Intelligent chassis stability control method based on low earth orbit satellite integrated space-ground network

Abstract

The application discloses a low-orbit satellite-based intelligent chassis stability control method for a ground-satellite integrated network, and calculates satellite data of a vehicle based on low-orbit satellite positioning; meanwhile, sensor data of the vehicle is collected. Based on lateral dynamic behavior of the vehicle, two degrees of freedom of lateral motion and yaw are considered, a state space expression of a vehicle dynamic model is constructed in combination with a dynamic equation of lateral displacement of the vehicle; an observation equation is established based on the satellite data and the sensor data of the vehicle; based on the vehicle dynamic model and the observation equation, a Kalman filtering method is used to dynamically correct lateral position, lateral velocity and yaw angular velocity of the vehicle, so that real-time and high-precision vehicle state estimation is obtained; based on the accurately estimated vehicle state, an MPC controller is designed, additional yaw torque required for realizing vehicle stability control is obtained through optimization calculation; and high-efficiency control on chassis stability of the vehicle is realized through optimization control on motor output.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of vehicle engineering and transportation engineering, and particularly relates to an intelligent chassis stability control method based on a low-orbit satellite space-ground integrated network, which makes up for the deficiency of vehicle perception and realizes high-performance intelligent chassis stability control. BACKGROUND

[0002] As the core of vehicle active safety, chassis stability control technology improves the safety and comfort of autonomous driving and the safety and maneuverability of manual driving through vehicle state perception and advanced control methods, ensuring the stability and controllability of vehicles under various driving conditions and reducing the risk of accidents.

[0003] Nowadays, with the development of automotive electronics technology, chassis stability control has gradually evolved from traditional ABS (anti-lock braking system), TCS (traction control system) and ESC (electronic stability control system) to a more intelligent and accurate control system. New technologies include active suspension systems that adjust suspension stiffness and damping in real time to adapt to different road conditions, dynamic torque distribution systems that optimize four-wheel drive power distribution to improve traction, and drive-by-wire chassis technology that significantly improves control response speed and accuracy. In addition, with the application of artificial intelligence and big data, the integration of the chassis domain with the intelligent driving domain, power domain and cabin domain has become an inevitable trend, which brings new opportunities for the chassis stability control system. Through multi-modal perception of driving environment and vehicle state, intelligent algorithms can realize real-time prediction of vehicle motion trajectory and accurate calculation of vehicle attitude, thereby actively adjusting braking, steering, suspension and power distribution, realizing longitudinal-horizontal-vertical multi-dimensional comprehensive coordination, and further improving the stability and driving safety of vehicles in complex road conditions.

[0004] However, the current technology route based on single vehicle intelligence is prone to perception failure of vehicle sensors in rain, snow, fog and other harsh weather conditions; vehicle sensors may not measure accurately due to obstruction in mixed traffic and complex urban road conditions; complex road conditions such as wet and slippery roads and potholes can affect the accuracy of vehicle speed and acceleration sensors; electromagnetic interference such as high-voltage power lines or communication signal towers can interfere with radar signals; in sharp turns, rapid acceleration or lateral tilting, the data of the inertial measurement unit may also be abnormal. In addition, extreme temperatures can reduce sensor performance, and sensor hardware aging, improper calibration or surface covered with mud and snow can also cause data distortion. These problems can seriously affect the normal operation of the vehicle's dynamic control and safety systems. SUMMARY

[0005] In order to solve the problems in the prior art, the application provides an intelligent chassis stability control method based on a low-orbit satellite space-ground integrated network, external correction is provided through satellite data, the influence of sensor data drift noise and the like on the system is reduced, high-frequency dynamics is provided through sensor data, the low-frequency defects of satellite data are compensated, the accuracy and reliability of state estimation are improved through fusion of satellite data and sensor data, and finally high-precision lateral velocity, yaw angular velocity and lateral position are output, more reliable input data for subsequent intelligent chassis lateral stability control is provided, and the intelligent driving demand in a complex environment is adapted.

[0006] The technical scheme adopted by the application is as follows:

[0007] The intelligent chassis stability control method based on the low-orbit satellite space-ground integrated network comprises the following steps:

[0008] Step 1: based on a low-orbit satellite positioning technology, satellite data of a vehicle is measured and calculated in real time, and sensor data of the vehicle is collected at the same time, the data types include lateral position, lateral velocity and yaw angular velocity of the vehicle;

[0009] Step 2: based on the lateral dynamic behavior of the vehicle, two degrees of freedom of lateral motion and yaw are considered, a state space expression of the vehicle dynamic model is constructed in combination with a dynamic equation of lateral displacement of the vehicle;

[0010] Step 3: an observation equation is established based on the satellite data and sensor data of the vehicle;

[0011] Step 4: based on the vehicle dynamic model and the observation equation, a Kalman filtering method is used for state prediction and update, the lateral position, lateral velocity and yaw angular velocity of the vehicle are dynamically corrected, and real-time and high-precision vehicle state estimation is obtained;

[0012] Step 5: based on the accurate vehicle state estimation, an MPC controller is designed, and an additional yaw moment required for realizing vehicle stability control is obtained through optimization calculation;

[0013] Step 6: in the MPC framework, a cost function and constraint condition are designed in combination with the lowest tire utilization principle, a torque distribution scheme of four-wheel motors is optimized and solved, motor output is optimized and controlled, and efficient control of the chassis stability of the vehicle is realized.

[0014] Further, the method for measuring and calculating the lateral position, lateral velocity and yaw angular velocity of the vehicle in step 1 based on the low-orbit satellite is as follows:

[0015]

[0016] wherein, represents the lateral position of the vehicle measured by the low-orbit satellite at the k th moment, denotes the lateral velocity estimated by the low earth orbit satellite at the kth time instant, denotes the lateral position of the vehicle measured by the satellite at the kth and (k-1)th time instant, respectively, and At denotes the time interval of the satellite data; denotes the yaw rate of the vehicle estimated by the satellite at the kth time instant, and denotes the heading angle of the vehicle measured by the satellite at the kth and (k-1)th time instant, respectively.

[0017] Further, the method of constructing the vehicle dynamics model in step 2 is as follows:

[0018] Step 2.1: Establish a two-degree-of-freedom vehicle dynamics model;

[0019] Step 2.2: Establish the lateral motion dynamics equation of the vehicle;

[0020] Step 2.3: Combine the two-degree-of-freedom vehicle dynamics model and the lateral motion dynamics equation, and write the vehicle dynamics model as a state space expression, denoted as:

[0021]

[0022] where x is a state vector composed of the lateral velocity, yaw rate, and lateral coordinate of the vehicle; A c is the system matrix; B c is the input matrix; and u is the control input.

[0023] Step 2.4: Discretize the state space model, denoted as:

[0024] x k+1 =A d x k +B d u k +w k

[0025] where x k and x k+1 denote the state quantity at the current time instant and the state quantity at the next time instant, respectively, A d denotes the discrete state matrix, B d denotes the discrete input matrix, u k denotes the control input, i.e., the front wheel steering angle, and w k is the process noise.

[0026] Further, the observation equation is used to describe the relationship between the observation data provided by the sensor and the satellite and the state vector, denoted as:

[0027] z k =Hx k +v k

[0028] where z k is the lateral velocity measured by low earth orbit satellites yaw rate ω sat , vehicle lateral position y sat and sensor measured vehicle lateral velocity yaw rate ω sen and lateral position y sen is the observation vector, denoted as H is the observation matrix, v k is the observation noise.

[0029] Further, the specific process of state prediction and update based on Kalman filtering in step 4 is as follows:

[0030] Step 4.1: Predict the state vector at the next time, denoted as:

[0031]

[0032] where denotes the corrected state vector at k-1 time, u k-1 denotes the input vector at k-1 time, the predicted state vector at k time; A d denotes the discrete state matrix, B d denotes the discrete input matrix;

[0033] Step 4.2: Predict the error covariance, denoted as:

[0034]

[0035] where denotes the predicted error covariance matrix at k time, P k-1 denotes the estimated error covariance matrix at k-1 time; Q k is the process noise covariance matrix;

[0036] Step 4.3: In the update stage, combine the observation values provided by the sensor and satellite to correct the predicted state vector and error covariance matrix, as follows:

[0037] First, calculate the Kalman gain:

[0038] K k = P k H T (HP k H T +R k ) -1

[0039] where P kdenotes the estimation error covariance matrix at time k, H is the observation matrix, and R k is the observation noise covariance matrix.

[0040] Next, the predicted state vector is corrected The corrected state vector is obtained

[0041]

[0042] wherein, denotes the corrected state vector at time k, denoted as The three elements in the vector are ω cor , y cor respectively denote the corrected lateral velocity, yaw rate and lateral position.

[0043] Finally, the covariance matrix is updated:

[0044]

[0045] wherein, I is the unit matrix, denotes the prediction error covariance matrix at time k.

[0046] Further, the method for designing the MPC controller in step 5 is as follows:

[0047] Step 5.1: Construct the spatial state equation of the system.

[0048] Step 5.2: Discretize the spatial state equation by using the Euler method.

[0049] Step 5.3: Combine the constraint conditions of the system on the target function to obtain the MPC cost function, which is expressed as follows:

[0050]

[0051] wherein, Q denotes the weight of the state vector, R denotes the weight of the output vector, N denotes the step length, y des (k+j) denotes the expected state vector, y(k+j) denotes the actual state vector, and u(k+j) denotes the input vector.

[0052] Further, the first time domain element obtained by the MPC controller optimization is the control output, denoted as respectively denote the optimal front wheel steering angle and additional yaw moment output.

[0053] Further, considering the lowest utilization rate of the tire, the target function is established as follows:

[0054]

[0055] wherein F xij , F zij represent the longitudinal force and vertical force of each wheel respectively, ij = fl, fr, rl, rr correspond to the left front wheel, right front wheel, left rear wheel, right rear wheel respectively, and μ represents the road adhesion coefficient.

[0056] Further, the constraint conditions include:

[0057] The longitudinal force of the tire needs to meet the demand of driving, denoted as:

[0058]

[0059] The longitudinal tire force meets the additional yaw moment demand, denoted as:

[0060]

[0061] wherein M is the vehicle body mass, F is the total longitudinal force, t f and t r are the widths of the front and rear axles of the vehicle respectively.

[0062] Further, based on the objective function and the constraint conditions, the additional yaw moment is determined Based on the additional yaw moment is distributed, and the additional yaw moment is converted into the driving force of the four wheels, denoted as:

[0063]

[0064] wherein, is the driving force of each wheel solved out, R eff is the effective rolling radius of the wheel.

[0065] The beneficial effects of the present application are:

[0066] (1) The present application provides a vehicle state correction framework that takes into account both local dynamic response (sensor data) and global consistency (satellite data), providing more comprehensive state information for vehicle control under complex conditions. Specifically, the present application uses low-orbit satellites as an external independent high-precision reference source, which can provide global consistent position information for vehicles, avoiding the drift, noise and other errors caused by internal vibration, temperature drift, interference or long-term operation of on-board sensors. At the same time, its global coverage capability enables it to provide stable position information in a variety of complex environments (such as urban canyons, mountainous areas and severe weather), making up for the functional deficiencies of sensors in special scenarios. On-board sensors have the advantage of high-frequency dynamic response, which can capture subtle dynamic changes of the vehicle in real time, especially when satellite data update frequency is low or signal is blocked, providing continuous short-term dynamic information support. Through the deep integration of the two, the global consistency of satellite data and the high dynamic sensitivity of sensor data complement each other, and this multi-source information synergy not only significantly improves the accuracy, real-time performance and robustness of vehicle state estimation, but also provides comprehensive technical support for intelligent driving, dynamic control and vehicle safety and stability under complex conditions.

[0067] (2) The present application fuses sensor data with high-frequency dynamic response capability but may drift and distort, and low-orbit satellite data that provides high-precision global reference information but has a low update frequency, dynamically adjusts the weight of the two, realizes the complementary advantages of the two, and finally outputs reliable vehicle state data, so that the vehicle stability controller realizes accurate dynamic control. BRIEF DESCRIPTION OF DRAWINGS

[0068] Figure 1 is an intelligent chassis overall stability control framework based on low-orbit satellite space-ground integrated network.

[0069] Figure 2 is a two-degree-of-freedom vehicle dynamics model. DETAILED DESCRIPTION

[0070] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application will be further described in detail below in combination with the drawings and examples. It should be understood that the specific examples described herein are only used to explain the present application and do not limit the present application.

[0071] Low-orbit satellites are becoming an important part of modern space technology due to their low latency, high positioning accuracy and economy, and are widely used in communication, navigation, earth observation and scientific research. Specifically, because of its proximity to the earth's surface, the signal propagation path is short, and its communication delay is generally less than 10 milliseconds, which is suitable for applications that require low latency; At the same time, the GNSS positioning of low-orbit satellites provides global position information, which can provide high-precision global position reference for vehicle motion; In addition, low-orbit satellites can also achieve continuous coverage of the earth's surface through constellation networking, which can provide reliable position information for vehicles, and the redundancy of the constellation makes the system more robust, even if a single satellite fails, other satellites can make up for the coverage.

[0072] Therefore, the present application constructs a low-orbit satellite earth-space integrated network enabled intelligent chassis stability control method, the overall framework of the control method of the present application is as shown in the figure, the control method comprises the following steps: Figure 1

[0073] Step 1: Based on low-orbit satellite positioning data, high-precision measurement of key dynamic parameters such as vehicle lateral position, lateral speed and yaw angular velocity is realized, which lays a foundation for subsequent modeling and control. The specific measurement process is as follows:

[0074] Low-orbit satellites can directly measure the lateral position y sat of the vehicle in the global coordinate system through GNSS positioning capability. Assuming that the lateral position is accurately measured by the satellite, the formula can be written as:

[0075]

[0076] wherein, y k represents the lateral position of the vehicle measured by the low-orbit satellite at the kth moment.

[0077] The satellite measures the lateral position y sat of the vehicle in the global coordinate system, and estimates the lateral speed by time difference combined with the time step Δt, and the lateral speed is denoted as:

[0078]

[0079] wherein, v k represents the lateral speed estimated by the low-orbit satellite at the kth moment, y k represents the lateral position of the vehicle measured by the satellite at the kth moment, y k-1 represents the lateral position of the vehicle measured by the satellite at the k-1th moment, and Δt represents the time interval of the satellite data.

[0080] The yaw angular velocity is the rate of change of the heading angle with respect to time, and the yaw angular velocity at the kth moment is estimated by the time difference formula, denoted as: ​

[0081]

[0082] where, denotes the vehicle yaw rate estimated by the satellite at the kth time instant, and denote the vehicle heading angle measured by the satellite at the kth and (k-1)th time instants, respectively, and Δt denotes the time interval of the satellite data.

[0083] Step 2: Based on the vehicle lateral dynamic behavior, considering both lateral and yaw motion degrees of freedom, and combining the lateral displacement dynamics equation, the state space expression of the vehicle dynamics model is constructed. The vehicle dynamics model is discretized to meet the needs of subsequent control algorithm design. The specific process is as follows:

[0084] Step 2.1: A two-degree-of-freedom vehicle dynamics model considering both lateral and yaw motion degrees of freedom is established, as shown in Figure 2

[0085]

[0086] where, M is the vehicle mass, v x is the longitudinal speed of the vehicle; v y is the lateral speed of the vehicle; is the lateral acceleration of the vehicle; ω is the yaw rate of the vehicle; is the yaw angular acceleration of the vehicle; a and b are the distances between the front and rear axles and the center of mass, respectively; I z is the moment of inertia of the vehicle about the z-axis; ΔM z is the additional yaw moment generated by the wheel longitudinal force; k1 and k2 are the side stiffness of the front and rear wheels, respectively; δ f is the front wheel steering angle.

[0087] Step 2.2: The lateral displacement dynamics equation is established:

[0088]

[0089] where, is the derivative of the lateral displacement.

[0090] Step 2.3: Combining the two-degree-of-freedom vehicle dynamics model and the lateral displacement dynamics equation, the vehicle dynamics model is written as a state space expression, denoted as:

[0091]

[0092] where, is the vehicle dynamics model; x is the state vector composed of the lateral speed, yaw rate, and lateral coordinate of the vehicle, denoted as x = [v​y ωy] T 。A c is the system matrix; B c is the input matrix; u is the control input; and x

[0093]

[0094] Step 2.4: Discretize the state space model, in the form of:

[0095] x k+1 = A d x k + B d u k + w k (7)

[0096] where x k and x k+1 represent the state quantity at the current time and the state quantity at the next time, respectively, A d represents the discrete state matrix, B d represents the discrete input matrix, u k represents the control input, i.e., the front wheel steering angle, w k is the process noise, and represents the dynamic influence of the model.

[0097] Step 3: Establish the observation equation, which describes the relationship between the observation data provided by the sensors and the satellite and the state vector, to provide a data association model for state estimation and error correction.

[0098] The observation equation is represented as follows:

[0099] z k = Hx k + v k (8)

[0100] where the observation vector ω sen , y sen represent the data measured by the vehicle sensors for the vehicle lateral speed, yaw rate, and lateral position, ω sat , y sat represent the real-time calculated vehicle lateral speed, yaw rate, and lateral position by the low-orbit satellite; H is the observation matrix, specifically v k is the observation noise.

[0101] Step 4: State prediction and update based on vehicle dynamics model and observation data using Kalman filter method. In the Kalman filter framework, the lateral position, lateral velocity and yaw rate of the vehicle are dynamically corrected to obtain real-time and high-precision vehicle state estimation.

[0102] The specific process of state prediction and update based on Kalman filter is as follows:

[0103] Step 4.1: First, predict the state vector at the next time:

[0104]

[0105] where, is the corrected state vector at the k-1 time, u k-1 is the input vector at the k-1 time, is the predicted state vector at the k time, calculated by the dynamics model, which represents the state prediction result without combining the observation value.

[0106] Step 4.2: Second, predict the error covariance, which represents the uncertainty of the predicted state:

[0107]

[0108] where, is the predicted error covariance matrix at the k time, P k-1 is the estimated error covariance matrix at the k-1 time; Q k is the process noise covariance matrix, which represents the unmodeled error in the model.

[0109] Step 4.3: In the update stage, combine the observation values provided by the sensors and satellites to correct the predicted state vector and error covariance matrix. Specifically as follows:

[0110] First, calculate the Kalman gain, which determines the weight of the predicted value and the observation value in the correction:

[0111]

[0112] where, K k is the Kalman gain matrix, P k is the estimated error covariance matrix at the k time, H is the observation matrix, R k is the observation noise covariance matrix, which describes the measurement error of the sensor and satellite data.

[0113] Next, correct the predicted state vector to obtain the corrected state vector

[0114]

[0115] wherein, denotes the corrected state vector at the kth moment, denoted as The three elements in the vector ω cor , y cor respectively represent the corrected lateral velocity, yaw rate and lateral position, z k is the observation vector at the kth moment, containing the observation values provided by sensors and satellites,

[0116] Finally update the covariance matrix:

[0117]

[0118] wherein, I is the unit matrix, denotes the prediction error covariance matrix at the kth moment.

[0119] Step 5: Based on the accurate estimation of the vehicle state, design a model predictive controller to obtain the additional yaw moment required to realize vehicle stability control through optimization calculation, so as to improve the stability and handling of the vehicle. The specific process is as follows:

[0120] After obtaining the corrected state of the vehicle ω cor and y cor , the following designs an MPC controller to calculate the real-time required front wheel steering angle and additional yaw moment.

[0121] First, the spatial state equation of the system is expressed as:

[0122]

[0123] wherein, x cor is the state vector, is the derivative of the state vector, y is the control output, u is the control input, A c is the state matrix, B c is the input matrix, C c is the output matrix, according to formula (6), x cor is the state vector composed of vehicle lateral velocity, yaw rate and lateral coordinate, denoted as

[0124] Discretize the above spatial state equation using Euler method, ignore high order terms, and assume that the matrix A is time invariant:

[0125]

[0126] wherein, x(k+1)cor is the discretized next time state variable; x(k) cor is the discretized current time state variable; matrix A = I + A c T s is the discretized current time input matrix; matrix B = B c T s , T s is the sampling time, matrix y(k) is the output vector, denoted as y(k) = [v y ω y] T , u(k) is the input vector, denoted as u(k) = [δ f ΔM z ] T

[0127] After continuous iteration, the following matrix form can be obtained:

[0128] Y(k) = ψx(k) cor + θU(k) (16)

[0129] where Y(k) is a matrix composed of N output vectors, denoted as

[0130] ψ is the state matrix of the system, denoted as

[0131] θ is the input matrix of the system, denoted as

[0132] U(k) is a matrix composed of N input vectors, denoted as

[0133] The goal of the MPC controller is to optimize the deviation between the state variable and the expected value within the prediction time domain, while minimizing the amplitude of the controller output as much as possible.

[0134] Combined with the constraint conditions of the system on the objective function, the MPC cost function can be obtained in the following form:

[0135]

[0136] where Q represents the weight of the state vector, R represents the weight of the output vector, N represents the step, y des (k+j) represents the expected state vector, y(k+j) represents the actual state vector, and u(k+j) represents the input vector.

[0137] Finally, it is converted into a quadratic programming problem, and the first time domain element obtained by solving is taken as the control output, that is The optimal front wheel steering angle and additional yaw moment output are represented, in which the additional yaw moment is a key adjustment parameter of vehicle chassis stability control, and plays a core role in dynamic performance optimization and driving safety improvement.

[0138] Step 6: In the MPC framework, a cost function and constraint condition are designed in combination with the principle of minimum tire utilization, and a torque distribution scheme of four-wheel motors is optimized and solved. By optimizing the motor output, efficient control of vehicle chassis stability is realized.

[0139] The following distribution of additional yaw moment needs to convert the additional yaw moment into the driving force of the four wheels, which requires torque distribution.

[0140] Considering the characteristics of the tire, considering the minimum tire utilization, the objective function is established as:

[0141]

[0142] Where F xij , F zij represent the longitudinal force and vertical force of each wheel, such as the left front wheel, the right front wheel, the left rear wheel, and the right rear wheel, and μ represents the road adhesion coefficient.

[0143] First, the longitudinal force of the tire needs to meet the driving demand:

[0144]

[0145] Where M is the vehicle mass, and the total longitudinal force F is the total longitudinal force to ensure the current vehicle longitudinal speed, which can be obtained according to the proportional control:

[0146] F = K p (v xdes -v x ) (20)

[0147] Where K p represents the proportional coefficient, v xdes represents the desired vehicle speed, and v x represents the actual vehicle speed.

[0148] Second, the longitudinal tire force meets the additional yaw moment demand:

[0149]

[0150] Where t f and t r are the widths of the front and rear axles of the vehicle.

[0151] Based on the objective function and constraint condition, the torque distribution of the four-wheel motors is determined Based on An additional yaw moment distribution is performed, and finally the torque of each motor is obtained:

[0152]

[0153] wherein, is the driving force of each wheel solved, R eff is the effective rolling radius of the wheel, wherein ij = fl, fr, rl, rr correspond to the four wheels respectively. The output torque of each motor solved completes the stability control of the vehicle chassis.

[0154] Kalman filtering is a recursive algorithm widely used in dynamic system state estimation. It combines the dynamic model of the system with the actual measurement data to achieve optimal prediction and correction of the state in the presence of noise. In the prediction phase, Kalman filtering predicts the current state and its uncertainty based on the state and control input at the previous time, using the dynamic equation of the system. Then, in the update phase, the algorithm fuses the measurement provided by the sensor with the prediction result, adjusts the state estimation and dynamically updates the confidence of the system. This method can track the dynamic changes of the system in real time, thereby improving the accuracy of the prediction and the robustness of the system.

[0155] The present application can provide reliable input data for the vehicle dynamics model based on multi-source data fusion and the precise state quantity of the vehicle after filtering correction, so that the control system can perceive the actual motion state of the vehicle in real time, thereby more accurately calculating the error of the vehicle deviating from the ideal state. In addition, this high-precision state information can significantly improve the response speed and robustness of the chassis stability control, so that the controller can quickly and accurately generate control instructions when facing complex road conditions, extreme working conditions (such as sharp turns, wet road surfaces) or sudden external disturbances (such as crosswinds, collision risks), optimizing the vehicle's yaw moment and lateral dynamic performance. At the same time, the precise vehicle state quantity also provides a reliable basis for the control algorithm (such as model predictive control), which helps to achieve optimized torque distribution, minimize tire utilization and wear, and ensure the best balance between stability, handling and economy, thereby further improving the driving safety of the vehicle.

[0156] The above examples are only used to illustrate the design idea and characteristics of the present application, and the purpose is to enable those skilled in the art to understand the content of the present application and implement it, and the protection scope of the present application is not limited to the above examples. Therefore, any equivalent changes or modifications made in accordance with the principles and design ideas disclosed by the present application are within the protection scope of the present application.

Claims

1. A method for intelligent chassis stability control based on a low-orbit satellite-space-ground integrated network, characterized in that, Includes the following steps: Step 1: Based on low-orbit satellite positioning technology, calculate the vehicle's satellite data in real time; at the same time, collect the vehicle's sensor data, including the vehicle's lateral position, lateral velocity, and yaw rate. Step 2: Based on the vehicle's lateral dynamics, considering the two degrees of freedom of lateral motion and yaw, and combining the dynamic equations of the vehicle's lateral displacement, construct the state-space expression of the vehicle dynamics model; Step 3: Establish observation equations based on satellite and sensor data of the vehicle; Step 4: Based on the vehicle dynamics model and observation equations, the Kalman filter method is used for state prediction and updating; the lateral position, lateral velocity and yaw rate of the vehicle are dynamically corrected to obtain real-time and high-precision vehicle state estimates; Step 5: Based on the accurately estimated vehicle state, design the MPC controller and obtain the additional yaw moment required to achieve vehicle stability control through optimization calculation; Step 6: Under the MPC framework, and in accordance with the principle of minimizing tire utilization, design the cost function and constraints, and optimize the torque distribution scheme of the four-wheel motors. By optimizing the control motor output, efficient control of vehicle chassis stability can be achieved.

2. The intelligent chassis stability control method based on a low-orbit satellite-space-ground integrated network according to claim 1, characterized in that, The method for calculating the vehicle's lateral position, lateral velocity, and yaw rate based on low-orbit satellites in step 1 is as follows: Directly measured lateral position of the vehicle in, This represents the lateral position of the vehicle measured by a low-orbit satellite at time k. This represents the lateral velocity estimated by the low-Earth orbit satellite at time k. Δt represents the lateral position of the vehicle measured by the satellite at time k and time (k-1), respectively, and Δt represents the time interval of the satellite data. This represents the vehicle's yaw rate estimated by the satellite at time k. and These represent the vehicle heading angles measured by the satellite at time k and time k-1, respectively.

3. The intelligent chassis stability control method based on a low-orbit satellite-space-ground integrated network according to claim 1, characterized in that, The method for constructing the vehicle dynamics model in step 2 is as follows: Step 2.1: Establish a two-degree-of-freedom vehicle dynamics model; Step 2.2: Establish the vehicle lateral displacement dynamic equations; Step 2.3: Combining the two-degree-of-freedom vehicle dynamics model and the lateral displacement dynamics equations, the vehicle dynamics model is written as a state-space expression, denoted as: Where x is a state vector consisting of the vehicle's lateral velocity, yaw rate, and x-coordinate; A c B is the system matrix; c The input matrix is ​​u; the control input is u. Step 2.4: Discretize the state-space model, denoted as: x k+1 =A d x k +B d u k +w k Where, x k x k+1 Let A represent the state quantity at the current time step and the state quantity at the next time step, respectively. d B represents the discrete state matrix. d Represents the discrete input matrix, u k This indicates the control input, specifically the front wheel steering angle, w. k This is process noise.

4. The intelligent chassis stability control method based on a low-orbit satellite-space-ground integrated network according to claim 1, characterized in that, The observation equation is used to describe the relationship between the observation data provided by the sensors and satellites and the state vector, denoted as: z k =Hx k +v k Among them, z k The lateral velocity is calculated by low-Earth orbit satellites. yaw rate ω sat , vehicle lateral position y sat and the vehicle's lateral speed measured by sensors yaw rate ω sen and horizontal position y sen The constructed observation vector is denoted as H is the observation matrix, v k To observe noise.

5. The intelligent chassis stability control method based on a low-orbit satellite-space-ground integrated network according to claim 1, characterized in that, The specific process of state prediction and update based on Kalman filtering in step 4 is as follows: Step 4.1: Predict the state vector at the next time step, denoted as: in, Let u represent the corrected state vector at time k-1. k-1 This represents the input vector at time k-1. The predicted state vector at time k; A d B represents the discrete state matrix. d Represents a discrete input matrix; Step 4.2: Prediction error covariance, denoted as: in, Let P represent the prediction error covariance matrix at time k. k-1 Q represents the estimation error covariance matrix at time k-1; k The process noise covariance matrix; Step 4.3: In the update phase, combining the observations provided by the sensors and satellites, the predicted state vector and error covariance matrix are corrected as follows: First, calculate the Kalman gain: K k =P k H T (HP k H T +R k ) -1 Among them, P k Let H be the estimation error covariance matrix at time k, H be the observation matrix, and R be the covariance matrix at time k. k To observe the noise covariance matrix; Next, the predicted state vector is corrected. Obtain the corrected state vector in, Let the corrected state vector at time k be denoted as . The three elements in the vector ω cor y cor These represent the corrected lateral velocity, yaw rate, and lateral position, respectively. Finally, update the covariance matrix: Where I is the identity matrix, Let represent the prediction error covariance matrix at time k.

6. The intelligent chassis stability control method based on a low-orbit satellite-space-ground integrated network according to claim 1, characterized in that, The method for designing the MPC controller in step 5 is as follows: Step 5.1: Construct the spatial state equations of the system; Step 5.2: Discretize the above spatial state equations using the Euler method; Step 5.3: Combining the system's constraints on the objective function, the MPC cost function is obtained, as follows: Where Q represents the weights of the state vector, R represents the weights of the output vector, N represents the step size, and y des (k+j) represents the desired state vector, y(k+j) represents the actual state vector, and u(k+j) represents the input vector.

7. The intelligent chassis stability control method based on a low-orbit satellite-space-ground integrated network according to claim 6, characterized in that, The first time-domain element obtained through optimization using the MPC controller is the control output, denoted as... These represent the optimal front wheel steering angle and the additional yaw moment output, respectively.

8. The intelligent chassis stability control method based on a low-orbit satellite-space-ground integrated network according to claim 7, characterized in that, Considering minimizing tire utilization, the objective function is established as follows: Among them, F xij F zij These represent the longitudinal and vertical forces of each wheel, respectively. ij = fl, fr, rl, rr correspond to the left front wheel, right front wheel, left rear wheel, and right rear wheel, respectively, and μ represents the road adhesion coefficient.

9. The intelligent chassis stability control method based on a low-orbit satellite-space-ground integrated network according to claim 8, characterized in that, The constraints include: The longitudinal force of the tire needs to meet the driving requirements, denoted as: The longitudinal tire force meets the additional yaw moment requirement, denoted as: Where M is the vehicle body mass, F is the total longitudinal force, and t f and t r These refer to the widths of the front and rear axles of the vehicle, respectively.

10. The intelligent chassis stability control method based on a low-orbit satellite-space-ground integrated network according to claim 9, characterized in that, Determined based on objective function and constraints based on The additional yaw moment is distributed and converted into driving force for the four wheels, denoted as: in, To calculate the driving force of each wheel, R eff This is the effective rolling radius of the wheel.

Images