Vehicle body attitude control method, system and vehicle body controller

Abstract

This application provides a vehicle body attitude control method, system, and body controller, relating to the fields of vehicle dynamic control and intelligent chassis control technology. The method includes acquiring attitude data during vehicle operation, extracting features from the attitude data to obtain fractal features; wherein the fractal features are used to characterize the spatial complexity of vehicle body attitude disturbances; determining the eddy current energy of the vehicle body attitude disturbances based on the fractal features; and adjusting the vehicle body attitude according to the eddy current energy. This application significantly improves the response speed and stability of vehicle body attitude control in nonlinear, multi-disturbance environments.

Description

Technical Field

[0001] This application relates to the field of vehicle dynamic control and intelligent chassis control technology, and in particular to a vehicle body attitude control method, system and body controller. Background Technology

[0002] In the field of vehicle dynamics control, vehicle attitude control is a key technology to ensure handling stability and ride comfort.

[0003] Existing control methods include proportional-integral-derivative control based on linear models, model predictive control, and data-driven methods.

[0004] However, these methods have the following technical problems: linear control models are difficult to accurately describe the nonlinear eddy current dynamic response of vehicles under conditions such as high-speed turning and emergency braking, resulting in a decrease in control accuracy. Summary of the Invention

[0005] The purpose of this application is to provide a vehicle body attitude control method, system, and vehicle body controller to alleviate the aforementioned technical problems existing in the prior art.

[0006] In a first aspect, the present invention provides a vehicle body posture control method, comprising: The vehicle's attitude data during driving is acquired, and features are extracted from the attitude data to obtain fractal features; among them, fractal features are used to characterize the spatial complexity of vehicle attitude perturbation. The eddy current energy of vehicle body attitude disturbance is determined based on fractal characteristics; The vehicle body posture is adjusted based on eddy current energy.

[0007] In an optional implementation, attitude data during vehicle movement is acquired, and feature extraction is performed on the attitude data to obtain fractal features, including: Acquire pitch, roll, and yaw angle data collected by the vehicle's body sensors during driving; Multi-scale decomposition was performed on the pitch, roll, and yaw angle data to obtain time-frequency characteristic coefficients at multiple scales. Based on the time-frequency characteristic coefficients, the box counting method is used to calculate the coverage number corresponding to each scale, and the fractal characteristics are determined according to the power law relationship between the coverage number and the scale.

[0008] In an optional implementation, the pitch, roll, and yaw angle data are decomposed into multiple scales to obtain time-frequency characteristic coefficients at multiple scales, including: Multi-resolution analysis of attitude data was performed to separate high-frequency components reflecting transient impacts and low-frequency components reflecting trend changes step by step. The corresponding detail coefficients and approximation coefficients are extracted at each decomposition level and used as time-frequency feature coefficients.

[0009] In an optional implementation, determining the eddy current energy of the vehicle body attitude disturbance based on fractal characteristics includes: The fractal features are mapped to vortex intensity parameters; where the larger the fractal feature, the higher the weight of the corresponding vortex intensity parameter. By embedding the vortex intensity parameter into the vehicle body energy flow equation, a vortex model characterizing the energy distribution and transfer characteristics of the vehicle body is constructed. The energy circulation equation is derived based on the eddy current model, and the energy circulation equation is solved to obtain the eddy current feedback gain matrix, which serves as a quantitative representation of eddy current energy.

[0010] In an optional implementation, mapping fractal features to vortex intensity parameters includes: Establish a nonlinear mapping function between fractal characteristics and vortex intensity parameters; The fractal features are converted into vortex intensity parameters at the corresponding scale using a nonlinear mapping function.

[0011] In an optional implementation, the energy circulation equation is derived based on the eddy current model, and the energy circulation equation is solved to obtain the eddy current feedback gain matrix, including: A state-space expression containing vortex intensity parameters is constructed based on the vortex model; Based on the state-space expression and energy conservation constraints, an energy circulation equation is generated. The energy circulation equation is stabilized and solved, and the eddy current feedback gain matrix is ​​output for feedback control.

[0012] In an optional implementation, adjusting the vehicle body posture based on eddy current energy includes: Real-time calculation of the tracking error between the current attitude data and the preset vehicle attitude reference model; The eddy current feedback gain matrix, which characterizes eddy current energy, is dynamically corrected based on the tracking error. The modified eddy current feedback gain matrix is ​​combined with the distribution characteristics of eddy current energy to generate vehicle body attitude control commands, which drive the suspension or steering actuators to complete attitude adjustment.

[0013] In an optional implementation, the method further includes: When the attitude data is drastically disturbed, abnormal patterns of vehicle body attitude are identified based on the abrupt change characteristics of fractal features. Based on the abnormal mode, the preset emergency control strategy is activated to generate emergency vehicle posture control commands for emergency adjustment of the vehicle posture.

[0014] The feature extraction module is used to acquire attitude data during vehicle driving, and to perform multi-scale feature extraction on the attitude data to obtain fractal features; wherein, the fractal features are used to characterize the spatial complexity of vehicle attitude perturbation. The eddy current energy determination module is used to determine the eddy current energy of the vehicle body attitude disturbance based on the fractal characteristics. The vehicle body attitude adjustment module is used to adjust the vehicle body attitude based on the eddy current energy.

[0015] Thirdly, the present invention provides a vehicle body controller, including a processor and a memory, wherein the memory stores computer-executable instructions that can be executed by the processor, and the processor executes the computer-executable instructions to implement the method of any of the foregoing embodiments.

[0016] The vehicle attitude control method, system, and vehicle controller provided in this application, by acquiring attitude data and extracting fractal features to characterize the spatial complexity of disturbances, can transform nonlinear, multi-scale disturbances that are difficult to model using traditional methods into quantifiable indicators, overcoming the deficiency of linear models in describing the dynamic response of complex eddy currents. Secondly, by determining eddy current energy based on fractal features, the energy transfer and dissipation processes within the vehicle body are incorporated into the control framework, solving the low-frequency oscillations and overshoot problems caused by insufficient modeling of eddy current feedback effects in existing technologies. Finally, by directly adjusting the vehicle attitude based on eddy current energy, adaptive matching between control gain and disturbance complexity is achieved, avoiding the limitation of purely data-driven methods lacking physical interpretability. In summary, this application significantly improves the response speed and stability of vehicle attitude control in nonlinear, multi-disturbance environments. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the specific embodiments of this application or the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0018] Figure 1 A flowchart of a vehicle body attitude control method provided in this application embodiment; Figure 2 A structural diagram of a vehicle body attitude control system provided in an embodiment of this application; Figure 3 This is a structural diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0020] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0021] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0022] This application provides a vehicle body attitude control method. See [link to relevant documentation] Figure 1 As shown, the method mainly includes the following steps: S110: Acquire attitude data during vehicle movement, extract features from the attitude data to obtain fractal features. These fractal features characterize the spatial complexity of vehicle attitude perturbations.

[0023] During vehicle operation, especially during high-speed cornering, emergency braking, or traversing bumpy roads, the vehicle body undergoes multi-dimensional attitude changes such as pitch, roll, and yaw. This step first acquires real-time vehicle attitude data using onboard sensors (such as a six-axis inertial measurement unit, suspension displacement sensors, etc.). Attitude data may include pitch angle, roll angle, yaw rate, and vertical acceleration.

[0024] After acquiring the raw attitude data, it is not directly used for control, but rather for feature extraction. This feature extraction differs from conventional filtering or noise reduction; instead, it employs fractal analysis methods capable of characterizing nonlinear, multi-scale perturbation patterns. In practice, feature extraction can involve multi-scale decomposition of the attitude data to obtain time-frequency characteristic coefficients at multiple scales. For example, using multi-resolution analysis, high-frequency components reflecting transient impacts and low-frequency components reflecting trend changes are separated level by level, and corresponding detail coefficients and approximation coefficients are extracted at each decomposition level as time-frequency characteristic coefficients.

[0025] In this scheme, fractal features are used to quantitatively describe the spatial complexity of vehicle body attitude disturbances. The more irregular and refined the disturbance, the higher the fractal dimension value. When determining fractal features, box counting can be used to calculate the coverage number corresponding to each scale. Specifically, the signal at each scale is divided into grids of different side lengths on the time-amplitude plane, the number of grids containing signal points is counted, and then the relationship between the grid side length and the coverage number is fitted in a double logarithmic coordinate system. The fractal features are determined based on the power-law relationship between the coverage number and the scale.

[0026] In this way, complex road surface excitations and vehicle body elastic deformation responses, which are originally difficult to model, can be transformed into quantifiable scalar or vector features, laying the foundation for subsequent eddy current energy calculations. Furthermore, this method does not rely on linear model assumptions, can adapt to various complex road surface inputs, and has a moderate computational load, allowing it to run in real time within the vehicle controller.

[0027] S120 determines the eddy current energy of vehicle body attitude disturbance based on fractal characteristics.

[0028] In one implementation, fractal features can be mapped to vortex intensity parameters. Specifically, a nonlinear mapping function is first established. The larger the fractal feature, the higher the weight of the corresponding vortex intensity parameter. The vortex intensity parameter characterizes the concentration of disturbance energy at the corresponding scale. For example, a higher fractal dimension indicates more complex disturbances and greater energy accumulation, thus the vortex intensity parameter takes a larger value; conversely, a lower dimension takes a smaller value. Then, the vortex intensity parameter is embedded into the vehicle body energy flow equation to construct an vortex model characterizing the energy distribution and transfer characteristics of the vehicle body.

[0029] Based on this, the energy circulation equation is derived from the eddy current model, and the equation is solved to obtain the eddy current feedback gain matrix, which serves as a quantitative representation of eddy current energy. Specifically, a state-space expression containing eddy current intensity parameters can be constructed based on the eddy current model, and then the energy circulation equation can be generated by combining energy conservation constraints. Finally, the equation is stabilized and solved to output the eddy current feedback gain matrix for feedback control. The dimension of this matrix corresponds to the degrees of freedom of control (e.g., roll, pitch, vertical), and its element values ​​are updated in real time with changes in fractal characteristics. In this way, the abstract disturbance complexity is transformed into control parameters with physical meaning.

[0030] The S130 adjusts its vehicle posture based on eddy current energy.

[0031] After obtaining the eddy current energy (e.g., represented by an eddy current feedback gain matrix), this step uses it for closed-loop control. Specifically, the current attitude data is compared with a preset vehicle attitude reference model (representing the ideal driving attitude) to calculate the tracking error. This error is then used to dynamically correct the eddy current feedback gain matrix, enabling control commands to respond in real-time to disturbance changes. The corrected gain matrix, combined with the distribution characteristics of the eddy current energy, generates specific vehicle attitude control commands, such as desired suspension damping force, air spring pressure, or anti-roll bar torque. These commands are sent to the suspension electronic control unit or steering actuator to achieve active adjustment of the vehicle attitude.

[0032] In one specific implementation, the tracking error between the current attitude data and the preset vehicle attitude reference model is calculated in real time. Based on this tracking error, the eddy current feedback gain matrix, used to characterize eddy current energy, is dynamically corrected. Then, the corrected eddy current feedback gain matrix, combined with the distribution characteristics of eddy current energy, generates vehicle attitude control commands, which drive the suspension or steering actuators to complete attitude adjustments. For example, during emergency steering, the damping force of the outer suspension increases while that of the inner suspension decreases, creating an anti-roll moment; during emergency braking, the damping of the front suspension increases while that of the rear suspension decreases appropriately, suppressing the "nose-diving" phenomenon. Compared to traditional fixed-gain control, the adjustment amount in this scheme automatically scales with changes in fractal characteristics and eddy current energy, achieving adaptive matching between control gain and disturbance complexity.

[0033] Furthermore, considering potential anomalies such as sensor failure and extreme road impacts during actual driving, this solution also provides redundant safety mechanisms. When attitude data experiences severe disturbances, abnormal vehicle attitude patterns can be identified based on the abrupt changes in fractal characteristics. For example, when the fractal dimension increases sharply within a short period, it can be determined that the vehicle is experiencing extreme conditions (such as obstacle avoidance or sideslip). Once an abnormal pattern is identified, the system immediately activates a preset emergency control strategy, generating emergency vehicle attitude control commands to perform emergency adjustments to the vehicle attitude. Emergency control strategies may include maximizing suspension damping, actively adjusting air spring pressure to raise the vehicle body, or even triggering electronic stability program intervention. These emergency commands have higher priority than conventional control, helping the vehicle maintain stability under dangerous conditions.

[0034] This application quantifies the spatial complexity of vehicle body disturbances by introducing fractal features and constructs an eddy current energy model based on these features, achieving physically interpretable control of the vehicle body's dynamic response. Compared to traditional linear control or purely data-driven methods, this scheme exhibits faster response speed and stronger robustness under nonlinear disturbance environments. Simulation and real-vehicle test results show that under emergency steering conditions, the vehicle roll angle is effectively suppressed; under emergency braking conditions, the pitch angle convergence time is significantly shortened; and when driving on random road surfaces, the root mean square value of vertical acceleration is significantly reduced, improving both overall vehicle comfort and stability.

[0035] For ease of understanding, the method provided in this application will be described in detail below.

[0036] The above-mentioned acquisition of vehicle attitude data during driving and feature extraction from this data to obtain fractal features can be implemented by first acquiring pitch, roll, and yaw angle data collected by vehicle sensors during driving. These data typically originate from a six-axis inertial measurement unit (IMU) mounted at the vehicle's center of gravity, with a sampling frequency of, for example, 500Hz. The required angle values ​​can be directly output or obtained through integration. To more comprehensively describe the vehicle's attitude, vertical acceleration and the travel changes of each suspension element can also be acquired simultaneously to assist in the verification of fractal features.

[0037] After obtaining the aforementioned angle data, the pitch, roll, and yaw angle data need to be decomposed into multiple scales to obtain time-frequency characteristic coefficients at multiple scales. In one specific implementation, wavelet transform can be used for multi-resolution analysis. For example, the Daubechies-4 (db4) wavelet basis can be selected to perform a 5-level decomposition of the signal. Each level of decomposition divides the signal into high-frequency detail coefficients and low-frequency approximation coefficients. The detail coefficients correspond to high-frequency components such as transient impacts and minor road bumps, while the approximation coefficients correspond to low-frequency trend components such as overall vehicle attitude changes.

[0038] Taking the vertical acceleration signal as an example, the frequency ranges and physical meanings corresponding to the five scales are as follows: Scale 1 corresponds to 25-50Hz, reflecting high-frequency vibrations caused by minor bumps in the road surface; Scale 2 corresponds to 12.5-25Hz, reflecting mid-frequency vibrations caused by wheel imbalance; Scale 3 corresponds to 6.25-12.5Hz, reflecting the main vertical vibration of the vehicle body; Scale 4 corresponds to 3.125-6.25Hz, reflecting pitch attitude-related vibrations; Scale 5 corresponds to 0-3.125Hz, reflecting slow changes in the vehicle's posture (such as the pitching trend when driving on a long slope).

[0039] In this way, the corresponding detail coefficients (d1 to d5) and approximation coefficients (a5) can be extracted at each decomposition level as time-frequency characteristic coefficients.

[0040] After obtaining the time-frequency characteristic coefficients, the box counting method is further used to calculate the coverage number corresponding to each scale, and the fractal characteristics are determined based on the power-law relationship between the coverage number and the scale. The specific calculation formula is as follows:

[0041] in, D f For fractal dimensions, N(ε) This represents the number of characteristic responses at scale ε (i.e., the number of grid cells containing signal points). In practice, the signal at each scale is divided into grids ε of different side lengths on the time-amplitude plane, and the number of grid cells containing signal points is counted. N(ε) The absolute value of the slope of the straight line fitted to ln(ε) and lnN(ε) in a log-log coordinate system is the fractal dimension. D f As a core indicator characterizing the complexity of disturbances under current operating conditions.

[0042] The basic operation of box counting is as follows: For the detail coefficients at each scale, select an effective segment (e.g., 1 second in duration, corresponding to 500 sampling points), remove the DC component, and plot it as a curve on the time-amplitude two-dimensional plane. Then, cover the curve with a square grid of side length ε, and count the number of grid points containing the curve. N(ε) By changing the value of ε (for example, gradually increasing it from 0.01g to 0.5g, taking a total of 20 different ε values), a series of (ε, N(ε) Data pairs. In a double logarithmic coordinate system, with lnε as the abscissa and ln... N(ε) If a linear fit is performed on the ordinate, and the absolute value of the slope of the fitted line is k, then the fractal dimension D at that scale is... f = 2 + k (the fractal dimension in a two-dimensional plane ranges from 1 to 2). The fractal dimension (D) at each scale... f1 To D f5 The overall fractal dimension D is obtained by weighting the perturbation energy. f D serves as a core indicator characterizing the complexity of disturbances under current operating conditions. For example, when smoothing a road surface... f Approximately 1.2, D on bumpy roads or during emergency turns f Approximately 1.7 to 1.8. This method can quantitatively characterize the degree of disturbance irregularity at different frequency scales, has strong ability to describe complex nonlinear and non-stationary disturbances, and has a moderate computational load, making it suitable for real-time vehicle operation.

[0043] When performing multi-scale decomposition on pitch, roll, and yaw angle data, a multi-resolution analysis approach can be used to progressively separate high-frequency components reflecting transient impacts from low-frequency components reflecting trend changes. For example, in each decomposition level, the original signal is decomposed into a high-frequency detail coefficient and a low-frequency approximation coefficient, and then the low-frequency approximation coefficient is used as the input for the next level of decomposition.

[0044] Taking a five-layer decomposition as an example, the first layer of detail coefficients corresponds to high-frequency vibrations above 25Hz (wheel rebound impacts caused by minor road surface unevenness), the second layer of detail coefficients corresponds to mid-frequency vibrations from 12.5Hz to 25Hz (periodic excitations caused by poor wheel dynamic balance), and the third to fifth layers of detail coefficients correspond to lower-frequency vertical principal vibrations of the vehicle body, pitch-related vibrations, and slow changes in overall attitude, respectively. The approximation coefficients retain the most prominent low-frequency trends in the signal, such as long-period attitude changes caused by vehicle acceleration, deceleration, or steering. In this way, the corresponding detail coefficients and approximation coefficients are extracted at each decomposition level as time-frequency characteristic coefficients. This multi-resolution analysis method preserves the temporal location information of the signal, accurately capturing the moment of disturbance occurrence and providing a reliable data foundation for the accurate calculation of fractal dimension.

[0045] After obtaining the fractal features, it is necessary to determine the eddy current energy of the vehicle body attitude disturbance based on these features. One approach is to first map the fractal features to eddy current intensity parameters. A fractal feature (e.g., the comprehensive fractal dimension) is a dimensionless complexity index, typically ranging from 1 to 2. A nonlinear mapping function (e.g., an exponential function or a piecewise linear function) can be established, with the core principle being: the larger the fractal feature, the higher the weight of the corresponding eddy current intensity parameter. For example, a fractal dimension close to 1.2 corresponds to a smooth road surface, with a smaller eddy current intensity parameter value; a fractal dimension close to 1.8 corresponds to severe bumps or emergency steering, with a larger eddy current intensity parameter value. This parameter characterizes the concentration of disturbance energy at the corresponding scale.

[0046] Next, the vortex intensity parameter is embedded into the vehicle body energy flow equation to construct an vortex model characterizing the energy distribution and transfer characteristics of the vehicle body. The vehicle body energy flow equation can be compared to the energy conservation equation in continuum mechanics. Treating the vehicle body as an elastic body, its attitude changes are accompanied by the generation, transfer, and dissipation of energy. The vortex feedback equation describes the relationship between local energy circulation and attitude deviation. A specific form of the vortex feedback equation is as follows:

[0047] in, Ec Local circulation flow (unit: J) reflects the degree of energy accumulation in a certain area of ​​the vehicle body (such as suspension connection points); t Ec This represents the rate of change of energy; a positive value indicates an increase in energy (increased attitude deviation), while a negative value indicates energy dissipation (attitude recovery). kroll and kpitch These are the roll and pitch energy coefficients, which can be obtained through vehicle calibration. ωroll and ω Pitch The yaw and pitch angular velocities are from the IMU; η This is the energy dissipation coefficient, with a default value of 0.3, and it varies with the fractal dimension D. f Dynamic adjustment (D) f The larger, η The smaller the size, the better the energy feedback. Fctrl The control force to be solved; vattitude The attitude change rate (e.g., the rate of change of the roll angle). The vortex intensity parameter is multiplied as a coefficient in the energy generation or transfer term, causing the response characteristics of the energy equation to adapt to the disturbance complexity.

[0048] Then, the energy circulation equation is derived based on the eddy current model, and solved to obtain the eddy current feedback gain matrix. In practical implementation, the finite element method can be used for solving. First, a simplified finite element model of the vehicle body is established. For example, the vehicle body is divided into 1000 tetrahedral elements, with a focus on refining the elements near the suspension connection points and the center of gravity of the vehicle body (mesh size 5mm), and the element size in other areas is 10-15mm. Boundary conditions include: acceleration and angular velocity data from the input IMU as loads (e.g., lateral acceleration of 1.2g corresponds to an emergency steering condition), and suspension displacement as constraints (reflecting road support). Then, the strain energy, kinetic energy and dissipated energy of each element are calculated, and the regions with a rate of change of energy greater than zero are extracted as energy accumulation areas (corresponding to the parts where the attitude deviation increases). The energy circulation intensity of these regions is calculated (unit: W / m²).

[0049] Finally, the control gain matrix K (a 3×3 matrix corresponding to the roll, pitch, and vertical directions) is dynamically adjusted based on the energy circulation intensity. The adjustment rule is as follows: for every 10 W / m² increase in energy circulation intensity, the gain matrix elements increase by 15%, ensuring faster energy accumulation and a stronger control force. Fctrl The larger.

[0050] In this way, an eddy current feedback gain matrix is ​​output for feedback control. The dimension of this matrix corresponds to the degrees of freedom of the control, and the element values ​​are updated in real time with the fractal characteristics. This scheme transforms the abstract disturbance complexity into physically meaningful control parameters, enabling the control system's response to match the actual energy dissipation requirements.

[0051] When mapping fractal features to vortex intensity parameters, a nonlinear mapping function can be established between the fractal features and the vortex intensity parameters. This nonlinear mapping function can be based on vehicle dynamics characteristics and actual control requirements. For example, the correspondence between fractal dimensions and ideal control gains under different operating conditions can be obtained through simulation or real-vehicle calibration. Then, a monotonically increasing nonlinear function (such as a power function, exponential function, or sigmoid function) can be fitted. Through this nonlinear mapping function, the fractal features are converted into vortex intensity parameters at the corresponding scale.

[0052] Since fractal features themselves may be obtained by weighted averaging of fractal dimensions at multiple scales, the mapping can also maintain multi-scale characteristics: map the fractal dimensions at each scale separately to obtain a set of vortex intensity parameters, and then synthesize a comprehensive vortex intensity parameter according to energy weights.

[0053] In this way, the contribution of disturbances at different scales to the control gain can be adjusted independently. This nonlinear mapping can more realistically reflect the nonlinear relationship between disturbance complexity and control requirements in actual physical processes. It grows gradually in the low-complexity region to avoid oversensitivity, and grows rapidly in the high-complexity region to ensure timely suppression of severe disturbances.

[0054] When deriving the energy circulation equation and solving it to obtain the eddy current feedback gain matrix based on the eddy current model, in addition to the finite element method, an algebraic method based on state-space expressions can also be used. Specifically: First, a state-space expression containing vortex intensity parameters is constructed based on the vortex model. State variables can be selected as vehicle roll angle, pitch angle, vertical displacement, and their derivatives. The vortex intensity parameters appear as time-varying coefficients in the system matrix. Specifically, the state variables... Where φ is the roll angle, θ is the pitch angle, and z is the vertical displacement of the vehicle's center of gravity. Control input u = [FLF, FRF, FLR, FRR] These represent the output forces of the left front, right front, left rear, and right rear suspension actuators, respectively. The system output y = [φ, θ, z] .

[0055] Introducing the vortex intensity parameter γ(D) f This parameter is determined by the fractal feature D. f Obtained through nonlinear mapping. The standard vehicle dynamics state equation is modified as follows:

[0056]

[0057] In this system matrix A(γ), the stiffness and damping terms are replaced with time-varying parameters related to γ: roll stiffness and pitch stiffness are multiplied by (1 + αγ), and roll damping and pitch damping are multiplied by (1 + αγ). βγ), where α and β are pre-calibrated coefficients.

[0058] Then, based on this state-space expression and energy conservation constraints, the energy circulation equation is generated. The total energy function is defined as follows:

[0059] Where M is the mass matrix, and K(γ) is the time-varying stiffness matrix containing vortex intensity parameters. The rate of energy change is obtained by subtracting the internal dissipated power from the external input power:

[0060] D(γ) is the time-varying damping matrix. Substituting the state-space expression and introducing Lagrange multipliers to handle constraints, we obtain the local energy circulation equation at each suspension connection point:

[0061] Where e i Let d be the local circulation flow at the i-th suspension point. i (γ) is the corresponding damping coefficient, κ ij This represents the energy exchange coefficient between adjacent points.

[0062] Finally, the energy circulation equation is stabilized and solved, outputting the eddy current feedback gain matrix for feedback control. The Lyapunov function V = x is constructed. P x + Σ_i e i ², requiring its derivative to be negative definite. Substituting into the state-space expression and energy circulation equation, we derive the equation containing P and K. v Linear matrix inequalities (LMI) for γ:

[0063] Where R is the weight matrix positively correlated with the vortex intensity parameter γ, and I is the identity matrix. Solving for the LMI (e.g., using the interior-point method) yields the positive definite matrix P, from which the feedback gain matrix can be calculated:

[0064] Within each control cycle, firstly, based on the current fractal dimension D... f Update γ, then update A(γ) and R, and then solve LMI to obtain the new K. vTo meet the 50Hz operating frequency of the vehicle controller, a gain table corresponding to different γ values ​​can be pre-solved offline. During actual vehicle operation, K is obtained in real time through table lookup and linear interpolation. v .

[0065] After obtaining the eddy current feedback gain matrix, the specific process of adjusting the vehicle attitude based on eddy current energy is as follows. First, the tracking error between the current attitude data and the preset vehicle attitude reference model is calculated in real time. The preset vehicle attitude reference model is an ideal response model pre-designed based on vehicle dynamics and driving intentions (e.g., a smooth response in comfort mode and a direct response in sport mode). The tracking error includes roll angle error, pitch angle error, and yaw angle error, etc. Then, the eddy current feedback gain matrix, which characterizes eddy current energy, is dynamically corrected based on this tracking error. The correction algorithm can adopt the gradient descent method or the recursive least squares method in adaptive control, and its update rule is as follows:

[0066] Among them, K t Let K be the feedback gain matrix at the current time. t-1 Let E be the gain matrix from the previous time step, Δt be the time step, and E be the gain matrix from the previous time step. v_ref E is the reference value for eddy current energy. v_meas This is the measured value of eddy current energy. E v / t represents the rate of change of eddy current energy. If the tracking error is large, it indicates that the current gain is insufficient to suppress disturbances, and the gain needs to be increased; if the tracking error is small or overshoot occurs, the gain should be appropriately reduced. This update rule enables the controller to adjust the gain matrix in real time based on the deviation between the measured energy and the reference energy, as well as the rate of energy change, achieving self-learning and rapid convergence.

[0067] Finally, the modified eddy current feedback gain matrix is ​​combined with the distribution characteristics of eddy current energy to generate vehicle attitude control commands, which then drive the suspension or steering actuators to complete attitude adjustments. The distribution characteristics of eddy current energy indicate the areas where energy accumulates (e.g., front or rear suspension, left or right side), and the control commands should prioritize suppressing disturbances in these energy accumulation areas. In practice, the control output signal can be a PWM wave (frequency 1kHz, duty cycle 0-100%), driving active suspension actuators (such as electromagnetic dampers or air spring solenoid valves). For roll control, the difference in damping force between the left and right suspensions is adjusted; for example, during emergency steering, the damping of the outer suspension increases while the damping of the inner suspension decreases to create an anti-roll moment. For pitch control, the damping force between the front and rear suspensions is adjusted; for example, during emergency braking, the damping of the front suspension increases while the damping of the rear suspension decreases appropriately to suppress the "nose-diving" phenomenon. For vertical control, the overall damping is adjusted according to the fractal dimension; a larger fractal dimension results in increased damping to reduce vibration transmission. This closed-loop adaptive control mechanism enables the control system to continuously optimize its own parameters, while also allocating differentiated commands based on energy distribution characteristics, thereby improving control accuracy and energy utilization efficiency.

[0068] Furthermore, this method also includes handling of abnormal situations. When the attitude data experiences severe disturbances, abnormal patterns in the vehicle's attitude are identified based on the abrupt change characteristics of fractal features. Severe disturbances can be defined as the rate of change of the attitude data exceeding a certain threshold, or the increment of the fractal feature over multiple consecutive control cycles exceeding a preset threshold. The abrupt change characteristic of the fractal feature refers to its value rapidly increasing within a short period of time (e.g., jumping from 1.3 to 1.8), which usually indicates that the vehicle is experiencing dangerous conditions such as emergency obstacle avoidance, sideslip, or rollover. By monitoring the rate of change of the fractal feature, anomalies can be identified faster than simply relying on attitude angle thresholds. Based on the identified abnormal patterns, a preset emergency control strategy is activated, generating emergency vehicle attitude control commands to perform emergency adjustments to the vehicle's attitude. The emergency control strategy has higher priority than conventional control, and its control objective shifts from comfort to safety.

[0069] For example, when a rollover risk is identified, emergency control strategies may include: maximizing the damping of the outer suspension and minimizing the damping of the inner suspension, while actively adjusting the air spring pressure to improve the vehicle's anti-roll capability. If necessary, the electronic stability program or anti-lock braking system may be requested to intervene. These instructions are sent to the actuators with the highest priority to ensure a rapid response in hazardous conditions.

[0070] Meanwhile, the system also incorporates safety constraints: actuator outputs have limits (such as the maximum damping force of electromagnetic shock absorbers and the maximum air volume of air springs) to prevent overload damage; when abnormal sensor data is detected (such as IMU signal loss), it switches to default PID control to ensure safety. This hierarchical control architecture balances the comfort of daily driving with the safety of extreme conditions, improving the redundancy level of the vehicle's active safety.

[0071] Based on the above method embodiments, this application also provides a vehicle body posture control system, see [link to relevant documentation]. Figure 2 As shown, it includes: The feature extraction module 210 is used to acquire attitude data during vehicle driving, and to perform multi-scale feature extraction on the attitude data to obtain fractal features; wherein, the fractal features are used to characterize the spatial complexity of vehicle attitude disturbance. Eddy current energy determination module 220 is used to determine the eddy current energy of the vehicle body attitude disturbance based on the fractal characteristics; The vehicle body posture adjustment module 230 is used to adjust the vehicle body posture based on the eddy current energy.

[0072] In one feasible implementation, the feature extraction module 210 is specifically used for: Acquire pitch, roll, and yaw angle data collected by the vehicle's body sensors during driving; Multi-scale decomposition was performed on the pitch, roll, and yaw angle data to obtain time-frequency characteristic coefficients at multiple scales. Based on the time-frequency characteristic coefficients, the box counting method is used to calculate the coverage number corresponding to each scale, and the fractal characteristics are determined according to the power law relationship between the coverage number and the scale.

[0073] In one feasible implementation, the feature extraction module 210 is further configured to: Multi-resolution analysis of attitude data was performed to separate high-frequency components reflecting transient impacts and low-frequency components reflecting trend changes step by step. The corresponding detail coefficients and approximation coefficients are extracted at each decomposition level and used as time-frequency feature coefficients.

[0074] In one feasible implementation, the above-mentioned eddy current energy determination module is specifically used for: The fractal features are mapped to vortex intensity parameters, where the larger the fractal feature, the higher the weight of the corresponding vortex intensity parameter; By embedding the vortex intensity parameter into the vehicle body energy flow equation, a vortex model characterizing the energy distribution and transfer characteristics of the vehicle body is constructed. The energy circulation equation is derived based on the eddy current model, and the energy circulation equation is solved to obtain the eddy current feedback gain matrix, which serves as a quantitative representation of eddy current energy.

[0075] In one feasible implementation, the above-mentioned eddy current energy determination module is further used for: Establish a nonlinear mapping function between fractal characteristics and vortex intensity parameters; The fractal characteristics are converted into vortex intensity parameters at the corresponding scale using a nonlinear mapping function, which are used to characterize the concentration of perturbation energy at the corresponding scale.

[0076] In one feasible implementation, the eddy current energy determination module is further configured to: A state-space expression containing vortex intensity parameters is constructed based on the vortex model; Based on the state-space expression and energy conservation constraints, an energy circulation equation is generated. The energy circulation equation is stabilized and solved, and the eddy current feedback gain matrix is ​​output for feedback control.

[0077] In one feasible implementation, the above-mentioned vehicle posture adjustment module is specifically used for: Real-time calculation of the tracking error between the current attitude data and the preset vehicle attitude reference model; The eddy current feedback gain matrix, which characterizes eddy current energy, is dynamically corrected based on the tracking error. The modified eddy current feedback gain matrix is ​​combined with the distribution characteristics of eddy current energy to generate vehicle body attitude control commands, which drive the suspension or steering actuators to complete attitude adjustment.

[0078] In one feasible implementation, the system further includes: an abnormal mode adjustment module, used for: When the attitude data is drastically disturbed, abnormal patterns of vehicle body attitude are identified based on the abrupt change characteristics of fractal features. Based on the abnormal mode, the preset emergency control strategy is activated to generate emergency vehicle posture control commands for emergency adjustment of the vehicle posture.

[0079] The system provided in this application embodiment has the same implementation principle and technical effects as the aforementioned method embodiment. For the sake of brevity, any parts not mentioned in the system embodiment can be referred to the corresponding content in the aforementioned method embodiment.

[0080] This application also provides an electronic device, such as... Figure 3 The diagram shows the structure of the electronic device 100, which includes a processor 31 and a memory 30. The memory 30 stores computer-executable instructions that can be executed by the processor 31. The processor 31 executes the computer-executable instructions to implement any of the methods described above.

[0081] exist Figure 3 In the illustrated embodiment, the electronic device further includes a bus 32 and a communication interface 33, wherein the processor 31, the communication interface 33, and the memory 30 are connected via the bus 32.

[0082] The memory 30 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 33 (which can be wired or wireless), such as the Internet, wide area network, local area network, metropolitan area network, etc. The bus 32 may be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The bus 32 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0083] Processor 31 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of processor 31 or by software instructions. Processor 31 can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the embodiments of this application can be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in the memory, and the processor 31 reads the information in the memory and, in conjunction with its hardware, completes the steps of the method in the aforementioned embodiment.

[0084] This application also provides a computer-readable storage medium storing computer-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions cause the processor to implement the above-described method. For specific implementation details, please refer to the foregoing method embodiments, which will not be repeated here.

[0085] The computer program products of the vehicle body attitude control method, system and vehicle body controller provided in the embodiments of this application include a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods described in the preceding method embodiments. For specific implementation, please refer to the method embodiments, which will not be repeated here.

[0086] Unless otherwise specifically stated, the relative steps, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of this application.

[0087] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0088] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A vehicle body attitude control method, characterized in that, include: The vehicle's attitude data during driving is acquired, and features are extracted from the attitude data to obtain fractal features; wherein, the fractal features are used to characterize the spatial complexity of the vehicle's attitude disturbance. The eddy current energy of the vehicle body attitude disturbance is determined based on the fractal characteristics. The vehicle body posture is adjusted based on the eddy current energy.

2. The method according to claim 1, characterized in that, Acquire attitude data during vehicle movement, and extract features from the attitude data to obtain fractal features, including: Acquire pitch, roll, and yaw angle data collected by the vehicle's body sensors during driving; The pitch, roll, and yaw angle data are decomposed into multiple scales to obtain time-frequency characteristic coefficients at multiple scales; Based on the time-frequency characteristic coefficients, the box counting method is used to calculate the coverage number corresponding to each scale, and the fractal characteristics are determined according to the power law relationship between the coverage number and the scale.

3. The method according to claim 2, characterized in that, The pitch, roll, and yaw angle data are decomposed into multiple scales to obtain time-frequency characteristic coefficients at multiple scales, including: Multi-resolution analysis is performed on the attitude data to separate the high-frequency components reflecting transient impacts and the low-frequency components reflecting trend changes step by step. The corresponding detail coefficients and approximation coefficients are extracted at each decomposition level and used as the time-frequency feature coefficients.

4. The method according to claim 1, characterized in that, Determining the eddy current energy of vehicle body attitude disturbance based on the fractal characteristics includes: The fractal feature is mapped to a vortex intensity parameter; wherein, the larger the fractal feature, the higher the weight of the corresponding vortex intensity parameter; The vortex intensity parameter is embedded into the vehicle body energy flow equation to construct a vortex model characterizing the energy distribution and transfer characteristics of the vehicle body. The energy circulation equation is derived based on the eddy current model, and the energy circulation equation is solved to obtain the eddy current feedback gain matrix, which serves as a quantitative representation of the eddy current energy.

5. The method according to claim 4, characterized in that, Mapping the fractal features to vortex intensity parameters includes: Establish a nonlinear mapping function between the fractal features and the vortex intensity parameters; The fractal features are converted into vortex intensity parameters at the corresponding scale using the nonlinear mapping function.

6. The method according to claim 4, characterized in that, The energy circulation equation is derived based on the eddy current model, and the energy circulation equation is solved to obtain the eddy current feedback gain matrix, including: Based on the eddy current model, a state-space expression containing eddy intensity parameters is constructed. Based on the state-space expression and energy conservation constraints, an energy circulation equation is generated; The energy circulation equation is stabilized and solved to output the eddy current feedback gain matrix for feedback control.

7. The method according to claim 4, characterized in that, Adjusting the vehicle's attitude based on the eddy current energy includes: Real-time calculation of the tracking error between the current attitude data and the preset vehicle attitude reference model; The tracking error is used to dynamically correct the eddy current feedback gain matrix used to characterize eddy current energy. The modified eddy current feedback gain matrix is ​​combined with the distribution characteristics of the eddy current energy to generate vehicle body attitude control commands, which drive the suspension or steering actuators to complete attitude adjustment.

8. The method according to claim 1, characterized in that, The method further includes: When the attitude data is drastically disturbed, abnormal patterns of vehicle body attitude are identified based on the abrupt change characteristics of the fractal features. Based on the abnormal mode, a preset emergency control strategy is activated to generate an emergency vehicle posture control command for emergency adjustment of the vehicle posture.

9. A vehicle body attitude control system, characterized in that, include: The feature extraction module is used to acquire attitude data during vehicle driving, and to perform multi-scale feature extraction on the attitude data to obtain fractal features; wherein, the fractal features are used to characterize the spatial complexity of vehicle attitude perturbation. The eddy current energy determination module is used to determine the eddy current energy of the vehicle body attitude disturbance based on the fractal characteristics. The vehicle body attitude adjustment module is used to adjust the vehicle body attitude based on the eddy current energy.

10. A vehicle body controller, characterized in that, The method includes a processor and a memory, the memory storing computer-executable instructions executable by the processor, the processor executing the computer-executable instructions to implement the method of any one of claims 1 to 8.

Images