Load prediction method and system for vehicle steering system

Abstract

The invention discloses a load prediction method and system for a vehicle steering system, and the method comprises the steps: obtaining the GPS track data of a user road vehicle based on a public data platform; based on the GPS track data, the turning curvature of the user road vehicle is calculated; constructing a transverse dynamic model of the vehicle based on the turning curvature; constructing a load calculation model of the vehicle steering system based on the transverse dynamic model; and completing prediction of the vehicle steering system load by using the load calculation model. Key input is provided for design, evaluation and reliability test of the steering system, meanwhile, a load basis is provided for constructing a steering system reliability test cyclic working condition associated with a user load, the prediction accuracy is improved, dependence on sensor installation and load analysis complexity is reduced, and the reliability of the steering system is improved. And the safety and the reliability of an automobile steering system can be improved.

Description

Technical Field

[0001] The present invention relates to the field of vehicle steering system load prediction, and particularly to a load prediction method and system for a vehicle steering system. Background Art

[0002] The steering system is an important part of an automobile, and its performance directly affects the handling stability of the automobile, which is crucial for ensuring vehicle driving safety and the personal safety of the driver. With the development of automobile technology, the steer-by-wire system (SBW) is considered to be one of the key technologies for realizing high-level autonomous driving in the future. It controls the output torque of the steering motor and the steering wheel angle according to the driver's instructions and the current vehicle operating state, so as to realize the steering function; it can achieve goals such as personalized driving, assisted driving, and driverless driving, and directly controls the precise control of the autonomous driving path and direction. Whether it is a traditional steering system or a common rack and pinion type steer-by-wire system, their global system loads mainly include the steering wheel angle, tie rod displacement, steering rod force, steering gear rack force, and steering motor torque. Therefore, accurately obtaining these system loads is the key input for carrying out the design and evaluation of the steering system.

[0003] Currently, constructing the load boundary and test specifications of vehicle key systems based on user operation data is one of the research directions that the automotive industry focuses on. However, due to factors such as the complexity of sensor installation and load analysis, and relevant regulatory restrictions, it is difficult to directly obtain these system loads based on user-operated vehicles. Therefore, predicting and extracting the key loads of the steering system that reflect the actual usage conditions of users provides loads for the reliability assessment of the steering system and bench tests, and provides a basis for constructing the reliability test cycle conditions of the steering system associated with user loads in the future. Summary of the Invention

[0004] To solve the technical problems in the above background, the present invention provides a load prediction method for a vehicle steering system, and the steps include:

[0005] Based on a public data platform, obtain the GPS trajectory data of user road vehicles;

[0006] Based on the GPS trajectory data, calculate the turning curvature of user road vehicles;

[0007] Based on the turning curvature, construct a lateral dynamics model of the vehicle;

[0008] Based on the lateral dynamics model, construct a load calculation model of the vehicle steering system;

[0009] Use the load calculation model to complete the prediction of the vehicle steering system load.

[0010] Preferably, the method for obtaining the GPS trajectory data includes: extracting the GPS trajectory data of the user's road vehicle through a public vehicle big data platform, including the time history data of longitude, latitude and vehicle speed; after collection, preprocessing the GPS trajectory data, and the steps include: setting the low-pass filter cut-off frequency to 0.5 Hz, and filtering the GPS trajectory data; using a threshold determination rule to determine the abnormality and invalidity of the user data, removing the invalid or abnormal data in the GPS trajectory data and then performing frame filling to obtain the effective vehicle GPS trajectory data.

[0011] Preferably, the method for frame filling for longitude, latitude and vehicle speed includes:

[0012] For the latitude lat, if |lat i -lat i-1 |>0.1, then lat i =(lat i-1 +lat i+1 ) / 2; if |lat i -lat i-1 |≤0.1, then lat i remains unchanged.

[0013] For the longitude lon, if |lon i -lon i-1 |>0.1, then lon i =(lon i-1 +lon i+1 ) / 2; if |lon i -lon i-1 |≤0.1, then lon i remains unchanged.

[0014] For the vehicle speed v, if |v i -v i-1 |>10, then v i =(v i-1 +v i+1 ) / 2; if |v i -v i-1 |≤10, then v i remains unchanged.

[0015] Preferably, the method for calculating the turning curvature includes: taking points of longitude and latitude and spline interpolation, judgment processing of vehicle speed, calculation of curvature and determination of the positive and negative of curvature;

[0016] Among them, the method for calculating the curvature and determining the positive and negative of curvature includes:

[0017] Using the Haversine formula to calculate the turning curvature, and the steps include:

[0018]

[0019] ρ = 1 / (R e ·b)

[0020] where a 1 represents an intermediate variable, which calculates a part of the radian between two points; dlat represents the adjacent latitude difference; dlon represents the adjacent longitude difference; b 1 represents the radian between two points; R e represents the radius of the earth; ρ represents the turning curvature;

[0021] Using the scalar form of the cross product in two-dimensional space as the basis for the relative direction of two vectors to determine the positive and negative of the curvature, the steps include:

[0022]

[0023] where and represent two unit vectors, which are used to calculate the cross product; lon represents the longitude; lat represents the latitude;

[0024] If the cross product cross_product > 0, then relative to is counterclockwise, judged to be to the left, that is, the curvature is positive; if the cross product cross_product < 0, then relative to is clockwise, judged to be to the right, that is, the curvature is negative.

[0025] Preferably, the constructed lateral dynamics model includes:

[0026]

[0027] where v represents the actual vehicle speed at the center of mass; ρ represents the radius of curvature; β represents the sideslip angle at the center of mass; ω r represents the yaw angular velocity of the vehicle; k 1 represents the front wheel sideslip stiffness; k 2 represents the rear wheel sideslip stiffness; I z represents the moment of inertia of the vehicle about the z-axis; represents the front wheel sideslip angle; represents the rear wheel sideslip angle.

[0028] Preferably, the constructed load calculation model includes: a steering resistance moment calculation model, a steering tie rod force calculation model, and a steering gear rack force calculation model.

[0029] The present invention also provides a load prediction system for a vehicle steering system, which is used to implement the above method and includes: a collection module, a calculation module, a first construction module, a second construction module, and a prediction module;

[0030] The collection module is used to obtain the GPS trajectory data of the user's road vehicle based on a public data platform;

[0031] The calculation module is used to calculate the turning curvature of the user's road vehicle based on the GPS trajectory data;

[0032] The first construction module is used to construct a lateral dynamics model of the vehicle based on the turning curvature;

[0033] The second construction module is used to construct a load calculation model of the vehicle steering system based on the lateral dynamics model;

[0034] The prediction module is used to complete the prediction of the load of the vehicle steering system by using the load calculation model.

[0035] Preferably, the working process of the collection module includes: extracting the GPS trajectory data of the user's road vehicle through a public vehicle big data platform, including the time history data of longitude, latitude, and vehicle speed; after collection, preprocessing the GPS trajectory data, and the steps include: setting the low-pass filter cut-off frequency to 0.5 Hz and filtering the GPS trajectory data; using a threshold determination rule to determine anomalies and invalidity of the user data, removing invalid or abnormal data from the GPS trajectory data and then filling frames to obtain the effective vehicle GPS trajectory data.

[0036] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0037] By combining the vehicle dynamics model and GPS trajectory data, the present invention proposes a method and system for quickly predicting the key loads of a vehicle steering system. The present invention provides key inputs for the design, evaluation, and reliability test of the steering system, and at the same time provides a load basis for constructing a reliability test cycle condition of the steering system associated with user loads, improves the prediction accuracy, reduces the dependence on sensor installation and load analysis complexity, and helps to improve the safety and reliability of the vehicle steering system. Description of the Drawings

[0038] In order to more clearly illustrate the technical solutions of the present invention, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings without creative efforts.

[0039] Figure 1 It is a flowchart of the method in the embodiment of the present invention;

[0040] Figure 2 It is a steering wheel angle load diagram predicted by using the extreme operating condition method to check the steering resistance moment in the embodiment of the present invention;

[0041] Figure 3 It is a tie rod force load diagram predicted by using the inverse method of model input parameters to check the steering resistance moment in the embodiment of the present invention;

[0042] Figure 4 It is a predicted curvature check GPS data diagram in the embodiment of the present invention;

[0043] Figure 5 It is a predicted curvature check data ratio diagram in the embodiment of the present invention;

[0044] Figure 6 It is a curvature relative error analysis data diagram of the predicted curvature check in the embodiment of the present invention;

[0045] Figure 7 It is a normal distribution diagram of the curvature relative error of the predicted curvature check in the embodiment of the present invention;

[0046] Figure 8 It is a steering system analysis diagram for establishing a load calculation model in the embodiment of the present invention;

[0047] Figure 9 It is a partial curvature data diagram in the embodiment of the present invention;

[0048] Figure 10 It is a partial key load diagram of the steering system in the embodiment of the present invention; where (a) represents the steering wheel angle; (b) represents the tie rod displacement; (c) represents the tie rod force; (d) represents the steering gear rack force; (e) represents the steering motor torque. Detailed implementation manners

[0049] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0050] To make the above objects, features, and advantages of the present invention more obvious and understandable, the present invention will be further described in detail below in conjunction with the accompanying drawings and specific implementation manners.

[0051] Embodiment 1

[0052] This embodiment discloses a load prediction method for a vehicle steering system, and the steps include:

[0053] S1. Based on the public data platform, obtain the GPS trajectory data of the user's road vehicle.

[0054] First of all, this embodiment needs to construct a GPS trajectory database of the user's road vehicle. The GPS trajectory data of the user's road vehicle can be extracted through the public vehicle big data platform, including the time history data of longitude, latitude and vehicle speed. The actual GPS driving trajectory data of the user's road vehicle covers the GPS driving trajectory data of vehicles in different regions and models. For the abnormal data caused by poor GPS signals, long-term idling, and severe acceleration and deceleration during the data collection process, the user's GPS data is cleaned to generate a new GPS vehicle driving trajectory after removing the abnormal or invalid GPS data.

[0055] In this embodiment, the low-pass filter cut-off frequency is set to 0.5 Hz to filter the GPS trajectory data; the threshold judgment rule is used to judge the abnormality and invalidity of the user data, and the invalid or abnormal data in the GPS trajectory data is removed and then frame filling is performed to obtain an available vehicle GPS trajectory database.

[0056] The above steps of frame filling include (for longitude, latitude and vehicle speed):

[0057] For the latitude lat, if |lat i -lat i-1 | > 0.1, then lat i = (lat i-1 + lat i+1 ) / 2; if |lat i -lat i-1 | ≤ 0.1, then lat i remains unchanged.

[0058] For the longitude lon, if |lon i -lon i-1 | > 0.1, then lon i = (lon i-1 + lon i+1 ) / 2; if |lon i -lon i-1 | ≤ 0.1, then lon i remains unchanged.

[0059] For the vehicle speed v, if |v i -v i-1 | > 10, then v i = (v i-1 + v i+1 ) / 2; if |v i -vi-1 If ≤ 10, then v i remains unchanged.

[0060] S2. Calculate the turning curvature of the user's road vehicle based on GPS trajectory data.

[0061] Based on the obtained available vehicle GPS trajectory database, calculate the turning curvature of the user's road vehicle. The calculation steps include: longitude and latitude point sampling and spline interpolation, vehicle speed judgment and processing, curvature calculation, and curvature sign determination, etc. The specific steps are as follows:

[0062] S201. Longitude and latitude point sampling and spline interpolation.

[0063] The number of intervals for longitude, latitude, and vehicle speed sampling is i, which forms longitude, latitude, and vehicle speed data of a new time history.

[0064] Perform spline interpolation on the data after the above longitude, latitude, and vehicle speed sampling to restore it to the original time history data. The expression for the number of interpolation points is:

[0065] n i = i (1)

[0066] Then perform spline interpolation on it. Taking vehicle speed as an example:

[0067] 1) Assume that the two known vehicle speed points are (t 1 , v 1 ) and (t 2 , v 2 );

[0068] 2) Construct a spline interpolation function S(t), on the interval [t 1 , t 2 :

[0069] S(t) = a + b(t - t 1 ) + c(t - t 1 ) 2 + d(t - t 1 ) 3 (2)

[0070] 3) According to the above spline interpolation conditions, the following system of equations can be listed:

[0071] a) S(t 1 ) = v 1 , that is, a = v 1 ;

[0072] b) S(t 2 ) = v 2 , substituting into the function gives v 1 + b(t 2 - t 1) + c(t 2 -t 1 ) 2 +d(t 2 -t 1 ) 3 =v 2 ;

[0073] 4) Determine the coefficients c and d by using the second derivatives at both endpoints being 0.

[0074] 5) Substitute c = 0 and d = 0 into v 1 +b(t 2 -t 1 )+c(t 2 -t 1 ) 2 +d(t 2 -t 1 ) 3 =v 2 to find

[0075] 6) Divide the interval |v i =i into i equal parts according to the set number of interpolation points n 1 ,v 2 |. The interpolation points are respectively Then substitute the above interpolation points into S(t) = v 1 +b(t - t 1 ) to obtain the interpolated vehicle speeds.

[0076] S202. Judgment and processing of vehicle speed.

[0077] Find the index of the position where the vehicle speed is less than 0 and make the vehicle speed positive.

[0078] S203. Calculation of curvature and determination of the sign of curvature.

[0079] Use the Haversine formula to calculate the turning curvature. The steps include:

[0080]

[0081] ρ = 1 / (R e ·b) (5)

[0082] In the formula, a 1 represents an intermediate variable and calculates a part of the radian between two points; dlat represents the adjacent latitude difference; dlon represents the adjacent longitude difference; b 1 represents the radian between two points; R e represents the radius of the earth; ρ represents the turning curvature.

[0083] The determination of the positive and negative of curvature is based on the scalar form of the cross product in a two-dimensional space, considering the relative direction of two vectors. The steps are as follows:

[0084]

[0085] In the formula, and represent two unit vectors used to calculate the cross product; lon represents longitude; lat represents latitude.

[0086] If the cross product cross_product > 0, then is counterclockwise relative to , and it is determined to be to the left, that is, the curvature is positive; if the cross product cross_product < 0, then is clockwise relative to , and it is determined to be to the right, that is, the curvature is negative.

[0087] The above steps finally obtain the turning curvature during vehicle driving.

[0088] S3. Based on the turning curvature, construct the lateral dynamics model of the vehicle.

[0089] Based on the turning curvature obtained above during vehicle driving, establish the lateral dynamics model of the vehicle to obtain the yaw rate and the sideslip angle of the center of mass. The constructed lateral dynamics model includes:

[0090]

[0091] In the formula, v represents the actual vehicle speed at the center of mass; ρ represents the radius of curvature; β represents the sideslip angle of the center of mass; ω r represents the yaw rate of the vehicle; k 1 represents the cornering stiffness of the front wheels; k 2 represents the cornering stiffness of the rear wheels; I z represents the moment of inertia of the vehicle about the z-axis; represents the front wheel sideslip angle; represents the rear wheel sideslip angle.

[0092] S4. Based on the lateral dynamics model, construct the load calculation model of the vehicle steering system.

[0093] The constructed load calculation model includes: calculation models for the steering resistance moment, the steering tie rod force, the steering gear rack force, etc., which are calculation models for the global loads of the steering system.

[0094] (1) Construction of the steering resistance moment calculation model:

[0095] Considering the dominant position of each self-aligning torque during steering, the main self-aligning torques are selected to establish a calculation model for the lateral force self-aligning torque and a calculation model for the gravity self-aligning torque. The calculation steps for the steering resistance torque are as follows:

[0096] M sw = M G + M y (11)

[0097] Among them, M G represents the gravity self-aligning torque; M y represents the lateral force self-aligning torque.

[0098] The calculation expression of M G is:

[0099] M G = G f · e z · sinθ · cosγ · sinδ (12)

[0100] e z = (d 偏 + r w tanθ) cosθ (13)

[0101] In the formula, G f represents the front axle load; e z represents the gravity self-aligning torque arm; θ represents the kingpin inclination angle; γ represents the caster angle; r w represents the dynamic rolling radius of the tire; d 偏 represents the kingpin offset.

[0102] The calculation expression of M y is:

[0103] M y = k · α 1 · (n R + n k ) (14)

[0104] Among them, α 1 represents the front wheel side slip angle; k represents the cornering stiffness; the shift distance n k formed by the kingpin ground contact point and the center of the tire footprint, the offset distance n R and the tire ground contact footprint length l 轮 The calculation expressions are:

[0105]

[0106] In the formula, r represents the wheel radius; D represents the tire outer diameter; Δ represents the radial deformation of the steering tire under the action of the front axle vertical load, which is determined by unit conversion through the following formula:

[0107]

[0108] K Δ = 0.0015B + 0.42 (18)

[0109] In the formula, C represents the coefficient, taking 1.5 (radial tire); B represents the tire section width; p 压 represents the tire inflation pressure; G 1 represents the rated load of a single tire.

[0110] A calculation model for the steering wheel angle and tie rod displacement is established, and the calculation formula is as follows:

[0111] l d = l sw ·(ψ s + ψ s ′) / i s (19)

[0112] ψ s ′ = l sw ·M sw / (l s ·K sd ·i s ) (20)

[0113] δ = M sw / K sl + l d / l s = arctan(ρL) (21)

[0114] In the formula, l d represents the tie rod displacement; l s represents the length of the steering link; l sw represents the length of the steering arm; i s represents the steering transmission ratio; δ represents the front wheel angle; ψ s represents the steering wheel angle; ψ s ′ represents the total moment feedback steering angle of the tire around the kingpin; M sw represents the total moment of the front wheel around the kingpin; K sd represents the torsional stiffness of the steering column; K sl represents the equivalent stiffness of the steering connection mechanism.

[0115] (2) Construction of the steering tie rod force calculation model:

[0116]

[0117] M H = M sw (23)

[0118] Among them, F 拉Indicates the steering tie rod force.

[0119] The calculation formula for the included angle η between the knuckle arm and the steering tie rod is expressed as:

[0120]

[0121] In the formula, φ represents the included angle between the knuckle arm and the steering tie rod; p represents the center distance between the two ball joints; s represents the rack stroke; h represents the installation distance from the rack axis to the bottom of the trapezoid; l represents the length of the knuckle arm; m represents the module of the steering gear; μ represents the helix angle μ = 14°; Represents the length of the other side of the triangle formed by the knuckle arm and the steering tie rod in the steering position.

[0122] Establish a calculation model for the output torque of the steering motor (since the motion state of the rack has little influence on the output torque of the steering motor, the motion state during rack movement is not considered), and the calculation formula is as follows:

[0123]

[0124] In the formula, T a Represents the output torque of the steering motor; r p Represents the pitch circle radius of the pinion; N represents the reduction ratio of the steering motor.

[0125] (3) Construction of the calculation model for the rack force of the steering gear:

[0126]

[0127] In the formula, F N Represents the rack force of the steering gear.

[0128] The calculation formula for the included angle ε between the steering tie rod and the rack shaft is expressed as:

[0129]

[0130] γ - α = φ + χ (31)

[0131]

[0132] In the formula, F N Represents the rack force of the rack and pinion steering gear; p represents the center distance between the two ball joints; d represents the distance between the kingpins; l represents the length of the knuckle arm; λ represents the pressure angle, λ = 20°; α represents the angle rotated by the knuckle arm during steering; χ represents the included angle between the other side of the triangle formed by the knuckle arm and the steering tie rod and the connection line between the left and right kingpins.

[0133] S5. Use the load calculation model to complete the prediction of the vehicle steering system load.

[0134] Calculate the key loads of the steering system using the constructed load calculation model, including the steering wheel angle, tie rod displacement, steering rod force, steering gear rack force, and steering motor torque.

[0135] Obtain the steering wheel angle and tie rod displacement loads. Substitute the yaw rate and center of mass sideslip angle calculated by the lateral dynamics model into the steering resistance torque calculation model, then obtain the steering resistance torque, and substitute it into the steering wheel angle and tie rod displacement calculation model to obtain this load.

[0136] Obtain the steering rod force load. Substitute the yaw rate and center of mass sideslip angle calculated by the lateral dynamics model into the steering resistance torque calculation model, then obtain the steering resistance torque, and substitute it into the steering rod force calculation model to obtain this load.

[0137] Obtain the steering gear rack force and steering motor torque loads. Substitute the steering rod force load calculated by the steering rod force calculation model into the steering gear rack force and steering motor torque calculation model to obtain this load.

[0138] Thus, the prediction of the load is completed. The method flow block diagram of this embodiment is as Figure 1 shown, and the steering system analysis diagram established by the constructed load calculation model is as Figure 8 shown.

[0139] Embodiment 2

[0140] For the effectiveness and accuracy of the present invention, this embodiment is specifically set as a comparative verification.

[0141] As Figure 2 shown, a sine cornering signal is constructed with the maximum turning curvature as the amplitude, and the uniform vehicle speed of 8 m / s is used as the input of the lateral dynamics model. Combining with the steering mechanics model, when the vehicle reaches the minimum turning radius of 3.473 m, the steering wheel angle is 522.5°, and the error compared with the steering wheel angle of 540° under the extreme condition is about 3.24%. As Figure 3 shown, 100 s of GPS and steering wheel angle data measured by urban users are intercepted, and the accuracy of the steering rod force is verified by reverse using the model input parameters, and the difference between the simulation and measured data is determined by comparing in the time domain.

[0142] As Figure 3 shown, 100 s of GPS and steering wheel angle data measured by urban users are intercepted, and the accuracy of the steering rod force is verified by reverse using the model input parameters, and the difference between the simulation and measured data is determined by comparing in the time domain; as Figure 4 shown, a circular runway with a diameter of 200 m is selected at a certain test site, and the actual GPS data is as shown; as Figure 5 shown, the curvature obtained by preprocessing the data of the 200 m diameter circular runway at the test site is compared with the actual turning curvature of 0.01.

[0143] As Figures 6 - 7 shown, further, in order to evaluate the uncertainty of curvature prediction, the relative error between the calculated curvature and the standard curvature after the vehicle enters the 200m circular runway is statistically analyzed. The relative error is determined to follow a normal distribution through the distribution goodness-of-fit test method. It can be seen that the relative error at the 95th percentile is within -6.29% to 11.77%, while the relative error at the 50th percentile is within 3%. It can be considered that the curvature prediction result is within the acceptable error range for engineering applications. This shows that the method for predicting curvature is feasible.

[0144] Figures 9 - 10 The predicted curvature data in the embodiment and the key loads of the steering system (steering wheel angle, tie rod displacement, steering rod force, steering gear rack force, and steering motor torque) predicted by the method proposed in the present invention.

[0145] Embodiment III

[0146] The present invention also provides a load prediction system for a vehicle steering system, including: a collection module, a calculation module, a first construction module, a second construction module, and a prediction module; the collection module is used to obtain the GPS trajectory data of the user's road vehicle based on a public data platform; the calculation module is used to calculate the turning curvature of the user's road vehicle based on the GPS trajectory data; the first construction module is used to construct a lateral dynamics model of the vehicle based on the turning curvature; the second construction module is used to construct a load calculation model of the vehicle steering system based on the lateral dynamics model; the prediction module is used to complete the prediction of the vehicle steering system load by using the load calculation model.

[0147] Next, in combination with this embodiment, it will be described in detail how the present invention solves technical problems in real life.

[0148] The collection module obtains the GPS trajectory data of the user's road vehicle based on a public data platform.

[0149] First of all, in this embodiment, it is necessary to construct a GPS trajectory database for the user's road vehicle. The GPS trajectory data of the user's road vehicle, including the time history data of longitude, latitude, and vehicle speed, can be extracted through a public vehicle big data platform. The actual GPS driving trajectory data of the user's road vehicle covers the GPS driving trajectory data of vehicles in different regions and models. For the abnormal data caused by poor GPS signals, long-term idling, and severe acceleration and deceleration during the data collection process, the user's GPS data is cleaned to generate a new GPS vehicle driving trajectory after excluding abnormal or invalid GPS data.

[0150] In this embodiment, the cut-off frequency of the low-pass filter is set to 0.5 Hz to filter the GPS trajectory data; the threshold judgment rule is used to judge the abnormality and invalidity of the user data, and the invalid or abnormal data in the GPS trajectory data is removed and then frame filling is performed to obtain an available vehicle GPS trajectory database.

[0151] The above-mentioned steps of frame filling include (for latitude and longitude and vehicle speed):

[0152] For the latitude lat, if |lat i -lat i-1 | > 0.1, then lat i = (lat i-1 + lat i+1 ) / 2; if |lat i -lat i-1 | ≤ 0.1, then lat i remains unchanged.

[0153] For the longitude lon, if |lon i -lon i-1 | > 0.1, then lon i = (lon i-1 + lon i+1 ) / 2; if |lon i -lon i-1 | ≤ 0.1, then lon i remains unchanged.

[0154] For the vehicle speed v, if |v i -v i-1 | > 10, then v i = (v i-1 + v i+1 ) / 2; if |v i -v i-1 | ≤ 10, then v i remains unchanged.

[0155] The calculation module calculates the turning curvature of the user's road vehicle based on the GPS trajectory data.

[0156] Based on the obtained available vehicle GPS trajectory database, calculate the turning curvature of the user's road vehicle. The calculation steps include: taking points of latitude and longitude and spline interpolation, judgment and processing of vehicle speed, calculation of curvature, and determination of the positive and negative of curvature, etc. The specific process includes:

[0157] Taking points of latitude and longitude and spline interpolation. The number of intervals for taking points of latitude, longitude and vehicle speed is i, so as to form the data of latitude, longitude and vehicle speed in the new time history.

[0158] Spline interpolation is performed on the data obtained by sampling the above longitude, latitude, and vehicle speed to restore it to the original time history data. The expression for the number of interpolation points is:

[0159] n i = i (33)

[0160] Then, spline interpolation is performed on it. Taking the vehicle speed as an example:

[0161] 1) Assume that the two known vehicle speed points are (t 1 , v 1 ) and (t 2 , v 2 );

[0162] 2) Construct the spline interpolation function S(t). On the interval [t 1 , t 2 :

[0163] S(t) = a + b(t - t 1 ) + c(t - t 1 ) 2 + d(t - t 1 ) 3 (34)

[0164] 3) According to the above spline interpolation conditions, the following system of equations can be listed:

[0165] a) S(t 1 ) = v 1 , that is, a = v 1 ;

[0166] b) S(t 2 ) = v 2 , substituting into the function gives v 1 + b(t 2 - t 1 ) + c(t 2 - t 1 ) 2 + d(t 2 - t 1 ) 3 = v 2 ;

[0167] 4) Use the second derivative at both endpoints to be 0 to determine the coefficients c and d.

[0168] 5) Substitute c = 0, d = 0 into v 1 + b(t 2 - t 1 ) + c(t 2 - t 1 ) 2 + d(t 2 - t 1) 3 = v 2 Find

[0169] 6) According to the set number of interpolation points n i = i, divide the interval |v 1 , v 2 | into i equal parts, and the interpolation points are respectively Then substitute the above interpolation points into S(t) = v 1 + b(t - t 1 ), and the inserted vehicle speed can be obtained.

[0170] Judgment and processing of vehicle speed. Find the index of the position where the vehicle speed is less than 0, and make the vehicle speed positive.

[0171] Calculation of curvature and determination of the positive and negative of curvature. Use the Haversine formula to calculate the turning curvature, and the steps include:

[0172]

[0173] ρ = 1 / (R e ·b) (37)

[0174] In the formula, a 1 represents an intermediate variable, calculating a part of the radian of two points; dlat represents the adjacent latitude difference; dlon represents the adjacent longitude difference; b 1 represents the radian between two points; R e represents the radius of the earth; ρ represents the turning curvature.

[0175] Use the scalar form of the cross product in two-dimensional space as the relative direction of two vectors as the basis for determining the positive and negative of curvature, and the steps include:

[0176]

[0177] In the formula, and represent two unit vectors, used to calculate the cross product; lon represents longitude; lat represents latitude.

[0178] If the cross product cross_product > 0, then is counterclockwise relative to , judge to the left, that is, the curvature is positive; if the cross product cross_product < 0, then is clockwise relative to , judge to the right, that is, the curvature is negative.

[0179] The above process finally obtains the turning curvature during vehicle driving.

[0180] The first construction module constructs a lateral dynamics model of the vehicle based on the turning curvature.

[0181] Based on the turning curvature obtained during vehicle driving, a lateral dynamics model of the vehicle is established to obtain the yaw rate and the sideslip angle of the center of mass. The constructed lateral dynamics model includes:

[0182]

[0183]

[0184] In the formula, v represents the actual vehicle speed at the center of mass; ρ represents the radius of curvature; β represents the sideslip angle of the center of mass; ω r represents the yaw rate of the vehicle; k 1 represents the cornering stiffness of the front wheels; k 2 represents the cornering stiffness of the rear wheels; I z represents the moment of inertia of the vehicle about the z-axis; represents the sideslip angle of the front wheels; represents the sideslip angle of the rear wheels.

[0185] The second construction module constructs a load calculation model of the vehicle steering system based on the lateral dynamics model.

[0186] The constructed load calculation model includes: calculation models of the steering resistance moment, the steering tie rod force, the steering gear rack force, etc., which are calculation models of the global loads of the steering system.

[0187] (1) Construction of the steering resistance moment calculation model:

[0188] Considering the dominant position of each self-aligning moment during steering, the main self-aligning moments are selected to establish a lateral force self-aligning moment calculation model and a gravity self-aligning moment calculation model. The calculation steps of the steering resistance moment are as follows:

[0189] M sw = M G + M y (43)

[0190] Among them, M G represents the gravity self-aligning moment; M y represents the lateral force self-aligning moment.

[0191] The calculation expression of M G is:

[0192] M G = G f · e z · sinθ · cosγ · sinδ (44)

[0193] e z = (d 偏+r w tanθ)cosθ (45)

[0194] In the formula, G f represents the front axle load; e z represents the moment arm of the gravity self-aligning torque; θ represents the kingpin inclination angle; γ represents the caster angle; r w represents the dynamic rolling radius of the tire; d 偏 represents the kingpin offset.

[0195] The calculation expression of M y is as follows:

[0196] M y = k·α 1 ·(n R + n k ) (46)

[0197] Among them, α 1 represents the front wheel side slip angle; k represents the cornering stiffness; the shift distance n k formed by the kingpin ground contact point and the center of the tire footprint, the offset n R and the tire ground contact footprint length l 轮 The calculation expressions are as follows:

[0198]

[0199] In the formula, r represents the wheel radius; D represents the tire outer diameter; Δ represents the radial deformation of the steering tire under the action of the front axle vertical load, and is determined by the following formula through unit conversion:

[0200]

[0201] K Δ = 0.0015B + 0.42 (50)

[0202] In the formula, C represents a coefficient, taking 1.5 (radial tire); B represents the tire section width; p 压 represents the tire air pressure; G 1 represents the single tire rated load.

[0203] Establish a calculation model for the steering wheel angle and the tie rod displacement. The calculation formula is as follows:

[0204] l d = l sw ·(ψ s + ψ s ′) / i s (51)

[0205] ψ s ′ = l sw ·M sw / (l s ·K sd ·i s ) (52)

[0206] δ = M sw / K sl +l d / l s = arctan(ρL) (53)

[0207] In the formula, l d represents the displacement of the tie rod; l s represents the length of the steering linkage; l sw represents the length of the steering rocker arm; i s represents the steering transmission ratio; δ represents the front wheel steering angle; ψ s represents the steering wheel steering angle; ψ s ' represents the total moment feedback steering angle of the tire around the kingpin; M sw represents the total moment of the front wheel around the kingpin; K sd represents the torsional stiffness of the steering column; K sl represents the equivalent stiffness of the steering connection mechanism.

[0208] (2) Construction of the tie rod force calculation model:

[0209]

[0210] M H = M sw (55)

[0211] Among them, F 拉 represents the tie rod force.

[0212] The calculation formula for the included angle η between the steering knuckle arm and the tie rod is expressed as:

[0213]

[0214] In the formula, φ represents the included angle between the steering knuckle arm and the tie rod; p represents the center distance between the two end ball joints; s represents the rack stroke; h represents the installation distance from the rack axis to the bottom edge of the trapezoid; l represents the length of the steering knuckle arm; m represents the module of the steering gear; μ represents the helix angle μ = 14°; represents the length of the other side of the triangle formed by the steering knuckle arm and the tie rod at the steering position.

[0215] Establish a calculation model for the output torque of the steering motor (since the motion state of the rack has little influence on the output torque of the steering motor, the motion state during rack movement is not considered), and the calculation formula is as follows:

[0216]

[0217] Where, T a represents the output torque of the steering motor; r p represents the pitch circle radius of the pinion; N represents the reduction ratio of the steering motor.

[0218] (3) Construction of the steering rack force calculation model:

[0219]

[0220] Where, F N represents the steering rack force of the steering gear.

[0221] The calculation formula for the included angle ε between the tie rod and the rack shaft is expressed as:

[0222]

[0223] γ - α = φ + χ (63)

[0224]

[0225] Where, F N represents the rack force of the rack and pinion steering gear; p represents the center distance between the two ball joints; d represents the distance between the kingpins; l represents the length of the steering knuckle arm; λ represents the pressure angle, λ = 20°; α represents the angle rotated by the steering knuckle arm during steering; χ represents the included angle between the other side of the triangle formed by the steering knuckle arm and the tie rod and the connection line between the left and right kingpins.

[0226] Finally, the prediction module uses the load calculation model to complete the prediction of the vehicle steering system load.

[0227] Use the constructed load calculation model to calculate the key loads of the steering system, including the steering wheel angle, tie rod displacement, tie rod force, steering rack force and steering motor torque.

[0228] Obtain the steering wheel angle and tie rod displacement loads. Combine the yaw rate and center of mass sideslip angle calculated by the lateral dynamics model and substitute them into the steering resistance torque calculation model, then obtain the steering resistance torque, and then substitute it into the steering wheel angle and tie rod displacement calculation model to obtain this load.

[0229] Obtain the tie rod force load. Combine the yaw rate and center of mass sideslip angle calculated by the lateral dynamics model and substitute them into the steering resistance torque calculation model, then obtain the steering resistance torque, and then substitute it into the tie rod force calculation model to obtain this load.

[0230] Obtain the steering rack force and steering motor torque loads. Combine the tie rod force load calculated by the tie rod force calculation model and substitute it into the steering rack force and steering motor torque calculation model to obtain this load.

[0231] Thus, the prediction of the load is completed. The flowchart of the method in this embodiment is as Figure 1 shown. The analysis diagram of the steering system established by the constructed load calculation model is as Figure 8 shown.

[0232] The embodiments described above are only descriptions of the preferred embodiments of the present invention, and do not limit the scope of the present invention. Without departing from the design spirit of the present invention, various deformations and improvements made by those of ordinary skill in the art to the technical solutions of the present invention shall fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for predicting the load of a vehicle steering system, characterized in that the steps include: Based on the public data platform, obtain the GPS trajectory data of the user's road vehicles; Calculating the turning curvature of the user's road vehicle based on the GPS trajectory data; Based on the turning curvature, constructing a lateral dynamics model of the vehicle; Based on the lateral dynamics model, a load calculation model of the vehicle steering system is constructed; The load calculation model is used to predict the load on the vehicle steering system.

2. The load prediction method for a vehicle steering system according to claim 1, characterized in that: The method for obtaining the GPS trajectory data includes: extracting the GPS trajectory data of the user's road vehicle through a public vehicle big data platform, including the time history data of the longitude and latitude and the vehicle speed; after collection, preprocessing the GPS trajectory data, the steps include: setting the low-pass filter cutoff frequency to 0.5 Hz, filtering the GPS trajectory data; using a threshold judgment rule to make abnormal and invalid judgments on the user data, eliminating the invalid or abnormal data in the GPS trajectory data and then filling the frame to obtain the valid vehicle GPS trajectory data.

3. The load prediction method for a vehicle steering system according to claim 2, characterized in that: Methods for interpolating frames for latitude, longitude and vehicle speed include: For latitude lat, if |lat i -lat i-1 |>0.1, then lat i =(lat i-1 +lat i+1 ) / 2; if |lat i -lat i-1 |≤0.1, then lat i remain unchanged; For longitude lon, if |lon i -lon i-1 |>0.1, then lon i =(lon i-1 +lon i+1 ) / 2; if |lon i -lon i-1 |≤0.1, then lon i remain unchanged; For vehicle speed v, if |v i -v i-1 |>10, then v i =(v i-1 +v i+1 ) / 2; if |v i -v i-1 |≤10, then v i Remain unchanged.

4. The load prediction method for a vehicle steering system according to claim 1, characterized in that: The method for calculating the curvature of the turning includes: taking points of longitude and latitude and spline interpolation, judging the vehicle speed, calculating the curvature, and judging whether the curvature is positive or negative; The calculation of curvature and the method of determining whether the curvature is positive or negative include: The Haversine formula is used to calculate the curvature of the turn. The steps include: ρ=1 / (R e ·b) In the formula, a1 represents the intermediate variable, which calculates part of the arc between two points; dlat represents the difference between adjacent latitudes; dlon represents the difference between adjacent longitudes; b1 represents the arc between two points; R e represents the radius of the earth; ρ represents the curvature of rotation; The scalar form of the cross product in two-dimensional space is used as the basis for judging the positive and negative curvature based on the relative directions of the two vectors. The steps include: In the formula, and Represents two unit vectors, used to calculate the cross product; lon represents longitude; lat represents latitude; If cross_product>0, then Relative to is counterclockwise, it is judged to be left, that is, the curvature is positive; if the cross product cross_product < 0, then Relative to It is clockwise, and is judged to be to the right, that is, the curvature is negative.

5. The load prediction method for a vehicle steering system according to claim 1, characterized in that: The constructed lateral dynamics model includes: Where v is the actual vehicle speed at the center of mass; ρ is the radius of curvature; β is the sideslip angle at the center of mass; ω is r represents the vehicle yaw rate; k1 represents the front wheel cornering stiffness; k2 represents the rear wheel cornering stiffness; I z represents the moment of inertia of the car around the z-axis; Indicates the front wheel slip angle; Indicates the rear wheel slip angle.

6. The load prediction method for a vehicle steering system according to claim 1, characterized in that: The constructed load calculation model includes: a steering resistance torque calculation model, a steering rod force calculation model and a steering gear rack force calculation model.

7. A load prediction system for a vehicle steering system, the system being used to implement the method according to any one of claims 1 to 6, characterized in that: include: A collection module, a calculation module, a first construction module, a second construction module and a prediction module; The acquisition module is used to obtain GPS track data of the user's road vehicle based on a public data platform; The calculation module is used to calculate the turning curvature of the user's road vehicle based on the GPS trajectory data; The first building module is used to build a lateral dynamics model of the vehicle based on the turning curvature; The second construction module is used to construct a load calculation model of the vehicle steering system based on the lateral dynamics model; The prediction module is used to complete the prediction of the vehicle steering system load using the load calculation model.

8. The load prediction system for a vehicle steering system according to claim 7, characterized in that: The workflow of the acquisition module includes: extracting the GPS trajectory data of the user's road vehicles through the public vehicle big data platform, including the time history data of the longitude and latitude and the speed; after the acquisition, preprocessing the GPS trajectory data, the steps include: setting the low-pass filter cutoff frequency to 0.5Hz, filtering the GPS trajectory data; using the threshold judgment rule to judge the user data as abnormal and invalid, eliminating the invalid or abnormal data in the GPS trajectory data and then filling the frame to obtain the valid vehicle GPS trajectory data.

Images