A sensor system for measuring three-dimensional forces and moments on a wheel during travel

Abstract

The present application belongs to the technical field of vehicle dynamic testing, and discloses a sensor system for measuring three-dimensional force and torque suffered by a wheel during driving. The system comprises a six-component force sensing acquisition module, a wheel posture deformation calculation module, a vehicle-mounted electromagnetic characteristic identification module, an embedded comprehensive correction operation module and a data output and communication module. The wheel posture deformation calculation module directly calculates real-time dynamic roll angle parameters and elastomer dynamic deformation variable parameters from digital signals output by the six-component force sensing acquisition module, and the embedded comprehensive correction operation module is based on the above parameters and dynamic electromagnetic interference equivalent parameters. The present application solves the problem of insufficient measurement accuracy caused by posture, deformation and electromagnetic interference coupling during dynamic driving of the wheel through a single sensor cooperating with an embedded progressive correction architecture.

Description

Technical Field

[0001] This invention relates to the field of vehicle dynamic testing technology, and specifically to a sensor system for measuring the three-dimensional forces and torques experienced by a wheel during driving. Background Technology

[0002] The three-dimensional forces and torques experienced by wheels during actual road travel are core parameters for vehicle dynamics analysis, chassis control, and safety assessment. Widely used six-component force sensors can directly sense the forces and torques borne by the wheels. However, under dynamic driving conditions, the sensor output signals are affected by the overlapping influence of various time-varying physical disturbances. Wheel roll attitude causes a shift in the sensor's force reference plane, introducing nonlinear measurement errors related to the roll angle; road undulations cause non-uniform dynamic deformation of the wheel's elastic body, and its stress distribution is closely related to the wheel's real-time attitude, with both contributing to coupling errors; changes in vehicle speed cause fluctuations in the intensity of electromagnetic interference in the onboard electrical environment, which intrudes into the sensor's analog acquisition circuit, contaminating the original mechanical signal. Existing measurement systems typically employ fixed-coefficient linear correction or single filtering algorithms, failing to construct an error compensation model that progressively correlates attitude shift, deformation coupling, and electromagnetic interference. This results in inherent limitations in the measurement accuracy of the three-dimensional forces and torques experienced by the wheels under varying driving conditions, making it difficult to meet the urgent need for high-fidelity testing of vehicle dynamic performance. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of existing technologies and to propose a sensor system for measuring the three-dimensional forces and torques experienced by a wheel during driving, comprising:

[0004] The six-component force sensing and acquisition module is used to acquire the original three-dimensional force signal and the original three-dimensional torque signal borne by the wheel in real time, and then convert the original three-dimensional force signal and the original three-dimensional torque signal into digital signals for output.

[0005] The wheel attitude deformation calculation module is electrically connected to the six-component force sensing and acquisition module. It is used to receive the digital signal and directly calculate the real-time dynamic roll angle parameter and the elastic body dynamic deformation parameter of the wheel from the strain distribution characteristics contained in the digital signal, and output the real-time dynamic roll angle parameter and the elastic body dynamic deformation parameter.

[0006] An embedded integrated correction and calculation module, electrically connected to the six-component force sensing and acquisition module and the wheel attitude deformation calculation module, is used to receive the digital signal, the real-time dynamic roll angle parameter, and the elastic body dynamic deformation parameter. Based on the real-time dynamic roll angle parameter, it calculates the attitude offset mechanical correction; based on the real-time dynamic roll angle parameter and the elastic body dynamic deformation parameter, it calculates the deformation coupling mechanical correction; and based on the vehicle-mounted electromagnetic interference parameter, it calculates the electromagnetic interference mechanical correction. By fusing the attitude offset mechanical correction, the deformation coupling mechanical correction, and the electromagnetic interference mechanical correction, it compensates the digital signal to generate corrected three-dimensional force data and corrected three-dimensional torque data.

[0007] The data output and communication module is electrically connected to the embedded integrated correction calculation module and is used to receive and output the corrected three-dimensional force data and the corrected three-dimensional torque data.

[0008] Preferably, it also includes an on-board electromagnetic feature recognition module;

[0009] The vehicle-mounted electromagnetic feature recognition module is electrically connected to the embedded integrated correction and calculation module. It is used to synchronously collect the real-time driving speed signal of the vehicle and the electromagnetic fluctuation signal of the vehicle electrical circuit. Based on the real-time driving speed signal of the vehicle and the electromagnetic fluctuation signal, it calculates the equivalent parameters of dynamic electromagnetic interference and outputs the equivalent parameters of dynamic electromagnetic interference as the vehicle-mounted electromagnetic interference parameters to the embedded integrated correction and calculation module.

[0010] More preferably, the embedded integrated correction calculation module includes an attitude correction calculation unit, a deformation coupling correction calculation unit, an electromagnetic interference correction calculation unit, and an integrated compensation unit;

[0011] The attitude correction calculation unit is electrically connected to the wheel attitude deformation calculation module and is used to call the preset attitude offset mechanical correction calculation formula. The attitude offset mechanical correction calculation formula uses the real-time dynamic roll angle parameter and the original three-dimensional force signal as variables. The attitude offset mechanical correction calculation formula includes nonlinear attitude disturbance component and linear lever arm offset component.

[0012] The deformation coupling correction calculation unit is electrically connected to the wheel attitude deformation calculation module and is used to call a preset attitude coupling deformation mechanical correction calculation formula. The attitude coupling deformation mechanical correction calculation formula uses the real-time dynamic roll angle parameter, the elastic body dynamic deformation parameter, the original three-dimensional force signal and the preset sensor elastic body reference length as variables. The attitude coupling deformation mechanical correction calculation formula includes the first-order coupling component of attitude and deformation and the higher-order nonlinear error component of deformation.

[0013] The electromagnetic interference correction calculation unit is used to call a preset electromagnetic interference mechanical correction calculation formula, which uses the vehicle electromagnetic interference parameters and the preset sensor equivalent internal resistance as variables.

[0014] The integrated compensation unit is electrically connected to the six-component force sensing and acquisition module, the attitude correction calculation unit, the deformation coupling correction calculation unit, and the electromagnetic interference correction calculation unit, respectively. It is used to perform subtraction calculations on the original three-dimensional force signal and the original three-dimensional torque signal with the attitude offset mechanical correction amount, the deformation coupling mechanical correction amount, and the electromagnetic interference mechanical correction amount, respectively, to generate the corrected three-dimensional force data and the corrected three-dimensional torque data.

[0015] More preferably, the nonlinear attitude disturbance component included in the attitude offset mechanical correction formula is characterized by the sine square function of the real-time dynamic tilt angle parameter, and the linear lever arm offset component is characterized by the product of the real-time dynamic tilt angle parameter and the original three-dimensional force signal, and the ratio of an attitude equivalent lever arm to the reference length of the sensor elastomer.

[0016] More preferably, the first-order coupling component of attitude and deformation included in the attitude coupling deformation mechanics correction formula is characterized by the product of the cosine function of the dynamic deformation parameter of the elastic body and the real-time dynamic roll angle parameter, and the higher-order nonlinear error component of deformation is characterized by the product of the square of the dynamic deformation parameter of the elastic body and the original three-dimensional force signal divided by the reference length of the sensor elastic body.

[0017] More preferably, the electromagnetic interference mechanical correction calculation formula is characterized by multiplying the equivalent voltage amplitude in the vehicle electromagnetic interference parameters, the preset electromagnetic signal force conversion coefficient, and the vehicle's real-time driving speed signal by the equivalent internal resistance of the sensor.

[0018] More preferably, the wheel attitude deformation calculation module includes a strain feature extraction unit and an attitude deformation mapping unit;

[0019] The strain feature extraction unit is used to extract the asymmetric strain component and alternating strain amplitude in the digital signal;

[0020] The attitude deformation mapping unit is electrically connected to the strain feature extraction unit and is used to determine the real-time dynamic tilt angle parameter according to the preset first mapping relationship and the distribution direction and magnitude of the asymmetric strain component, and to determine the elastic body dynamic deformation parameter according to the preset second mapping relationship and the fluctuation characteristics of the alternating strain amplitude.

[0021] More preferably, the vehicle-mounted electromagnetic feature recognition module includes a vehicle speed signal interface, an electromagnetic acquisition front-end, and an interference parameter solver;

[0022] The vehicle speed signal interface is used to access and acquire the real-time driving speed signal of the vehicle;

[0023] The electromagnetic acquisition front end is used to acquire the electromagnetic fluctuation signal of the vehicle electrical circuit;

[0024] The interference parameter solver is electrically connected to the vehicle speed signal interface and the electromagnetic acquisition front end, respectively. It is used to perform real-time amplitude-frequency characteristic analysis of the electromagnetic wave signal in relation to the real-time vehicle speed signal, and extract the electromagnetic interference characteristic amplitude related to the dynamic change of vehicle speed as the equivalent parameter of the dynamic electromagnetic interference.

[0025] More preferably, the six-component force sensing acquisition module is the only hardware sensing module in the sensor system used to acquire the mechanical signals of the wheel.

[0026] Further preferably, it also includes an in-vehicle embedded processor, which integrates the wheel attitude deformation calculation module and the embedded comprehensive correction calculation module.

[0027] The vehicle-mounted embedded processor establishes electrical connections with the six-component force sensing acquisition module and the data output and communication module through its internal high-speed bus interface.

[0028] The technical effects include:

[0029] The core innovation of the sensor system provided by this invention lies in the fact that the wheel attitude deformation calculation module directly calculates the real-time dynamic roll angle parameters and elastic body dynamic deformation parameters from the raw digital signals of the six-component force sensor. Using these as input, the embedded integrated correction calculation module progressively performs attitude offset mechanical correction, attitude and deformation coupling mechanical correction, and electromagnetic interference mechanical correction, ultimately fusing the three types of corrections to compensate for the original measurement data. This solution incorporates three physical interference factors—wheel roll attitude, elastic body deformation, and vehicle-mounted electromagnetic interference—into a unified progressive correction architecture, breaking through the limitations of existing technologies that only perform single-dimensional or fixed-coefficient corrections, and solving the problem of decreased measurement accuracy caused by multi-dimensional physical interference coupling under dynamic driving conditions. Attached Figure Description

[0030] Figure 1 A schematic diagram of the onboard hardware deployment architecture for a vehicle three-dimensional force and torque sensing system;

[0031] Figure 2 This is a schematic diagram of the hardware hierarchy of the six-component force sensing and acquisition module. Detailed Implementation

[0032] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0033] Please see Figures 1-2 This invention relates to a sensor system for measuring the three-dimensional forces and torques experienced by a wheel during driving. Deployed on a vehicle, the system performs online measurements of the dynamic loads on the wheels during vehicle operation and outputs error-compensated force and torque data for use in vehicle dynamics analysis, chassis electronic control, or durability assessment. The sensor system comprises a six-component force sensing and acquisition module, a wheel attitude deformation calculation module, an onboard electromagnetic feature recognition module, an embedded integrated correction and calculation module, and a data output and communication module. These modules work collaboratively through clearly defined electrical connections.

[0034] The six-component force sensing and acquisition module is the hardware module in the system that directly bears the wheel load and generates mechanical signals. This module is rigidly connected to the wheel hub via bolts, flanges, or adapters. When the wheel is in motion, longitudinal, lateral, and vertical forces from the road surface, as well as moments about the three coordinate axes, act simultaneously on the elastic body structure inside the module. The elastic body is made of alloy steel or titanium alloy and has multiple strain-sensitive areas machined on it. Each sensitive area has a metal foil strain gauge or semiconductor strain gauge attached, and these strain gauges form multiple independent measuring bridges in the form of a Wheatstone bridge. Different bridges correspond to different force or moment components, and the specific distribution of the bridges is arranged according to the decoupling design principle, so that each bridge has a high output for the target component while the output for other components is close to zero. When the wheel bears a load, the elastic body undergoes elastic deformation, and each bridge outputs an analog voltage signal proportional to the strain. The amplitude of these analog voltage signals is typically in the microvolt to millivolt range. The six-component force sensing and acquisition module internally includes a signal conditioning circuit, which consists of an instrumentation amplifier, an anti-aliasing low-pass filter, and an analog-to-digital converter connected in sequence. The instrumentation amplifier amplifies the weak bridge differential signal, with its gain set by a precision resistor, while simultaneously suppressing common-mode interference. The cutoff frequency of the anti-aliasing low-pass filter is determined based on the wheel's highest rotational frequency and the required measurement bandwidth, for example, set between 500Hz and 2000Hz, to attenuate aliasing noise and high-frequency components of mechanical resonance. The analog-to-digital converter converts the analog signal into a multi-bit digital signal at a sampling rate of no less than 1kHz, with a quantization bit depth of 16 or 24 bits, thereby outputting the original three-dimensional force signal and the original three-dimensional torque signal in digital form. This digital signal is simultaneously transmitted to the wheel attitude deformation calculation module and the embedded integrated correction calculation module via a serial peripheral interface or a parallel data bus. It should be noted that the six-component force sensing acquisition module is the only hardware sensing module in the sensor system used to acquire wheel mechanical signals. The system does not contain independent inclinometers, accelerometers, displacement sensors, or other auxiliary mechanical sensing devices; all wheel attitude and elastic body deformation information is calculated from the output signal of this module.

[0035] The wheel attitude deformation calculation module is electrically connected to the six-component force sensing and acquisition module. It receives the digital signals and calculates the real-time dynamic roll angle parameters and elastic body dynamic deformation parameters of the wheel from the strain distribution characteristics contained in the digital signals. This module internally includes a strain feature extraction unit and an attitude deformation mapping unit. The strain feature extraction unit reads multi-channel strain digital signals in real time, with each channel corresponding to the strain value at a specific measuring point within the six-component force sensor. These measuring points are symmetrically arranged circumferentially and axially along the elastic body, forming a spatial strain vector. The strain feature extraction unit decomposes the spatial strain vector into symmetrical and asymmetrical strain components using a symmetrical component decomposition algorithm. The symmetrical strain components reflect the symmetrical load borne by the wheel, while the asymmetrical strain components are mainly related to the bending effect caused by wheel roll. Specifically, when the wheel generates a roll angle, the tensile and compressive states experienced by the measuring points near and far from the vehicle body are different, exhibiting an asymmetrical distribution pattern. The amplitude of the asymmetrical strain component has a monotonic mapping relationship with the magnitude of the roll angle, and its sign indicates the roll direction. The strain feature extraction unit further filters the strain signal of a specific channel using a bandpass filter. The center frequency of this bandpass filter is locked to the wheel rotation frequency or an integer multiple thereof, extracting the alternating strain amplitude. The alternating strain amplitude originates from the periodic impact of road surface unevenness on the wheel. This impact causes the elastic body to undergo dynamic bending deformation synchronized with the wheel rotation. The magnitude of the alternating strain amplitude directly reflects the degree of dynamic deformation of the elastic body at that moment. The attitude deformation mapping unit internally stores a first mapping relationship and a second mapping relationship established through pre-calibration. The first mapping relationship is a multidimensional lookup table or a family of regression functions, whose inputs are the direction and amplitude parameters of the asymmetric strain components, and whose output is the real-time dynamic roll angle parameter. The directional parameter is represented by the azimuth angle or sign of the asymmetric strain component in the elastic body cross-sectional coordinate system, and the amplitude parameter is the root mean square value or peak value of the asymmetric strain component. The calibration process involves mounting the wheel on a tumbler-adjustable test bench, adjusting the tumble angle in 0.5° increments from the negative tumble limit to the positive tumble limit, and recording the asymmetric strain data at each angle. A first mapping relationship is formed through interpolation or curve fitting. The second mapping relationship maps the fluctuation characteristics of the alternating strain amplitude to the dynamic deformation parameters of the elastic body. The fluctuation characteristics can be represented by the peak-to-peak value, root mean square value, or envelope average value of the alternating strain amplitude over several wheel rotation cycles. The unit of the dynamic deformation parameters of the elastic body is length. During calibration, a contact displacement gauge or laser displacement sensor is used to directly measure the deformation of the elastic body under stress, and the alternating strain amplitude is recorded simultaneously to establish a mapping model from the strain dimension to the length dimension. This model can be a linear proportional relationship or a quadratic polynomial relationship. The wheel attitude deformation calculation module outputs real-time dynamic roll angle parameters and elastic body dynamic deformation parameters, and simultaneously transmits these two parameters to the embedded integrated correction calculation module.

[0036] The vehicle-mounted electromagnetic feature recognition module is electrically connected to the embedded integrated correction and calculation module. It is used to synchronously acquire real-time vehicle speed signals and electromagnetic fluctuation signals from the vehicle's electrical circuits, and calculates equivalent parameters of dynamic electromagnetic interference based on these two signals. This module includes a vehicle speed signal interface, an electromagnetic acquisition front-end, and an interference parameter solver. The vehicle speed signal interface connects to the vehicle's CAN bus or wheel speed sensor signal lines, parsing timestamped vehicle speed data frames from CAN messages, or periodically measuring and converting the square wave signals from the wheel speed sensors using a pulse counter, outputting a real-time vehicle speed signal in meters per second. The electromagnetic acquisition front-end includes at least one inductive probe, either a near-field magnetic probe or a capacitive coupling clip, which holds or is close to the six-component force sensor signal harness, power harness, or cables near high-power electrical equipment to pick up electromagnetic fluctuation signals generated by devices such as motor controllers, DC-DC converters, and ignition systems. The electromagnetic acquisition front-end includes a preamplifier and an anti-aliasing filter. The preamplifier amplifies the weak induced voltage signal to a suitable amplitude, and the cutoff frequency of the anti-aliasing filter is set according to the electromagnetic interference frequency band to be analyzed, for example, from 100kHz to 10MHz. The interference parameter solver receives the amplified and filtered electromagnetic wave signal and the vehicle's real-time driving speed signal, and analyzes the electromagnetic wave signal in the time-frequency domain. The interference parameter solver uses short-time Fourier transform or continuous wavelet transform to convert the electromagnetic wave signal into a time spectrum, and identifies characteristic frequency components that are correlated with the vehicle's real-time driving speed signal in the time spectrum. Specifically, changes in vehicle speed cause changes in the drive motor speed, inverter switching frequency, and their harmonic frequencies. These frequency components exhibit a trajectory that changes synchronously with the vehicle speed in the time spectrum. The interference parameter solver calculates the power spectral density or envelope amplitude of the characteristic frequency component in a specific frequency band, and extracts the amplitude of this component as the equivalent voltage amplitude in the dynamic electromagnetic interference equivalent parameters. The equivalent voltage amplitude has voltage dimensions and represents the magnitude of the noise voltage injected in the equivalent series by electromagnetic interference in the sensor signal circuit. This parameter is strongly correlated with vehicle speed and changes dynamically as the vehicle speed increases or decreases. It is transmitted to the embedded integrated correction calculation module in real time.

[0037] The embedded integrated correction calculation module is electrically connected to the six-component force sensing and acquisition module, the wheel attitude deformation calculation module, and the vehicle-mounted electromagnetic feature recognition module, and is the core module for performing error compensation calculations in the system. This module consists of an attitude correction calculation unit, a deformation coupling correction calculation unit, an electromagnetic interference correction calculation unit, and an integrated compensation unit. The attitude correction calculation unit receives real-time dynamic roll angle parameters from the wheel attitude deformation calculation module and raw three-dimensional force signals from the six-component force sensing and acquisition module, and calls the internally stored attitude offset mechanical correction calculation formula to calculate the attitude offset mechanical correction amount. The deformation coupling correction calculation unit receives real-time dynamic roll angle parameters, elastic body dynamic deformation parameters, and raw three-dimensional force signals, and reads the sensor elastic body reference length parameters pre-stored in the register, and calls the attitude coupling deformation mechanical correction calculation formula to calculate the deformation coupling mechanical correction amount. The electromagnetic interference correction calculation unit receives the equivalent voltage amplitude and real-time vehicle speed signal from the dynamic electromagnetic interference equivalent parameters, as well as the pre-stored sensor equivalent internal resistance parameters and electromagnetic signal force conversion coefficient parameters, and calls the electromagnetic interference mechanical correction calculation formula to calculate the electromagnetic interference mechanical correction amount. The integrated compensation unit contains an arithmetic logic path consisting of adders and subtractors. It subtracts the attitude offset mechanical correction, deformation coupling mechanical correction, and electromagnetic interference mechanical correction in parallel from the original three-dimensional force signal and the original three-dimensional torque signal, respectively, to generate corrected three-dimensional force data and corrected three-dimensional torque data. The corrected data is then sent to the data output and communication module through the module output port.

[0038] The theoretical basis, physical meaning, and parameter determination method of the attitude deviation mechanical correction calculation formula based on the attitude correction calculation unit will now be explained. This calculation formula is expressed as follows:

[0039]

[0040] In the formula The calculated attitude deviation mechanical correction has the following dimensions: That is, the dimension of force. This is the raw three-dimensional force signal, provided by the six-component force sensing acquisition module. It represents the force measurement value directly output by the sensor without any correction, and its dimensions are also [dimensions missing]. . The attitude influence coefficient is dimensionless. This coefficient comprehensively reflects the overall influence of various factors such as the sensor's elastomer structure, strain gauge arrangement, and material properties on the tilt interference sensitivity. The value of is determined through a static roll calibration experiment. During the experiment, a wheel simulation device is mounted on a calibration bench capable of precisely adjusting the roll angle. A constant, known standard force is applied under multiple roll angle conditions, and the raw force signals output by the six-component force sensors are recorded. The force measurement error at each roll angle is calculated, and the error data is fitted with the roll angle and a nonlinear model to minimize the sum of squared residuals to determine . The value. The real-time dynamic roll angle parameters for the wheels are provided by the wheel attitude deformation calculation module, and the unit is radians. Since radians are a derived unit of length-to-length ratio, this is important to note. It is considered dimensionless in dimensional analysis. As the attitude equivalent force arm, the dimension is Its physical meaning is the offset distance of the equivalent point of force transmission from the tire contact surface to the sensor elastomer relative to the sensor reference force surface when the wheel has a unit lateral tilt angle. The calculation can be performed by establishing a geometric relationship model between the tire contact imprint and the geometric center of the sensor elastic body, or by extracting the displacement trajectory of the contact force application point under different roll angles through finite element simulation. The reference length of the sensor elastomer, with dimensions of This parameter is a characteristic dimension of the elastic body under stress and deformation, such as the effective length of the strain measurement beam on the elastic body, the thickness of the thinnest section of the elastic body, or the distance from the center of the strain gauge to the fixed end of the elastic body. The specific value is determined by the sensor design drawings and stored in the module's non-volatile memory.

[0041] The formula consists of two terms to be added together. The first term... The physical mechanism of the nonlinear attitude disturbance component lies in the fact that wheel roll transforms the stress state of the sensor's elastic body from a uniform stress distribution under a single load to a composite stress distribution with an additional bending moment. The local strain sensed by the strain gauges on the elastic body no longer maintains its original linear relationship with the applied force, but instead exhibits a nonlinear deviation related to the roll angle. This nonlinear deviation changes slowly when the roll angle is small, but increases rapidly after the roll angle increases to a certain extent, showing an accelerated growth or saturation trend. (Sine square function) exist When the value is small, the value is small and the change is gradual. Approaching a right angle, the coefficient closes to 1, which can fit the physical law that the transmission efficiency of an elastic body decreases as the tilt angle increases. The amplitude of this nonlinear term is scaled proportionally to accommodate differences in sensor structures. From a dimensional perspective, Dimensions are , Dimensionless Dimensionless, the dimension remains the same after multiplying the three. This meets the dimensional requirements of the force correction quantity. (Second item) Characterizing the linear force arm offset component. When the wheel tilts, the line of action of the tire contact force shifts relative to the geometric center of the sensor elastomer. This shift can be approximated as the equivalent force arm. with roll angle The product of the two forces, the additional torque generated by the offset, is partially perceived by the sensor as a spurious force component. This spurious force component is different from the original force. It is directly proportional to the ratio of the offset to the reference length. Reference length The introduction of this term serves to make it dimensionless, allowing it to be directly added to the first term numerically. In dimensional analysis... Dimensionless Dimensions are , Dimensions are , Dimensions are ,ratio Dimensionless, the overall dimension of this term is Attitude deviation mechanical correction amount By unifying the errors from both nonlinear interference and linear lever arm offset into a single calculation formula, and subtracting them from the original force signal during comprehensive compensation, the measurement deviation caused by the individual effect of wheel roll can be eliminated.

[0042] Next, the attitude coupling deformation mechanics correction calculation formula based on the deformation coupling correction calculation unit will be explained in detail. This calculation formula is expressed as follows:

[0043]

[0044] In the formula This is a deformation coupling mechanical correction quantity, with dimensions of . The comprehensive coefficient for deformation influence has the following dimensions: This coefficient maps the dynamic deformation parameters of the elastic body to force errors, and its physical essence is the manifestation of the stiffness characteristics of the elastic body in the error propagation path. The determination method is as follows: Under zero roll angle, radial or lateral forced deformation of different amplitudes is applied to the wheel using a vibrator or hydraulic actuator. Simultaneously, a high-precision displacement sensor records the deformation, a standard force sensor records the actual input force, and a six-component force sensor outputs the original force signal. The rate of change of force error relative to deformation is calculated, and linear regression is performed to obtain the result. . The dynamic deformation parameter of the elastic body has dimensions of Provided by the wheel attitude deformation calculation module, it represents the maximum deformation displacement of the key stress-bearing parts of the elastic body under road surface excitation. The roll angle parameter is a real-time dynamic parameter and is dimensionless. This is the original three-dimensional force signal. This is the reference length of the sensor's elastomer.

[0045] The first term of the calculation formula This describes the first-order coupling effect between attitude and deformation. Road surface undulations cause dynamic deformation of the wheel's elastic body. This deformation leads to a readjustment of the stress distribution within the elastic body, with increased strain in some areas and decreased strain in others. This stress disturbance is detected by strain gauges on the sensor and manifests as spurious force changes. The direction of the deformation's influence on the stress distribution is closely related to the wheel's real-time roll attitude: when the roll angle is zero, the direction of the stress change caused by deformation is perpendicular to the main sensing surface of the sensor, maximizing the interference with the strain gauge output; as the roll angle increases, the spatial attitude of the elastic body relative to the load direction changes, the angle between the deformation direction and the sensor's sensing surface increases, and the projection of the deformation stress onto the sensing axis decreases according to a cosine law. Therefore, a first-order coupling effect is introduced... This term is used to quantify the posture constraint effect. Dimensions With deformation Dimensions Multiply to get This ensures that the term has the dimension of force. The second term... This describes the higher-order nonlinear error components of an elastic body under large deformation. When the deformation of the elastic body exceeds its linear elastic range or the geometric nonlinear effect cannot be ignored, the relationship between deformation and force error is no longer a simple proportional relationship, but a quadratic or higher-order dependence appears. This quadratic term describes the growth trend of the error contribution in the form of the square of the deformation, multiplied by the original force signal. To reflect the amplification effect of the load level on nonlinear errors, divide by the reference length. To achieve dimensional normalization. (Molecules) The dimensions are , divided by The dimensions are then restored to This meets the requirement of dimensional consistency for the correction amount. The sum of the first and second terms is the total mechanical correction under the coupling effect of attitude and deformation. This correction removes the measurement error caused by the combined effects of wheel roll posture and elastic body deformation from the original force signal.

[0046] The electromagnetic interference (EMI) correction calculation formula based on the EMI correction calculation unit is explained in detail below. This formula is expressed as follows:

[0047]

[0048] In the formula This is a mechanical correction quantity for electromagnetic interference, with dimensions of... This represents the false force output caused by electromagnetic interference. The equivalent voltage amplitude in the dynamic electromagnetic interference equivalent parameters is provided by the vehicle-mounted electromagnetic feature recognition module, and its dimensions are... , which is a voltage dimension, represents the magnitude of the interference voltage that is equivalently superimposed on the electromagnetic noise in the sensor signal circuit. The electromagnetic signal force conversion coefficient has the following dimensions: This coefficient establishes the conversion relationship between electromagnetic interference current and sensor spurious force output. The determination requires a specialized electromagnetic compatibility (EMC) calibration experiment: In an anechoic chamber or GTEM chamber, the six-component force sensor system is placed in a radiated electromagnetic field of known intensity, or a conducted interference signal of known amplitude and frequency is injected into the sensor's power and signal lines through a coupling-decoupling network. The force output offset of the sensor under zero load is recorded, and the ratio between this offset and the injected interference parameter is calculated to obtain the required value. . The real-time speed signal of the vehicle, with dimensions of . The equivalent internal resistance of the sensor has dimensions of , which is the dimension of resistance, represents the Thevenin equivalent resistance as seen from the sensor signal output terminal into the bridge. This parameter can be provided by the sensor manufacturer or measured using a precision impedance analyzer within the sensor's operating frequency range.

[0049] The physical mechanism of this calculation lies in the fact that after vehicle-mounted electromagnetic interference couples to the sensor signal line through conduction or radiation, a common-mode or differential-mode interference voltage is formed between the signal line shielding layer and the core wire, or between the signal line and the reference ground. This interference voltage drives an interference current to flow through the sensor bridge internal resistance and the signal conditioning circuit input impedance. The voltage drop generated by the interference current across the input impedance is amplified by the instrumentation amplifier and superimposed on the actual strain signal, resulting in a shift in the force signal output. The amplitude of the interference current is proportional to the interference voltage. Inversely proportional to the total loop impedance, in the simplified model the total loop impedance is mainly composed of the sensor's equivalent internal resistance. Furthermore, high-power devices in a vehicle's electrical system operate in different states at different vehicle speeds. For example, the switching frequency harmonic energy of the drive motor and its inverter in an electric vehicle increases at high speeds, leading to an increase in interference voltage and interference current coupled to the sensor circuit. Therefore, the vehicle speed... A calculation formula is introduced to dynamically reflect the modulation effect of vehicle speed on electromagnetic interference levels. Electromagnetic signal force conversion coefficient. It acts as a bridge, converting electromagnetic quantities into mechanical quantities. Dimensional verification is performed on the numerator. The dimensions are

[0050]

[0051] denominator The dimensions are Dividing the two yields

[0052]

[0053] It has the same dimensions as force. Electromagnetic interference mechanical correction quantity. It is the third independent correction item in the comprehensive compensation. Subtracting this correction amount from the original force signal can eliminate the measurement error introduced by vehicle speed-related electromagnetic interference.

[0054] The integrated compensation unit, as the final component of the embedded integrated correction and calculation module, is implemented in hardware consisting of multiple data paths, a latch group, and multiple parallel subtractors. The integrated compensation unit simultaneously receives the original three-dimensional force signal, the original three-dimensional torque signal, the attitude offset mechanical correction, the deformation coupling mechanical correction, and the electromagnetic interference mechanical correction, completing the subtraction operation within a single clock cycle. For the three-dimensional force data, the integrated compensation unit performs the calculation. For three-dimensional torque data, the integrated compensation unit adopts an isomorphic calculation formula. ,in This is the original three-dimensional torque signal. , , These represent the torque corrections for attitude, deformation coupling, and electromagnetic interference, respectively. Their calculation formulas correspond to those for force correction, only replacing the force variables with torque variables and adjusting the coefficients to their corresponding values ​​in torque dimensions. The output of the integrated compensation unit is the corrected three-dimensional force data and the corrected three-dimensional torque data.

[0055] The data output and communication module receives the corrected three-dimensional force and torque data and transmits them to other electronic control units or data acquisition devices in the vehicle via the vehicle communication protocol. This module includes a CAN controller and CAN transceiver, or an Ethernet MAC and physical layer transceiver. The CAN controller encapsulates the corrected data into data frames according to a preset CAN identifier and data field format. The period of the data frame can be set to 5 milliseconds, 10 milliseconds, or 20 milliseconds. If vehicle Ethernet is used, the data is packaged into UDP or TCP packets according to the SOME / IP or DoIP protocol. The data output and communication module connects to the vehicle's backbone network via a twisted-pair cable or vehicle Ethernet cable to achieve real-time transmission of measurement data to the entire vehicle system.

[0056] To further refine the implementation details of the wheel attitude deformation calculation module, the strain feature extraction unit performs calculations based on the symmetrical layout characteristics of the strain gauges inside the six-component force sensor when extracting asymmetric strain components. Taking an elastic body structure with eight measuring points as an example, the eight measuring points are divided into four symmetrical pairs, each pair arranged opposite each other along the diameter of the elastic body. When the wheel has no roll, the two measuring points in a symmetrical pair produce approximately equal strain values ​​under the same load; when the wheel rolls, the additional bending moment caused by the roll results in one measuring point in the symmetrical pair being under tension and the other under compression, thus generating a strain difference. The strain feature extraction unit calculates the strain difference of each symmetrical pair sequentially, takes the symmetrical pair with the largest absolute value of the strain difference as the main symmetrical pair for determining the roll direction, determines the sign of the roll angle based on the sign of the strain difference, and outputs the amplitude of the strain difference as the asymmetric strain component. When extracting the alternating strain amplitude, the strain feature extraction unit simultaneously acquires the wheel rotation frequency signal, which can come from the wheel speed sensor or the fundamental frequency obtained by frequency analysis of the six-component force signal. The center frequency of the bandpass filter is adjusted in real time according to the rotation frequency of the wheel. For example, the first or second harmonic component is extracted. The passband width is set to ±10% or ±2Hz of the center frequency. The envelope value of the filtered signal is the alternating strain amplitude.

[0057] The lookup table structure within the attitude deformation mapping unit is constructed using a large number of calibration data points. Taking the first mapping relationship as an example, during calibration, the roll angle is scanned from -10° to +10° at 0.2° intervals, and the amplitude of the asymmetric strain component is recorded at each angle, forming a data table corresponding to the asymmetric strain amplitude and the roll angle. In actual use, for the asymmetric strain amplitude falling between two adjacent grids in the data table, the corresponding roll angle parameter is calculated through linear interpolation or cubic spline interpolation. The calibration of the second mapping relationship is similar. By applying dynamic deformation of different amplitudes to the wheel, the correspondence between the alternating strain amplitude and the deformation is recorded, and a mapping table is established or the coefficients of a polynomial function are fitted.

[0058] During time-frequency analysis, the interference parameter solver of the vehicle-mounted electromagnetic feature recognition module performs sliding window processing on the electromagnetic fluctuation signal. A Hanning or Hamming window is selected as the window function to reduce spectral leakage. The window length must balance frequency and time resolution; for example, a window length of 128 milliseconds or 256 milliseconds is used, with a sliding step size of half the window length. Within each window, the interference parameter solver calculates the power spectral density and tracks characteristic frequencies related to vehicle speed on the power spectrum, such as sixth- or twelfth-order motor frequency harmonics. The amplitude of the power spectral density at that frequency or the integrated energy within a narrow band near that frequency is used as the interference characteristic amplitude at that moment. After low-pass filtering, the interference characteristic amplitude is output as the equivalent voltage amplitude. The cutoff frequency of the low-pass filter can be set between 1Hz and 5Hz to avoid high-frequency jitter in the correction amount.

[0059] In utilizing driving excitation feature data, the wheel attitude deformation calculation module continuously monitors the time series of real-time dynamic roll angle parameters and elastic body dynamic deformation parameters. When the rate of change of the roll angle parameter exceeds a preset threshold or the peak-to-peak value of the deformation parameter exceeds a preset limit within a short time window, the module determines that the wheel is passing over a road bump or depression, and extracts the time location, peak amplitude, and duration of this event as driving excitation feature data. This data is transmitted to the deformation coupling correction calculation unit through an internal communication channel, which then calculates the comprehensive coefficient of deformation influence based on this data. Make temporary adjustments or enable additional transient compensation items. For example, under high-impact conditions, you can... Multiply by a reinforcement factor of 1.2 to 1.5 to compensate for the local nonlinear hardening effect of the elastomer caused by impact.

[0060] In terms of physical implementation, the wheel attitude deformation calculation module, the vehicle-mounted electromagnetic feature recognition module, and the embedded integrated correction calculation module can all be executed by the same vehicle-mounted embedded processor. This processor uses a 32-bit or 64-bit automotive-grade microcontroller or field-programmable gate array (FPGA), internally integrating a central processing unit, a digital signal processing unit, on-chip random access memory (RAM), and flash memory. The three modules are deployed in the processor as software tasks or hardware IP cores. Data is transferred between tasks via shared memory. The real-time operating system is responsible for task scheduling and interrupt management, ensuring that after the six-component digital signals arrive, the wheel attitude deformation calculation task is executed first, followed by the correction calculation task, and finally the data output task. The latency of the entire processing pipeline is determined and controllable.

[0061] At the reference length of the sensor elastomer For 20 mm, attitude equivalent arm The overall influence coefficient of attitude is 4 mm. With a roll angle of 0.35, In this situation, The value is approximately The magnitude of the nonlinear component is approximately The linear lever arm component is approximately The sum of the two factors results in an attitude deviation mechanical correction of approximately two percent of the original force signal. This is in the deformation influence comprehensive coefficient. Deformation roll angle At this time, the first-order coupling term contributes approximately 0.56N, and the nonlinear term contributes approximately 0.01N. The electromagnetic interference correction contributes little under normal operating conditions, but in harsh electromagnetic environments such as high-power inverter output and long unshielded signal lines, [the contribution is significant]. The voltage can reach several volts, and the correction can be increased to several Newtons to tens of Newtons. The above values ​​are only used to indicate the typical magnitude of each correction item. For different sensor models and vehicle vehicles, the relevant parameters need to be set according to the actual calibration results.

[0062] Through the above implementation method, this sensor system uses a single six-component force sensor as the core of mechanical sensing, supplemented by a wheel posture deformation calculation module and an on-board electromagnetic feature recognition module. It extracts real-time wheel posture and deformation information, as well as electromagnetic interference characteristics, online from the sensor's raw digital signals. The embedded integrated correction calculation module progressively calculates various mechanical correction quantities in the order of posture offset correction, posture and deformation coupling correction, and electromagnetic interference correction. Signal compensation is completed in one go in the integrated compensation unit, outputting the corrected three-dimensional force and three-dimensional torque data. Each calculation formula in the entire correction calculation chain is based on the principles of mechanics and electromagnetism, and each parameter has a clear physical meaning and dimension. The correction logic directly corresponds to the real physical interference sources present during dynamic wheel movement. After calibration and experimental verification, it can adapt to a wide range of changes in roll angle, deformation amplitude, and vehicle speed, achieving reliable measurement of dynamic loads.

[0063] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A sensor system for measuring three-dimensional forces and moments on a wheel during travel, characterized by include: The six-component force sensing and acquisition module is used to acquire the original three-dimensional force signal and the original three-dimensional torque signal borne by the wheel in real time, and then convert the original three-dimensional force signal and the original three-dimensional torque signal into digital signals for output. The wheel attitude deformation calculation module is electrically connected to the six-component force sensing and acquisition module. It is used to receive the digital signal and directly calculate the real-time dynamic roll angle parameter and the elastic body dynamic deformation parameter of the wheel from the strain distribution characteristics contained in the digital signal, and output the real-time dynamic roll angle parameter and the elastic body dynamic deformation parameter. An embedded integrated correction and calculation module, electrically connected to the six-component force sensing and acquisition module and the wheel attitude deformation calculation module, is used to receive the digital signal, the real-time dynamic roll angle parameter, and the elastic body dynamic deformation parameter. Based on the real-time dynamic roll angle parameter, it calculates the attitude offset mechanical correction; based on the real-time dynamic roll angle parameter and the elastic body dynamic deformation parameter, it calculates the deformation coupling mechanical correction; and based on the vehicle-mounted electromagnetic interference parameter, it calculates the electromagnetic interference mechanical correction. By fusing the attitude offset mechanical correction, the deformation coupling mechanical correction, and the electromagnetic interference mechanical correction, it compensates the digital signal to generate corrected three-dimensional force data and corrected three-dimensional torque data. The data output and communication module is electrically connected to the embedded integrated correction calculation module and is used to receive and output the corrected three-dimensional force data and the corrected three-dimensional torque data.

2. The sensor system of claim 1, wherein, It also includes an on-board electromagnetic feature recognition module; The vehicle-mounted electromagnetic feature recognition module is electrically connected to the embedded integrated correction and calculation module. It is used to synchronously collect the real-time driving speed signal of the vehicle and the electromagnetic fluctuation signal of the vehicle electrical circuit. Based on the real-time driving speed signal of the vehicle and the electromagnetic fluctuation signal, it calculates the equivalent parameters of dynamic electromagnetic interference and outputs the equivalent parameters of dynamic electromagnetic interference as the vehicle-mounted electromagnetic interference parameters to the embedded integrated correction and calculation module.

3. The sensor system of claim 1, wherein, The embedded integrated correction calculation module includes an attitude correction calculation unit, a deformation coupling correction calculation unit, an electromagnetic interference correction calculation unit, and an integrated compensation unit; The attitude correction calculation unit is electrically connected to the wheel attitude deformation calculation module and is used to call the preset attitude offset mechanical correction calculation formula. The attitude offset mechanical correction calculation formula uses the real-time dynamic roll angle parameter and the original three-dimensional force signal as variables. The attitude offset mechanical correction calculation formula includes nonlinear attitude disturbance component and linear lever arm offset component. The deformation coupling correction calculation unit is electrically connected to the wheel attitude deformation calculation module and is used to call a preset attitude coupling deformation mechanical correction calculation formula. The attitude coupling deformation mechanical correction calculation formula uses the real-time dynamic roll angle parameter, the elastic body dynamic deformation parameter, the original three-dimensional force signal and the preset sensor elastic body reference length as variables. The attitude coupling deformation mechanical correction calculation formula includes the first-order coupling component of attitude and deformation and the higher-order nonlinear error component of deformation. The electromagnetic interference correction calculation unit is used to call a preset electromagnetic interference mechanical correction calculation formula, which uses the vehicle electromagnetic interference parameters and the preset sensor equivalent internal resistance as variables. The integrated compensation unit is electrically connected to the six-component force sensing and acquisition module, the attitude correction calculation unit, the deformation coupling correction calculation unit, and the electromagnetic interference correction calculation unit, respectively. It is used to perform subtraction calculations on the original three-dimensional force signal and the original three-dimensional torque signal with the attitude offset mechanical correction amount, the deformation coupling mechanical correction amount, and the electromagnetic interference mechanical correction amount, respectively, to generate the corrected three-dimensional force data and the corrected three-dimensional torque data.

4. The sensor system of claim 3, wherein, The nonlinear attitude disturbance component included in the attitude deviation mechanical correction formula is characterized by the sine square function of the real-time dynamic tilt angle parameter, and the linear lever arm offset component is characterized by the product of the real-time dynamic tilt angle parameter and the original three-dimensional force signal, as well as the ratio of an attitude equivalent lever arm to the reference length of the sensor elastomer.

5. The sensor system of claim 3, wherein, The first-order coupling component of attitude and deformation included in the attitude coupling deformation mechanics correction formula is characterized by the product of the dynamic deformation parameter of the elastic body and the cosine function of the real-time dynamic roll angle parameter. The higher-order nonlinear error component of deformation is characterized by the product of the square of the dynamic deformation parameter of the elastic body and the original three-dimensional force signal divided by the reference length of the sensor elastic body.

6. The sensor system of claim 3, wherein, The electromagnetic interference mechanical correction calculation formula is characterized by multiplying the equivalent voltage amplitude in the vehicle electromagnetic interference parameters, the preset electromagnetic signal force conversion coefficient, and the real-time vehicle speed signal by the equivalent internal resistance of the sensor.

7. The sensor system according to claim 1, characterized in that, The wheel attitude deformation calculation module includes a strain feature extraction unit and an attitude deformation mapping unit; The strain feature extraction unit is used to extract the asymmetric strain component and alternating strain amplitude in the digital signal; The attitude deformation mapping unit is electrically connected to the strain feature extraction unit and is used to determine the real-time dynamic tilt angle parameter according to the preset first mapping relationship and the distribution direction and magnitude of the asymmetric strain component, and to determine the elastic body dynamic deformation parameter according to the preset second mapping relationship and the fluctuation characteristics of the alternating strain amplitude.

8. The sensor system of claim 2, wherein, The vehicle-mounted electromagnetic feature recognition module includes a vehicle speed signal interface, an electromagnetic acquisition front-end, and an interference parameter solver. The vehicle speed signal interface is used to access and acquire the real-time driving speed signal of the vehicle; The electromagnetic acquisition front end is used to acquire the electromagnetic fluctuation signal of the vehicle electrical circuit; The interference parameter solver is electrically connected to the vehicle speed signal interface and the electromagnetic acquisition front end, respectively. It is used to perform real-time amplitude-frequency characteristic analysis of the electromagnetic wave signal in relation to the real-time vehicle speed signal, and extract the electromagnetic interference characteristic amplitude related to the dynamic change of vehicle speed as the equivalent parameter of the dynamic electromagnetic interference.

9. The sensor system of claim 1, wherein, The six-component force sensing acquisition module is the only hardware sensing module in the sensor system used to acquire the mechanical signals of the wheel.

10. The sensor system of claim 1, wherein, The vehicle-mounted embedded processor is internally integrated with the wheel posture deformation solving module and the embedded comprehensive correction operation module. The vehicle-mounted embedded processor is electrically connected with the six-component force sensing and collecting module and the data output and communication module through an internal high-speed bus interface.

Images