Vehicle brake assist dynamic control system based on multi-source data fusion

Abstract

This invention provides a vehicle brake assist dynamic control system based on multi-source data fusion. The system comprises a distributed sensing module, a dynamic environmental compensation module, a multi-source fusion decision module, and an electronic assist execution module. A humidity sensor array directly measures the humidity in the tire-road contact area, reducing misjudgments caused by traditional rain sensors due to the difference between the windshield and the road surface. Its dynamic pre-judgment mechanism, based on an improved D-S evidence theory algorithm within the multi-source fusion decision module, adjusts brake assist before wheels slip, enabling proactive intervention and adjustment.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent automobile safety control, and in particular to a vehicle brake assist dynamic control system based on multi-source data fusion. Background Art

[0002] Current vehicle braking control relies heavily on the driver's subjective judgment. On slippery, icy, and snowy roads with low adhesion, manual operation can easily lead to uneven braking force distribution or delayed response, resulting in extended braking distances and even skidding accidents. While anti-lock braking systems and electronic stability programs (ESPs) employ wheel speed monitoring to intervene after a skid, their response mechanisms are lagging, preventing them from adjusting braking force in time when the tires experience critical slip. Summary of the Invention

[0003] In view of this, the present invention provides a vehicle brake assist dynamic control system based on multi-source data fusion to solve the technical defects existing in the prior art.

[0004] Specifically, the present invention provides a vehicle brake assist dynamic control system based on multi-source data fusion, comprising:

[0005] The distributed sensing module is configured as an array of miniaturized capacitive humidity sensors embedded in the tire's inner wall using three-dimensional curved surface bonding technology. The humidity sensor array is arranged in a spiral topology, and the spacing between adjacent cells is dynamically adjusted to a preset value.

[0006] A dynamic environmental compensation module, configured as a MEMS three-axis slope sensor array and an air pressure compensation unit mounted at the four corners of the vehicle chassis;

[0007] a multi-source fusion decision module configured to receive data from the distributed sensing module and the dynamic environment compensation module, wherein the multi-source fusion decision module implements a three-level data processing architecture: time domain synchronization and spatial alignment of the original signal layer, parameter gradient change rate calculation of the feature extraction layer, and an improved DS evidence theory algorithm of the decision layer;

[0008] The electronic power-assistance execution module includes a linear electromagnetic booster with pressure-displacement dual closed-loop control.

[0009] In some embodiments, the humidity sensor adopts a double-layer silicone packaging structure, the inner layer is a high-temperature resistant epoxy resin protection circuit substrate, and the outer layer is an elastic silicone layer with a groove design;

[0010] The humidity sensor array forms a self-repairing electrical connection with the hub transmitting module through a flexible conductive adhesive;

[0011] A six-axis inertia compensation unit is installed inside the wheel hub to eliminate the interference of centrifugal force caused by tire rotation.

[0012] In some embodiments, the air pressure compensation unit of the dynamic environment compensation module processes tire internal air pressure fluctuation data using a Kalman filter algorithm to generate an air pressure change impact component;

[0013] The slope sensor uses a MEMS gyroscope and accelerometer composite structure to form a spatial attitude perception network, and generates a three-dimensional road slope angle by fusing data from four-corner position sensors.

[0014] In some embodiments, the multi-source fusion decision module sets three decision strategy libraries: rainy day mode, snowy mode, and long downhill mode;

[0015] The humidity change rate of each tire contact area is calculated using the humidity sensor array data, and the slope hazard coefficient is generated by combining the slope sensor data;

[0016] Generate energy consumption load index through real-time monitoring data of battery management system;

[0017] The improved DS evidence theory algorithm run by the decision layer performs uncertainty reasoning on the humidity change rate, the slope risk coefficient and the energy consumption load index.

[0018] In some embodiments, the master cylinder of the electronic power-assistance actuator module adopts a dual-chamber redundant design and a built-in Hall effect displacement sensor;

[0019] The power assist characteristic curve is dynamically adjusted through the output signal of the multi-source fusion decision module, supporting two adjustment modes: millisecond-level gradient change and preset mode jump.

[0020] In some embodiments, the step of generating dynamic power assist map parameters includes: collecting contact surface microscopic water film thickness distribution data at a sampling frequency of 100 Hz through a tire sensor network, and generating a moisture distribution matrix for each tire after filtering through a six-axis inertia compensation unit;

[0021] Packaging and transmitting the humidity distribution matrix and tire internal temperature and pressure data to a chassis hub via near field communication;

[0022] A three-dimensional road model is established using slope sensor array data to calculate and generate the effective slope angle of each tire contact area;

[0023] Performing spatiotemporal alignment processing on the humidity distribution matrix, the effective slope angle, and the battery load data through a central decision unit;

[0024] Generate dynamic power map parameters through multi-dimensional feature fusion algorithm.

[0025] In some embodiments, the multi-dimensional feature fusion includes: generating a regional humidity weight coefficient by analyzing the difference in humidity data between the front wheel guide area and the rear wheel brake area;

[0026] Generate axle load distribution correction factors through vehicle pitching moment calculation;

[0027] The energy consumption reduction coefficient is generated by evaluating the system's sustainable working time.

[0028] In some embodiments, the deceleration build-up rate and tire slip ratio change parameters are monitored in real time via wheel speed sensors;

[0029] The dynamic power assist map parameters are closed-loop optimized using a recursive learning algorithm to generate weight adjustments.

[0030] In some embodiments, the calculation formula for generating the dynamic power map parameters is:

[0031] )

[0032] where w i is the spatial weighting coefficient of each tire humidity sensor, which comes from the differential analysis of the front wheel guide area and the rear wheel braking area by the multi-source fusion decision module; ΔH i is the humidity change rate of the i-th tire contact area, which is obtained through the tire embedded sensor network with a sampling frequency of 100 Hz; α is the slope compensation factor, which is obtained by calibrating the three-axis slope sensor array of the dynamic environmental compensation module; θ j is the pitch angle component output by the three-axis slope sensor, which is measured by the MEMS gyroscope composite structure at the four corners of the chassis; β is the energy consumption attenuation coefficient, which is dynamically adjusted according to the instantaneous load state of the electronic power-assisted execution module; P k is the tire pressure value at the kth sampling point, which is monitored in real time by the air pressure compensation unit on the inner side of the wheel hub; P0 is the standard air pressure value, which is stored in the system preset parameter library; T k is the tire temperature value at the corresponding sampling point, which is synchronously collected by the capacitive humidity sensor array.

[0033] In some embodiments, the weight adjustment calculation formula for calculating the weight adjustment amount is:

[0034]

[0035] Where γ is the learning rate, which is dynamically adjusted by exponential decay method based on historical braking data; is the error gradient of the mth weight parameter, which comes from the pressure-displacement dual closed-loop control feedback of the electronic power-assistance execution module; S n is the actual slip rate at the nth time point, which is calculated using the wheel speed sensor and the road surface model; To predict the slip rate, it is derived from the DS evidence theory algorithm of the multi-source fusion decision module; η is the momentum factor, which is related to the vehicle load parameter and the remaining power of the battery system; Ap is the average deceleration of the p-th braking process, measured after eliminating the centrifugal force interference through the six-axis inertia compensation unit; is the historical average deceleration, stored in the system self-calibration database; σ A is the standard deviation of deceleration, which is obtained through statistics of the latest 50 braking processes.

[0036] At least one embodiment of the present invention utilizes a distributed sensing module, a dynamic environmental compensation module, a multi-source fusion decision module, and an electronic power assist execution module. This allows for direct measurement of humidity in the tire-road contact area via a humidity sensor array, reducing misjudgments caused by traditional rain sensors due to differences between the windshield and the road surface. Its dynamic pre-judgment mechanism, based on an improved DS evidence theory algorithm within the multi-source fusion decision module, can adjust brake assist before wheels slip, enabling proactive intervention and adjustment of brake assist. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a structural block diagram of a vehicle brake assist dynamic control system based on multi-source data fusion provided by the present invention. DETAILED DESCRIPTION

[0038] The following description sets forth many specific details to facilitate a thorough understanding of this specification. However, this specification can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the scope of this specification. Therefore, this specification is not limited to the specific implementations disclosed below.

[0039] The terms used in one or more embodiments of this specification are for the purpose of describing specific embodiments only and are not intended to limit one or more embodiments of this specification. The singular forms of "a" and "the" used in one or more embodiments of this specification and the appended claims are also intended to include plural forms, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more associated listed items. The modifications of "one" and "a plurality" mentioned in this disclosure are illustrative and not restrictive, and those skilled in the art should understand that unless the context clearly indicates otherwise, it should be understood as "one or more".

[0040] It should be understood that although the terms first, second, etc. may be used to describe various information in one or more embodiments of this specification, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of this specification, the first may also be referred to as the second, and similarly, the second may also be referred to as the first. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0041] First, the terms involved in one or more embodiments of this specification are explained.

[0042] DS Evidence Theory: Dempster / Shafer, Evidence Theory.

[0043] MEMS: Micro Electromechanical System.

[0044] See also Figure 1 , Figure 1 A structural block diagram of a vehicle brake assist dynamic control system based on multi-source data fusion provided according to some embodiments of the present specification is shown, and the vehicle brake assist dynamic control system based on multi-source data fusion includes: a distributed sensing module, which is configured to be composed of a miniaturized capacitive humidity sensor array embedded in the inner wall of the tire through three-dimensional surface fitting technology, and the humidity sensor array is arranged in a spiral topology structure, and the spacing between adjacent units is dynamically adjusted to a preset value; a dynamic environment compensation module, which is configured to be a MEMS three-axis slope sensor array and an air pressure compensation unit installed at the four corners of the vehicle chassis; a multi-source fusion decision module, which is configured to receive data from the distributed sensing module and the dynamic environment compensation module, and the multi-source fusion decision module executes a three-level data processing architecture: time domain synchronization and spatial alignment of the original signal layer, parameter gradient change rate calculation of the feature extraction layer, and improved DS evidence theory algorithm of the decision layer; an electronic power assist execution module, which includes a linear electromagnetic booster with pressure-displacement dual closed-loop control, and the electronic power assist execution module.

[0045] The distributed sensing module utilizes a miniaturized capacitive humidity sensor array, embedded directly into the tire's inner wall using three-dimensional surface bonding technology, forming a detection structure that perfectly conforms to the tire's curved surface. The sensor units are arranged in a spiral topology to provide full circumferential coverage of the tire, and the spacing between them can be dynamically adjusted based on tread wear. For example, the spacing between adjacent units can be dynamically adjusted to a preset value of 8-15cm. The humidity sensor array is a sensor cluster composed of multiple groups of humidity detection units arranged in a specific topology, capable of simultaneously monitoring the humidity distribution in different areas of the tire's contact surface.

[0046] Three-dimensional surface bonding technology is a flexible packaging process that adapts a rigid sensor array to the tire's curved surface, ensuring constant contact pressure between the detection unit and the tread. A spiral topology, where sensors are arranged in a spiral pattern along the tire's circumference, better captures differences in friction characteristics across different tread areas compared to a parallel arrangement. The MEMS three-axis slope sensor is a microelectromechanical system-based inclination measurement unit that can simultaneously detect angle changes along the X, Y, and Z axes with an accuracy of ±0.5 degrees.

[0047] The dynamic environmental compensation module consists of a MEMS triaxial slope sensor array mounted at each corner of the vehicle chassis. This array monitors changes in the vehicle's pitch and roll angles in real time and works in conjunction with the air pressure compensation unit to mitigate the effects of altitude fluctuations on braking force. A redundant design ensures that basic functionality is maintained even in the event of a single point of failure. The air pressure compensation unit automatically adjusts braking parameters based on altitude fluctuations, eliminating the effects of atmospheric pressure on the hydraulic system.

[0048] The multi-source fusion decision module receives tire moisture and chassis environment data and performs three levels of processing: the raw signal layer synchronizes the time domain of multiple sensors and aligns spatial coordinates; the feature extraction layer calculates the parameter gradient change rate to identify dangerous trends; and the decision layer uses the improved DS evidence theory algorithm to fuse multi-dimensional evidence and generate braking strategies. The improved DS evidence theory algorithm is an enhanced evidence fusion method that improves decision reliability in rainy conditions by incorporating moisture credibility weights and slope risk factors.

[0049] The electronic power-assistance actuator module integrates a linear electromagnetic booster with dual closed-loop pressure-displacement control, translating decision-making instructions into precise braking force. A dual-chamber design ensures that electromagnetic actuation can maintain partial braking force even in the event of hydraulic failure.

[0050] At least one embodiment of the present invention utilizes a distributed sensing module, a dynamic environmental compensation module, a multi-source fusion decision module, and an electronic power assist execution module. This allows for direct measurement of humidity in the tire-road contact area via a humidity sensor array, reducing misjudgments caused by traditional rain sensors due to differences between the windshield and the road surface. Its dynamic pre-judgment mechanism, based on an improved DS evidence theory algorithm within the multi-source fusion decision module, can adjust brake assist before wheels slip, enabling proactive intervention and adjustment of brake assist.

[0051] In some embodiments, the humidity sensor adopts a double-layer silicone packaging structure, the inner layer is a high-temperature resistant epoxy resin protection circuit substrate, and the outer layer is an elastic silicone layer with a groove design; the humidity sensor array forms a self-repairing electrical connection with the wheel hub transmitting module through a flexible conductive adhesive; a six-axis inertia compensation unit is installed on the inside of the wheel hub to eliminate the interference of the centrifugal force of tire rotation.

[0052] The double-layer silicone encapsulation structure comprises a composite protective system composed of two polymer layers. The inner layer, a high-temperature-resistant epoxy resin, encapsulates the sensor circuit substrate, providing basic insulation protection. The outer layer is an elastic silicone layer with a grooved surface that enhances strain tolerance and aids in water drainage and heat dissipation. The groove design refers to microscopic drainage channels etched into the outer silicone layer, which accelerates water drainage and increases contact friction between the sensor surface and the tire surface. Flexible conductive adhesive, a conductive and flexible adhesive, provides the electrical connection between the sensor array and the wheel hub transmitter module, capable of withstanding the cyclical bending stresses generated by tire rotation. The wheel hub transmitter module, an electronic unit with integrated wireless transmission capabilities, is mounted inside the wheel hub and is responsible for transmitting sensor data to the onboard control system. It features a vibration-resistant design. The six-axis inertial compensation unit, a composite sensor module integrating a three-axis accelerometer and a three-axis gyroscope, detects centrifugal force and vibration interference generated by tire rotation in real time and uses an algorithm to compensate for any interference with the humidity signal. Tire rotation centrifugal force interference is the inertial force generated by the high-speed rotation of the tires when the vehicle is driving. It may cause sensor signal offset or poor connector contact, which is a systematic error source.

[0053] In some embodiments, the air pressure compensation unit of the dynamic environment compensation module processes the tire internal pressure fluctuation data through the Kalman filter algorithm to generate an air pressure change impact component; the slope sensor uses a MEMS gyroscope and accelerometer composite structure to form a spatial attitude perception network, and generates a three-dimensional road slope angle through the fusion of four-corner position sensor data.

[0054] The Kalman filter algorithm is a recursive optimization and estimation algorithm that uses the statistical characteristics of noise to correct tire pressure measurements in real time, separating the effective signal from the random noise components in the pressure fluctuations. The pressure change impact component is a standardized parameter output after algorithmic processing. It quantitatively represents the correction required for braking force distribution due to tire pressure fluctuations and serves as the basis for compensation for the electronic power steering system. The MEMS gyroscope is an angular velocity sensor based on microelectromechanical systems. It detects the rotational motion of each axis of the vehicle and complements the accelerometer data to form a complete attitude solution. The spatial attitude perception network is a distributed measurement system composed of four groups of MEMS sensors. It reconstructs the three-dimensional spatial position of the vehicle body relative to the road surface through a spatiotemporal synchronization algorithm. The three-dimensional road slope angle is a vector parameter calculated by fusing data from the four corner sensors. It contains multi-dimensional road geometry characteristics such as longitudinal slope angle, lateral roll angle, and compound curvature angle.

[0055] The air pressure compensation unit significantly improves tire pressure monitoring's anti-interference capabilities through Kalman filtering, preventing brake system misadjustments caused by transient air pressure fluctuations. The spatial attitude perception network detects road slopes under all operating conditions. Four-corner data fusion effectively eliminates single-point measurement errors, providing accurate three-dimensional road characteristic parameters for the electronic power steering system. This integrated solution ensures optimal braking force distribution across various vehicle altitudes, slopes, and load conditions.

[0056] In some embodiments, the multi-source fusion decision module sets up three decision strategy libraries: rainy day mode, snow mode, and long downhill mode; calculates the humidity change rate of each tire contact area through humidity sensor array data, and generates a slope hazard coefficient in combination with slope sensor data; generates an energy consumption load index through real-time monitoring data of the battery management system; and the improved DS evidence theory algorithm run by the decision layer performs uncertainty reasoning on the humidity change rate, the slope hazard coefficient, and the energy consumption load index.

[0057] The decision strategy library is a set of preset standardized control parameters that includes mapping relationships between core parameters such as braking force distribution and torque output under different road conditions, and supports dynamic calling and online updates. The humidity change rate is the differential calculation result of the tire contact surface humidity data, reflecting the rate of change of the road water film thickness and used to predict the trend of tire grip attenuation. The slope hazard coefficient is a composite indicator that integrates the three-dimensional slope angle and road curvature. It quantitatively assesses the threat level of the slope to vehicle stability, with special attention to the brake thermal decay caused by continuous downhill. The energy consumption load index is a dynamic parameter output by the battery management system, which represents the matching degree between the current remaining capacity of the power system and the energy consumption demand of the drive / brake system.

[0058] A multi-mode strategy library enables precise identification of operating conditions in extreme weather conditions. The coordinated calculation of humidity change rate and slope hazard factor provides early warning of hydroplaning and brake failure risks. The energy consumption load index dynamically balances power system requirements with safety requirements. An improved DS algorithm effectively resolves sensor data conflicts, enabling the system to maintain optimal braking performance in complex scenarios such as heavy rain, ice, and snow.

[0059] In some embodiments, the master cylinder of the electronic power-assistance execution module adopts a dual-chamber redundant design and a built-in Hall effect displacement sensor; the power-assistance characteristic curve is dynamically adjusted through the output signal of the multi-source fusion decision module, supporting two adjustment methods: millisecond-level gradient changes and preset mode jumps.

[0060] The dual-chamber redundant design is two independent pressure chambers connected in parallel inside the master cylinder. When any chamber fails, the basic braking performance can still be maintained, which meets the requirements of the ASIL-D functional safety level. The Hall effect displacement sensor is a non-contact displacement detection device based on the principle of magnetic field change. It provides real-time feedback on the movement stroke of the master cylinder piston with an accuracy of up to micron level. The power assist characteristic curve describes the nonlinear mapping relationship between the brake pedal stroke and the output hydraulic pressure. The foot feel adaptation under different working conditions is achieved by adjusting the slope of the curve. The millisecond-level gradient change is the smooth transition adjustment capability of the brake assist intensity, and the response time is less than the mechanical delay of the conventional hydraulic system. The preset mode jump is to instantly switch to the pre-stored power assist parameters according to the instructions of the decision module, which is suitable for the fast switching needs of snow / sports and other modes.

[0061] The dual-chamber design significantly improves the braking system's fault tolerance, while Hall sensors enable high-precision closed-loop control of piston displacement. Dynamic power curve adjustment ensures a linear feel during emergency braking, while maintaining comfortable feedback during everyday driving. Millisecond-level response ensures seamless mode switching, and a pre-set jump function allows for instant adaptation to unexpected road conditions. The overall system balances safety and driving quality.

[0062] In some embodiments, the steps of generating dynamic power assist map parameters include: collecting contact surface microscopic water film thickness distribution data at a sampling frequency of 100 Hz through a tire sensor network, and generating a moisture distribution matrix for each tire after filtering processing by a six-axis inertia compensation unit; packaging the moisture distribution matrix and the tire internal temperature and pressure data to transmit to a chassis hub through near-field communication; establishing a three-dimensional road model through slope sensor array data, and calculating and generating the effective slope angle of each tire contact area; performing spatiotemporal alignment processing on the moisture distribution matrix, the effective slope angle and the battery load data through a central decision unit; and generating dynamic power assist map parameters through a multi-dimensional feature fusion algorithm.

[0063] The tire sensor network is a distributed monitoring system comprised of microsensors embedded in the tire tread. It utilizes MEMS technology to simultaneously collect multiple parameters, including water film thickness and temperature. The microscopic water film thickness distribution measures the spatial gradient of water film thickness at the tire contact patch, reflecting local water accumulation depth and drainage capacity, and is used to predict areas at risk of hydroplaning. The moisture distribution matrix stores quantified moisture values ​​for each tire contact patch in a two-dimensional array, mapping location coordinates to corresponding water film thickness. Near-field communication (NFC) is a short-range wireless data transmission protocol that enables low-latency data exchange between tire sensors and the chassis system. The chassis hub is the central communication node in the vehicle chassis, responsible for data aggregation and protocol conversion from multiple sensor sources. The slope sensor array is a distributed array of tilt measurement devices that constructs a three-dimensional road surface geometric model through spatial differential calculations. The effective slope angle is the slope component of the tire contact patch in the direction of actual force applied, representing the true tilt angle after compensation for vehicle posture. The central decision unit (CDU) is an onboard high-performance computing module that enables spatiotemporal synchronization and feature correlation analysis of heterogeneous multi-source data. Spatiotemporal alignment unifies the timestamps and spatial coordinate systems of data from different sensors, eliminating the effects of transmission delays and installation location differences. Multidimensional feature fusion is a cross-domain data association algorithm that maps parameters such as humidity, slope, and load into a unified control decision vector. The dynamic power assist map parameter is a three-dimensional lookup table of braking system parameters, with the horizontal axis representing the humidity / slope / load combination and the vertical axis representing the optimal braking force distribution solution.

[0064] High-frequency humidity sampling and inertia compensation ensure accurate road water film detection, while near-field communication enables lossless transmission of tire data. Three-dimensional slope modeling accurately reflects actual load conditions, while spatiotemporal alignment eliminates multi-sensor coordination errors. Dynamic power assist mapping parameters enable the braking system to automatically adapt to complex road conditions such as heavy rain, ice, and snow, significantly improving braking efficiency and directional stability on slippery roads.

[0065] In some embodiments, the multi-dimensional feature fusion includes: generating a regional humidity weight coefficient through humidity data difference analysis between the front wheel guide area and the rear wheel braking area; generating an axle load distribution correction factor through vehicle pitching moment calculation; and generating an energy consumption attenuation coefficient through system sustainable working time evaluation.

[0066] The front wheel guidance zone is the specific area within the front wheel contact patch that provides primary guidance during steering. Its humidity data directly impacts steering response characteristics. The rear wheel braking zone is the key area within the rear wheel contact patch that prioritizes braking. Humidity changes play a decisive role in vehicle stability control. The regional humidity weight coefficient is a dynamic parameter that quantifies the impact of humidity differences between the front and rear wheels, reflecting the varying contributions of different areas to vehicle handling stability. The pitching moment is the longitudinal rotational torque generated by the shift in center of gravity during acceleration and braking, directly impacting the dynamic load distribution ratio between the front and rear axles. The axle load distribution correction factor is a load distribution parameter dynamically adjusted based on the real-time pitching moment to compensate for axle load transfer effects during braking. The system's sustainable operating time is a predicted value for how long the battery system will maintain safe braking performance under current operating conditions. The energy consumption decay coefficient is a dynamic parameter that characterizes the rate at which the system's sustainable operating capacity decreases, reflecting the changing trend in the matching of the battery's remaining energy with the braking demand.

[0067] Regional humidity weighting enables differentiated control of steering and braking areas, while axle load correction factors effectively suppress brake nodding. An energy attenuation coefficient warning function extends the system's reliable operation under extreme operating conditions, and a three-dimensional fusion algorithm ensures optimal load distribution and energy utilization during braking in heavy rain.

[0068] In some embodiments, the deceleration build-up rate and tire slip rate change parameters are monitored in real time by wheel speed sensors; the dynamic power assist map parameters are closed-loop optimized by a recursive learning algorithm to generate weight adjustment amounts.

[0069] The deceleration build-up rate reflects the gradient of the acceleration at which the vehicle's speed decreases during braking and is used to evaluate the dynamic response characteristics of the braking system. The tire slip rate variation parameter is a quantitative indicator that characterizes the dynamic evolution of the relative slip between the tire and the road surface. It is calculated by using the percentage relationship between the wheel speed difference and the vehicle speed. Closed-loop optimization is an iterative process that continuously adjusts control parameters based on real-time feedback data, forming a complete control loop of monitoring, calculation, and adjustment. The weight adjustment is a correction value that reflects the importance of different control parameters in the algorithm and is used to dynamically balance the influence of each input variable.

[0070] High-frequency monitoring of wheel speed sensors enables millisecond-level capture of braking dynamics, while slip parameter trend analysis can predict tire lock risk. A recursive learning algorithm's closed-loop optimization mechanism enables the system to continuously evolve, with dynamic weight adjustment ensuring optimal control strategy adaptation under varying operating conditions. This solution significantly improves directional stability and ABS system response accuracy during emergency braking.

[0071] In some embodiments, the dynamic power map parameter K a The generation formula is:

[0072] )

[0073] where w i is the spatial weighting coefficient of each tire humidity sensor, which comes from the differential analysis of the front wheel guide area and the rear wheel braking area by the multi-source fusion decision module; ΔH i is the humidity change rate of the i-th tire contact area, which is obtained through the tire embedded sensor network with a sampling frequency of 100 Hz; α is the slope compensation factor, which is obtained by calibrating the three-axis slope sensor array of the dynamic environmental compensation module; θ j is the pitch angle component output by the three-axis slope sensor, which is measured by the MEMS gyroscope composite structure at the four corners of the chassis; β is the energy consumption attenuation coefficient, which is dynamically adjusted according to the instantaneous load state of the electronic power-assisted execution module; P k is the tire pressure value at the kth sampling point, which is monitored in real time by the air pressure compensation unit on the inner side of the wheel hub; P0 is the standard air pressure value, which is stored in the system preset parameter library; T k is the tire temperature value at the corresponding sampling point, which is synchronously collected by the capacitive humidity sensor array.

[0074] The spatial weighting coefficient is a weight parameter that reflects the impact of different tire areas on the braking stability of the entire vehicle, and is dynamically adjusted according to the tire position and functional zoning. The slope compensation factor is a parameter that is dynamically adjusted according to the vehicle's tilt state and is used to correct the braking force distribution ratio under slope conditions. The pitch angle component is the projection component of the vehicle's longitudinal tilt angle in the three-dimensional coordinate system, reflecting the degree of center of gravity offset when driving on a slope. The energy consumption attenuation coefficient is a dynamic parameter that represents the matching degree of the system's available energy with the braking demand, and affects the linear release curve of the braking force. The standard air pressure value is the tire baseline pressure parameter preset by the system, which serves as a reference benchmark for air pressure anomaly detection.

[0075] Dynamic power assist mapping parameters integrate multi-dimensional environmental parameters to precisely distribute braking force. Humidity change rate monitoring enables the system to proactively predict the risk of aquaplaning. Slope compensation ensures directional stability during braking on slopes, and optimized energy reduction coefficients balance braking performance with system protection. A standardized pressure reference enhances the sensitivity of tire pressure anomaly detection, while a spatial weighting factor prioritizes braking responsiveness on the steering wheel. The overall solution significantly improves braking safety margins in complex road conditions.

[0076] In some embodiments, the weight adjustment calculation formula for calculating the weight adjustment amount is:

[0077]

[0078] Where γ is the learning rate, which is dynamically adjusted by exponential decay method based on historical braking data; is the error gradient of the mth weight parameter, which comes from the pressure-displacement dual closed-loop control feedback of the electronic power-assistance execution module; S nis the actual slip rate at the nth time point, which is calculated using the wheel speed sensor and the road surface model; To predict the slip rate, it is derived from the DS evidence theory algorithm of the multi-source fusion decision module; η is the momentum factor, which is related to the vehicle load parameter and the remaining power of the battery system; A p is the average deceleration of the p-th braking process, measured after eliminating the centrifugal force interference through the six-axis inertia compensation unit; is the historical average deceleration, stored in the system self-calibration database; σ A is the standard deviation of deceleration, which is obtained through statistics of the latest 50 braking processes.

[0079] The learning rate is a key factor in controlling the algorithm's parameter update step size and determines the system's sensitivity to error feedback. The error gradient reflects the partial derivative of the weight parameter's influence on the system's overall error, guiding parameter optimization. The actual slip rate is the relative slip ratio between the tire and the road surface, measured directly by the wheel speed sensor. The predicted slip rate is a theoretically optimal slip rate reference value estimated by a multi-source data fusion algorithm. The momentum factor is a dynamic parameter that adjusts the algorithm's inertia, balancing the influence of current data with historical experience. Average deceleration is the average rate of decrease in vehicle speed during a single braking event. The historical average deceleration is the statistical average of multiple braking decelerations recorded by the system. The deceleration standard deviation represents a statistical measure of the degree of fluctuation in braking performance and reflects braking stability.

[0080] A recursive learning algorithm improves system adaptability through continuous self-optimization, while error gradient guidance ensures correct parameter updates. Slip ratio deviation detection enables precise adjustment of anti-lock control, while the momentum factor maintains control stability during sudden load changes. Deceleration statistical analysis automatically identifies brake performance degradation, while historical data reference enables the system to identify operating conditions. This overall solution significantly enhances the intelligence and continuous optimization capabilities of the braking system.

[0081] The preferred embodiments disclosed above are intended only to help illustrate this specification. The optional embodiments do not exhaustively describe all details, nor do they limit the invention to the specific embodiments described. Obviously, many modifications and variations are possible based on the content of the present invention. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize this specification. This specification is limited only by the claims and their full scope and equivalents.

Claims

1. A vehicle brake assist dynamic control system based on multi-source data fusion, characterized in that: include: The distributed sensing module is configured as a miniaturized capacitive humidity sensor array embedded in the inner wall of the tire using a three-dimensional curved surface bonding technology, wherein the humidity sensor array is arranged in a spiral topology; A dynamic environmental compensation module, configured as a MEMS three-axis slope sensor array and an air pressure compensation unit mounted at the four corners of the vehicle chassis; The multi-source fusion decision module is configured to receive data from the distributed sensing module and the dynamic environment compensation module. The multi-source fusion decision module implements a three-level data processing architecture: time domain synchronization and spatial alignment of the original signal layer, parameter gradient change rate calculation of the feature extraction layer, and an improved DS evidence theory algorithm of the decision layer to calculate the dynamic power map parameters. The calculation formula for calculating the dynamic power map parameters is: ) in, is the dynamic power map parameter, wi is the spatial weighting coefficient of each tire humidity sensor, which comes from the differential analysis of the front wheel guide area and the rear wheel braking area by the multi-source fusion decision module; ΔH i is the humidity change rate of the i-th tire contact area, which is obtained through the tire embedded sensor network with a sampling frequency of 100 Hz; α is the slope compensation factor, which is obtained by calibrating the three-axis slope sensor array of the dynamic environmental compensation module; θ j is the pitch angle component output by the three-axis slope sensor, which is measured by the three-axis slope sensor; β is the energy consumption attenuation coefficient, which is generated by evaluating the sustainable working time of the system; P k is the tire pressure value at the kth sampling point, which is monitored in real time by the air pressure compensation unit on the inner side of the wheel hub; P0 is the standard air pressure value, which is stored in the system preset parameter library; T k is the tire temperature value at the corresponding sampling point, which is synchronously collected by the capacitive humidity sensor array; The electronic power assist execution module includes a linear electromagnetic booster with pressure-displacement dual closed-loop control. The electronic power assist execution module dynamically adjusts the power assist characteristic curve through the output signal of the multi-source fusion decision module.

2. The system according to claim 1, wherein: The humidity sensor adopts a double-layer silicone packaging structure, the inner layer is a high-temperature resistant epoxy resin protection circuit substrate, and the outer layer is an elastic silicone layer with a groove design; The humidity sensor array forms a self-repairing electrical connection with the hub transmitting module through a flexible conductive adhesive; A six-axis inertia compensation unit is installed on the inner side of the wheel hub to eliminate the interference of centrifugal force of tire rotation.

3. The system according to claim 1, wherein: The air pressure compensation unit of the dynamic environment compensation module processes the tire internal air pressure fluctuation data through the Kalman filter algorithm to generate an air pressure change impact component; The slope sensor uses a MEMS gyroscope and accelerometer composite structure to form a spatial attitude perception network, and generates a three-dimensional road slope angle by fusing data from four-corner position sensors.

4. The system according to claim 1, wherein: The multi-source fusion decision module sets three decision strategy libraries: rainy day mode, snowy mode and long downhill mode; The humidity change rate of each tire contact area is calculated using the humidity sensor array data, and the slope hazard coefficient is generated using the slope sensor data; Generate energy consumption load index through real-time monitoring data of battery management system; The improved DS evidence theory algorithm run by the decision layer performs uncertainty reasoning on the humidity change rate, the slope risk coefficient and the energy consumption load index.

5. The system according to claim 1, wherein: The master cylinder of the electronic power-assistance actuator module adopts a dual-chamber redundant design and a built-in Hall effect displacement sensor; The electronic power assist execution module supports two adjustment modes: millisecond-level gradient change and preset mode jump.

6. The system according to claim 2, characterized in that: The steps for generating dynamic power map parameters include: The tire sensor network collects the microscopic water film thickness distribution data on the contact surface at a sampling frequency of 100Hz, and after filtering and processing by the six-axis inertia compensation unit, generates the moisture distribution matrix of each tire; Packaging and transmitting the humidity distribution matrix and tire internal temperature and pressure data to a chassis hub via near field communication; A three-dimensional road model is established using slope sensor array data to calculate and generate the effective slope angle of each tire contact area; The humidity distribution matrix, the effective slope angle and the battery load data are subjected to spatiotemporal alignment processing by a central decision unit.

7. The system according to claim 6, characterized in that: Real-time monitoring of deceleration build-up rate and tire slip rate change parameters through wheel speed sensors; The dynamic power assist map parameters are closed-loop optimized using a recursive learning algorithm to generate weight adjustments.

8. The system according to claim 7, characterized in that The weight adjustment calculation formula for calculating the weight adjustment amount is: in, is the weight adjustment amount, γ is the learning rate, which is dynamically adjusted by exponential decay method based on historical braking data; is the error gradient of the mth weight parameter, which comes from the pressure-displacement dual closed-loop control feedback of the electronic power-assistance execution module; S n is the actual slip rate at the nth time point, which is calculated using the wheel speed sensor and the road surface model; To predict the slip rate, it is derived from the DS evidence theory algorithm of the multi-source fusion decision module; η is the momentum factor, which is related to the vehicle load parameter and the remaining power of the battery system; A p is the average deceleration of the p-th braking process, measured after eliminating the centrifugal force interference through the six-axis inertia compensation unit; is the historical average deceleration, stored in the system self-calibration database; σ A is the standard deviation of deceleration, which is obtained through statistics of the latest 50 braking processes.

Images