A vehicle control method and system based on a central computing platform and regional controllers

Abstract

The application relates to a whole vehicle control method and system based on a central computing platform and a regional controller, the method comprising the following steps: collecting sensor data of a target vehicle through the regional controller, and uploading the sensor data to the central computing platform; the central computing platform analyzes and optimizes the sensor data based on a physical model and an optimization algorithm, and obtains optimal brake pressures of wheel cylinders, wherein the physical model comprises influences of the sensor data on a tire-ground friction coefficient and influences of weight transfer on loads of each tire; corresponding control instructions are generated according to the optimal brake pressures of the wheel cylinders, and are sent to the regional controller to control the target vehicle, and execution state feedback is obtained and fed back to the regional controller. The application realizes centralized control and dynamic scheduling of whole vehicle functions, and reduces complexity and integration difficulty of the system.

Description

Technical Field

[0001] This invention relates to the field of intelligent connected vehicles and vehicle control technology, and in particular to a vehicle control method and system based on a central computing platform and a regional controller. Background Technology

[0002] As the automotive industry moves towards electrification, intelligentization, and connectivity, traditional distributed electronic and electrical architectures can no longer meet the ever-increasing functional demands. In existing technologies, vehicle control systems typically employ a distributed architecture, with each electronic control unit (ECU) operating independently, leading to the following specific problems:

[0003] High system complexity: In traditional distributed architectures, each functional module (such as engine control, body control, chassis control, etc.) requires an independent ECU, resulting in a large number of ECUs (usually exceeding 100) and high system complexity. For example, patent CN102529905A proposes a vehicle control system based on a distributed architecture, but it has a large number of ECUs and is difficult to integrate.

[0004] Long development cycle and high cost: Each ECU needs to be developed, tested and verified independently, resulting in a long development cycle and high cost. For example, patent CN110155001A proposes a vehicle control system based on a central computing unit, but it does not solve the design problem of the regional controller, resulting in still high development costs.

[0005] Difficulty in Achieving Cross-Domain Collaborative Control: In traditional distributed architectures, ECUs communicate with each other via low-speed buses such as CAN and LIN, resulting in low communication efficiency and making it difficult to achieve cross-domain collaborative control. For example, patent CN112487123A proposes an SOA-based automotive software architecture, but it is not integrated with a centralized architecture, making it difficult to achieve cross-domain collaborative control.

[0006] High coupling between software and hardware: In traditional architectures, software and hardware are highly coupled, making it difficult to support rapid iteration and functional upgrades. For example, patent CN111245825A proposes a remote diagnostic system based on vehicle-cloud collaboration, but it does not solve the problem of decoupling software and hardware.

[0007] Insufficient communication bandwidth: Traditional vehicle networks (such as CAN and LIN) have limited bandwidth, making it difficult to meet the large data transmission requirements of applications such as autonomous driving. For example, patent CN110155001A proposes a vehicle control system based on a central computing unit, but it does not solve the problem of insufficient communication bandwidth.

[0008] Inadequate functional safety and information security design: In traditional architectures, functional safety and information security design are usually carried out independently, failing to achieve integrated safety design. For example, patent CN112487123A proposes an SOA-based automotive software architecture, but it does not address the integration of functional safety and information security.

[0009] As can be seen from the above, existing technologies have not solved the problems of high system complexity, long development cycle, and high cost of distributed architectures, nor have they proposed effective solutions to achieve cross-domain collaborative control, decoupling of software and hardware, and combination of functional safety and information security. As a result, they cannot play their corresponding role in scenarios that require rapid response, such as the pressure increase and decrease function in tire braking.

[0010] When tire pressure is high, tire deformation is smaller, allowing for a faster response to braking commands and more rapid transmission of braking force to the ground, thus shortening braking distance and improving vehicle braking efficiency. Simultaneously, with increased vehicle load or on slippery surfaces, appropriate inflation reduces the tire's contact area with the ground, increasing pressure per unit area, improving tire grip and adhesion, and enhancing vehicle stability during braking. Conversely, during braking, appropriate depressurization increases the tire's contact area with the ground, increasing friction and improving braking performance. This is especially true on soft or uneven surfaces, where depressurized tires adapt better to the road surface, increasing the actual contact area and enhancing grip. However, the independent operation of each ECU in a distributed architecture makes rapid tire depressurization difficult to coordinate during complex calculations, hindering timely and efficient tire pressure adjustments. Summary of the Invention

[0011] The purpose of this invention is to provide a vehicle control method and system based on a central computing platform and a regional controller. Through centralized, software-based, and service-oriented design, it solves the problems of high system complexity, long development cycle, high cost, and difficulty in achieving cross-domain collaborative control in traditional distributed architectures.

[0012] To achieve the above objectives, the present invention provides the following solution:

[0013] A vehicle control method based on a central computing platform and a regional controller includes:

[0014] Sensor data of the target vehicle is collected by the regional controller and uploaded to the central computing platform.

[0015] The central computing platform analyzes and optimizes the sensor data based on physical models and optimization algorithms to obtain the optimal braking pressure for each wheel cylinder. The physical model includes the influence of the sensor data on the coefficient of friction between the tire and the ground, as well as the influence of weight transfer on the load of each tire.

[0016] Based on the optimal braking pressure of each wheel cylinder, corresponding control commands are generated and sent to the area controller to control the target vehicle, and the execution status is obtained and fed back to the area controller.

[0017] Optionally, the sensor data collected from the target vehicle includes:

[0018] The system collects the vehicle's speed information using a speed sensor, tire temperature information using a temperature sensor, vehicle acceleration information using an acceleration sensor, and vehicle steering angle information using a steering angle sensor.

[0019] Optionally, the physical model includes: a tire-ground friction model, a load distribution model, and a lateral load transfer model;

[0020] The tire-ground friction model is as follows:

[0021] ;

[0022] The load allocation model is as follows:

[0023] ;

[0024] The lateral load transfer is as follows:

[0025] ;

[0026] Where μ is the coefficient of friction, T is the tire temperature, λ is the slip ratio, μ0 is the maximum coefficient of friction at room temperature, and k T T is the temperature decay coefficient. opt For the target temperature, λ c Here, m is the critical slip ratio, L is the vehicle weight, h is the wheelbase, lr is the distance from the rear axle to the center of gravity, and a is the center of gravity height. x For longitudinal acceleration, F z,front For the front wheel vertical load, The load on the inner and outer tires during steering, t is the track width, and a is the load on the inner and outer tires. y This is lateral acceleration.

[0027] Optionally, the central computing platform analyzes and optimizes the sensor data based on physical models and optimization algorithms, including:

[0028] The target braking force is calculated based on the vehicle weight and deceleration target of the target vehicle, and the single-wheel braking force is distributed according to the physical model to obtain the single-wheel target braking force of each wheel.

[0029] Based on the slip ratio, tire temperature, and steering angle, a comprehensive pressure control constraint including safety boundaries and compensation mechanisms is generated;

[0030] The Model Predictive Control (MPC) algorithm is used to establish a predictive model that includes actuator mapping. The single-wheel target braking force is used as the tracking target, and the comprehensive pressure control constraint is used as the boundary condition. The wheel cylinder pressure sequence in the future time domain is solved by rolling optimization.

[0031] Extract the current time value from the wheel cylinder pressure sequence to obtain the optimal braking pressure for each wheel cylinder.

[0032] Optionally, the target braking force for each wheel is:

[0033] ;

[0034] in, The target braking force for the i-th wheel, Let be the vertical load on the i-th wheel. To calculate the vertical load of the j-th wheel in the summation formula, For target braking force, μ i (T,λ) and μ j (T,λ) represent the real-time friction coefficients of the i-th and j-th tires at temperature T and slip ratio λ, respectively.

[0035] Optionally, generating a comprehensive pressure control constraint that includes safety boundaries and compensation mechanisms based on slip ratio, tire temperature, and steering angle includes: calculating slip ratio deviation based on the real-time slip ratio and the target slip ratio;

[0036] The pressure correction is calculated using a PID controller based on the slip ratio deviation.

[0037] The dynamic target pressure is obtained based on the baseline reference pressure and the pressure correction amount.

[0038] If tire temperature T > maximum temperature T max Temperature compensation is then performed, limiting the upper limit of braking pressure to obtain temperature safety boundary constraints.

[0039] Based on the steering angle, steering compensation coefficient, and baseline outer front wheel braking pressure, calculate the corrected optimal braking pressure for the outer front wheel:

[0040] The comprehensive pressure control constraint is obtained based on the dynamic target pressure, the temperature safety boundary constraint, and the corrected optimal braking pressure of the outer front wheel.

[0041] Optionally, the prediction model including the actuator mapping is:

[0042] ;

[0043] in, Indicates when predicting pressure is applied At that time, the expected braking force calculated by the model; A caliper Let f(T,λ) represent the effective area of ​​the caliper, and let f(T,λ) represent the temperature T-slip ratio λ compensation function. For the steering angle, a y This is lateral acceleration.

[0044] The present invention also provides a vehicle control system based on a central computing platform and a regional controller, comprising:

[0045] The data acquisition and uploading module is used to acquire sensor data of the target vehicle through the area controller and upload the sensor data to the central computing platform;

[0046] The decision planning and instruction generation module is used by the central computing platform to analyze and optimize the sensor data based on the physical model and optimization algorithm to obtain the optimal braking pressure of each wheel cylinder. The physical model includes the influence of the sensor data on the coefficient of friction between the tire and the ground, as well as the influence of weight transfer on the load of each tire.

[0047] The instruction issuance and execution module is used to generate corresponding control instructions based on the optimal braking pressure of each wheel cylinder, and issue them to the area controller to control the target vehicle, and obtain the execution status feedback to the area controller.

[0048] Optionally, the system further includes:

[0049] The cross-domain collaborative control module is used by the central computing platform to call services from different domains through SOA architecture to achieve collaborative control;

[0050] The vehicle-cloud collaboration module is used for communication between the target vehicle and the cloud server;

[0051] The functional safety and information security module is used to integrate encryption and authentication information security technologies to ensure data transmission security.

[0052] The beneficial effects of this invention are as follows: This invention provides a vehicle control system and method based on a central computing platform and a regional controller. Through centralized, software-based, and service-oriented design, it solves the problems of high system complexity, long development cycle, high cost, and difficulty in achieving cross-domain collaborative control in traditional distributed architectures. The system includes a central computing platform, a regional controller, an SOA architecture, an AUTOSAR Adaptive platform, a vehicle-cloud collaboration module, and a functional safety and information security module. It has advantages such as simplified system architecture, improved software development and iteration efficiency, realization of cross-domain collaborative control, and support for future functional expansion. Attached Figure Description

[0053] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 This is a flowchart of a vehicle control method based on a central computing platform and a regional controller according to an embodiment of the present invention;

[0055] Figure 2 This is a framework diagram of a vehicle control system based on a central computing platform and a regional controller, according to an embodiment of the present invention.

[0056] Figure 3 This is a comparison chart of vehicle trajectory tracking capabilities under emergency braking conditions on curves, according to an embodiment of the present invention. Detailed Implementation

[0057] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0058] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0059] like Figure 1 As shown, this embodiment proposes a vehicle control method based on a central computing platform and a regional controller, including:

[0060] Sensor data from the target vehicle is collected by the regional controller and uploaded to the central computing platform.

[0061] The central computing platform analyzes and optimizes sensor data based on physical models and optimization algorithms to obtain the optimal braking pressure for each wheel cylinder. The physical model includes the influence of sensor data on the coefficient of friction between the tire and the ground, as well as the influence of weight transfer on the load of each tire.

[0062] Based on the optimal braking pressure of each wheel cylinder, corresponding control commands are generated and sent to the area controller to control the target vehicle, and the execution status is obtained and fed back to the area controller.

[0063] Furthermore, the collection of sensor data from the target vehicle includes:

[0064] The system collects the vehicle's speed information using a speed sensor, tire temperature information using a temperature sensor, vehicle acceleration information using an acceleration sensor, and vehicle steering angle information using a steering angle sensor.

[0065] Specifically, data collection and uploading:

[0066] Area Controller: The area controller is responsible for collecting data from various sensors in the vehicle, including but not limited to speed sensors, temperature sensors, and pressure sensors. These sensors are distributed in different parts of the vehicle, such as the wheels, engine, and battery. The area controller includes: speed sensors, temperature sensors, acceleration sensors, and steering angle sensors.

[0067] Speed ​​sensor: Monitors the rotational speed of the front wheels in real time, providing information on the vehicle's speed.

[0068] Temperature sensor: Monitors tire temperature.

[0069] Accelerometer sensor: Monitors the vehicle's acceleration and provides dynamic information about the vehicle.

[0070] Steering angle sensor: Monitors the vehicle's steering angle and provides information on the vehicle's direction of travel.

[0071] The central computing platform includes:

[0072] Data fusion: Time synchronization and state alignment of heterogeneous sensor data from different regional controllers are performed to construct a comprehensive motion state vector of the vehicle at the same moment.

[0073] Intelligent Decision Engine: Based on machine learning algorithms, it analyzes the vehicle's comprehensive status information in real time to generate the optimal brake pressure adjustment command. Specifically, based on a physics model, it considers the effects of speed, temperature, acceleration, and steering on the tire-road friction coefficient, as well as the impact of weight transfer on the load of each tire. It calculates the maximum available braking force for each tire based on the load and friction coefficient, distributes the braking force to maximize the total braking force, and avoids locking or overheating of individual tires. The pressure is adjusted in real time, possibly through table lookup or online calculation, combined with feedback control to correct errors, thereby generating the optimal brake pressure adjustment command, which corresponds to the steps in S2.

[0074] Service orchestration: Call the services of the area controller to achieve precise control of the front wheel braking pressure.

[0075] The central computing platform employs high-performance computing units such as multi-core processors, GPUs, and AI acceleration chips to support core applications such as autonomous driving and smart cockpits. It communicates with area controllers via high-speed Ethernet, supporting low-latency, high-bandwidth data transmission. Each area controller is equipped with a microcontroller and communication module, responsible for sensor data acquisition and actuator control within its area, supporting multiple communication protocols such as CAN, LIN, and Ethernet.

[0076] Furthermore, the physical models include: tire-ground friction model, load distribution model, and lateral load transfer model;

[0077] The tire-ground friction model is as follows:

[0078] ;

[0079] The load distribution model is as follows:

[0080] ;

[0081] Lateral load transfer is as follows:

[0082] ;

[0083] Where μ is the coefficient of friction, T is the tire temperature, λ is the slip ratio, μ0 is the maximum coefficient of friction at room temperature, and k T T is the temperature decay coefficient. opt For the target temperature, λ c Here, m is the critical slip ratio, L is the vehicle weight, h is the wheelbase, lr is the distance from the rear axle to the center of gravity, and a is the center of gravity height. x For longitudinal acceleration, F z,front For the front wheel vertical load, The load on the inner and outer tires during steering, t is the track width, and a is the load on the inner and outer tires. y This is lateral acceleration.

[0084] Furthermore, the central computing platform analyzes and optimizes sensor data based on physical models and optimization algorithms, including:

[0085] Calculate the target braking force based on the target vehicle's weight and deceleration target, and distribute the single-wheel braking force according to the target braking force and the physical model;

[0086] The target braking force is calculated based on the vehicle weight and deceleration target of the target vehicle, and the single-wheel braking force is distributed according to the physical model to obtain the single-wheel target braking force of each wheel.

[0087] Based on the slip ratio, tire temperature, and steering angle, a comprehensive pressure control constraint including safety boundaries and compensation mechanisms is generated;

[0088] The Model Predictive Control (MPC) algorithm is used to establish a predictive model that includes actuator mapping. The single-wheel target braking force is used as the tracking target, and the comprehensive pressure control constraint is used as the boundary condition. The wheel cylinder pressure sequence in the future time domain is solved by rolling optimization.

[0089] Extract the current time value from the wheel cylinder pressure sequence to obtain the optimal braking pressure for each wheel cylinder.

[0090] Furthermore, the target braking force for each wheel includes:

[0091] ;

[0092] in, The target braking force for the i-th wheel, Let be the vertical load on the i-th wheel. To calculate the vertical load of the j-th wheel in the summation formula, For target braking force, μ i (T,λ) and μ j (T,λ) represent the real-time friction coefficients of the i-th and j-th tires at temperature T and slip ratio λ, respectively.

[0093] Furthermore, based on the slip ratio, tire temperature, and steering angle, generating a comprehensive pressure control constraint that includes safety boundaries and compensation mechanisms includes: calculating the slip ratio deviation based on the real-time slip ratio and the target slip ratio;

[0094] The pressure correction is calculated using a PID controller based on the slip ratio deviation.

[0095] The dynamic target pressure is obtained based on the baseline reference pressure and the pressure correction amount.

[0096] If tire temperature T > maximum temperature T max Temperature compensation is then performed, limiting the upper limit of braking pressure to obtain temperature safety boundary constraints.

[0097] Based on the steering angle, steering compensation coefficient, and baseline outer front wheel braking pressure, calculate the corrected optimal braking pressure for the outer front wheel:

[0098] The comprehensive pressure control constraint is obtained based on the dynamic target pressure, the temperature safety boundary constraint, and the corrected optimal braking pressure of the outer front wheel.

[0099] Furthermore, the prediction model including the actuator mapping is:

[0100] ;

[0101] in, Indicates when predicting pressure is applied At that time, the expected braking force calculated by the model; Indicates the MPC algorithm at the 1st... A is the wheel cylinder pressure prediction variable for each prediction step length; caliper Let f(T,λ) represent the effective area of ​​the caliper, and let f(T,λ) represent the temperature T-slip ratio λ compensation function. For the steering angle, a y This is lateral acceleration.

[0102] Specifically, decision planning and instruction generation:

[0103] Central Computing Platform: The central computing platform receives sensor data uploaded by the regional controllers. This data is preprocessed and analyzed to assess the vehicle's current status and operating environment.

[0104] Generating Control Commands: The central computing platform makes decisions and plans based on data, generating corresponding control commands, including but not limited to power control, braking control, and steering control. For example, based on vehicle speed and road conditions, the central computing platform can generate commands to adjust engine power or brake pressure.

[0105] The physics-based model considers the effects of speed, temperature, acceleration, and steering on the tire-ground friction coefficient, as well as the impact of weight transfer on the load of each tire. It calculates the maximum available braking force for each tire based on the load and friction coefficient, distributes the braking force to maximize the total braking force, and avoids locking or overheating of individual tires. The pressure is adjusted in real time, possibly through table lookup or online calculation, combined with feedback control to correct errors, thereby generating the optimal braking pressure adjustment command.

[0106] Speed ​​(v): Affects kinetic energy and braking distance, and needs to be monitored in real time.

[0107] Tire temperature (T): obtained by infrared sensor or embedded thermocouple. High temperature may reduce the coefficient of friction.

[0108] Acceleration:

[0109] Longitudinal acceleration (a) x ): Reflects braking intensity and weight transfer.

[0110] Lateral acceleration (a) y Centrifugal force during steering affects tire load distribution.

[0111] Steering angle (δ) and steering rate: determine the lateral force required by the tires.

[0112] 1. Establish dynamic models, including: tire-ground friction model and load distribution model.

[0113] Friction coefficient (μ): related to temperature and slip ratio; the tire-ground friction model is as follows:

[0114] ;

[0115] Where μ0: the maximum coefficient of friction at room temperature, k T Temperature decay coefficient, λ: slip ratio, λ c Critical slip ratio, T opt The target temperature.

[0116] Weight transfer: Longitudinal acceleration causes changes in the load on the front and rear axles. The load distribution model is as follows:

[0117] ;

[0118] Where m: vehicle weight, L: wheelbase, h: center of gravity height, lr: distance from rear axle to center of gravity.

[0119] Lateral load transfer: Difference in load between the inner and outer tires during steering:

[0120] ;

[0121] Where t is the wheel track.

[0122] 2. Calculate the target braking force:

[0123] From the deceleration target a req Decide:

[0124] ;

[0125] Among them, deceleration target a req The required deceleration is determined by the depth to which the driver depresses the brake pedal, which directly reflects the intensity of the deceleration demand.

[0126] Single-wheel braking force distribution: Distributed according to load and coefficient of friction, with priority given to tires under high load.

[0127] ;

[0128] Where, μ i (T,λ) and μ j (T,λ) represent the real-time friction coefficients of the i-th and j-th tires at tire temperature T and slip ratio λ, respectively. Let i be the target braking force for the i-th wheel.

[0129] For example, for the front left wheel ( :

[0130] ;

[0131] (If it's a left turn, add to the outer wheel) Inner wheel reduction ).

[0132] Substitute into the allocation formula, and finally, calculate... Precisely distribute braking force :

[0133] .

[0134] 3. Stress control strategy:

[0135] The central computing platform uses the single-wheel target braking force calculated in step 2 as a reference, and combines it with real-time collected slip ratio, tire temperature, and steering angle data to generate comprehensive pressure control constraints that include safety boundaries and compensation mechanisms, specifically including:

[0136] (1) Closed-loop correction control based on slip ratio:

[0137] The single-wheel target braking force is converted into a basic reference pressure based on the physical parameters of the braking actuator. The conversion formula is (in The single-wheel target braking force obtained in step 2, (This refers to the effective working area of ​​the brake caliper). This value is based solely on load calculations, followed by pressure correction calculated using a PID algorithm. To obtain dynamic target pressure The specific adjustments are as follows:

[0138] (a) Definition of deviation: The slip ratio deviation is calculated in real time by the central computing platform. ,in For real-time slip ratio, The target slip ratio (the target slip ratio is a preset value or a lookup value based on road surface characteristics, generally 10% to 20%).

[0139] (b) Regulation process: when When the slip ratio is too high and the wheels tend to lock up, the PID controller outputs a positive correction value. This reduces the dynamic target pressure, and the command execution mechanism reduces the braking pressure; when When the real-time slip ratio is too small and the braking force is insufficient, the PID controller outputs a negative correction value, which increases the dynamic target pressure and the command execution mechanism increases the braking pressure.

[0140] (c) Calculate the pressure regulation using a PID algorithm: ,in These are the proportional, integral, and differential coefficients, respectively.

[0141] (2) Temperature safety boundary constraints:

[0142] If the tire temperature is real-time In this case, temperature compensation is performed, limiting the upper limit of braking pressure and reducing the maximum allowable pressure to prevent overheating.

[0143] ;

[0144] Among them, T max For the maximum temperature, T safe For safe temperature, P0 is the reference pressure value.

[0145] (3) Steering Coordinated Feedforward Compensation:

[0146] When the vehicle is steering and braking, increase the pressure on the outer front wheel to suppress understeer:

[0147] ;

[0148] in, It is the steering compensation coefficient, used to quantify the adjustment intensity of steering and lateral acceleration on braking pressure. To achieve the corrected optimal braking pressure for the outer front wheel, The base is the braking pressure on the outer front wheel. For the steering angle, a y This is lateral acceleration.

[0149] 4. Real-time optimization algorithm:

[0150] A model predictive control (MPC) algorithm is employed to establish a closed-loop dynamic optimization solution process. In this process, the algorithm uses the single-wheel target braking force obtained in step 2 as the basis for the solution. To track the target (i.e., the ideal value to be achieved), the comprehensive pressure control constraints generated in step 3 (including temperature limit, PID correction direction, and steering compensation lower limit) are transformed into hard constraint boundaries. The MPC solver will search for an optimal pressure sequence within the safe range defined by the above "constraint boundaries" so that the actual braking force is as close as possible to the "tracking target".

[0151] The algorithm predicts within the time domain ( Find the optimal pressure sequence and take the first value in the sequence as the optimal braking pressure. The output is then passed to step 5 for execution.

[0152] The specific implementation steps are as follows:

[0153] (1) Establish a prediction model that includes actuator mapping:

[0154] The central computing platform is based on the current time. Using sensor data, a prediction model is established based on the "tire-ground friction model" and "load distribution model" as described in the claims. Specifically, the prediction model integrates the physical mapping relationship between pressure and braking force.

[0155] ;

[0156] in, : Indicates when the predicted pressure is applied At that time, the expected braking force calculated by the model; The MPC algorithm in the first stage A is the wheel cylinder pressure prediction variable for each prediction step length; caliper : Effective area of ​​the caliper, f(T,λ): Temperature-slip ratio compensation function can be calculated based on μ(T,λ) and temperature and slip ratio.

[0157] This mapping ensures that the algorithm can predict pressure based on the input. By working backward, the actual braking force of the vehicle in the future can be derived. This allows for a precise physical correspondence within the model.

[0158] (2) Set the optimization objective function: The solver aims to find a set of optimal pressure sequences. This makes the cost function Minimum:

[0159] ;

[0160] The formula parameters are explained as follows:

[0161] (Cost function value): Represents the overall objective value of the optimization problem at the current moment. The core task of the MPC algorithm is to find the solution that minimizes the J value by continuously adjusting the stress sequence.

[0162] N (Prediction Time Domain): Represents the number of time steps the MPC algorithm predicts into the future (corresponding to "predicting the vehicle state in the next few seconds" in the manual).

[0163] k (prediction step size variable): The iterative variable in the summation symbol, representing the k-th time point in the prediction time domain.

[0164] (Optimization variable): Represents the predicted wheel cylinder pressure applied to the wheel at the k-th future time.

[0165] (Previous time pressure): Represents the pressure value at the previous time step at time k, used to calculate the rate of change of pressure.

[0166] (Predicted braking force function): Based on a physical model, predicts the braking force function when pressure P is applied. k The actual braking force generated at that time.

[0167] (Target braking force for a single wheel): The target value of the ideal braking force calculated in step 2.

[0168] (Predicted temperature rise function): Based on a thermodynamic model, predict the increase in tire temperature when pressure Pk is applied.

[0169] (Braking performance weight): The larger this value, the more the system prioritizes ensuring that the actual braking force closely matches the target value. .

[0170] (Tire protection weight): The larger this value is, the more priority the system gives to limiting temperature rise.

[0171] (Stability weight): The larger this value is, the more the system prioritizes smooth pressure changes and prevents actuator oscillations.

[0172] (3) Applying constraints: During the solution process, the strategy in step 3 is transformed into hard constraints, and solutions that do not meet the requirements are eliminated: a) Safety upper bound constraint: Predicted pressure at any time. The calculated temperature limit must not be exceeded, i.e. b) Cooperative lower bound constraint: Predicting pressure under steering conditions. The steering coordination feedforward compensation requirements that need to be met, namely c) Slip ratio constraint: In conjunction with the PID correction direction, limit the deviation between the pressure adjustment direction and the slip ratio. The elimination direction is consistent.

[0173] (4) Scrolling optimization and output:

[0174] After solving the problem, only the first value in the optimal pressure sequence is extracted. Define it as the optimal braking pressure at the current moment. It is used to guide the final pressure build-up of electro-hydraulic systems or brake-by-wire systems.

[0175] 5. Based on the optimal braking pressure of each wheel cylinder This generates the corresponding control commands.

[0176] 6. Further, the issuance and execution of instructions:

[0177] High-speed Ethernet: The central computing platform sends generated control commands to the area controllers via high-speed Ethernet. High-speed Ethernet ensures that commands can be transmitted quickly and accurately, reducing communication latency.

[0178] The area controller performs operations: After receiving control commands, the area controller performs specific operations, such as controlling the motor speed, adjusting the braking pressure, and adjusting the steering angle, to ensure that the vehicle operates according to the decisions of the central computing platform.

[0179] Furthermore, cross-domain collaborative control:

[0180] SOA Architecture: Service-Oriented Architecture (SOA) supports dynamic invocation and composition of services, enabling cross-domain collaborative control. The central computing platform can invoke services from different domains, such as the powertrain, chassis, and body systems, to achieve collaborative control. For example, the central computing platform can simultaneously invoke services from both the powertrain and chassis systems to achieve collaborative control of power output and braking systems, improving vehicle performance and safety.

[0181] Furthermore, vehicle-cloud collaboration:

[0182] Vehicle-to-Cloud Collaboration Module: This module enables interconnectivity between the vehicle and its external environment, supporting remote control and data sharing. Through this module, the vehicle can communicate with a cloud server to remotely start the vehicle, check its status, and perform software upgrades. For example, users can remotely start the vehicle and check battery level and mileage using a mobile application. Utilizing technologies such as 5G and V2X, it achieves interconnectivity between vehicles, between vehicles and infrastructure, and between vehicles and the cloud. It supports remote diagnostics, OTA upgrades, and data sharing.

[0183] Furthermore, functional safety and information security:

[0184] Functional Safety and Information Security Module: This module ensures the safe and reliable operation of the system. It includes, but is not limited to, program rollback mechanisms, data encryption, authentication, fault diagnosis and handling, etc. For example, if an OTA upgrade fails, the system will automatically trigger a rollback mechanism to restore the software version to the stable version before the upgrade. Data encryption and authentication ensure the security and integrity of data transmission, preventing malicious attacks and data leaks. It adopts ASIL D level functional safety design to ensure the reliability and security of the system. It integrates information security technologies such as encryption and authentication to prevent data leaks and malicious attacks.

[0185] The method in this embodiment has the following effects:

[0186] Improved system integration: Through the collaborative work of the central computing platform and regional controllers, the centralization of computing resources and the distribution of regional control are realized, which simplifies the system architecture and improves control efficiency; centralized control and dynamic scheduling of vehicle functions are realized, reducing system complexity and integration difficulty.

[0187] Improved communication efficiency: The use of high-speed Ethernet improves the efficiency and reliability of data transmission, ensuring that control commands can be issued and executed quickly and accurately.

[0188] Support for cross-domain collaborative control: The SOA architecture supports dynamic service invocation and composition, enabling cross-domain collaborative control and improving vehicle performance and safety. Through deep integration of the central computing platform and the SOA architecture, deep collaborative control between the powertrain and chassis systems is achieved, providing a novel technological path for the development of intelligent vehicles. This innovative solution not only enhances vehicle performance and safety but also improves the user's driving experience, demonstrating broad application prospects and significant economic benefits.

[0189] Through intelligent decision-making and collaborative control via a central computing platform, the powertrain and chassis systems can achieve precise matching based on different driving scenarios and vehicle conditions, improving acceleration, braking performance, and handling stability, thus providing drivers with a smoother and more comfortable driving experience. In different driving scenarios, the central computing platform can optimize power output, reduce unnecessary energy consumption, improve fuel economy, and lower operating costs. In dangerous driving scenarios such as slippery roads and emergency braking, the central computing platform can respond quickly, enhancing the stability control of the chassis system while appropriately adjusting power output to effectively prevent vehicle loss of control and improve driving safety. The central computing platform monitors the status of the powertrain and chassis systems in real time, enabling timely detection of potential faults and uploading fault information to the cloud via the vehicle-cloud collaboration module for remote diagnosis and early warning, providing strong support for vehicle maintenance and upkeep.

[0190] The intelligent decision engine learns the driver's driving habits and preferences to generate personalized control strategies, providing a customized driving experience. Through the vehicle-cloud collaboration module, users can remotely control vehicle functions such as starting, stopping, and adjusting the air conditioning, while simultaneously obtaining real-time vehicle status information, such as battery level and mileage, enhancing user convenience and satisfaction.

[0191] Supports vehicle-cloud collaboration: The vehicle-cloud collaboration module enables interconnection between the vehicle and the external environment, supports remote control and data sharing, and improves the user experience.

[0192] Ensuring system safety: The functional safety and information security modules ensure the safe and reliable operation of the system, guaranteeing that the vehicle operates safely and reliably under various working conditions.

[0193] The method of this embodiment will be further explained below with reference to a specific example:

[0194] Vehicle control methods based on a central computing platform and regional controllers include:

[0195] Data collection and uploading:

[0196] The area controller collects front wheel speed and temperature data and uploads it to the central computing platform via high-speed Ethernet.

[0197] Data acquisition is the starting point of the entire control process, ensuring that the central computing platform can obtain real-time status information of the vehicle.

[0198] Decision planning and instruction generation:

[0199] Data reception and decision planning are crucial steps in generating control commands, ensuring that the central computing platform can generate appropriate control commands based on the vehicle's real-time status. The intelligent decision engine, based on the fused vehicle status information and combined with preset driving strategies and machine learning algorithms, identifies the current driving scenario, such as urban congestion, highway cruising, or slippery road surfaces. Based on the identified driving scenario and vehicle status, the intelligent decision engine generates optimal control commands, including power output adjustment, brake pressure regulation, and steering angle control.

[0200] The central computing platform receives data uploaded by the regional controller 1, makes decisions and plans based on the data, and generates instructions to control the front wheel braking pressure.

[0201] Command issuance and execution:

[0202] The central computing platform sends generated control commands to the area controllers via high-speed Ethernet. Upon receiving the commands, the area controllers execute specific operations, adjusting the front wheel brake pressure. Specifically, the area controllers convert the control commands into specific brake pressure adjustment operations, controlling the solenoid valves of the front wheel braking system. The area controllers then transmit the control commands to the front wheel braking system via CAN or LIN bus, achieving precise adjustment of the front wheel brake pressure.

[0203] Front wheel brake pressure adjustment can be specifically as follows:

[0204] 1) Conventional braking:

[0205] During normal braking, the ABS system is not activated. No current flows through the solenoid coil, the solenoid valve is in the "boost" position, the master cylinder and wheel cylinders are directly connected, and the wheel cylinder pressure increases or decreases with the master cylinder pressure.

[0206] 2) Pressure holding process:

[0207] When the wheel speed sensor sends a lock-up warning signal, the ECU sends a small holding current to the solenoid coil, and the solenoid valve is in the "pressure holding" position, so that the braking pressure in the wheel cylinder remains constant.

[0208] 3) Decompression process:

[0209] If a wheel lock-up signal still exists after the "maintain pressure" command is issued, the ECU will supply a maximum current to the solenoid coil, the solenoid valve will be in the "depressurization" position, and the brake fluid in the wheel cylinder will flow into the reservoir through the solenoid valve, thus reducing the wheel cylinder pressure.

[0210] 4) Pressurization process:

[0211] When the pressure drops and the wheels accelerate too quickly, the ECU cuts off the current to the solenoid valve, reconnecting the master cylinder and wheel cylinders. The high-pressure brake fluid in the master cylinder then re-enters the wheel cylinders, increasing the braking pressure.

[0212] During the execution of control commands, the front wheel braking system feeds back the execution status to the area controller in real time.

[0213] The area controller then uploads this feedback information to the central computing platform via high-speed Ethernet.

[0214] Based on the feedback of the execution status and combined with the real-time operating data of the vehicle, the central computing platform dynamically adjusts the control strategy to achieve precise control and continuous optimization of the vehicle status.

[0215] Cross-domain collaborative control:

[0216] The central computing platform uses SOA architecture to call services from the powertrain and chassis systems, enabling coordinated control of the powertrain and chassis systems and improving vehicle performance and safety.

[0217] Cross-domain collaborative control is a key step in expanding the functionality of a central computing platform, improving the overall performance and safety of vehicles by collaboratively controlling systems in different domains.

[0218] Based on the control commands generated by the intelligent decision engine, services from the powertrain and chassis systems are invoked. For example, the powertrain service is invoked to adjust engine power, and the chassis service is invoked to adjust brake pressure. Upon receiving the invocation commands, the powertrain and chassis systems execute the corresponding operations, achieving coordinated control. During execution, each system provides real-time feedback on its execution status, and the central computing platform dynamically adjusts the control strategy based on the feedback information to ensure optimal control performance.

[0219] During the execution of control commands, the powertrain and chassis systems feed back the execution status to the central computing platform in real time.

[0220] Control strategy optimization: The central computing platform dynamically adjusts the control strategy based on the feedback execution status and the real-time operation data of the vehicle, so as to achieve precise control and continuous optimization of the vehicle status.

[0221] Vehicle-cloud collaboration:

[0222] Through the vehicle-cloud collaboration module, users can remotely start the vehicle, check the vehicle status, and perform software upgrades.

[0223] Functional safety and information security:

[0224] The functional safety and information security module ensures the safe and reliable operation of the system through program rollback mechanisms, data encryption, authentication, fault diagnosis and handling, and other means.

[0225] Figure 3This demonstrates the trajectory tracking effect of a vehicle during emergency braking in a curve. The horizontal axis represents the vehicle's longitudinal displacement (m), and the vertical axis represents the vehicle's lateral displacement (m). The gray dashed line represents the ideal target driving trajectory. Existing technical solution: Without the introduction of steering coordination control, due to the front axle load transfer, the vehicle exhibits significant understeer, and the actual trajectory deviates to the outside of the curve (e.g., Figure 3 (The middle arrow indicates 'severe head push'). Invention solution: Based on formula... The steering coordination compensation control actively increases the braking pressure on the outer front wheel, generating a self-aligning torque. Analysis conclusion: Comparison shows that the actual trajectory of this invention closely follows the ideal trajectory, significantly reducing lateral deviation, proving that this control strategy can effectively improve the vehicle's lateral stability and tracking ability under complex operating conditions.

[0226] like Figure 2 As shown, this embodiment also provides a vehicle control system based on a central computing platform and a regional controller, including:

[0227] The data acquisition and upload module is used to acquire sensor data of the target vehicle through the area controller and upload the sensor data to the central computing platform.

[0228] The decision planning and instruction generation module is used by the central computing platform to analyze and optimize sensor data based on physical models and optimization algorithms to obtain the optimal braking pressure for each wheel cylinder. The physical model includes the influence of sensor data on the coefficient of friction between the tire and the ground, as well as the influence of weight transfer on the load of each tire.

[0229] The instruction issuance and execution module is used to generate corresponding control instructions based on the optimal braking pressure of each wheel cylinder, and issue them to the area controller to control the target vehicle, and obtain the execution status feedback to the area controller.

[0230] Furthermore, the system also includes:

[0231] The cross-domain collaborative control module is used by the central computing platform to call services from different domains through SOA architecture to achieve collaborative control;

[0232] The vehicle-cloud collaboration module is used for communication between the target vehicle and the cloud server;

[0233] The functional safety and information security module is used to integrate encryption and authentication information security technologies to ensure data transmission security.

[0234] This system achieves centralized computing resources and distributed regional control through the collaborative work of a central computing platform and regional controllers, simplifying the system architecture and improving control efficiency. It also enables centralized control and dynamic scheduling of vehicle functions, reducing system complexity and integration difficulty. Deep integration of the central computing platform with SOA architecture enables deep collaborative control of the powertrain and chassis systems, enhancing vehicle driving performance and safety.

[0235] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A vehicle control method based on a central computing platform and a regional controller, characterized in that, include: Sensor data from the target vehicle is collected by the regional controller and uploaded to the central computing platform. The central computing platform analyzes and optimizes the sensor data based on physical models and optimization algorithms to obtain the optimal braking pressure for each wheel cylinder. The physical model includes the influence of the sensor data on the coefficient of friction between the tire and the ground, as well as the influence of weight transfer on the load of each tire. The central computing platform analyzes and optimizes the sensor data based on physical models and optimization algorithms, including: The target braking force is calculated based on the vehicle weight and deceleration target of the target vehicle, and the single-wheel braking force is distributed according to the physical model to obtain the single-wheel target braking force of each wheel. Based on slip ratio, tire temperature, and steering angle, a comprehensive pressure control constraint including safety boundaries and compensation mechanisms is generated. The Model Predictive Control (MPC) algorithm is used to establish a predictive model that includes actuator mapping. The single-wheel target braking force is used as the tracking target, and the comprehensive pressure control constraint is used as the boundary condition. The wheel cylinder pressure sequence in the future time domain is solved by rolling optimization. Extract the current time value from the wheel cylinder pressure sequence to obtain the optimal braking pressure for each wheel cylinder; Based on the optimal braking pressure of each wheel cylinder, corresponding control commands are generated and sent to the area controller to control the target vehicle, and the execution status is obtained and fed back to the area controller.

2. The vehicle control method based on a central computing platform and a regional controller according to claim 1, characterized in that, The sensor data collected from the target vehicle includes: The system collects the vehicle's speed information using a speed sensor, tire temperature information using a temperature sensor, vehicle acceleration information using an acceleration sensor, and vehicle steering angle information using a steering angle sensor.

3. The vehicle control method based on a central computing platform and a regional controller according to claim 1, characterized in that, The physical models include: tire-ground friction model, load distribution model, and lateral load transfer model; The tire-ground friction model is as follows: ； The load allocation model is as follows: ； The lateral load transfer is as follows: ； Where μ is the coefficient of friction, T is the tire temperature, and λ is the slip ratio. This is the maximum coefficient of friction at room temperature. The temperature decay coefficient is For the target temperature, Where m is the critical slip ratio, L is the vehicle weight, and h is the center of gravity height. This is the distance from the rear axle to the center of gravity. For longitudinal acceleration, For the front wheel vertical load, The load on the inner and outer tires during steering, where t is the track width. This is lateral acceleration.

4. The vehicle control method based on a central computing platform and a regional controller according to claim 1, characterized in that, The target braking force for each wheel is: ； in, The target braking force for the i-th wheel, Let be the vertical load on the i-th wheel. To calculate the vertical load of the j-th wheel in the summation formula, For target braking force, and Let T and λ represent the real-time friction coefficients of the i-th and j-th tires at temperature T and slip ratio λ, respectively.

5. The vehicle control method based on a central computing platform and a regional controller according to claim 1, characterized in that, Based on the slip ratio, tire temperature, and steering angle, generating a comprehensive pressure control constraint that includes safety boundaries and compensation mechanisms includes: calculating the slip ratio deviation based on the slip ratio and the target slip ratio; The pressure correction is calculated using a PID controller based on the slip ratio deviation. The dynamic target pressure is obtained based on the baseline reference pressure and the pressure correction amount. If tire temperature T > maximum temperature Temperature compensation is then performed, limiting the upper limit of braking pressure to obtain temperature safety boundary constraints. Based on the steering angle, steering compensation coefficient, and baseline outer front wheel braking pressure, calculate the corrected optimal outer front wheel braking pressure: The comprehensive pressure control constraint is obtained based on the dynamic target pressure, the temperature safety boundary constraint, and the corrected optimal braking pressure of the outer front wheel.

6. The vehicle control method based on a central computing platform and a regional controller according to claim 1, characterized in that, The prediction model that includes the actuator mapping is: ； in, Indicates when predicting pressure is applied At that time, the expected braking force calculated by the model; A caliper Let f(T,λ) represent the effective area of ​​the caliper, and let f(T,λ) represent the temperature T-slip ratio λ compensation function. For the steering angle, a y For lateral acceleration, It is the steering compensation coefficient.

7. A vehicle control system based on a central computing platform and a regional controller, used to implement the method described in any one of claims 1-6, characterized in that, include: The data acquisition and uploading module is used to acquire sensor data of the target vehicle through the area controller and upload the sensor data to the central computing platform; The decision planning and instruction generation module is used by the central computing platform to analyze and optimize the sensor data based on the physical model and optimization algorithm to obtain the optimal braking pressure of each wheel cylinder. The physical model includes the influence of the sensor data on the coefficient of friction between the tire and the ground, as well as the influence of weight transfer on the load of each tire. The instruction issuance and execution module is used to generate corresponding control instructions based on the optimal braking pressure of each wheel cylinder, and issue them to the area controller to control the target vehicle, and obtain the execution status feedback to the area controller.

8. The vehicle control system based on a central computing platform and a regional controller according to claim 7, characterized in that, The system also includes: The cross-domain collaborative control module is used by the central computing platform to call services from different domains through SOA architecture to achieve collaborative control; The vehicle-cloud collaboration module is used for communication between the target vehicle and the cloud server; The functional safety and information security module is used to integrate encryption and authentication information security technologies to ensure data transmission security.

Images