EMB wheel speed signal redundancy processing method based on multi-source data and related equipment thereof

Abstract

The application provides a kind of EMB wheel speed signal redundancy processing method based on multi-source data and its related equipment, it is related to electronic mechanical brake system technical field, this method is by real-time monitoring wheel speed sensor power supply voltage, when voltage drops to first threshold, current harmonic analysis module and vehicle kinematic observer are activated synchronously, and historical data buffer is established, and dual redundancy estimation is started.The current module uses permanent magnet synchronous motor current frequency domain mapping model to analyze the first wheel speed of electric angular velocity acquisition;Kinematic observer combines yaw rate, longitudinal acceleration and steering angle, based on the second wheel speed of dynamic model output.Fully system constructs confidence evaluation matrix, according to the time domain consistency and frequency domain stability of two signal dynamic distribution weight, generate high-precision fusion wheel speed signal.When voltage drops to second threshold, replace failure sensor signal with fusion signal to control EMB actuator brake.The scheme effectively solves the safety hazard of wheel speed loss caused by single line power supply failure.

Description

Technical Field

[0001] This invention relates to the field of electromechanical braking system technology, and in particular to an EMB wheel speed signal redundancy processing method based on multi-source data and related equipment. Background Technology

[0002] In the field of electromechanical braking systems, wheel speed signal acquisition generally relies on Hall effect sensors combined with independent power supply modules. However, this hardware architecture suffers from a major failure source: single-wire power supply. Power supply failure becomes the primary cause of signal loss. Existing technologies have significant limitations in redundancy solutions. Hardware-level dual-power backup significantly increases wiring complexity, while software-level single-signal replacement solutions based on motor speed calculations have large estimation errors under low-traction road conditions. Furthermore, traditional solutions risk simultaneous failure of redundant systems in environments with strong electromagnetic interference, and the protective lock-up mechanism employed after signal interruption leads to an abnormal increase in braking distance. Real-world testing shows that braking performance degrades under normal road conditions, failing to meet safety requirements under extreme operating conditions. Summary of the Invention

[0003] The purpose of this invention is to provide an EMB wheel speed signal redundancy processing method and related equipment based on multi-source data, in order to solve one or more technical problems existing in the prior art, and at least provide a beneficial option or create conditions that enable high-precision redundancy estimation and seamless switching of wheel speed signals by fusing dual data of motor current harmonic analysis and vehicle kinematic observation under extreme working conditions such as wheel speed sensor power failure or signal interruption, thereby ensuring the continuous and stable operation of the electromechanical braking system and driving safety.

[0004] On the one hand, this application provides a method for redundant processing of EMB wheel speed signals based on multi-source data, the method comprising the following steps: Step S100: Monitor the power supply voltage of the wheel speed sensor in real time. When the power supply voltage drops to the first threshold, activate the current harmonic analysis module and the vehicle kinematics observer, and establish a historical data buffer. Step S200: The current harmonic analysis module establishes a frequency domain mapping model of the current and speed of the permanent magnet synchronous motor, analyzes the electric angular velocity by real-time monitoring of the phase current harmonic components, and estimates the first wheel speed signal based on the electric angular velocity. In step S300, the vehicle kinematics observer acquires the vehicle's yaw rate, longitudinal acceleration, and steering angle, and estimates the second wheel speed signal based on the vehicle dynamics model. Step S400: Construct a confidence evaluation matrix, dynamically allocate weight coefficients based on the temporal consistency and frequency domain stability of the first wheel speed signal and the second wheel speed signal, and output the fused wheel speed signal; Step S500: When the power supply voltage drops to the second threshold, the fused wheel speed signal is used to replace the failed wheel speed sensor signal, and the EMB actuator is controlled to brake.

[0005] Furthermore, in step S100, the first threshold is higher than the second threshold; When the power supply voltage drops to the first threshold, the current phase current data is collected by the motor controller, and background harmonic features are pre-extracted based on the phase current data to generate a corresponding pre-loaded analytical model. Set the computation window length of the preloaded parsing model to cover the maximum mechanical response delay duration of the EMB actuator.

[0006] Furthermore, in step S200, the specific process of resolving the electrical angular velocity by real-time monitoring of the phase current harmonic components includes: Establish a frequency domain mapping model of current and speed for a permanent magnet synchronous motor; The current direct-axis current component and quadrature-axis current component of the motor are obtained, the harmonic signals of the direct-axis current component and quadrature-axis current component in a preset specific frequency band are extracted, and the harmonic signals are integrated to obtain the target harmonic energy value. Based on the target harmonic energy value, the rate of change of the spectral energy distribution of the direct-axis current component and the quadrature-axis current component with time is calculated, and the rate of change is input into a preset observer for integration to output the analytically obtained electric angular velocity.

[0007] Furthermore, in step S300, the specific process of estimating the second wheel speed signal based on the vehicle dynamics model includes: A state equation with multiple input sources is constructed, wherein the state equation is a function of the estimated wheel speed equal to the yaw rate, longitudinal acceleration and steering angle plus a road adhesion coefficient compensation term; The road surface adhesion coefficient compensation term is used to correct the slippage effect of tires on low-adhesion roads. The vehicle kinematics observer employs an extended Kalman filter to perform state estimation based on the state equation and the road adhesion coefficient compensation term, in order to output the second wheel speed signal.

[0008] Furthermore, in step S400, the specific process of constructing the confidence assessment matrix and dynamically assigning weight coefficients includes: The confidence scores of current analysis and kinematic observation are acquired in real time, and the basic weights of current analysis and kinematic observation are calculated by combining the preset first dynamic adjustment coefficient and second dynamic adjustment coefficient. The changes in the current analysis signal and the changes in the kinematic observation signal are monitored in real time. When the absolute value of the difference between the two is greater than a preset threshold, a weight attenuation mechanism is triggered to attenuate and correct the first dynamic adjustment coefficient and / or the second dynamic adjustment coefficient. Based on the first dynamic adjustment coefficient and / or the second dynamic adjustment coefficient after attenuation correction, the current analysis basis weight and the kinematic observation basis weight are recalculated, and the recalculated results are used as the final fusion weights.

[0009] Furthermore, in step S500, replacing the failed wheel speed sensor signal with the fused wheel speed signal specifically includes a three-level fault tolerance mechanism: Early warning phase: When the power supply voltage drops to the first threshold, the data fusion module is pre-started and a historical data buffer covering the mechanical response delay of the EMB actuator is established; Switching phase: When the supply voltage drops to the second threshold, the signal source switching is completed within a single pulse width modulation cycle, and a hexadecimal status code is sent via the CAN bus; Minimum protection phase: If both the current analysis signal and the kinematic observation signal fail, the minimum safe braking force is calculated based on the vehicle dynamics model, and the electronic parking brake linkage control protocol is activated.

[0010] Furthermore, the method also includes control logic after fault switching: During the switching phase, when the absolute value of the deviation between the first wheel speed signal and the second wheel speed signal is detected to be greater than a preset safety threshold, a fusion smoothing filtering mechanism is triggered to perform amplitude limiting processing on the fused wheel speed signal based on the data in the historical buffer. During the baseline protection phase, when it is determined that the confidence levels of both the current analysis signal and the kinematic observation signal are lower than the preset failure threshold, the electronic parking brake linkage control protocol is activated, and a message containing the target deceleration parameters is sent to the vehicle chassis controller to provide redundant braking torque using the electronic parking brake system.

[0011] On the other hand, this application provides an EMB wheel speed signal redundancy processing system based on multi-source data, comprising: The voltage monitoring module is used to monitor the voltage status of the wheel speed sensor power supply module; The current harmonic analysis module, whose input is connected to the current sampling module of the motor controller, is used to perform sensorless speed estimation based on the phase current spectrum characteristics of the permanent magnet synchronous motor. The vehicle kinematics observer, whose input is connected to an inertial measurement module and a steering angle sensor, is used to estimate the vehicle's motion state based on multi-source sensor data; The multi-source data fusion electronic control module is used to receive the output signals of the current harmonic analysis module and the vehicle kinematics observer, and to perform the fusion wheel speed signal as described above. The safety control module is used to generate braking control commands based on the fused wheel speed signals when the wheel speed sensor signals fail.

[0012] On the other hand, this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the aforementioned EMB wheel speed signal redundancy processing method based on multi-source data.

[0013] On the other hand, this application provides a vehicle equipped with the aforementioned EMB wheel speed signal redundancy processing system based on multi-source data.

[0014] The beneficial effects of this invention are as follows: This application provides a redundant processing method for EMB wheel speed signals based on multi-source data. This method monitors the wheel speed sensor supply voltage in real time and simultaneously activates the current harmonic analysis module and the vehicle kinematics observer when it drops to a first threshold, and establishes a historical data buffer. A dual redundancy estimation mechanism is initiated at the initial stage of power supply anomalies. The current harmonic analysis module uses a frequency domain mapping model of permanent magnet synchronous motor current and rotational speed to analyze the electrical angular velocity through phase current harmonic components to obtain the first wheel speed signal. The vehicle kinematics observer outputs a second wheel speed signal based on yaw rate, longitudinal acceleration, and steering angle combined with a vehicle dynamics model. A confidence evaluation matrix is ​​then constructed, dynamically allocating weight coefficients based on the temporal consistency and frequency domain stability of the two signals to generate a high-precision fused wheel speed signal. Finally, when the supply voltage further drops to a second threshold, this fused signal directly replaces the failed sensor signal to control the EMB actuator braking. This effectively solves the safety hazard of complete wheel speed signal loss due to single-wire power supply failure in traditional solutions, significantly improving the fault tolerance and driving reliability of the electromechanical braking system under extreme conditions. This application also provides related equipment for the above method. The beneficial effects of the related equipment are similar to those of the above method, and will not be described in detail here.

[0015] Other features and advantages of this application will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the application. The objectives and other advantages of this application may be realized and obtained by means of the structures particularly pointed out in the description, claims and drawings. Attached Figure Description

[0016] The accompanying drawings are provided to further understand the technical solutions of the present invention and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the technical solutions of the present invention, and do not constitute a limitation on the technical solutions of the present invention.

[0017] Figure 1 This is a flowchart of the EMB wheel speed signal redundancy processing method based on multi-source data provided in this application; Figure 2 This is a structural diagram of the EMB wheel speed signal redundancy processing system based on multi-source data provided in this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0019] The present application will be further described below with reference to the accompanying drawings and specific embodiments. The described embodiments should not be considered as limitations on the present application, and all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of the present application.

[0020] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0022] In automotive chassis electronic control systems, wheel speed signals are core input parameters for active safety functions such as anti-lock braking systems (ABS) and electronic stability programs (ESCs). Their accuracy and real-time performance directly determine vehicle stability during emergency braking or driving on low-traction surfaces. With the widespread adoption of drive-by-wire chassis technology, electromechanical braking systems are gradually replacing traditional hydraulic braking as a development trend. However, this system is highly dependent on electrical signals; any abnormality in the wheel speed signal will directly lead to the failure of the braking control strategy, causing serious safety accidents. Therefore, how to obtain highly reliable wheel speed signals even in the event of sensor hardware failure or power supply anomalies is a core problem that urgently needs to be solved in current EMB technology development.

[0023] Existing wheel speed signal acquisition technologies primarily rely on Hall effect sensors or magnetoelectric sensors mounted at the wheel hub, along with a separate power supply module and signal transmission harness to transmit data to the electronic control module. While this hardware-based direct measurement solution can meet accuracy requirements under normal operating conditions, its reliability faces significant challenges in the complex electromagnetic and mechanical vibration environments of automobiles.

[0024] Especially for electromechanical braking systems, wheel speed sensors typically employ a single-wire power supply and signal transmission architecture. Open circuits, short circuits, or voltage drops in the power supply line are the primary causes of signal loss. Furthermore, the sensor probe and signal gear ring are susceptible to interference from contaminants such as mud, water, and metal shavings, leading to changes in the air gap and resulting in signal distortion. Existing technologies typically employ hardware redundancy or simple software compensation strategies to address these risks, but these have revealed significant shortcomings in practical applications.

[0025] The main drawbacks of existing technologies lie primarily in the high cost and low integration of hardware redundancy solutions. To prevent power supply failures, traditional redundancy designs often employ dual power supplies or dual sensor backups. This not only multiplies the complexity of wiring harness layout but also significantly increases system costs, typically raising the material costs of a single vehicle's braking system. With the trend towards lightweight and integrated vehicles, this approach of simply piling on hardware to achieve reliability is no longer sufficient to meet the needs of OEMs.

[0026] Secondly, existing software estimation methods lack accuracy and are limited in application scenarios. Some existing technologies attempt to calculate wheel speed using the rotational speed signal of the drive motor, but when the vehicle is slipping on a low-traction surface or in a non-drive wheel position, there is a huge difference in slip rate between the motor speed and the actual wheel speed, resulting in a large estimation error. This cannot meet the control requirements of anti-lock braking systems for millisecond-level response and high-precision signals.

[0027] Furthermore, existing technologies have logical flaws in their fault-tolerance mechanisms for extreme operating conditions. When an abnormal wheel speed signal is detected, traditional control strategies typically employ a conservative protective lock-up mechanism, which directly exits the closed-loop control of the electromechanical braking system and issues an alarm, requesting driver intervention. This "one-size-fits-all" approach leads to an abnormal increase in braking distance. Real-world testing data shows that under normal road conditions, the instantaneous drop in braking performance can easily trigger rear-end collisions.

[0028] Meanwhile, in environments with strong electromagnetic interference, traditional single signal sources or simple dual-path backup systems are prone to simultaneous interference and failure. They lack a confidence assessment mechanism based on multi-source data fusion and cannot smoothly transition and compensate in the early stages of signal quality degradation, resulting in drastic fluctuations in the vehicle dynamic control system and seriously affecting driving smoothness and safety.

[0029] To address the aforementioned issues, this application provides a redundant EMB wheel speed signal processing method based on multi-source data. This technical solution monitors the power supply voltage of the wheel speed sensor in real time. When the voltage drops to a first threshold, it immediately activates the current harmonic analysis module and the vehicle kinematics observer, and establishes a historical data buffer. It uses a frequency domain mapping model of permanent magnet synchronous motor current and rotational speed to analyze the phase current harmonic components to obtain the electric angular velocity, thereby estimating the first wheel speed signal. Simultaneously, it combines the vehicle's yaw rate, longitudinal acceleration, and steering angle to estimate the second wheel speed signal based on a dynamic model. Then, it constructs a confidence evaluation matrix and dynamically allocates weight coefficients based on the temporal consistency and frequency domain stability of the two signals to generate a fused wheel speed signal. Finally, when the power supply voltage further drops to a second threshold, it directly uses this fused signal to replace the failed sensor signal to control the electromechanical brake actuator for braking. This achieves high-precision redundant estimation and seamless switching of multi-source data fusion, effectively solving the safety hazard of signal loss caused by single-wire power supply failure.

[0030] Specifically, this application aims to ensure the operational safety and stability of vehicles under complex conditions through multi-dimensional algorithmic innovation and logical coupling. Firstly, at the underlying signal acquisition level, the system employs a sensorless speed estimation method based on the phase current spectrum characteristics of a permanent magnet synchronous motor, namely, the current harmonic analysis algorithm. The innovation of this algorithm lies in abandoning the traditional FFT peak detection method and instead adopting a specific frequency band harmonic energy integration strategy, thereby more accurately capturing the motor's operating state and providing a reliable raw data foundation for subsequent control.

[0031] Building upon this foundation, to address the potential limitations of a single signal source, the system constructs a multi-source data confidence assessment mechanism. This mechanism establishes a two-dimensional assessment matrix encompassing temporal consistency and frequency domain stability indicators. Using this matrix as the core protection focus, a set of fuzzy inference rules for dynamically adjusting confidence weights is designed. This rule set can detect changes in data quality in real time, ensuring the system always trusts the most reliable information source. Complementing this assessment mechanism is a dynamic weight allocation logic, the core of which lies in designing a nonlinear weight decay function based on signal difference rate. To further enhance robustness, this logic introduces historical data trend analysis as a weight correction factor. This innovative approach allows the system to not only respond to the current state but also smoothly transition based on historical trends, avoiding drastic weight jumps caused by instantaneous interference.

[0032] Finally, as the ultimate safety safeguard, the system implements a fault-tolerance control architecture. This architecture achieves a deep coupling design between a three-level fault-tolerance mechanism and the vehicle dynamics model. Its core protection focus lies in the minimum safe braking force calculation model in the backup phase. That is, in extreme fault conditions, the system can still calculate the minimum braking force required to maintain vehicle stability based on the dynamics model, thereby preventing the vehicle from losing control and ensuring a safe landing. These four links are interconnected and together constitute a complete closed-loop protection system from signal estimation, quality assessment, dynamic adjustment to fault backup.

[0033] First, the EMB wheel speed signal redundancy processing method based on multi-source data provided in the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0034] Reference Figure 1 The implementation process of the EMB wheel speed signal redundancy processing method based on multi-source data provided in this application embodiment includes, but is not limited to, the following steps.

[0035] In step S100, the power supply voltage of the wheel speed sensor is monitored in real time. When the power supply voltage drops to the first threshold, the current harmonic analysis module and the vehicle kinematics observer are activated, and a historical data buffer is established.

[0036] In step S100, a fault warning and multi-source data synchronous startup mechanism is established. By monitoring the power supply voltage of the wheel speed sensor in real time, the system can capture early characteristics of voltage drop before complete hardware failure. Once the voltage drops to a preset first threshold, the current harmonic analysis module and vehicle kinematics observer are immediately activated. This tiered triggering strategy avoids the waste of computing power caused by running complex algorithms all the time. At the same time, establishing a historical data buffer can provide a benchmark reference for subsequent signal analysis, ensuring that the control system still has reliable initial state data during the transition period when sensor signals are completely lost, thereby gaining valuable response time for subsequent redundancy estimation.

[0037] In step S200, the current harmonic analysis module establishes a frequency domain mapping model of the current and speed of the permanent magnet synchronous motor, analyzes the electric angular velocity by real-time monitoring of the phase current harmonic components, and estimates the first wheel speed signal based on the electric angular velocity.

[0038] In step S200, a first redundant estimation path is constructed using the internal physical characteristics of the drive system to extract the speed information implicit in the motor current signal. This step establishes a frequency domain mapping model between the current and speed of the permanent magnet synchronous motor, and utilizes the strong coupling relationship between the inherent harmonic components in the phase current and the rotor speed to directly analyze the electrical angular velocity without the need for external position sensors. This method not only eliminates the dependence on traditional wheel speed sensors, but also allows for high-precision speed feedback using the existing current sampling circuit of the motor controller when the sensor power supply is abnormal. This enables the estimation of the first wheel speed signal, which reflects the true motion state of the drive wheels, providing the system with internal observation data based on the power source perspective.

[0039] In step S300, the vehicle kinematics observer acquires the vehicle's yaw rate, longitudinal acceleration, and steering angle, and estimates the second wheel speed signal based on the vehicle dynamics model.

[0040] In step S300, a second redundant estimation path based on the vehicle's macroscopic motion state is constructed, achieving state inversion from the whole vehicle level to the wheel level through the vehicle dynamics model. This step acquires key motion parameters such as the vehicle's yaw rate, longitudinal acceleration, and steering angle, and uses the vehicle's two- or more-degree-of-freedom dynamic equations to calculate the theoretical wheel speeds of the vehicle under a specific trajectory. The significance of this process lies in introducing a third type of data source independent of the drive motor and wheel speed sensors. Even in complex conditions such as vehicle slippage or motor failure, the motion trend of the wheels can be deduced from the vehicle's inertial measurement unit data, thereby estimating the second wheel speed signal and ensuring the independence and diversity of the redundant signal in terms of physical mechanism.

[0041] Step S400: Construct a confidence evaluation matrix, dynamically allocate weight coefficients based on the temporal consistency and frequency domain stability of the first and second round speed signals, and output the fused round speed signal.

[0042] In step S400, multi-source information fusion technology addresses the issue of insufficient reliability of a single estimation source. By constructing a confidence evaluation matrix, the system can analyze the temporal consistency and frequency domain stability of the first and second wheel speed signals in real time, identifying which estimation method is more accurate under the current operating conditions. For example, during severe acceleration and deceleration, the dynamic model is given higher weight, while during stable driving, the optimal fused wheel speed signal is output by dynamically allocating weight coefficients based on the analytical results of the motor current. This adaptive fusion strategy effectively smooths out the estimation error of a single algorithm and significantly improves the robustness and accuracy of redundant signals in complex dynamic environments.

[0043] Step S500: When the power supply voltage drops to the second threshold, the fused wheel speed signal replaces the failed wheel speed sensor signal, and the EMB actuator is controlled to brake.

[0044] In step S500, the system's safety baseline under extreme fault conditions is established, achieving closed-loop fault tolerance from signal monitoring to braking control. When the supply voltage further deteriorates to the second threshold, meaning the wheel speed sensor is completely inoperable, the system decisively uses the fused high-confidence wheel speed signal to replace the failed sensor signal, directly controlling the electromechanical brake actuator to apply brakes. This step completely solves the problem of traditional systems directly exiting control after sensor failure, ensuring that the electromechanical braking system still possesses complete anti-lock braking and stability control capabilities under extreme conditions such as power supply failures, maximizing vehicle driving safety.

[0045] In some embodiments of this application, the first threshold is higher than the second threshold; step S100 specifically includes the following sub-steps: Step S110: When the power supply voltage drops to the first threshold, the current phase current data is collected by the motor controller, and the background harmonic features are pre-extracted based on the phase current data to generate the corresponding pre-loaded analytical model.

[0046] In step S110, the "hot start" and operating condition adaptation of the redundancy algorithm are implemented. This step utilizes the voltage difference between the first and second thresholds as a safety buffer period. Before the sensor signal completely fails, it captures the specific electromagnetic environment characteristics of the motor under the current load and speed, separates the background noise from the effective signal characteristics, and constructs a dedicated analytical model. This effectively avoids the computational delay and adaptation error caused by temporarily loading a general model at the moment of emergency braking or signal loss. It ensures that once a fault state is entered, the system can immediately call the algorithm model that best matches the current operating condition for high-precision calculation, significantly improving the response speed and accuracy of redundancy estimation.

[0047] Step S120: Set the computation window length of the preloaded parsing model to cover the maximum mechanical response delay duration of the EMB actuator.

[0048] In step S120, it is ensured that the redundant signal is strictly synchronized with the physical action of the braking actuator in the time dimension. Because there is an inherent mechanical hysteresis in the electromechanical braking system from receiving control commands to the caliper generating actual clamping force, if the calculation window for signal estimation is too short, it will cause frequent changes in control commands, leading to actuator oscillation. By setting a calculation window that covers the maximum mechanical response delay, the system can smoothly process the estimated data within a complete mechanical response cycle, filtering out high-frequency noise interference, ensuring that the output fused wheel speed signal accurately reflects the actual action trend of the actuator, thereby avoiding misjudgments by the anti-lock braking system due to signal fluctuations and improving the smoothness and stability of the braking process.

[0049] In some embodiments of this application, step S200, which involves analyzing the electrical angular velocity of the phase current harmonic components in real time, includes the following specific steps: Step S210: Establish a frequency domain mapping model of current and speed for the permanent magnet synchronous motor.

[0050] In step S210, a mathematical bridge is constructed between the physical signal and the motion state, laying the theoretical foundation for subsequent sensorless speed measurement. During operation, the stator current of a permanent magnet synchronous motor not only contains the fundamental component used to generate torque, but also inevitably includes higher-order harmonics caused by the motor's structure (such as stator slotting and rotor magnet distribution) and inverter switching operations. These harmonic components are not random noise, but rather have a strictly linear coupling relationship with the motor's rotational speed.

[0051] By establishing a frequency domain mapping model, the system can clearly identify which specific frequency harmonics carry the real rotational speed information, thereby establishing a functional relationship between the current spectrum characteristics and the electric angular velocity at the mathematical level. This makes it possible to "decode" the rotational speed from complex current signals, which is a prerequisite for achieving high-precision signal redundancy estimation.

[0052] Step S220: Obtain the current direct-axis current component and quadrature-axis current component of the motor, extract the harmonic signals of the direct-axis current component and quadrature-axis current component in a preset specific frequency band, and perform energy integration calculation on the harmonic signals to obtain the target harmonic energy value.

[0053] In step S220, the effective rotational speed characteristic signal is accurately extracted from the strong noise background, and its intensity is quantified. In actual vehicle-mounted electrical grid environments, current signals are highly susceptible to electromagnetic interference, and direct analysis of the original waveform often results in significant errors. By decomposing the current vector into direct-axis and quadrature-axis components and filtering them using the specific frequency band determined in step S210, the system can shield fundamental waves and other interference noise unrelated to rotational speed. The subsequent energy integration operation transforms the instantaneous, volatile harmonic amplitude into a more stable energy index with physical statistical significance. This process effectively smooths high-frequency jitter in the signal, highlights the main characteristics, ensures that the data relied upon for subsequent calculations has sufficient signal-to-noise ratio and reliability, and prevents jumps in rotational speed estimation caused by signal glitches.

[0054] Step S230: Calculate the rate of change of the spectral energy distribution of the direct-axis current component and the quadrature-axis current component with time based on the target harmonic energy value, and input the rate of change into a preset observer for integration to output the analytically obtained electric angular velocity.

[0055] Step S230 is a crucial step in realizing the final calculation of physical quantities from signal characteristics. It utilizes dynamic system state observation technology to eliminate estimation errors and output a smooth rotational speed result. The rate of change of the spectral energy distribution over time reflects the dynamic trend of the motor's operating state, directly reflecting the transient process of acceleration or deceleration. However, simple rate of change data often contains accumulated errors and is not intuitive enough.

[0056] Therefore, a pre-defined observer (such as a Romberg observer or a sliding mode observer) is introduced to perform integral calculations and state reconstruction on the rate of change. The observer can utilize a feedback correction mechanism to compensate for the effects of model parameter deviations and external disturbances in real time, transforming discrete, noisy rate of change data into continuous, accurate, and phase-delay-free electrical angular velocity signals. This not only completes the conversion from frequency domain energy to time domain rotational speed but also ensures that the output first wheel speed signal closely follows the vehicle's actual driving state in terms of dynamic response.

[0057] In some embodiments of this application, the motor current harmonic analysis technology employed is used to establish a frequency domain mapping model of the current and speed of a permanent magnet synchronous motor, and to analyze the electric angular velocity by real-time monitoring of the phase current harmonic components. Specifically, the system utilizes the direct-axis current component of the motor. and cross-axis current components The rate of change of the spectral energy distribution over time is used to deduce the angular velocity. That is, satisfying Relationship, This represents the spectral energy distribution. Furthermore, to ensure the accuracy of the estimation, a special anti-interference algorithm was designed to effectively eliminate the impact of high-frequency noise generated by inverter switching operations on signal analysis, thereby ensuring the acquisition of pure characteristic signals even in complex electromagnetic environments.

[0058] The innovative effects of this technology are mainly reflected in its extremely high system redundancy and excellent estimation accuracy. Firstly, it can directly utilize the existing current sampling signals within the motor controller for wheel speed analysis even under extreme conditions such as external power failure, maintaining the acquisition of critical data without relying on additional physical sensors. Secondly, experimental verification shows that this analytical method exhibits excellent performance across a wide speed range of 100 to 1000 RPM, with an analytical accuracy controllable within ±2 RPM, providing high-precision data support for vehicle dynamics control.

[0059] In some embodiments of this application, step S300, the specific process of estimating the second wheel speed signal based on the vehicle dynamics model, includes: constructing a state equation for multi-source input, wherein the state equation is a function of the estimated wheel speed equal to the yaw rate, longitudinal acceleration, and steering angle plus a road surface adhesion coefficient compensation term.

[0060] Specifically, through step S300 above, a vehicle-level motion observation channel independent of the drive system is constructed, and the theoretical speed of the wheels is inferred by fusing the macroscopic dynamic state of the vehicle. This step is no longer limited to local data of a single wheel or motor, but treats the vehicle as a rigid body system, and uses three key physical quantities that can characterize the overall motion trend of the vehicle—yaw rate, longitudinal acceleration, and steering angle—to establish a multi-source input state equation.

[0061] The equation defines the estimated wheel speed as a function of the three physical quantities mentioned above and introduces a road adhesion coefficient compensation term. This means that the system not only considers the vehicle's geometric motion but also fully accounts for the influence of tire-ground contact characteristics on wheel speed. This estimation method based on a dynamic model can use objective motion data provided by the vehicle's inertial measurement unit to calculate the theoretical wheel speed under current road conditions when the drive motor slips, idles, or the sensor signal is severely distorted. This provides the system with a highly valuable second wheel speed signal, ensuring the independence and diversity of redundant signals in terms of physical mechanisms and effectively compensating for the limitations of a single signal source under extreme conditions.

[0062] In some embodiments of this application, the road surface adhesion coefficient compensation term is used to correct the slippage effect of the tire on the low-adhesion road surface, solve the deviation problem between the theoretical motion model and the real physical environment, and especially eliminate the estimation error caused by tire slippage under low-adhesion road surface conditions.

[0063] In an ideal vehicle dynamics model, wheel speed usually has a strict linear relationship with vehicle speed. However, in actual driving, when a vehicle is on a low-traction surface such as ice, snow, wet, or gravel, the friction limit between the tire and the ground is reduced, making it very easy for wheel to slip or lock up and skid. This results in a significant difference between the actual wheel speed and the theoretical speed calculated based on the rigid body motion of the vehicle.

[0064] Introducing a road adhesion coefficient compensation term means that the system can dynamically adjust the correction parameters in the estimation algorithm based on real-time monitored road friction characteristics. When a decrease in adhesion is detected, this compensation term automatically performs nonlinear correction on the estimation results, eliminating spurious velocity components caused by excessive tire slippage. This ensures that the estimated second wheel speed signal accurately reflects the actual motion state of the wheel under current grip conditions, greatly improving the robustness and reliability of the redundant algorithm under extreme road conditions.

[0065] In some embodiments of this application, the vehicle kinematics observer employs an extended Kalman filter to perform state estimation based on the state equation and road adhesion coefficient compensation term, thereby outputting a second wheel speed signal. In this way, an advanced nonlinear optimal estimation algorithm is used to solve the problems of strong coupling and nonlinearity in the vehicle dynamics system.

[0066] Since vehicle driving is a highly dynamic and noisy complex process, simple linear calculations cannot accurately describe the true relationship between yaw rate, longitudinal acceleration and wheel speed. However, the extended Kalman filter has the excellent ability to handle nonlinear systems. It can optimally fuse the theoretical prediction values ​​in the state equation with the real-time correction values ​​provided by the road adhesion coefficient compensation term through a recursive prediction and update mechanism.

[0067] In each iteration, the filter calculates the predicted covariance and updates the Kalman gain, thereby extracting the state estimate closest to the true value from sensor data containing measurement noise. This not only effectively filters out high-frequency interference and random errors in the sensor signal, but also adaptively compensates for dynamic characteristics not modeled in the model, ultimately outputting a smooth second wheel speed signal with high dynamic response characteristics. This ensures that the system can still provide accurate wheel speed references when the vehicle is under drastic changes in operating conditions or when there are deviations in the sensor data.

[0068] In some embodiments of this application, the technical architecture of the vehicle kinematics observer adopts a multi-source input fusion strategy to collect the yaw rate of the inertial measurement unit. Longitudinal acceleration and steering angle As the core state variable, the system constructs its state equation framework through functional relationships. To estimate wheel speed Furthermore, a road adhesion coefficient compensation term is specifically introduced to correct estimation errors caused by tire slippage, thereby adapting to complex road conditions. In addition, an improved Kalman filter is employed at the algorithm level, capable of dynamically correcting observation errors based on real-time data, ensuring the convergence and accuracy of state estimation.

[0069] The innovative design significantly improves the vehicle's perception and stability under extreme conditions. Specifically, it reduces the friction coefficient. On surfaces with low adhesion (less than 0.3), the observer's error was successfully controlled within 3%, ensuring data reliability in skidding scenarios. Simultaneously, the system exhibits strong robustness, supporting stable output under conditions of maximum lateral acceleration of 0.5g, effectively addressing the dynamic challenges of severe vehicle maneuvering or high-speed cornering.

[0070] In some embodiments of this application, step S400, which involves constructing the confidence assessment matrix and dynamically assigning weight coefficients, includes the following specific steps: Step S410: Real-time acquisition of current analysis confidence and kinematic observation confidence, and calculation of current analysis basic weight and kinematic observation basic weight by combining the preset first dynamic adjustment coefficient and second dynamic adjustment coefficient.

[0071] In step S410, an initial trust benchmark for multi-source signal fusion is established. Since the reliability of different redundancy estimation paths varies under different operating conditions, this step scientifically allocates the initial fusion ratio based on the system's objective evaluation of the quality of the two current signals, combined with prior experience (i.e., dynamic adjustment coefficients) of the vehicle's operating environment or sensor characteristics. This confidence-based weight allocation mechanism ensures that, under normal driving conditions, the system prioritizes the use of more accurate and stable signal sources, laying a reasonable starting point for subsequent dynamic adjustments.

[0072] Step S420: Monitor the changes in the current analysis signal and the kinematic observation signal in real time. When the absolute value of the difference between the two is greater than a preset threshold, trigger the weight attenuation mechanism to attenuate and correct the first dynamic adjustment coefficient and / or the second dynamic adjustment coefficient.

[0073] In step S420, an abnormal conflict detection and adaptive degradation defense line is constructed. In actual complex working conditions, if there is a significant divergence between the two estimated signals, it indicates that at least one signal may have been severely interfered with or the model may have diverged. The weight decay mechanism introduced at this time can quickly identify this instability and weaken the influence of that signal by reducing the value of the corresponding dynamic adjustment coefficient. This proactive defensive correction strategy effectively prevents erroneous signals from dominating the fusion result, avoids the braking control system from malfunctioning due to receiving sudden changes or false wheel speed data, and significantly improves the safety of the system under extreme boundary conditions.

[0074] Step S430: Based on the first dynamic adjustment coefficient and / or the second dynamic adjustment coefficient after attenuation correction, recalculate the current analysis basis weight and the kinematic observation basis weight, and use the recalculated results as the final fusion weight.

[0075] In step S430, the closed-loop dynamic fault-tolerant decision-making is completed and a highly reliable fused wheel speed signal is output. After the aforementioned abnormal conflict detection and attenuation correction, the system has automatically eliminated unreliable signal components. The recalculation at this point is equivalent to a real-time self-correction and optimization of the fusion strategy. Using this corrected result as the final fusion weight ensures that the output fused wheel speed signal is always dominated by the most reliable data source. This allows the system to provide accurate, smooth, and seamless control input to the electromechanical brake actuator even under hardware failures or harsh operating conditions, perfectly achieving the high fault tolerance goal of multi-source data redundancy processing.

[0076] In some embodiments of this application, an adaptive weighted fusion algorithm is employed, wherein the core of the dynamic weight allocation logic lies in constructing a confidence evaluation matrix, which is achieved through the formula... To quantify the overall credibility of the system. Among them, This represents the current resolution. Represents the kinematic observation of position, while and This is a coefficient that is dynamically adjusted based on real-time operating conditions. Based on this, the system is designed with sophisticated fuzzy control rules to handle complex scenarios: when the current analyzes the change in angular velocity... With change in kinematic angular velocity When the absolute value of the difference exceeds the preset first threshold, the algorithm will automatically activate the weight decay mechanism to reduce the impact of the conflict signal; and when the power supply voltage drops to the second threshold, the system will be forced to switch to fusion mode to ensure that the core functions can still be maintained when the power supply is unstable.

[0077] This algorithm demonstrates superior performance in practical applications, primarily in terms of response speed and stability. Firstly, it achieves millisecond-level dynamic weight adjustment response, with the adjustment time controlled within 20ms, enabling rapid adaptation to drastic changes in vehicle status. Secondly, in extreme scenarios involving signal conflicts, the algorithm significantly optimizes output quality, improving output stability by 40%. This effectively avoids control jitter or misjudgment caused by fluctuations in a single signal source, providing a solid guarantee for smooth vehicle operation.

[0078] In some embodiments of this application, step S500, which replaces the failed wheel speed sensor signal with the fused wheel speed signal, specifically includes a three-level fault tolerance mechanism: In step S510, during the early warning phase, when the power supply voltage drops to the first threshold, the data fusion module is pre-started, and a historical data buffer with a length covering the mechanical response delay of the EMB actuator is established.

[0079] In step S510, the valuable time window during the initial stage of voltage drop is utilized to achieve seamless hot backup and state prediction of the system. The first threshold serves as a warning line, indicating that the system is about to enter an unstable state. Activating the data fusion module in advance at this time can avoid the calculation delay caused by the system's cold start at the moment of failure. At the same time, establishing a historical data buffer covering the mechanical response delay of the electromechanical brake actuator ensures that the control algorithm can backtrack and call up historical state data that matches the mechanical action time at the moment of signal switching. This proactive preparation eliminates control oscillations that may be caused by the asynchrony between data processing and mechanical action, ensuring that the redundant system is in a standby state ready to take over control at any time before the sensor completely fails, laying a solid data foundation for subsequent smooth switching.

[0080] In step S520, during the switching phase, when the power supply voltage drops to the second threshold, the signal source switching is completed within a single pulse width modulation cycle, and a hexadecimal status code is sent via the CAN bus.

[0081] In step S520, extremely fast and transparent fault isolation and information synchronization are achieved. The second threshold typically represents the critical voltage at which the sensor can no longer maintain normal operation. At this point, the system must cut off its dependence on the failed physical sensor within a very short time. Limiting the signal source switching to a single pulse width modulation cycle means that the switching process is almost instantaneous for the upper-level control algorithm, effectively avoiding brake pressure fluctuations caused by signal interruptions or jumps.

[0082] Meanwhile, sending specific hexadecimal status codes via the CAN bus is to broadcast the current fault level and operating mode change to the vehicle controller and other relevant subsystems. This real-time information exchange ensures coordinated response at the vehicle level, preventing misjudgments in other systems due to changes in the braking system's status, thus simultaneously guaranteeing vehicle safety at both the low-level execution and high-level communication levels.

[0083] In step S530, during the safety phase, if both the current analysis signal and the kinematic observation signal fail, the minimum safe braking force is calculated based on the vehicle dynamics model, and the electronic parking brake linkage control protocol is activated.

[0084] In step S530, controlled stopping of the vehicle is achieved under extreme, completely blind conditions. When both the analytical signal based on motor current and the observation signal based on vehicle dynamics fail simultaneously, it means that the system has lost its ability to directly and indirectly estimate wheel speed. In this situation, blind braking may lead to loss of vehicle control. Therefore, the system instead uses the vehicle dynamics model to calculate the minimum braking force required to maintain vehicle stability, avoiding wheel lock-up or skidding. At the same time, the electronic parking brake linkage control protocol is activated, utilizing the independent power supply or mechanical self-locking characteristics typically found in electronic parking brake systems to assist in deceleration. This cross-system linkage strategy ensures that even in extreme cases where the main braking control system completely loses its sensing capability, the vehicle can still decelerate and stop in the safest way allowed by physical limits, maximizing occupant safety.

[0085] In some embodiments of this application, the above method further includes control logic after fault switching: (1) During the switching phase, when the absolute value of the deviation between the first wheel speed signal and the second wheel speed signal is detected to be greater than the preset safety threshold, the fusion smoothing filtering mechanism is triggered. Based on the data in the historical buffer, the fusion wheel speed signal is subjected to amplitude limiting processing to solve the step impact problem that may occur at the moment of signal source switching and to ensure the dynamic stability of the control system.

[0086] Since the first wheel speed signal typically originates from physical sensors, while the second wheel speed signal comes from an estimation model, there are inherent differences in their physical characteristics and response frequencies, making them prone to numerical jumps during fault switching. By introducing a fusion smoothing filter mechanism and combining it with historical buffer data, the system can use the stable state before the fault as a benchmark to reasonably limit and smoothly transition the signal after the abrupt change. This processing method effectively filters out high-frequency noise and logic errors caused by inconsistent signal sources, prevents brake pressure oscillations or false triggering of the anti-lock braking system caused by drastic fluctuations in wheel speed signals, and achieves a seamless and smooth transition from fault signals to redundant signals.

[0087] (2) During the backup phase, when the confidence levels of both the current analysis signal and the kinematic observation signal are lower than the preset failure threshold, the electronic parking brake linkage control protocol is activated, and a message containing the target deceleration parameters is sent to the vehicle chassis controller. The electronic parking brake system provides redundant braking torque to build the ultimate safety barrier under the perception failure of the entire system, ensuring that the vehicle still has controllable deceleration capability under extreme conditions.

[0088] When both current analysis and kinematic observation, the two main redundant estimation paths, are deemed unreliable, it means that the vehicle has fallen into a "blind driving" state where wheel speed cannot be accurately determined. At this point, continuing to rely on conventional braking may pose a risk of lock-up and loss of control.

[0089] Therefore, the system sends specific target deceleration parameters to the chassis controller via cross-domain communication, triggering the electronic parking brake system, which is typically independent of the service braking system. Utilizing the high redundancy and mechanical locking characteristics unique to the electronic parking brake system, it provides auxiliary braking torque. This strategy not only avoids dependence on failed wheel speed signals but also forces the vehicle to safely decelerate and stop within physical limits through multi-actuator collaborative operation, maximizing the safety of personnel and the vehicle.

[0090] In some embodiments of this application, the three-level fault-tolerance mechanism of the fault-switching control logic ensures the safety and continuity of the system under abnormal operating conditions through a graded response strategy. During the early warning phase, when the supply voltage drops to threshold A, the system pre-starts the data fusion module to prepare for takeover and establishes a historical data buffer of no less than 100ms to retain critical time window information for subsequent processing. After entering the switching phase, if the supply voltage further drops to threshold B, the system requires a rapid and seamless switching of the signal source within a single PWM cycle, while simultaneously sending a 0x5F3 message to the CAN bus to inform other nodes of the current status code. In the most extreme fallback phase, i.e., when both signal sources fail, the system calculates the minimum safe braking force required to maintain vehicle stability based on the vehicle dynamics model and immediately activates the EPB linkage control protocol as a final safety line to prevent vehicle loss of control.

[0091] In some embodiments of this application, the signal switching timing is divided into stages as follows: the signal switching process is precisely divided into four consecutive time windows to ensure a smooth and reliable system response. In the T0-T1 stage (0-20ms), the system primarily monitors the wheel speed sensor power supply voltage in real time and simultaneously performs current harmonic background analysis to preload the model, laying the foundation for subsequent operations. Then, in the T1-T2 stage (20-50ms), once the voltage drops to the first threshold, the system immediately activates the IMU data channel and initiates a historical data caching mechanism with a window length of not less than 100ms to retain critical transition information. Next, in the T2-T3 stage (50-100ms), when the voltage further drops to the second threshold, the system triggers a dynamic weighted fusion algorithm to process multi-source data and simultaneously sends a warning signal with status code 0x5F3 to the CAN bus to report the anomaly. Finally, in the T3-T4 stage (100-150ms), the system completes the final seamless signal switching and updates the brake controller input source flag, marking the end of the entire switching process and the establishment of a new steady state.

[0092] In some embodiments of this application, in an example scenario of power outage when crossing a speed bump, the test setup is that the vehicle is traveling at 60 km / h on a concrete road, and the rear wheels experience a sudden power failure when crossing the speed bump. Under this condition, the system exhibits excellent responsiveness, rapidly switching to the fused signal mode within 50ms, ensuring that the wheel speed signal fluctuation rate is controlled within 2%, effectively avoiding the failure problem of traditional solutions under such impacts, and reducing the probability of triggering EPB linkage by 80%. In the strong electromagnetic interference environment test of Scenario 2, the system was verified under stringent parameters of 200V / m radiated field strength and 20-1000MHz frequency band. The results show that the number of signal interruptions of this solution is 0, significantly better than the traditional solution's performance of 3-5 times per minute, and the maximum phase deviation is strictly limited to within 0.5rad, demonstrating its extremely strong anti-interference performance.

[0093] In some embodiments of this application, the system defines key message interaction mechanisms in the design of the CAN communication framework. The message with message ID 0x5F3 is primarily used to transmit the redundancy mode status; its high two bits (Bits 0-1) are defined as the signal source flag, while the next two bits (Bits 2-3) indicate the confidence level. Another message with ID 0x6A1 is responsible for brake controller feedback, specifically containing key information such as the fusion signal acquisition status. Furthermore, the system has established clear DTC code rules to assist in fault diagnosis: code C1201 represents current harmonic resolution error, C1202 indicates kinematic observer out-of-step, and C1203 is specifically used to mark weight allocation conflicts. These codes together constitute the system's underlying monitoring logic.

[0094] Secondly, refer to Figure 2 This application provides an EMB wheel speed signal redundancy processing system based on multi-source data. The system's workflow begins with three parallel input modules: a voltage monitoring module monitors the power supply status in real time, while a current harmonic analysis module and a vehicle kinematics observer provide two independent wheel speed estimation signals based on motor current characteristics and a vehicle dynamics model, respectively. These three signals are then fed into a multi-source data fusion electronic control module, where confidence assessment and dynamic weight allocation are performed to generate a high-precision fused wheel speed signal. The fused signal is transmitted to a safety control module, where a three-level fault-tolerant mechanism is used for final safety verification and braking force decision-making. Finally, the control command is sent to the EMB actuator to complete the braking action, thus constructing a complete closed-loop control link from signal perception and fusion processing to safe execution.

[0095] In some embodiments of this application, a voltage monitoring module is used to monitor the voltage status of the wheel speed sensor power supply module. The core function of the voltage monitoring module is to act as an early warning system for the entire redundancy processing system, accurately capturing early characteristics of power supply anomalies by continuously monitoring the voltage status of the wheel speed sensor power supply module. This module can identify voltage drop trends before complete hardware failure, thereby gaining valuable response time for the early activation and initialization of subsequent multi-source data estimation algorithms, ensuring that the system can seamlessly take over control the moment a fault occurs.

[0096] In some embodiments of this application, the input of the current harmonic analysis module is connected to the current sampling module of the motor controller, and is used to perform sensorless speed estimation based on the phase current spectrum characteristics of the permanent magnet synchronous motor. This module deeply mines the speed information hidden in the motor current signal, eliminating the dependence on traditional external position sensors, and can still provide internal observation data reflecting the true motion state of the drive wheels even when the sensor power supply is abnormal.

[0097] In some embodiments of this application, the input end of the vehicle kinematics observer is connected to an inertial measurement module and a steering angle sensor for estimating the vehicle's motion state based on multi-source sensor data. This module utilizes key motion parameters such as the vehicle's yaw rate and longitudinal acceleration, combined with a dynamic model, to calculate the theoretical speed of the wheels under current road conditions. This ensures the independence of redundant signals in terms of physical mechanisms and effectively compensates for the limitations of a single signal source under extreme conditions.

[0098] In some embodiments of this application, the multi-source data fusion electronic control module receives the output signals from the current harmonic analysis module and the vehicle kinematics observer, and performs the fused wheel speed signal as described above. This module dynamically allocates weighting coefficients and performs conflict detection and smoothing filtering by evaluating the confidence and time-frequency stability of the two estimated signals in real time, ultimately outputting an optimal wheel speed reference value that combines high accuracy and robustness, thus solving the problem of insufficient reliability of a single estimation source.

[0099] In some embodiments of this application, the safety control module is used to generate braking control commands based on the fused wheel speed signals when the wheel speed sensor signals fail. This module achieves closed-loop fault tolerance from the underlying signal monitoring to the upper-level braking control, ensuring that even under extreme fault conditions, the electromechanical braking system can still obtain reliable wheel speed feedback and maintain complete anti-lock braking and stability control functions, maximizing driving safety.

[0100] Furthermore, embodiments of this application provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the aforementioned EMB wheel speed signal redundancy processing method based on multi-source data.

[0101] Furthermore, embodiments of this application provide a vehicle equipped with the aforementioned EMB wheel speed signal redundancy processing system based on multi-source data.

[0102] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards of the relevant countries and regions. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirects to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data for the proper functioning of the embodiments of this application obtained.

[0103] In some alternative embodiments, the functions / operations mentioned in the block diagrams may not occur in the order shown in the operation diagrams. For example, depending on the functions / operations involved, two consecutively shown blocks may actually be executed substantially simultaneously, or the blocks may sometimes be executed in reverse order. Furthermore, the embodiments presented and described in the flowcharts of this application are provided by way of example to provide a more comprehensive understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is changed and sub-operations described as part of a larger operation are executed independently.

[0104] Furthermore, although this application is described in the context of functional modules, it should be understood that, unless otherwise stated, one or more of the functions and / or features may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in a separate physical device or software module. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding this application. Rather, given the properties, functions, and internal relationships of the various functional modules in the apparatus disclosed herein, the actual implementation of the module will be understood within the scope of ordinary skill of an engineer. Therefore, those skilled in the art can implement the application set forth in the claims using ordinary skill. It is also understood that the specific concepts disclosed are merely illustrative and are not intended to limit the scope of this application, which is determined by the full scope of the appended claims and their equivalents.

[0105] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several programs to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0106] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequential list of executable programs for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, a program execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can retrieve and execute a program from or in conjunction with such a program execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can mean any means that can contain, store, communicate, propagate, or transmit a program for use by or in conjunction with a program execution system, apparatus, or device.

[0107] More specific examples (a non-exhaustive list) of computer-readable media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpreting, or, if necessary, processing in a suitable manner, and then stored in computer memory.

[0108] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable program execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0109] In the foregoing description of this specification, the reference to terms such as "one embodiment / implementation," "another embodiment / implementation," or "certain embodiments / implementations," etc., indicates that a specific feature, structure, material, or characteristic described in connection with an embodiment or example is included in an embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0110] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

[0111] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of the present invention.

Claims

1. A method for redundant processing of EMB wheel speed signals based on multi-source data, characterized in that, The method includes the following steps: Step S100: Monitor the power supply voltage of the wheel speed sensor in real time. When the power supply voltage drops to the first threshold, activate the current harmonic analysis module and the vehicle kinematics observer, and establish a historical data buffer. Step S200: The current harmonic analysis module establishes a frequency domain mapping model of the current and speed of the permanent magnet synchronous motor, analyzes the electric angular velocity by real-time monitoring of the phase current harmonic components, and estimates the first wheel speed signal based on the electric angular velocity. In step S300, the vehicle kinematics observer acquires the vehicle's yaw rate, longitudinal acceleration, and steering angle, and estimates the second wheel speed signal based on the vehicle dynamics model. Step S400: Construct a confidence evaluation matrix, dynamically allocate weight coefficients based on the temporal consistency and frequency domain stability of the first wheel speed signal and the second wheel speed signal, and output the fused wheel speed signal; Step S500: When the power supply voltage drops to the second threshold, the fused wheel speed signal is used to replace the failed wheel speed sensor signal, and the EMB actuator is controlled to brake.

2. The method according to claim 1, characterized in that, In step S100, the first threshold is higher than the second threshold; When the power supply voltage drops to the first threshold, the current phase current data is collected by the motor controller, and background harmonic features are pre-extracted based on the phase current data to generate a corresponding pre-loaded analytical model. Set the computation window length of the preloaded parsing model to cover the maximum mechanical response delay duration of the EMB actuator.

3. The method according to claim 1, characterized in that, In step S200, the specific process of analyzing the electrical angular velocity by real-time monitoring of the phase current harmonic components includes: Establish a frequency domain mapping model of current and speed for a permanent magnet synchronous motor; The current direct-axis current component and quadrature-axis current component of the motor are obtained, the harmonic signals of the direct-axis current component and quadrature-axis current component in a preset specific frequency band are extracted, and the harmonic signals are integrated to obtain the target harmonic energy value. Based on the target harmonic energy value, the rate of change of the spectral energy distribution of the direct-axis current component and the quadrature-axis current component with time is calculated, and the rate of change is input into a preset observer for integration to output the analytically obtained electric angular velocity.

4. The method according to claim 1, characterized in that, In step S300, the specific process of estimating the second wheel speed signal based on the vehicle dynamics model includes: A state equation with multiple input sources is constructed, wherein the state equation is a function of the estimated wheel speed equal to the yaw rate, longitudinal acceleration and steering angle plus a road adhesion coefficient compensation term; The road surface adhesion coefficient compensation term is used to correct the slippage effect of tires on low-adhesion roads. The vehicle kinematics observer employs an extended Kalman filter to perform state estimation based on the state equation and the road adhesion coefficient compensation term, in order to output the second wheel speed signal.

5. The method according to claim 1, characterized in that, In step S400, the specific process of constructing the confidence assessment matrix and dynamically allocating weight coefficients includes: The confidence scores of current analysis and kinematic observation are acquired in real time, and the basic weights of current analysis and kinematic observation are calculated by combining the preset first dynamic adjustment coefficient and second dynamic adjustment coefficient. The changes in the current analysis signal and the changes in the kinematic observation signal are monitored in real time. When the absolute value of the difference between the two is greater than a preset threshold, a weight attenuation mechanism is triggered to attenuate and correct the first dynamic adjustment coefficient and / or the second dynamic adjustment coefficient. Based on the first dynamic adjustment coefficient and / or the second dynamic adjustment coefficient after attenuation correction, the current analysis basis weight and the kinematic observation basis weight are recalculated, and the recalculated results are used as the final fusion weights.

6. The method according to claim 1, characterized in that, In step S500, replacing the failed wheel speed sensor signal with the fused wheel speed signal specifically includes a three-level fault tolerance mechanism: Early warning phase: When the power supply voltage drops to the first threshold, the data fusion module is pre-started and a historical data buffer covering the mechanical response delay of the EMB actuator is established; Switching phase: When the supply voltage drops to the second threshold, the signal source switching is completed within a single pulse width modulation cycle, and a hexadecimal status code is sent via the CAN bus; Minimum protection phase: If both the current analysis signal and the kinematic observation signal fail, the minimum safe braking force is calculated based on the vehicle dynamics model, and the electronic parking brake linkage control protocol is activated.

7. The method according to claim 6, characterized in that, The method also includes control logic after fault switching: During the switching phase, when the absolute value of the deviation between the first wheel speed signal and the second wheel speed signal is detected to be greater than a preset safety threshold, a fusion smoothing filtering mechanism is triggered to perform amplitude limiting processing on the fused wheel speed signal based on the data in the historical buffer. During the baseline protection phase, when it is determined that the confidence levels of both the current analysis signal and the kinematic observation signal are lower than the preset failure threshold, the electronic parking brake linkage control protocol is activated, and a message containing the target deceleration parameters is sent to the vehicle chassis controller to provide redundant braking torque using the electronic parking brake system.

8. A redundant processing system for EMB wheel speed signals based on multi-source data, characterized in that, include: The voltage monitoring module is used to monitor the voltage status of the wheel speed sensor power supply module; The current harmonic analysis module, whose input is connected to the current sampling module of the motor controller, is used to perform sensorless speed estimation based on the phase current spectrum characteristics of the permanent magnet synchronous motor. The vehicle kinematics observer, whose input is connected to an inertial measurement module and a steering angle sensor, is used to estimate the vehicle's motion state based on multi-source sensor data; A multi-source data fusion electronic control module is used to receive the output signals of the current harmonic analysis module and the vehicle kinematics observer, and to perform the method described in any one of claims 1 to 7 to calculate the fused wheel speed signal; The safety control module is used to generate braking control commands based on the fused wheel speed signals when the wheel speed sensor signals fail.

9. An electronic device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the EMB wheel speed signal redundancy processing method based on multi-source data as described in any one of claims 1 to 7.

10. A vehicle, characterized in that, The vehicle is equipped with the EMB wheel speed signal redundancy processing system based on multi-source data as described in claim 8.

Images