Speed correction method, vehicle navigation method and storage medium

By compensating the odometer speed using a pre-trained error compensation model, the problem of degradation of odometer measurement accuracy is solved, the accuracy and robustness of the navigation system are improved, and the efficient operation of the navigation system is ensured.

CN119085690APending Publication Date: 2024-12-06UNICORE COMM INC
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
CN202411434394.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-14
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

In actual applications, odometers are susceptible to dynamic factors, resulting in a decrease in measurement accuracy and significant deviations in navigation system position estimation, affecting system performance and reliability.

Method used

The pre-trained error compensation model is used to compensate for the forward and lateral velocities, and the slip rate and side slip rate are effectively controlled through the n-order polynomial model to improve the accuracy and robustness of the navigation system.

Benefits of technology

Through error compensation, the accuracy of the navigation system and vehicle control performance are significantly improved, the overall accuracy and reliability of the navigation system are improved, and the stability of the measurement accuracy is ensured.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119085690A_ABST
    Figure CN119085690A_ABST
Patent Text Reader

Abstract

The invention discloses a speed correction method, a vehicle navigation method and a storage medium, error compensation is performed on at least one of speedometer speed and lateral speed participating in dead reckoning, and effective control and compensation are performed on at least one of slip rate and sideslip rate through an n-degree polynomial. And the precision of the navigation system and the control performance of the vehicle are improved, so that the overall precision and reliability of the navigation system are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to but is not limited to vehicle navigation technology, and in particular to a speed correction method, a vehicle navigation method and a storage medium. Background Art

[0002] Integrated navigation systems are widely used in mobile platforms such as autonomous driving, drones, and robots to provide accurate location information and motion state estimation. Odometry is one of the common sensors, which estimates the moving distance of the device by measuring wheel rotation or step count. However, the data of the odometer is easily affected by various dynamic factors in practical applications, resulting in a decrease in measurement accuracy. These dynamic errors may come from sliding, vibration, speed changes, and complex interference from the external environment.

[0003] If dynamic errors are not effectively compensated, the errors in the odometer measurement results will accumulate over time, eventually leading to significant deviations in the position estimate of the navigation system. Due to the influence of dynamic errors, the navigation system will find it difficult to maintain high-precision position estimates during long-term or high-speed motion, which will affect the overall performance and reliability of the system and cannot guarantee measurement accuracy.

[0004] How to compensate for the dynamic error of the odometer is a technical problem that needs to be solved urgently to improve the accuracy and robustness of the integrated navigation system. Summary of the invention

[0005] The present application provides a speed correction method, a vehicle-mounted navigation method and a storage medium, which can compensate for the dynamic error of an odometer and improve the accuracy and robustness of a combined navigation system.

[0006] The present application provides a speed correction method, including:

[0007] According to a pre-trained error compensation model, at least one of the forward speed and the lateral speed is compensated to obtain a compensated speed;

[0008] Among them, the error compensation model is obtained by training the coefficients of the n-order polynomial using the value of the preset parameter during the travel as the input of the n-order polynomial and the actual speed during the travel as the output of the n-order polynomial, where n is an integer greater than or equal to 2.

[0009] The embodiment of the present application also provides a vehicle navigation method, including:

[0010] The dead reckoning is performed using the speed compensated by any of the speed correction methods described above to obtain the position of the vehicle.

[0011] An embodiment of the present application further provides a computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions are used to execute any of the methods described above.

[0012] An embodiment of the present application further provides a computer device, including a memory and a processor, wherein the memory stores the following instructions executable by the processor: for executing the steps of any of the methods described above.

[0013] In the embodiment of the present application, error compensation is performed on at least one of the odometer speed and the lateral speed involved in dead reckoning, and effective control of at least one of the slip rate and the sideslip rate is achieved through an n-order polynomial, thereby improving the accuracy of the navigation system and the control performance of the vehicle, thereby improving the overall accuracy and reliability of the navigation system.

[0014] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The accompanying drawings are used to provide further understanding of the technical solution of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the technical solution of the present application and do not constitute a limitation on the technical solution of the present application.

[0016] Figure 1 This is a schematic diagram showing the effect of the original speed output by the odometer on the positioning accuracy of the navigation system;

[0017] Figure 2 Schematic diagram of the flow of the speed correction method in the embodiment of the present application;

[0018] Figure 3 This is a schematic diagram of the effect of the compensated odometer speed on the positioning accuracy of the navigation system in an embodiment of the present application;

[0019] Figure 4 It is a schematic diagram of simulation analysis of lateral error compensation in an embodiment of the present application;

[0020] Figure 5 Schematic diagram of the structure of the vehicle navigation device in the embodiment of the present application. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the present application more clear, the embodiments of the present application will be described in detail below with reference to the accompanying drawings. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other arbitrarily without conflict.

[0022] In order to facilitate understanding of the present application, the present application will be described more fully below with reference to the relevant drawings. Embodiments of the present application are provided in the drawings. However, the present application can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present application more thorough and comprehensive.

[0023] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit this application.

[0024] It is understood that the terms "first" and "second" used in this application are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0025] It can be understood that the “connection” in the following embodiments should be understood as “electrical connection”, “communication connection”, etc. if the connected circuits, modules, units, etc. have electrical signals or data transmission between each other.

[0026] When used herein, the singular forms "a", "an", and "said / the" may also include plural forms, unless the context clearly indicates otherwise. It should also be understood that the terms "include / comprise" or "have" and the like specify the presence of stated features, wholes, steps, operations, components, parts, or combinations thereof, but do not exclude the possibility of the presence or addition of one or more other features, wholes, steps, operations, components, parts, or combinations thereof. At the same time, the term "and / or" used in this specification includes any and all combinations of the relevant listed items.

[0027] In vehicle navigation, when the satellite navigation receiver cannot locate the vehicle, it uses the information such as the vehicle wheel speed (or pulse) and steering wheel angle measured by the odometer, combined with the vehicle heading and attitude measured by inertial navigation, to calculate the vehicle's real-time position at each moment, which is the dead reckoning algorithm. The odometer wheel speed error is one of the main sources of dead reckoning error, and its accuracy determines the performance of dead reckoning. In order to meet the high-precision positioning requirements of the vehicle, it is necessary to improve the odometer wheel speed accuracy as much as possible.

[0028] In the odometer wheel speed measurement, it is assumed that the speed v of the wheel center satisfies: v = ω p r, where ω pis the angular rate of wheel rotation, which can be obtained through a rotary encoder; r is the tire rotation radius, which is usually obtained through experimental measurement and is assumed to remain unchanged during vehicle driving. However, in actual vehicle driving, the tire rotation radius r is not constant. It is affected by many factors, such as vehicle load, road conditions, tire pressure, etc. These factors lead to the actual tire rotation radius r e This is inconsistent with the ideal tire rotation radius r, which affects the measurement of the vehicle's forward speed. In actual situations, the vehicle's forward speed should be expressed as: v = ω p r e .

[0029] In actual operation, the actual rotation radius r of the wheel when rolling e It is often larger than the statically measured tire rotation radius r, and the actual rotation radius r when the wheel rolls e It is related to multiple factors, such as tires, normal forces, road surfaces, and torques of the transmission or braking systems. Specifically, the material, structure, and degree of wear of the tire will affect its deformation under different loads and speeds, thereby affecting the actual turning radius. The vehicle weight and dynamic load distribution will change the degree of deformation of the tire. The greater the normal force, the more obvious the compression deformation of the tire, resulting in a decrease in the turning radius. The friction coefficient and roughness of different road surfaces will affect the contact and deformation of the tire with the ground, thereby affecting the actual turning radius. When the vehicle accelerates, decelerates, or brakes, the torque applied to the wheel by the transmission or braking system will change the force state of the tire, thereby affecting the rolling radius of the tire. Due to the complexity of these factors, it is difficult for the sensor system inside the vehicle to directly and accurately obtain the real-time actual tire turning radius r e ,This results in that in actual driving process, the speed measurement based on the wheel speed often has errors and cannot accurately reflect the real speed of the vehicle.

[0030] In the ideal tire rotation model, it is usually assumed that the contact surface of the tire is almost motionless with the road surface, so the tread linear speed is the forward speed of the vehicle. However, in the actual vehicle driving process, the tire and the road surface are not completely rolling, but there is a phenomenon of micro-slipping. This phenomenon is caused by the tire rubber block sliding slightly relative to the road surface and then moving like a track. Therefore, every time the wheel rotates, the tread pattern block in contact with the road surface does not roll purely on the surface, which causes the tread speed to be inconsistent with the forward speed of the vehicle. This difference between the tread speed and the forward speed of the vehicle is called the slip rate. The slip rate reflects the proportion of the sliding component of the tire during movement, and its size has an important impact on the control and safety of the vehicle. A slip rate of 0% means that the wheel is in a pure rolling state. At this time, the tread speed is consistent with the forward speed of the vehicle, the lateral adhesion coefficient is the largest, the vehicle's anti-skid ability is the strongest, and the directional stability is the best, especially when braking. A slip rate of 100% means that the wheel is completely locked and in a pure sliding state. At this time, the wheel no longer rolls, the tread and the road surface slide completely, and the vehicle's control and grip are reduced to a minimum.

[0031] In actual driving, no matter what the road conditions are, the tires will slip to a certain extent. For example, on dry roads, the slip rate is usually maintained in the range of 10%-15%, at which time the tires can exert the best grip. However, once the slip rate exceeds this range, the friction of the tires will gradually decrease, and the friction will be reduced to the minimum when the slip rate reaches 100%. Therefore, in order to maintain the best grip and braking effect of the tires, the slip rate must be controlled within an appropriate range.

[0032] Since the slip rate cannot be zero during actual driving, there must be an error between the tread speed and the forward speed of the vehicle. This error will accumulate in the dead reckoning based on tire speed integration (i.e., estimating the position by accumulating the tire speed), resulting in a non-negligible error between the estimated position and the actual position of the vehicle. Therefore, in practical applications, it is necessary to effectively control and compensate the slip rate to improve the accuracy of the navigation system and the control performance of the vehicle. However, the related art does not provide a solution for correcting the above error from the perspective of integrated navigation.

[0033] Similarly, lateral speed is equally important in tire dynamics, especially when the vehicle is turning or the lateral force is large. Due to the existence of lateral speed, the tire not only slips in the longitudinal direction, but also in the lateral direction, which is called the sideslip rate. The sideslip rate is similar to the slip rate and reflects the proportion of the lateral sliding of the tire. When the vehicle turns, the tire does not roll completely in the direction of the vehicle body, but a certain side slip occurs. A smaller side slip rate can improve the handling and stability of the vehicle, while a larger side slip rate may cause the vehicle to understeer or oversteer, thereby affecting the safety of the vehicle. Therefore, in certain practical applications, especially in vehicle dynamic control and advanced driver assistance systems (such as vehicle stability control systems ESC), it is necessary to effectively control and compensate for the sideslip rate to improve the accuracy of the navigation system and the control performance of the vehicle. However, the related art does not provide a solution for correcting the above errors from the perspective of combined navigation.

[0034] Figure 1 This is a schematic diagram showing the effect of the original speed output by the odometer on the positioning accuracy of the navigation system. Taking a certain test data as an example, Figure 1 Medium span Represents high-precision reference speed, v od Represents the original speed output by the odometer, with a high-precision reference speed v span The original speed v output by the odometer od The ratio is used as the first indicator to measure the stability of the scale factor λ, that is, λ = v span / v od Then, when the odometer speed is accurate, the first indicator λ should be a fixed value. During navigation, the first indicator λ is positively correlated with the inertial positioning error, that is, the fluctuation range of the first indicator λ is the inertial positioning error performance (unit: ‰D, D is the inertial recursive mileage).

[0035] Figure 1 In the figure, the horizontal axis is time t, in seconds (s), the left vertical axis is the first indicator λ, and the right vertical axis is the original speed v output by the odometer od Curve 11 shows the change of the first index λ over time, and curve 12 shows the original speed v output by the odometer od Changes over time. Figure 1 As shown in the figure, with the 7m / s line as the boundary, when the speed is greater than 7m / s, the first indicator λ is relatively stable, fluctuating by about 1‰ (such as 3400-3500s), but when the speed is less than 7m / s, the value of the first indicator λ fluctuates violently, with a peak-to-peak value of about 7‰ (such as 3150-3200s), and in extreme cases, the peak-to-peak value reaches 14‰. If the positioning accuracy required for vehicle navigation is 3‰D (D is the mileage), when the speed is less than 7m / s, the positioning accuracy decrease caused by the odometer measurement error alone exceeds the indicator requirement.

[0036] In high-precision vehicle navigation, the errors caused by tire radius changes and longitudinal slip will directly affect the accuracy of the odometer speed, thereby affecting the accuracy of navigation. In order to compensate for these nonlinear errors and improve the overall accuracy and reliability of the navigation system, the embodiment of the present application provides a speed correction method, which improves the accuracy and robustness of the integrated navigation system by compensating for the dynamic error of the odometer, and provides technical support for ensuring the accuracy of navigation. Figure 2 Schematic diagram of the flow of the speed correction method in the embodiment of the present application. Figure 2 As shown, it may include:

[0037] Step 200: Compensate at least one of the forward speed and the lateral speed according to a pre-trained error compensation model to obtain a compensated speed.

[0038] Among them, the error compensation model is obtained by training the coefficients of the n-order polynomial using the value of the preset parameter during the movement as the input of the n-order polynomial and the actual speed during the movement as the output of the n-order polynomial, where n is an integer greater than or equal to 2.

[0039] In one exemplary embodiment, step 200 may include:

[0040] According to the pre-trained odometer error compensation model, the original speed output by the odometer is compensated, and the compensated odometer speed is used as the forward speed.

[0041] Among them, the odometer error compensation model is used to compensate for the original speed output by the odometer during driving; the odometer error compensation model uses the original speed output by the odometer during driving as the input of a pre-set n-order polynomial, and the actual speed of the vehicle during actual driving as the output of the n-order polynomial, and trains the coefficients of the n-order polynomial, where n is an integer greater than or equal to 2.

[0042] In an exemplary embodiment, under an offline condition, before step 200, the following may also be included:

[0043] The original forward speed output by the odometer in different driving scenarios is collected by the inertial navigation system under test, and the high-precision reference speed measured by the high-precision integrated navigation system is used as the real forward speed during the driving process;

[0044] The collected original forward speed output by the odometer is used as the input of a preset n-order polynomial, and the high-precision reference speed is used as the output of the n-order polynomial. The coefficients of the n-order polynomial are fitted to obtain the coefficients of the n-order polynomial, and then the odometer error compensation model is obtained, where n is an integer greater than or equal to 2.

[0045] In the embodiment of the present application, under offline conditions, the coefficients of the n-order polynomial are trained using the original speed of the odometer output reflecting different driving scenarios and the actual speed of the vehicle in actual driving, so as to realize nonlinear modeling of the odometer error and obtain the odometer error compensation model. Since the nonlinear error caused by many factors such as tire radius change and longitudinal slip is taken into account in the offline odometer error nonlinear modeling stage, the odometer error compensation model provided by the embodiment of the present application can better adapt to the nonlinear characteristics of the odometer error in the actual vehicle-mounted process, thereby helping to improve the navigation positioning accuracy.

[0046] In an exemplary embodiment, the set n-order polynomial may be a quadratic polynomial, as shown in formula (1):

[0047]

[0048] In this embodiment, the compensated odometer speed v ′ od Represented as a raw speed v output by the odometer od In formula (1), v ′ od represents the compensated odometer speed, v od is the original speed output by the odometer, k0, k1, k2…k n are the coefficients in the odometer error compensation, which can be obtained by parameter fitting through measured data.

[0049] In an exemplary embodiment, the set n-order polynomial may be a quadratic polynomial, a cubic polynomial, etc., for example:

[0050] In an exemplary embodiment, training the preset n-order polynomial coefficients to obtain the odometer error compensation model may include:

[0051] The original speed output by the odometer is collected by the inertial navigation system under test, such as the inertial measurement unit (IMU); the actual speed of the vehicle during actual driving is measured by a high-precision integrated navigation system (such as SPAN), that is, the high-precision reference speed v span As the real speed of the vehicle in actual driving;

[0052] The collected odometer speed is used as the original speed v output by the odometer in the n-order polynomial such as the quadratic polynomial od , high-precision reference speed v span As an n-degree polynomial such as the compensated odometer speed v in the quadratic polynomial ′ od, perform parameter fitting on an n-order polynomial such as a quadratic polynomial to obtain parameters of the n-order polynomial such as a quadratic polynomial; taking the quadratic polynomial in formula (1) as an example, the parameters include: a first coefficient k0, a second coefficient k1, and a third coefficient k2;

[0053] Substitute the parameters of the obtained n-order polynomial such as the quadratic polynomial into the n-order polynomial such as the quadratic polynomial to obtain the odometer error compensation model.

[0054] In one embodiment, data collection is performed in an open scene to avoid the influence of occlusion on the signal, and to ensure that the high-precision benchmark is in the optimal state of accuracy to ensure the accuracy of the data. The collected data includes various driving scenes such as high speed (such as vehicle speed greater than 40km / h), low speed (such as vehicle speed less than 10km / h), and turning. The data collection time for each scene is not less than a preset time, such as 1 minute, to ensure the diversity and sufficiency of the data, which is conducive to fully stimulating the various errors of the odometer so as to fully calibrate its errors.

[0055] In one embodiment, the least square method may be used to fit the parameters of an n-order polynomial such as a quadratic polynomial to minimize the compensated odometer speed v ′ od The difference from the true speed. This can include:

[0056] The collected odometer speed is used as the input of an n-order polynomial such as a quadratic polynomial, that is, the original speed v output by the odometer od , the actual speed of the vehicle in actual driving is used as the output of the n-order polynomial such as the quadratic polynomial, that is, the compensated odometer speed v ′ od , use the least squares method to fit the parameters of the n-order polynomial such as the quadratic polynomial. Taking the quadratic polynomial in formula (1) as an example, fit the first compensation coefficient k0, the second compensation coefficient k1, and the third compensation coefficient k2 to find the predicted value of the n-order polynomial such as the quadratic polynomial, that is, the compensated odometer speed v ′ od The parameter value that has the smallest difference from the actual speed of the vehicle during actual driving.

[0057] In one exemplary embodiment, step 200 may include:

[0058] The lateral velocity is compensated according to a pre-trained lateral velocity error compensation model to obtain a compensated lateral velocity.

[0059] Among them, the lateral velocity error compensation model is used to compensate for the lateral velocity. The lateral velocity error is obtained by fitting the parameters of the n-order polynomial, with the heading angular velocity and forward velocity during the movement as the input of the n-order polynomial and the actual lateral velocity during the movement as the output of the n-order polynomial.

[0060] In an exemplary embodiment, before step 200, the following steps may also be included:

[0061] The functional relationship between the lateral velocity of the vehicle and known parameters (such as heading angular velocity, vehicle speed, etc.) is established, as shown in formula (2):

[0062]

[0063] in, represents the lateral velocity; v x represents the forward speed of the vehicle, ω represents the heading angular velocity, a, b, c are coefficients, and m, n, p are numbers greater than or equal to 1.

[0064] In one embodiment, formula (2) can be expressed in one of the following forms:

[0065]

[0066] In some exemplary instances, the lateral velocity error is obtained by performing parameter fitting on the n-order polynomial, with the wheel slip angle, heading angular rate, center of mass slip angle, vehicle body parameters (such as deflection stiffness, mass, moment of inertia, etc.), and forward velocity as inputs of the n-order polynomial, and the actual lateral velocity during the moving process as the output of the n-order polynomial.

[0067] In an exemplary embodiment, before step 200, the following steps may also be included:

[0068] First, the functional relationship between the lateral velocity of the vehicle and known parameters (including vehicle parameters, wheel slip angle, heading angular rate, vehicle speed, etc.) is established, as shown in formula (3):

[0069]

[0070] in, represents the lateral velocity; v x represents the forward speed of the vehicle, ω represents the angular velocity of the heading, a, b, c are coefficients, m, n, p are numbers greater than or equal to 1; C f and C r Respectively represent the deflection stiffness of the front and rear axles, m represents the vehicle mass, I f and l r Respectively represent the distance from the center of mass to the front and rear axes, l z represents the yaw moment of inertia, δ frepresents the front wheel turning angle. f is a fitting function that represents the relationship between the lateral velocity of the vehicle and the known parameters.

[0071] In one embodiment, formula (3) can be expressed as:

[0072] It should be noted that the order of each parameter in formula (3) is not limited, and the order of each parameter can be determined according to the fitting situation, for example, the parameter with the smallest fitting residual is determined to have an order of 1. In one embodiment, formula (3) can be simplified as

[0073] In an exemplary embodiment, the specific form of the function f depends on the surface fitting result of the actual vehicle test data, so the function f is not a fixed function, but an empirical model obtained through the fitting process. In an embodiment, an algorithm such as the least squares method can be used in the fitting to adjust the shape of the function so that the function f can best describe the relationship between the input parameters and the lateral velocity error.

[0074] In one embodiment, the heading angular rate ω can be measured by an inertial measurement unit (IMU); the forward velocity v x Can be replaced by odometer speed.

[0075] Next, all variable data related to formula (2) or formula (3) during the actual vehicle test are collected. In one embodiment, high-precision benchmark collection can be used.

[0076] Then, the collected data is fitted using surface fitting, and the appropriate fitting order is selected according to the fitting residual. The fitting formula is shown in formula (4):

[0077]

[0078] In one embodiment, the fitting may be performed using an algorithm such as least squares, and the specific implementation will not be described in detail here, nor will it limit the protection scope of the present application.

[0079] Finally, in the actual navigation process, the real-time parameters are substituted into formula (4), and the lateral velocity error compensation value, i.e., the compensated lateral velocity, is calculated according to the fitting results.

[0080] Step 200 provides a speed correction method, wherein the original speed output by the odometer is the speed compensated by the odometer error compensation model obtained by training an n-order polynomial, and the compensated speed is the speed after the nonlinear error caused by the tire radius change and the longitudinal slip is corrected, thereby providing an accurate odometer speed for the subsequent dead reckoning, thereby providing a technical guarantee for ensuring the accuracy of navigation; or, the lateral speed is the speed compensated and corrected by the pre-trained lateral speed error compensation model, thereby providing an accurate odometer speed for the subsequent dead reckoning, thereby providing a technical guarantee for ensuring the accuracy of navigation.

[0081] Based on the speed correction method provided in step 200, the embodiment of the present application further provides a vehicle navigation method, further comprising step 201: performing dead reckoning using the compensated speed to obtain the position of the vehicle.

[0082] In an exemplary embodiment, step 201 may include: performing dead reckoning using the compensated odometer speed to obtain the position of the vehicle. Specifically, it may include:

[0083] Using the above compensated odometer speed v ′ od , perform dead reckoning according to formula (4):

[0084]

[0085] In formula (5), Represents the transformation matrix from the IMU coordinate system to the navigation coordinate system, Represents the transformation matrix from the vehicle system to the IMU coordinate system, and Both can be obtained from inertial navigation. pos k Represents the dead reckoning position information at the previous moment, i.e., time k, pos k+1 Indicates the dead reckoning position information at the current time, i.e., time k+1. ′ od Represents the speed of the odometer after compensation. Speed ​​vector Indicates that the vehicle's motion is primarily in the direction of the vehicle's forward motion, assuming that the vehicle has no lateral or vertical velocity.

[0086] In an exemplary embodiment, step 201 may include: performing dead reckoning using the compensated lateral velocity to obtain the position of the vehicle. Specifically, it may include:

[0087] Using the above compensated lateral speed Dead reckoning is performed according to formula (6):

[0088]

[0089] In formula (5), v od Indicates the original speed output by the odometer.

[0090] In an exemplary embodiment, step 201 may include: performing dead reckoning using the compensated odometer speed and lateral speed to obtain the position of the vehicle. Specifically, it may include:

[0091] Using the above compensated lateral speed Dead reckoning is performed according to formula (7):

[0092]

[0093] In formula (5), v ′ od Indicates the speed of the odometer after compensation.

[0094] In an in-vehicle navigation method provided by an embodiment of the present application, the odometer speed involved in dead reckoning is the speed compensated by an odometer error compensation model obtained by training an n-order polynomial, so that the compensation for the original speed output by the odometer fully copes with the change of nonlinear errors and better compensates for these nonlinear errors, thereby improving the overall accuracy and reliability of the navigation system.

[0095] In an on-vehicle navigation method provided by an embodiment of the present application, the lateral speed involved in dead reckoning is the speed compensated by a lateral speed error compensation model, thereby achieving effective control and compensation of the sideslip rate, improving the accuracy of the navigation system and the control performance of the vehicle, thereby improving the overall accuracy and reliability of the navigation system.

[0096] An on-vehicle navigation method provided in an embodiment of the present application performs error compensation on both the odometer speed and the lateral speed involved in dead reckoning, thereby achieving effective control and compensation of the slip rate and the sideslip rate at the same time, improving the accuracy of the navigation system and the control performance of the vehicle, thereby improving the overall accuracy and reliability of the navigation system.

[0097] The present application also provides a computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions are used to execute any of the methods described above.

[0098] The present application further provides a computer device, including a memory and a processor, wherein the memory stores the following instructions executable by the processor: for executing the steps of any of the methods described above.

[0099] Figure 3 This is a schematic diagram showing the effect of the compensated odometer speed on the positioning accuracy of the navigation system in an embodiment of the present application. Figure 3 Mediumspan Represents high-precision reference speed, v od Represents the original speed output by the odometer, v ′ od represents the compensated odometer speed. In one embodiment, the high-precision reference speed v span The original speed v output by the odometer od The ratio is used as the first indicator to measure the stability of the scale factor λ, that is, λ = v span / v od In another embodiment, a high-precision reference speed v span and the compensated odometer speed v ′ od The ratio is used as the second indicator to measure the stability of the calibration coefficient λ ′ , that is, ′ =v span / v ′ od .

[0100] Figure 3 In the figure, the horizontal axis is time t, in seconds (s), and the vertical axis on the left is the first index λ and the second index λ ′ , the vertical axis on the right is the original speed v output by the odometer od and compensated odometer speed v ′ od Curve 31 represents the change of the first index λ over time, and curve 32 represents the odometer speed v od Curve 33 represents the second index λ over time. ′ Changes over time, such as Figure 3 According to the test data shown in the middle curve 33, when the vehicle speed is less than 7m / s, the second index λ ′ The value fluctuation range (maximum value 1.01529, maximum value 1.01187, minimum value 1.00905) is reduced from 7‰ (extreme value 14‰) to 2.8‰ (extreme value 6‰). It can be seen that the odometer correction method provided in the embodiment of the present application has an obvious compensation effect on the original speed output by the odometer, and the accuracy remains basically unchanged when the vehicle speed is greater than 7m / s.

[0101] Figure 4 Schematic diagram of simulation analysis of lateral error compensation in the embodiment of the present application, such as Figure 4 As shown in the figure, the simulation analysis results of lateral error compensation based on a certain turning accuracy test data are shown. Figure 4The trajectory without lateral slip compensation and after lateral slip compensation by the method provided in the embodiment of the present application is compared in FIG. 41, curve 41 represents the trajectory after the lateral slip compensation method provided in the embodiment of the present application, and curve 42 is the corresponding reference data trajectory after compensation; curve 51 is the trajectory without lateral slip compensation, and curve 52 is the corresponding reference data trajectory without compensation. Figure 4 The results show that after using the lateral error compensation in the embodiment of the present application, the position step caused by the lateral velocity is significantly suppressed, and the compensated inertial recursive trajectory is smoother than the uncompensated trajectory.

[0102] The obtained lateral speed error compensation model is substituted into the combined navigation algorithm, as shown in Table 1. Table 1 shows an embodiment of the turning recursive error statistics of the vehicle-mounted test. According to the test results of 12 intersection turns in the vehicle-mounted test, the average error after compensation is reduced by 55% to 60%, and the maximum position error is no more than 0.35m, which meets the accuracy requirements of inertial recursion for intelligent driving.

[0103]

[0104]

[0105] Table 1

[0106] It can be seen from Table 1 that the position error after compensation is greatly reduced compared with that before compensation, and the error range is controlled within 0.35 meters, which greatly improves the navigation accuracy of the system, especially during turning.

[0107] Figure 5 A schematic diagram of the structure of the vehicle navigation device provided in the embodiment of the present application is shown in FIG. Figure 5 As shown, it at least includes: a compensation module and a calculation module; wherein,

[0108] A compensation module, used to compensate the forward speed and / or lateral speed according to a pre-trained error compensation model to obtain a compensated speed; wherein the error compensation model is obtained by training the coefficients of the n-order polynomial using the value of the preset parameter during the travel as the input of the n-order polynomial and the actual speed during the travel as the output of the n-order polynomial, where n is an integer greater than or equal to 2;

[0109] The computing module is used to perform dead reckoning using the compensated odometer speed to obtain the vehicle's position.

[0110] In one exemplary embodiment, the compensation module is used to:

[0111] According to the pre-trained odometer error compensation model, the original speed output by the odometer is compensated to obtain the compensated odometer speed; wherein the odometer error compensation model is used to compensate the original speed output by the odometer; the odometer error compensation model is obtained by training the coefficients of the n-order polynomial by taking the original speed output by the odometer during the driving process as the input of a pre-set n-order polynomial and the actual speed of the vehicle during the actual driving as the output of the n-order polynomial, where n is an integer greater than or equal to 2;

[0112] In an exemplary embodiment, a training module may also be included, for:

[0113] Under offline conditions, the original speed output by the odometer in different driving scenarios is collected through the inertial navigation system under test, such as the inertial measurement unit (IMU), and the high-precision reference speed v is obtained by high-precision integrated navigation system such as SPAN measurement. span As the real speed of the vehicle in actual driving;

[0114] The original speed of the odometer output is collected as the input of the preset n-order polynomial, and the high-precision reference speed is used as the output of the n-order polynomial. The n-order polynomial is fitted with parameters and the coefficients of the n-order polynomial are obtained to obtain the odometer error compensation model.

[0115] In an exemplary embodiment, the n-th degree polynomial may be a quadratic polynomial, or other higher degree polynomials.

[0116] In the vehicle-mounted navigation device provided by the embodiment of the present application, the odometer speed involved in dead reckoning is the speed compensated by the odometer error compensation model obtained by n-order polynomial training, so that the compensation for the original speed output by the odometer fully copes with the change of nonlinear errors and better compensates for these nonlinear errors, thereby improving the overall accuracy and reliability of the navigation system.

[0117] In one exemplary embodiment, the compensation module is used to:

[0118] The lateral speed is compensated according to a pre-trained lateral speed error compensation model to obtain the compensated lateral speed; wherein the lateral speed error compensation model takes the wheel slip angle, heading angular velocity, vehicle body parameters, and forward speed during the moving process as the input of the n-order polynomial, takes the actual lateral speed during the moving process as the output of the n-order polynomial, and obtains the lateral speed by parameter fitting of the n-order polynomial.

[0119] In an exemplary embodiment, a training module may also be included, for:

[0120] Establish the functional relationship between the lateral velocity of the vehicle and known parameters (including vehicle parameters, wheel slip angle, heading angular rate, vehicle speed, etc.); collect all variable data related to the established functional relationship during the actual vehicle test; use surface fitting to fit the collected data. In the actual navigation process, calculate the lateral velocity error compensation value, that is, the compensated lateral velocity.

[0121] In the vehicle-mounted navigation device provided in the embodiment of the present application, the lateral speed involved in dead reckoning is the speed compensated by the lateral speed error compensation model, thereby achieving effective control and compensation of the sideslip rate, improving the accuracy of the navigation system and the control performance of the vehicle, thereby improving the overall accuracy and reliability of the navigation system.

[0122] In one exemplary embodiment, the compensation module is used to:

[0123] According to a pre-trained odometer error compensation model, the original speed output by the odometer is compensated to obtain a compensated odometer speed; wherein the odometer error compensation model is used to compensate the original speed output by the odometer; the odometer error compensation model is obtained by training the coefficients of the n-order polynomial by taking the original speed output by the odometer during the driving process as the input of a pre-set n-order polynomial and the actual speed of the vehicle during the actual driving as the output of the n-order polynomial, where n is an integer greater than or equal to 2; and

[0124] The lateral speed is compensated according to a pre-trained lateral speed error compensation model to obtain the compensated lateral speed; wherein the lateral speed error compensation model takes the wheel slip angle, heading angular velocity, vehicle body parameters, and forward speed during the moving process as the input of the n-order polynomial, takes the actual lateral speed during the moving process as the output of the n-order polynomial, and obtains the lateral speed by parameter fitting of the n-order polynomial.

[0125] The vehicle-mounted navigation device provided in the embodiment of the present application performs error compensation on the odometer speed and lateral speed involved in dead reckoning, thereby achieving effective control and compensation of the slip rate and sideslip rate at the same time, improving the accuracy of the navigation system and the control performance of the vehicle, thereby improving the overall accuracy and reliability of the navigation system.

[0126] Although the embodiments disclosed in this application are as above, the contents described are only embodiments adopted to facilitate understanding of this application and are not intended to limit this application. Any technician in the field to which this application belongs can make any modifications and changes in the form and details of implementation without departing from the spirit and scope disclosed in this application, but the scope of patent protection of this application shall still be based on the scope defined in the attached claims.

Claims

1. A speed correction method, characterized in that: include: According to a pre-trained error compensation model, at least one of the forward speed and the lateral speed is compensated to obtain a compensated speed; Among them, the error compensation model is obtained by training the coefficients of the n-order polynomial using the value of the preset parameter during the travel as the input of the n-order polynomial and the actual speed during the travel as the output of the n-order polynomial, where n is an integer greater than or equal to 2.

2. The speed correction method according to claim 1, wherein: The forward speed is compensated according to the pre-trained error compensation model, including: According to a pre-trained odometer error compensation model, the original speed output by the odometer is compensated to obtain the compensated odometer speed as the forward speed; Among them, the odometer error compensation model is obtained by training the coefficients of the n-order polynomial by using the original speed output by the odometer during the driving process as the input of a preset n-order polynomial and the actual forward speed during the driving process as the output of the n-order polynomial, where n is an integer greater than or equal to 2.

3. The speed correction method according to claim 2, further comprising: Under offline conditions, the original forward speed output by the odometer in different driving scenarios is collected through the inertial navigation system under test, and the high-precision reference speed measured by the high-precision integrated navigation system is used as the real forward speed during the driving process; The collected original forward speed output by the odometer is used as the input of the n-order polynomial, and the high-precision reference speed is used as the output of the n-order polynomial. The n-order polynomial is parameter-fitted to obtain the coefficients of the n-order polynomial, and then the odometer error compensation model is obtained.

4. The speed correction method according to claim 2 or 3, wherein: The n-degree polynomial is as follows: In the above formula, v ′ od represents the compensated odometer speed, v od represents the original speed output by the odometer, k0, k1, k2…k n is the coefficient.

5. The speed correction method according to claim 1, wherein: The compensating the lateral velocity according to the pre-trained error compensation model includes: Compensating the lateral velocity according to a pre-trained lateral velocity error compensation model to obtain a compensated lateral velocity; Among them, the lateral velocity error compensation model is obtained by fitting the parameters of the n-order polynomial by taking the heading angular velocity and forward velocity during the movement as the input of the n-order polynomial and the actual lateral velocity during the movement as the output of the n-order polynomial.

6. The speed correction method according to claim 5, wherein: The functional relationship is: in, represents the lateral velocity; v x represents the forward speed of the vehicle, ω represents the heading angular velocity, a, b, c are coefficients, and m, n, p are numbers greater than or equal to 1.

7. The speed correction method according to claim 3 or 5, wherein: The parameter fitting is performed on the n-order polynomial using the least square method to minimize the difference between the compensated speed and the true speed.

8. A vehicle navigation method, characterized in that: include: The dead reckoning is performed using the speed compensated by the speed correction method described in any one of claims 1 to 7 to obtain the position of the vehicle.

9. A computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions are used to execute the method according to any one of claims 1 to 8.

10. A computer device comprising a memory and a processor, wherein: The memory stores the following instructions executable by the processor: used to execute the steps of the method described in any one of claims 1-8.

Citation Information

Patent Citations

  • Method and device for calibrating odometer based on strapdown inertial navigation / satellite integrated navigation system

    CN106595715A

  • Doppler velocimeter speed correction method and autonomous underwater vehicle navigation error correction method

    CN107765032A

  • Navigation method and device

    CN112859132A

  • Urban rail vehicle running gear area array temperature measurement optimization method based on speed compensation

    CN116578808A

  • Improved MPC trajectory tracking control method for autonomous vehicle based on neural network

    CN118289016A