present invention has low cost, simple operation method, and can realize high-precision road slope measurement without relying on professional surveying and mapping tools.

3. The method of the present invention provides a quantitative idea of the road gradient accuracy, and the user can select different GPS numbers to measure the gradient according to the requirements for the gradient accuracy.

Description of drawings

Fig. 1 is the flow chart of the inventive method;

Figure 2a is the data before alignment of the road section elevation data;

Figure 2b is the data after alignment of the road section elevation data;

Figure 3a shows the fitting slope results and accuracy before the alignment of the road section elevation data;

Figure 3b shows the fitting slope results and accuracy after alignment of the road section elevation data;

Figure 4 is the gradient results for the tested route.

Detailed ways

The present invention will be described in further detail below with reference to the embodiments and the accompanying drawings, but the embodiments of the present invention are not limited thereto.

Example

Taking the actual road emission test of a motor vehicle as an example where the slope data of the tested route needs to be known and high accuracy is required, as shown in FIG.

S1. Use multiple GPS receivers to collect data, specifically:

One or more portable GPS receivers are placed on the vehicle, and the vehicle repeatedly drives through the test road for many times, and collects all the recorded data of the GPS receiver. The data collected by each GPS receiver includes the longitude, latitude and elevation data of the road.

In this embodiment, four portable GPSs are placed on the front end of the test vehicle to be fixed, and the vehicle travels normally along the vehicle emission test road, and repeatedly travels through the road section 4 times, and a total of 16 route trajectory data are collected.

S2. Calculate the cumulative distance of the route track, and correct the reference actual road length, specifically:

Assuming that the number of GPS receivers is m, and the vehicle travels through the test road repeatedly n times, then according to the latitude and longitude data of the GPS receiver, m*n route trajectories are obtained, and the cumulative driving distance of the route trajectory is calculated based on the second-by-second speed of each GPS trajectory, The average of the cumulative distance of m*n routes is the cumulative distance of the route. The formula for calculating the cumulative distance of the route is as follows:

Among them, d is the cumulative distance of the route track, i is the i-th GPS, 1≤i≤n, j is the j-th test, 1≤j≤n, l _i,j is the accumulation of the i-th GPS and the j-th test distance.

Correction with reference to the actual road length is as follows:

By inserting a consistent starting point and end point into the cumulative distance of each route track, it is corrected to the true distance, and the correction factor of the cumulative distance of each route track is expressed as follows:

θ _i,j =d/l _i,j .

Among them, θ _i,j is the cumulative distance correction factor of the ith GPS jth test.

In this embodiment, the above-mentioned 16 trajectories are corrected for the road cumulative distance. The cumulative distances of the 16 trajectories are shown in the following table, and the obtained average cumulative distance is 35.3 km and the standard deviation is ±0.7%.

S3. Test road division, specifically:

The number of data points in the road section also determines the accuracy of the obtained road gradient. In order to ensure the number of data points retained in each road segment, the length of the divided road segment must be determined first. The number of data points in each sub-section depends on the data recording frequency, vehicle speed and the length of the respective section. Under the data recording frequency of 1Hz of the GPS used, taking a typical highway speed of 100km h ^-1 as an example, at a speed of 0.1km There are only 3 data points for a single GPS in the road segment, but based on n operations of m GPS, there are 3*m*n data points on the length of the road segment with a speed of 0.1km at 100km h ^-1 ; if the length of the sub-road segment If it is too long, the more data points it obtains, the higher the slope accuracy will be, but if the branch section is too long, the actual change in the branch section will be averaged, which will lead to an underestimation of the true slope change; if the branch section length is too short, the internal Too few data points may lead to inaccurate estimation of the slope; generally, the number of data points per road segment needs to be greater than 20 to meet the needs of slope accuracy calculation.

The formula for calculating the number of data points is as follows:

Among them, number is the number of data points in the sub-section, Δl _i,j is the length of the road section of the i-th GPS jth test; Δv _i,j is the average speed of the road section of the i-th GPS jth test, and H is the GPS data record frequency.

In this embodiment, the division length is 0.1 km, and the test road is divided into 353 sections. Because the higher the driving speed, the fewer data points on each sub-section, and if the maximum driving speed is 120km/h, there are at least 48 data points in each sub-section.

S4. Align the elevation data of the sub-sections, specifically:

First of all, multiple pieces of data in each sub-section are combined into a data set. Due to the random errors in the absolute elevations of different GPS, it is necessary to align the acquired elevation data of each sub-section. Calculate the average altitude for each trip along the road segment. The average elevation is calculated for the fused data of each road segment as the reference absolute elevation, and then the absolute elevation reference value is subtracted from the elevation of each data point, and the relative elevation difference between each data point remains unchanged.

In another embodiment, the elevation data obtained from GPS B and GPS C are taken as an example, as shown in Fig. 2a and Fig. 2b.

S5. Calculate the road gradient of the test road, specifically:

Calculate the slope of each road segment for the road segment, and use the aligned elevation data points of each road segment to perform a linear regression fit to estimate the slope to be measured. The slope of the fitted straight line is multiplied by 100 to represent the road slope as a percentage. The slope information of the test road is obtained by connecting the slopes of all road sections; the standard deviation of the fitted straight line slope is the accuracy of the obtained road slope. The number of data points in a segment may vary depending on the speed of the vehicle, and the method can be used to calculate grades at any location when an appropriate segment length is chosen.

In this embodiment, the slopes and their standard deviations of all the sub-sections of the test road are obtained by calibrating different trajectory data points on each sub-section, and then fitting the test road. The results are shown in FIG. Slope ranges from -8%-6% with accuracy within ±0.5%.

In another embodiment, as shown in Fig. 3a and Fig. 3b, the slope accuracy obtained through the alignment of the elevation data is improved from ±4.6% to ±0.75%.

In addition, a high-resolution centimeter-level differential GPS commercial gradient meter is used to obtain road gradient data for comparison with the results obtained in this implementation. The comparison results show that the gradient results obtained by the method provided by the present invention are similar to the gradient results obtained by high-resolution centimeter-level GPS, which confirms the reliability of the method of the present invention.

The gradient measurement method of the present invention can be easily and accurately obtained through GPS data analysis, is not affected by the type of the tested road and the driving state of the vehicle, and can obtain more accurate gradient results by correcting the accumulated distance and elevation.

It should also be noted that, in this specification, terms such as "comprising", "comprising" or any other variation thereof are intended to encompass non-exclusive inclusion, such that a process, method, article or device comprising a series of elements not only includes Those elements, but also other elements not expressly listed or inherent to such a process, method, article or apparatus. Without further limitation, an element qualified by the phrase "comprising a..." does not preclude the presence of additional identical elements in a process, method, article or apparatus that includes the element.

The above description of the disclosed embodiments enables any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Thus, the present invention is not intended to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims (4)

1. A road gradient estimation method based on a portable GPS is characterized by comprising the following steps:

collecting data using one or more portable GPS receivers, each GPS receiver collecting data including longitude and latitude and elevation data of a road;

preprocessing the collected data, calculating the accumulated distance of each route track according to the longitude and latitude data collected by each GPS, and correcting by referring to the actual road length;

the calculation of the cumulative distance of the route trajectory is specifically:

assuming that the number of the GPS receivers is m, the vehicle repeatedly runs through the test road n times, then m × n route tracks are obtained according to the longitude and latitude data of the GPS receivers, the accumulated running distance of the route tracks is calculated based on the speed of each GPS track every second, and the average of the accumulated distances of the m × n route tracks is the accumulated distance of the route tracks, and the formula is as follows:

wherein d is the accumulated distance of the route track, i represents the ith GPS, i is more than or equal to 1 and less than or equal to m, j represents the jth test, j is more than or equal to 1 and less than or equal to n, l_i,jThe cumulative distance of the jth test of the ith GPS;

the step of correcting the reference actual road length is to correct the accumulated distance of each route track into a real distance by inserting a consistent starting point and an end point, and a correction factor of the accumulated distance of each route track is expressed as follows:

θ_i,j＝d/l_i,j

wherein, theta_i,jThe accumulated distance correction factor for the jth test of the ith GPS;

the method comprises the steps of dividing a test road into road sections, determining road section division length by considering the number of data points in the road sections, and dividing the test road into a plurality of divided road sections according to the determined length, wherein the data points are data recorded by a GPS receiver in the test;

the specific step of determining the road section segmentation length is as follows:

the data point quantity influences the precision of the tested road gradient, so that the road section segmentation length needs to ensure the quantity of data points of each segmented road section, and the quantity of the data points depends on the data recording frequency, the vehicle speed and the length of the segmented road section;

the longer the length of the branch section is, the more the number of data points is, the higher the gradient accuracy is obtained, but the longer the branch section is, the actual change in the section can be averaged, and the underestimation of the real gradient change can be caused; if the length of the branch section is too short, the estimated gradient is inaccurate due to too few data points in the branch section;

aligning the elevation data of the branch sections, combining a plurality of groups of GPS data related to each branch section into a data set, and aligning each group of elevation data to solve the problem of average elevation data error; the method specifically comprises the following steps:

fusing multiple groups of GPS data acquired by each shunt segment, merging the multiple groups of GPS data into a data set, calculating an average elevation of all elevation data on a test road as a reference absolute elevation, and subtracting the reference absolute elevation from the elevation of each data point, wherein the relative elevation difference between the data points is unchanged;

calculating the road gradient of a test road and quantifying the precision, performing the slope calculation of each branch section, performing linear regression by using the aligned elevation data of each branch section to estimate the road gradient, and deducing the precision of the measured road gradient according to the standard error of the slope estimation of the linear regression; the method specifically comprises the following steps:

and calculating the road section gradient of the sub road sections, carrying out unary linear regression fitting on the aligned elevation data points of each sub road section to estimate the road gradient, wherein the percent of the tested road gradient is obtained by multiplying the gradient of a fitting straight line by 100, and the gradient of all the sub road sections is connected to obtain the gradient information of the tested road.

2. The portable GPS-based road gradient estimation method according to claim 1, wherein the calculation formula of the number of data points is as follows:

wherein, the number is the number of data points in the branch segment, and is Δ l_i,jThe length of the road section tested for the jth GPS; Δ v_i,jAnd H is the average speed of the road section tested by the ith GPS at the jth time, and the data recording frequency of the GPS.

3. The portable GPS-based road grade estimation method of claim 1, wherein the number of data points per divided sub-section is greater than 20.

4. The portable GPS based road grade estimation method of claim 1 wherein the accuracy of the road grade is the standard error of the slope of the fitted line.

Images