A vehicle multi-path signal intelligent acquisition and cooperative control system

Abstract

The application relates to the field of vehicle electronic control, and discloses a vehicle multi-path signal intelligent acquisition and cooperative control system, which comprises a multi-path signal acquisition module, a working condition recognition module, an acquisition scheduling module, a signal verification module, a signal reconstruction module and a cooperative control module. The acquisition scheduling module determines the acquisition trigger time of the associated signals according to the transmission time delay of each signal, so that the actual generation time of the associated signals falls within the same time window. The signal verification module performs cross-consistency verification on the associated signals, and distinguishes the transmission time delay drift and signal abnormality. The signal reconstruction module reconstructs the vehicle state quantity corresponding to the marked signal based on the unmarked measured associated signals. The cooperative control module generates a cooperative control instruction, and updates the transmission time delay based on the generation time and the arrival time of the response signal, thereby forming a closed loop of acquisition, verification, control and feedback updating.

Description

Technical Field

[0001] This invention relates to the field of vehicle electronic control technology, specifically to a vehicle multi-channel signal intelligent acquisition and collaborative control system. Background Technology

[0002] As vehicles become more electrified and intelligent, multiple signals, such as vehicle speed, accelerator pedal opening, braking, steering angle, motor speed, and battery status, need to be collected during vehicle operation. These signals are then coordinated and controlled by multiple vehicle actuators, such as drive motors, brake actuators, and steering actuators. These signals include both sensor signals transmitted via hardwired lines and control unit status signals transmitted via the vehicle bus. The rate of change, transmission path, and transmission delay of each signal are different.

[0003] In existing technologies, a unified clock is typically used to add timestamps to each signal and signals are collected at fixed intervals. During signal verification, the abnormality of a signal is usually determined by comparing the values ​​between redundant signals or by model estimation results. Abnormal signals are then isolated or replaced by estimated values. Some vehicle network systems also measure signal propagation time and detect clock synchronization status to determine whether sensor data is valid, and separately detect the response of the execution unit after the control command is issued.

[0004] However, existing technologies typically do not adjust the acquisition trigger time according to the transmission delay of each signal, which means that signals within the same time window may not correspond to the same physical time of the vehicle; when comparing signals directly, transmission delay drift is easily misjudged as signal abnormality; when there are only two related signals and they are inconsistent, it is also difficult to determine the abnormal one; if the reconstructed signal continues to participate in the verification, it will also cause error propagation.

[0005] Therefore, the present invention provides a vehicle multi-channel signal intelligent acquisition and collaborative control system. Summary of the Invention

[0006] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0007] The technical solution adopted by this invention to solve its technical problem is: a vehicle multi-channel signal intelligent acquisition and collaborative control system, comprising: The multi-channel signal acquisition module is used to acquire multiple signals from the vehicle according to the acquisition cycle of each signal and add timestamps; The operating condition identification module is used to identify the vehicle's operating condition based on multiple signals; The acquisition and scheduling module is used to determine the acquisition trigger time of each associated signal according to the transmission delay of each signal, so that the actual generation time of each associated signal falls into the same time window; the associated signal is a signal that represents the state of the same vehicle or the state of the same vehicle component. The signal verification module is used to subtract the transmission delay of the corresponding acquisition channel from the timestamp to obtain the actual generation time, group the associated signals within the same time window according to the acquisition channel and perform cross-consistency verification; if the associated signal group of the same acquisition channel is offset in the same direction relative to other associated signal groups, it is determined that the transmission delay drift of the acquisition channel is determined and corrected according to the time offset of the corresponding change point; if the deviation of the vehicle status quantity of one signal after unified characterization exceeds the allowable deviation range determined by the measurement error among three or more associated signals, it is marked as an abnormal deviation signal. The signal reconstruction module is used to call the pre-established signal mapping relationship according to the operating conditions, and reconstruct the corresponding vehicle state quantity based on the unmarked measured associated signal corresponding to the deviation abnormal signal, and replace the deviation abnormal signal. The collaborative control module is used to generate collaborative control commands carrying target state quantities based on unlabeled signals and reconstructed vehicle state quantities, and send them to the corresponding vehicle execution unit. During the response period after the command is issued, the response signal is included in the corresponding time window for verification according to its actual generation time, and the transmission delay of the corresponding acquisition channel is updated according to the difference between the generation time and the arrival time carried by the response signal. The corrected or updated transmission delay is used to determine the acquisition trigger time and calculate the actual generation time.

[0008] Preferably, the multi-channel signal acquisition module includes multiple acquisition channels, and each signal is acquired through a corresponding acquisition channel; the multiple signals include vehicle sensor signals, bus signals, and response signals of vehicle execution units; each acquisition channel adds a timestamp to the acquired signal based on the same clock source, and the timestamp is the time when the signal arrives at the acquisition channel; each vehicle execution unit writes the generation time to the generated response signal based on the same clock source.

[0009] Preferably, the operating conditions include starting condition, acceleration condition, cruising condition, deceleration condition and idling condition; The operating condition identification module identifies the operating condition based on the vehicle speed signal, accelerator pedal opening signal, and brake pedal opening signal; After the system is powered on, the operating condition identification module continuously identifies the operating conditions. During the period before the initial operating condition is confirmed, the operating condition migration is not triggered. When two consecutive identical identification results appear for the first time, the operating condition corresponding to the identification result is confirmed as the initial operating condition. When the initial operating condition has been confirmed, if two adjacent identification results are identical, the operating condition corresponding to the identification result is confirmed as the current operating condition of the vehicle. If two adjacent identification results are inconsistent, the originally confirmed operating condition remains unchanged until two adjacent identification results are identical.

[0010] Preferably, the method for determining the acquisition trigger time of the acquisition scheduling module is as follows: For each associated signal, determine its acquisition time within the same time window, and subtract the transmission delay of the corresponding acquisition channel from each acquisition time to obtain the acquisition trigger time; the width of the same time window is determined according to the maximum value in the acquisition period of each associated signal.

[0011] Preferably, the signal verification module performs verification according to the following steps: S501. Subtract the transmission delay of the corresponding acquisition channel from the timestamp to obtain the actual generation time, and select the associated signals whose actual generation time falls within the same time window. S502. When each associated signal represents the same vehicle state quantity, the vehicle state quantity represented by each associated signal is directly compared; when each associated signal represents the state of the same vehicle component, one signal is mapped to the vehicle state quantity represented by another signal according to the signal mapping relationship before comparison. S503. When the deviation of the vehicle state quantity represented by the two signals being compared is greater than the sum of the measurement errors of the two signals, and the signals within one group of associated signals are consistent with each other, and the corresponding change points of the associated signals in this group are offset in the same direction in time relative to the corresponding change points of other associated signal groups, it is determined that the transmission delay drift of the acquisition channel corresponding to the associated signals in this group is corrected according to the time offset of the corresponding change points. S504. Recalculate the actual generation time based on the corrected transmission delay; if the verification deviation is not greater than the sum of measurement errors, confirm that the corresponding signal is normal; if the verification deviation is greater than the sum of measurement errors and there are more than three associated signals, mark the one with deviations from all other associated signals that exceed the allowable deviation range as a deviation abnormal signal. When there are two associated signals or when the abnormal signal cannot be determined based on the consistency of the associated signals, the expected vehicle state quantity, which is converted from the target state quantity carried by the reference coordinated control command through the signal mapping relationship, is used as the benchmark, and the one with the deviation exceeding the allowable deviation range is marked as a deviation abnormal signal.

[0012] Preferably, the signal verification module further includes: The fluctuation components obtained by high-pass filtering each signal are divided into m frequency bands. The power of each frequency band and the total power are calculated. When the total power is equal to zero, the signal is marked as a fluctuation abnormal signal. When the total power is not equal to zero, the power distribution is calculated and compared with the power distribution of the normal noise characteristics corresponding to the operating condition. When the deviation exceeds the allowable deviation range determined by the calibration result of the corresponding operating condition, the signal is marked as a fluctuation abnormal signal. Once any signal is marked, the acquisition scheduling module adjusts the acquisition period of that signal and its associated signals to a preset verification acquisition period, which is shorter than the acquisition period used before the signal was marked. If the verification results of that signal are all abnormal within a preset number of consecutive time windows, the abnormal state of that signal is confirmed to continue, and the corresponding mark is retained. The signal reconstruction module then reconstructs the corresponding vehicle state quantity. If the verification results of that signal are all normal within a preset number of consecutive time windows, the previous mark is determined to be caused by a transient disturbance, and the corresponding mark is cleared. After the verification is completed, the acquisition period of that signal and its associated signals is restored to the acquisition period used before the signal was marked.

[0013] Preferably, the signal reconstruction module calculates the unmarked measured associated signals according to the invoked signal mapping relationship to obtain the vehicle state quantity corresponding to the deviation abnormal signal or fluctuation abnormal signal that maintains the marking; when the reconstructed vehicle state quantity replaces the corresponding marked signal, a reconstruction identifier is added to the reconstructed vehicle state quantity. The vehicle state quantity with the reconstruction identifier does not participate in cross-consistency verification, nor is it used as the input for the reconstruction of other marked signals.

[0014] Preferably, the signal reconstruction module establishes and invokes the signal mapping relationship in the following way: The representative conversion coefficients for each operating condition partition are predetermined. Operating conditions whose differences between the corresponding conversion coefficients and the representative conversion coefficients are within the allowable fluctuation range are classified into the same operating condition partition. Operating conditions within the same operating condition partition share the corresponding signal mapping relationship. When an operating condition is migrated to another operating condition partition, the alternative mapping relationship corresponding to the operating condition partition to which the migrated operating condition belongs is called. During the migration transition period, the same vehicle state quantity is reconstructed in parallel using the original signal mapping relationship and the alternative mapping relationship. When the deviation between the two reconstruction results is within the allowable deviation range determined according to the measurement error, the alternative mapping relationship is enabled. When the deviation still exceeds the allowable deviation range at the end of the migration transition period, the alternative mapping relationship is not enabled, the reconstructed vehicle state quantity is marked as invalid, and the operating condition is reviewed. The migration transition period is a preset period calculated from the start of the operating condition migration.

[0015] Preferably, the collaborative control module generates collaborative control instructions and determines the execution status according to the following steps: S901. Determine the target driving state based on the unlabeled signal and the reconstructed vehicle state quantity, and generate cooperative control instructions corresponding to each vehicle execution unit. The cooperative control instructions carry the target state quantity. The vehicle execution unit includes a drive motor, a brake actuator, and a steering actuator. S902. The response signal and the signal representing the same vehicle state quantity or the same vehicle component state among the multiple signals are used as associated signals. After unified representation, cross-consistency verification is performed. When the deviation between the response signal and the other associated signals exceeds the allowable deviation range, the execution link of the corresponding vehicle execution unit is determined to be abnormal. S903. When the response signal enters the response range determined by the measurement error of the corresponding response signal based on the target state quantity during the response period, the execution is deemed normal; otherwise, the execution is deemed abnormal.

[0016] Preferably, the collaborative control module determines performance degradation and updates transmission latency in the following way: When the response signal enters the response range, the reference rise time stored in advance according to the type of the coordinated control instruction is retrieved; when the rise time of the response signal exceeds the allowable time range determined according to the reference rise time and normal response fluctuation, the corresponding vehicle execution unit performance is determined to be degraded and the number of degradations is accumulated. The response time period corresponding to the subsequent coordinated control instruction is extended according to the preset step size based on the number of degradations. The response signal carries the generation time written by the vehicle execution unit based on the same clock source; the difference between the time when the response signal arrives at the corresponding acquisition channel and the generation time is used as the measured delay of the corresponding acquisition channel, and the transmission delay of the corresponding acquisition channel is updated after being weighted and fused with the current transmission delay; the fusion weight of the measured delay is less than the fusion weight of the current transmission delay, and is determined according to the reliability of the corresponding acquisition channel. The reliability is generated according to the number of transmission delay corrections within a preset statistical period and the number of times the mark is maintained after verification.

[0017] This invention provides a vehicle multi-channel signal intelligent acquisition and collaborative control system. It has the following beneficial effects: 1. This invention determines the acquisition trigger time of associated signals according to the transmission delay of each signal, so that associated signals of different transmission paths correspond to the same physical time of the vehicle, avoiding signal time misalignment caused by transmission delay differences, and providing a unified time reference for subsequent cross-consistency verification.

[0018] 2. This invention can identify the same-direction offset of the entire group of associated signals as the transmission delay drift of the acquisition channel, and identify the deviation of a single signal as a signal abnormality, thereby distinguishing between transmission abnormalities and signal faults, avoiding mislabeling and reconstructing normal signals, and ensuring that the signals involved in control are more reliable.

[0019] 3. This invention utilizes the response signal after the issuance of the collaborative control command to verify the execution link, and updates the transmission delay online based on the response result, so that the acquisition trigger time and the actual generation time can be synchronously corrected with changes in bus load and device aging, forming a closed loop of acquisition, verification, control and feedback update. Attached Figure Description

[0020] Figure 1 This is a system flowchart of the present invention; Figure 2 This is a schematic diagram of the cross-consistency check of the present invention; Figure 3 This is a schematic diagram of the collaborative control and transmission delay update process of the present invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] like Figures 1 to 3 As shown, this invention proposes a vehicle multi-channel signal intelligent acquisition and collaborative control system, including a multi-channel signal acquisition module, used to acquire multiple vehicle signals according to the acquisition cycle of each signal and add timestamps; The operating condition identification module is used to identify the current operating condition of the vehicle based on the multi-channel signals; The acquisition and scheduling module is used to determine the acquisition trigger time of each associated signal according to the transmission delay of each signal, so that the actual generation time of each associated signal falls into the same time window. The associated signal is a signal that represents the state of the same vehicle or the state of the same vehicle component. The signal verification module is used to subtract the transmission delay of the corresponding acquisition channel from the timestamp to obtain the actual generation time. It groups the associated signals whose actual generation time falls within the same time window according to the acquisition channel and performs cross-consistency verification. If the associated signal group corresponding to the same acquisition channel has the same time offset as the associated signal group corresponding to other acquisition channels, it is determined that the transmission delay of the acquisition channel is drifted, and the transmission delay of the corresponding acquisition channel is corrected according to the time offset between the corresponding change points. If, among three or more associated signals, the deviation between the vehicle status quantity of one signal after unified characterization and the other signals exceeds the allowable deviation range, the signal is marked as an abnormal deviation signal. The signal reconstruction module is used to call the pre-established signal mapping relationship according to the operating conditions, and reconstruct the corresponding vehicle state quantity based on the unmarked measured associated signal corresponding to the deviation abnormal signal, so as to replace the deviation abnormal signal. The cooperative control module is used to generate cooperative control commands carrying target state quantities based on unlabeled signals and reconstructed vehicle state quantities, and send them to the corresponding vehicle execution unit. During the response period after the command is issued, the response signals collected by the multi-channel signal acquisition module are included in the time window to which their actual generation time belongs for cross-consistency verification. The transmission delay of the corresponding acquisition channel is updated according to the difference between the generation time carried by the response signal and the time when the response signal arrives at the corresponding acquisition channel. The corrected or updated transmission delay is used to determine the acquisition trigger time and calculate the actual generation time.

[0023] As an optional embodiment, the signals to be collected during vehicle operation come from diverse sources and have different transmission paths: one type is sensor signals such as vehicle speed and accelerator pedal opening transmitted via hard wires, and the other type is bus signals sent by various electronic control units via the vehicle bus. If each acquisition channel times independently, the timing of each signal lacks a unified reference, and subsequent timing alignment and cross-checking cannot be achieved; if each signal is acquired using a unified period, it is difficult to take into account signals with different rates of change.

[0024] Therefore, this application adopts an acquisition structure that uses multi-channel one-to-one correspondence acquisition, unified clock source timing, and preset acquisition period according to the rate of change, as detailed below: The multi-channel signal acquisition module includes multiple acquisition channels. Each signal is acquired through its corresponding acquisition channel, meaning that each signal is connected through a fixed acquisition channel to avoid signal mixing.

[0025] The multiple signals include vehicle sensor signals and bus signals, as well as response signals from vehicle actuators: sensor signals are signals output by on-board sensors via hard-wired interfaces, including vehicle speed signals, accelerator pedal opening signals, braking signals, steering angle signals, motor speed signals, and battery temperature signals; bus signals are signals sent by the vehicle's electronic control units via the on-board bus, including status signals of each electronic control unit transmitted via CAN bus or on-board Ethernet.

[0026] In some implementations, the system includes a unified clock source. Each acquisition channel adds a timestamp to the acquired signal based on the same clock source. The timestamp represents the moment the corresponding signal arrives at the acquisition channel. The unified clock source also provides a clock reference to each vehicle execution unit. When a vehicle execution unit generates a response signal, it writes the generation time of the response signal into the response signal. The acquisition channel records the arrival time of the response signal. Since the generation time and arrival time are determined based on the same clock source, the difference between the two can characterize the transmission delay of the acquisition channel corresponding to the response signal.

[0027] Transmission delay is the time elapsed from signal generation to arrival at the acquisition channel, hence the following relationship exists: ;in, The actual time when the signal is generated. The timestamp is the moment the signal arrives at the acquisition channel. This represents the transmission delay of the corresponding acquisition channel for that signal, in milliseconds (ms). The timestamps of each signal are determined based on the same clock source, making the times of each signal comparable. This makes the calculation of the "same time window" and the "actual generation time" meaningful.

[0028] Furthermore, the acquisition period for each signal is preset according to the corresponding signal's rate of change. Specifically, during the calibration phase, the change in each signal per unit time is counted as the rate of change of that signal. Signals with larger rates of change are assigned shorter acquisition periods, while signals with smaller rates of change are assigned longer acquisition periods. For example, the motor torque signal has a larger rate of change and is assigned a shorter acquisition period, while the battery temperature signal has a smaller rate of change and is assigned a longer acquisition period. The preset acquisition period serves as the initial acquisition period for each signal, and subsequently, the acquisition scheduling module determines the acquisition trigger time for each signal based on its transmission delay.

[0029] As an optional embodiment, during vehicle operation, the operating conditions change continuously with driving operations, and normal noise characteristics, signal mapping relationships, etc. are all called according to the operating conditions. Once the operating condition identification is wrong, the downstream modules will call the wrong parameters in a series. At the same time, bumps, pedal vibrations, etc. will cause instantaneous signal fluctuations. If the operating conditions are immediately transferred according to the single identification result, it will lead to frequent changes in operating conditions and repeated switching of parameters of each module.

[0030] Therefore, this application explicitly enumerates the operating conditions and identifies them according to three categories of clearly defined signals: vehicle speed signal, accelerator pedal opening signal, and braking signal, as detailed below: Operating conditions include starting, acceleration, cruising, deceleration, and idling. Each operating condition is defined by a combination of vehicle speed signal, accelerator pedal opening signal, and braking signal. In some implementations, the braking signal is a brake pedal opening signal, used to characterize the degree of braking operation by the driver; when the brake pedal opening signal is within a preset braking opening range, the braking state is determined to be a braking state; when the brake pedal opening signal is not within the braking opening range, the braking state is determined to be a non-braking state.

[0031] Stable vehicle speed signal and stable accelerator pedal opening signal mean that the changes in the corresponding signals within adjacent recognition cycles are within a stable range determined according to the measurement error and normal fluctuations of the signal path; the stable range is written during the vehicle calibration phase. Idle speed opening is the accelerator pedal opening corresponding to the vehicle's power unit being in an idling state; the idle speed opening is written during the vehicle's factory calibration. The starting condition is the condition in which the vehicle changes from a stationary state to a moving state. This corresponds to the vehicle speed signal changing from zero to non-zero, the accelerator pedal opening signal increasing, and the braking state being non-braking within two adjacent recognition cycles. Acceleration condition is the condition in which the power demand of the vehicle increases while driving, corresponding to an increase in the accelerator pedal opening signal, a non-braking state, and an increase in the vehicle speed signal. Cruise mode is the condition in which the vehicle is driving stably, corresponding to a stable accelerator pedal opening signal, a non-braking state, and a stable vehicle speed signal. Deceleration condition is a condition in which the vehicle's power demand is reduced. The corresponding braking state is the braking state, or the accelerator pedal opening signal is reduced to the idle speed opening and the vehicle speed signal is reduced. The idling condition is when the vehicle is stationary and the power unit is running, corresponding to a vehicle speed signal of zero, an accelerator pedal opening signal of idle speed, and a non-braking state.

[0032] In some implementations, the operating condition identification module reads the vehicle speed signal, accelerator pedal opening signal and brake pedal opening signal from multiple signals according to a preset identification cycle, and obtains the identification result according to the definition relationship of each operating condition.

[0033] The identification cycle is determined during the vehicle calibration phase: the interval between changes in operating conditions caused by driving operations is statistically analyzed, and the identification cycle is set to be less than the interval, so that a change in operating conditions can be covered by two adjacent identification processes, avoiding missed identification of operating conditions between two adjacent identifications.

[0034] After the system is powered on, before the initial operating condition is determined, the operating condition identification module will take the first identification result as the temporary operating condition. When the next identification result is consistent with the temporary operating condition, the operating condition corresponding to the two consecutive consistent identification results will be determined as the initial operating condition. When the next identification result is inconsistent with the temporary operating condition, the next identification result will be taken as the new temporary operating condition, and identification will continue until two consecutive consistent identification results are obtained. The system power-on refers to the vehicle supplying power to the multi-signal intelligent acquisition and collaborative control system, enabling the various functional modules in the system to enter a working state.

[0035] Before the initial operating conditions are determined, the operating condition identification module does not output the operating condition transition results. The signal verification module and the signal reconstruction module call the corresponding normal noise characteristics and signal mapping relationships according to the temporary operating conditions, but do not execute the operating condition transition record and transition processing.

[0036] After determining the initial operating condition, if two adjacent identification results are consistent, the operating condition corresponding to that identification result is confirmed as the current operating condition of the vehicle; if two adjacent identification results are inconsistent, the original confirmed operating condition is maintained until two adjacent identification results are consistent, and then the operating condition corresponding to that identification result is confirmed as the current operating condition of the vehicle.

[0037] For example, if the first identification result after the system is powered on is cruise mode and the second identification result is also cruise mode, cruise mode is determined as the initial operating mode; if the first identification result is cruise mode and the second identification result is acceleration mode, the initial operating mode is temporarily determined and acceleration mode is used as the new temporary operating mode until two consecutive consistent identification results appear.

[0038] After determining the initial operating condition, if the current identification result is cruise condition and the next identification result is also cruise condition, the current operating condition of the vehicle is confirmed as cruise condition; if the current identification result is cruise condition and the next identification result is acceleration condition, the original confirmed cruise condition is maintained and identification continues until two consecutive consistent identification results are obtained before the operating condition is migrated.

[0039] As an optional embodiment, the transmission paths of the related signals are not the same: some are transmitted directly via hardwired lines, while others are forwarded via the vehicle bus, and the transmission delays of each signal are different. If the signals are triggered for acquisition at the same time, the actual generation times of the signals with different transmission delays will be sequential. The state of the vehicle at the same physical moment will not be acquired within the same time window, and subsequent cross-consistency checks will misjudge this "time misalignment" as a signal anomaly.

[0040] Therefore, this application advances the corresponding acquisition trigger time according to the transmission delay of each signal, so that the actual generation time of each associated signal is aligned to the same time window, as follows: The acquisition and scheduling module generates the same time window sequentially according to the preset scheduling cycle. The starting point of each time window is the start time of the corresponding scheduling cycle. The width of the same time window is determined according to the maximum value in the acquisition cycle of each associated signal. The scheduling cycle is set according to the width of the same time window so that adjacent time windows are connected sequentially.

[0041] For each associated signal, the acquisition scheduling module determines its acquisition time within the same time window according to the acquisition period of the associated signal. When the acquisition period of the associated signal is equal to the width of the same time window, the associated signal has one acquisition time within the same time window. When the acquisition period of the associated signal is less than the width of the same time window, the associated signal has multiple acquisition times within the same time window, and the time interval between adjacent acquisition times is the acquisition period of the associated signal.

[0042] For signals with acquisition periods shorter than the width of the same time window, the acquisition scheduling module determines multiple acquisition times within that time window according to the acquisition period of the signal, and advances each acquisition time by the transmission delay of the corresponding acquisition channel to obtain the acquisition trigger time for each acquisition. Each acquisition trigger time satisfies the following: ,in The acquisition trigger time is the r-th acquisition time corresponding to a correlated signal within the same time window. This refers to the r-th acquisition time of the associated signal within the same time window. This represents the transmission delay of the corresponding acquisition channel for this signal, in milliseconds (ms).

[0043] For example, a set of associated signals includes a first signal transmitted via the vehicle bus and a second signal transmitted via hardwired wiring. The first signal has a sampling period of 10ms and a transmission delay of 5ms, while the second signal has a sampling period of 5ms and a transmission delay of 2ms. The starting point of the same time window is 100ms, and the width is 10ms. The sampling time of the first signal within this time window is 100ms, and the corresponding sampling trigger time is 95ms. The sampling times of the second signal within this time window are 100ms and 105ms, respectively, and the corresponding sampling trigger times are 98ms and 103ms. Therefore, the signal with the longer transmission delay is triggered earlier, and the signal with the shorter transmission delay is triggered relatively later, ensuring that the actual generation times of all associated signals fall within the same time window.

[0044] In some implementations, the initial value of the transmission delay of each signal is the calibration value written at the time of vehicle manufacturing. Specifically, during the calibration phase, a test signal with a known generation time is input to each acquisition channel, and the arrival time of the test signal at the acquisition channel is recorded. The difference between the arrival time and the generation time is the transmission delay of the corresponding acquisition channel for that signal, and this is written into storage as the calibration value. During vehicle operation, the transmission delay of each signal is updated online by the collaborative control module (see the collaborative control module section for the update method). The updated transmission delay is used by the acquisition scheduling module to determine the acquisition trigger time and by the signal verification module to calculate the actual generation time, in order to reflect the transmission delay drift caused by changes in bus load and device aging.

[0045] As an optional embodiment, each signal is transmitted through different acquisition channels, and its arrival time includes its own transmission delay. If the signals are directly compared without processing, the difference in transmission paths will introduce time misalignment. Furthermore, the representation forms of various related signals are not consistent; some can be directly compared, while others require a unified representation form before comparison. In addition, after a deviation is found during comparison, it is necessary to distinguish the root cause of the deviation: since the same group of related signals shares an acquisition channel, if the transmission delay of that channel drifts, the entire group of signals will show a unidirectional offset. In this case, the deviation originates from the time delay drift rather than an abnormality in the signal itself. If it is treated as a signal abnormality and reconstructed, the truly usable signal will be discarded, and the time delay drift will not be eliminated, and the alignment will continue to be misaligned.

[0046] Therefore, the signal verification module of this application verifies each signal in four steps: time alignment and selection, classification and comparison, deviation attribution and correction, and retesting and marking, as follows: In S501, the timestamp of each signal is subtracted from the transmission delay of the corresponding acquisition channel to obtain the actual generation time (satisfying the aforementioned...). (Relationship), select each associated signal that falls within the same time window at the actual time of generation as the comparison object; after alignment, each associated signal corresponds to the state of the vehicle at the same physical time, eliminating the time misalignment caused by transmission delay differences; The signal verification module groups the associated signals according to the acquisition channel. Associated signals acquired through the same acquisition channel constitute an associated signal group. The associated signal group with consistent signals within the group is used as the reference group. When there are multiple reference groups, the group with the most associated signals is used as the reference group. If the reference group cannot be determined, the verification is switched to the verification based on the expected vehicle state quantity. In S502, the comparison is classified according to the representation form: if each associated signal represents the same vehicle state quantity, such as the left and right front wheel speed signals both representing vehicle speed, the same vehicle state quantity represented by each associated signal is directly compared; if each associated signal represents the state of the same vehicle component, such as the motor speed signal and the vehicle speed signal both representing the transmission system state, they cannot be directly compared because the represented state quantities are different. First, one of the signals is mapped to the vehicle state quantity represented by the other signal according to the signal mapping relationship, and then the comparison is made.

[0047] The signal mapping relationship is a pre-established correspondence between associated signals and vehicle state variables, including conversion coefficients and conversion methods between signals and vehicle state variables. This is established and stored in the system based on the vehicle's structural parameters before it leaves the factory. This allows the signal verification module to standardize the representation of the two signals before comparison, and allows the signal reconstruction module to call upon the signals when reconstructing the corresponding vehicle state variables from the associated signals. For example, the motor speed signal and vehicle speed satisfy the following relationship: Where v is the vehicle speed in km / h, n is the vehicle speed signal representing the vehicle state quantity, k is the motor speed signal value in r / min, and k is a conversion factor determined by the transmission ratio of the transmission system and the wheel radius, the dimensions of which correspond to the units of v and n.

[0048] According to this correspondence, the motor speed signal can be mapped to the vehicle speed represented by the vehicle speed signal. Once the vehicle state quantities represented by the two signals are consistent, they can be compared. Conversely, the motor speed can also be obtained by mapping the vehicle speed signal. In S503, the deviation between the two vehicle state quantities compared in step S502 is calculated, and the sum of the measurement errors corresponding to the measurement accuracy of the two signals is taken as the allowable deviation. When the deviation of the vehicle state quantities represented by the two signals compared is greater than the allowable deviation, and the signals within a group of associated signals are consistent with each other, and the corresponding change points of the associated signals are offset in the same direction in time relative to the other associated signal groups, the transmission delay drift of the acquisition channel corresponding to the associated signals is determined, and the transmission delay of the corresponding acquisition channel is corrected according to the time offset between the corresponding change points. ; ,in and The vehicle state quantity is represented by a unified method for analyzing the two signals. and The measurement error corresponds to the two signals; This is the actual deviation. This is the allowable deviation amount; The corresponding change point refers to the signal change position when the same vehicle state quantity represented by each associated signal undergoes the same change, including the rising edge, falling edge, or extreme point of the signal; the time offset between corresponding change points is determined according to the difference in the actual time of occurrence of the corresponding change points.

[0049] And the corrected relationship is: , For the corrected transmission delay, The transmission delay before correction. This represents the time offset between corresponding points of change, all in milliseconds (ms). The sign is determined according to the direction of the signal group being advanced or delayed; If there is no corresponding change point available for comparison within the same time window, the transmission delay correction will not be performed for this time. The correction will be performed when the corresponding change point appears in a subsequent time window.

[0050] In S504, the actual generation time is recalculated based on the corrected transmission delay, and compared again in the manner of S502: if the deviation of the comparison is not greater than the sum of the measurement errors, it indicates that the deviation is indeed caused by the delay drift and has been eliminated, and no abnormality is marked; otherwise, it indicates that the deviation originates from the abnormality of the signal itself, and the one that deviates from the other associated signals is marked as a deviation abnormal signal; when there are more than three associated signals, the one that represents the vehicle state quantity that is inconsistent with the other signals is marked.

[0051] When only two related signals are involved in the comparison, and the deviation between the vehicle state quantities of the two related signals after unified characterization exceeds the allowable deviation range, it is impossible to determine the abnormal signal based on majority consensus. Therefore, the signal verification module retrieves the reference cooperative control command issued in the previous time window, converts the target state quantity carried by the reference cooperative control command into the expected vehicle state quantity through a signal mapping relationship, and uses the expected vehicle state quantity as the comparison benchmark. The signal that deviates from the expected vehicle state quantity is marked as an abnormal deviation signal, while the other signal is not marked as abnormal and continues to participate in subsequent verifications.

[0052] The previous time window is the time window that is located before and adjacent to the current time window in chronological order; the reference cooperative control instruction is a cooperative control instruction generated based on the verified signal other than the two associated signals, so that the desired vehicle state quantity is independent of the two associated signals to be verified, and avoids using the control target generated by the signal to be verified to back verify the signal to be verified.

[0053] When there is no reference coordinated control command in the previous time window, both associated signals are marked as pending confirmation signals. During the pending confirmation period, the two associated signals do not participate in the generation of coordinated control commands, and the coordinated control module uses the vehicle status variables verified in the previous time window for vehicle control. When one of the associated signals is not marked after cross-consistency verification in a subsequent time window, the pending confirmation mark of that associated signal is removed, and that associated signal is restored to participate in the generation of coordinated control commands.

[0054] For example, when the steering angle signal and the actual steering angle signal of the steering actuator are inconsistent, the target steering angle carried by the reference coordinated control command issued in the previous time window is converted into the desired steering angle through the signal mapping relationship, and the path that deviates from the desired steering angle is marked as a deviation anomaly signal. Thus, when two related signals are inconsistent, it is not necessary to treat both signals as anomalies, and the verification chain is not interrupted.

[0055] As an optional implementation, signal anomalies do not always manifest as numerical deviations: in signal jamming faults, the signal value remains at a certain value and does not change, and the deviation from the associated signal may still be within the allowable deviation, which cannot be detected by deviation comparison alone; however, the fluctuation components of signal changes caused by various faults will inevitably differ from normal fluctuations. In addition, regardless of whether the abnormal signal is due to deviation or fluctuation, the marking result of a single verification may be caused by transient disturbances: vehicle bumps, electromagnetic interference, etc., may cause the signal to be briefly abnormal within a certain time window. If reconstruction is started directly according to the single marking, normal signals will be discarded; if they are ignored directly, real faults may be missed.

[0056] Therefore, this application first identifies wave-type anomalies based on the power distribution of wave components, and then performs encrypted observation verification on the labeled results, as follows: The signal verification module performs high-pass filtering on each signal to remove trend components and obtain fluctuation components. The fluctuation components are then divided into m frequency bands in the frequency domain. The power of each frequency band and the sum of the power of each frequency band are calculated. The sum of the power of each frequency band is taken as the total power of the fluctuation components.

[0057] When the total power of the fluctuation component is equal to zero, it means that there is no fluctuation component in the corresponding time window of the signal. The power distribution is no longer calculated, and the signal is directly marked as a fluctuation abnormal signal.

[0058] When the total power of the wave components is not equal to zero, the power proportion of each frequency band is calculated to obtain the power distribution of the wave components. : ;in Let be the power of the fluctuation component in the i-th frequency band. Let be the power percentage of the fluctuation component in the i-th frequency band, m be the preset number of frequency bands, i = 1, 2, ..., m, and j be the frequency band number when summing the power. Retrieve the power distribution of normal noise characteristics corresponding to the operating conditions. Calculate the deviation between the power distribution of the fluctuation component and the power distribution of the normal noise characteristic. : ;in The power proportion of normal noise features in the i-th frequency band, and the power distribution of normal noise features. During the calibration phase, the normal signals under the corresponding operating conditions are established. When the deviation D exceeds the allowable deviation range determined by the calibration results of the corresponding operating condition, the signal is marked as an abnormal fluctuation signal. The upper limit of the allowable deviation range is the allowable deviation amount. Allowable deviation Determined according to the deviation distribution of normal signals under the corresponding operating conditions during the calibration phase.

[0059] In some implementations, normal noise characteristics are taken from a noise characteristic library established during the calibration phase according to various operating conditions, including the power distribution of fluctuation components. The deviation D is the sum of the absolute values ​​of the differences in the power proportions of each frequency band, which can reflect the overall offset of the power distribution across the entire frequency band: when the signal is stuck, the total power of the fluctuation component is equal to zero, the signal verification module does not calculate the power proportion, and directly marks the signal as a fluctuation abnormal signal; when the signal is only subject to transient interference, the power distribution of the fluctuation component usually changes only in a few frequency bands, and the deviation D does not exceed the allowable deviation range, thus avoiding the transient interference being mistakenly marked as a fluctuation abnormal signal; Once any signal is marked, the acquisition scheduling module adjusts the acquisition period of that signal and its associated signals to a preset re-verification acquisition period. This re-verification acquisition period is shorter than the acquisition period used before the signal was marked. The reason for adjusting the acquisition period of associated signals as well is that re-verification requires re-performing cross-consistency checks. The marked signal and its associated signals must be re-aligned and observed with denser sampling within the same time window. Adjusting only the acquisition period of the marked signal is insufficient to complete the intra-group comparison.

[0060] After adjusting the acquisition cycle, the signal verification module verifies the signal within a preset number of consecutive time windows, where the preset number is the number of verification windows written during the calibration phase. If the verification results for the signal are all abnormal within the preset number of consecutive time windows, the abnormal state of the signal is confirmed to persist, and the corresponding marker is retained. The signal reconstruction module then reconstructs the corresponding vehicle state quantity according to the invoked signal mapping relationship. If the verification results for the signal are all normal within the preset number of consecutive time windows, the previous marker is determined to have been caused by a transient disturbance, and the corresponding marker is cleared. The signal then resumes participation in cross-consistency verification and the generation of cooperative control commands.

[0061] After the retest is completed, regardless of whether the corresponding mark is retained or cleared, the acquisition scheduling module restores the acquisition period of the signal and its associated signals to the acquisition period used before the signal was marked.

[0062] As an optional embodiment, if a signal is marked as a deviation abnormal signal or a fluctuation abnormal signal and is directly discarded, the vehicle state quantity it represents will be missing, and the cooperative control command will lose its basis. If a reconstructed value is used instead, the reconstructed value is, after all, a calculated result. If it is used to verify or reconstruct other signals, the calculation error will propagate and amplify along the verification chain. Therefore, this application reconstructs the vehicle state quantity from the associated signal according to the signal mapping relationship of the called signal, and adds a reconstruction identifier to the reconstruction result for isolated use, as follows: During reconstruction, the signal reconstruction module calculates the unmarked measured associated signals corresponding to the marked signals according to the called signal mapping relationship, obtains the vehicle state quantity corresponding to the deviation anomaly signal, and uses the reconstructed vehicle state quantity to replace the deviation anomaly signal to participate in the generation of cooperative control commands.

[0063] For example, both the motor speed signal and the vehicle speed signal characterize the state of the vehicle's transmission system. When the motor speed signal is marked as an abnormal deviation signal, according to the aforementioned... The corresponding relationship is obtained by calculating the corresponding motor speed from the vehicle speed signal, which is used as the reconstructed vehicle state quantity to replace the original motor speed signal in subsequent coordinated control.

[0064] When a marked signal corresponds to more than one associated signal, the corresponding vehicle state quantity is calculated separately for each associated signal. When there are mutually consistent calculation results, the mutually consistent calculation results are taken as the reconstructed vehicle state quantity. When there are no mutually consistent calculation results, the calculation results are weighted and synthesized according to the reciprocal of the measurement error of each associated signal. The specific method is selected according to the signal configuration.

[0065] When all associated signals corresponding to the marked signal are also marked, and there are no unmarked measured associated signals, the signal reconstruction module does not output the reconstructed vehicle state quantity and marks the corresponding vehicle state quantity as invalid; the cooperative control module uses the vehicle state quantity verified in the previous time window to maintain the current control; when there is no vehicle state quantity verified in the previous time window, the cooperative control module calls the pre-stored safety target state quantity to generate safety control instructions until any associated signal is restored to an unmarked measured signal.

[0066] The safety target state quantity is the target state quantity written at the time of vehicle factory calibration, which is used to limit the operating range of the drive motor, brake actuator and steering actuator when the corresponding vehicle state quantity is invalid.

[0067] In some implementations, when the signal reconstruction module outputs the reconstructed vehicle state quantity, a reconstruction identifier is attached to the reconstructed vehicle state quantity. The reconstruction identifier is used to record that the vehicle state quantity comes from reconstruction rather than actual measurement. The vehicle state quantity with the reconstruction identifier does not participate in cross-compliance verification, that is, it is not used as a benchmark for comparing other signals - because it is itself an estimated value, using it to verify other signals will pass the estimation error to other signals; it also does not participate in the reconstruction of other abnormal signals, that is, it is not used as input for reconstruction calculation - using the estimated value to back-calculate other vehicle state quantities will amplify the error step by step; therefore, cross-compliance verification and signal reconstruction always use the measured signal as the benchmark.

[0068] Once the previously marked signal has been restored to normal after retesting and passed the cross-consistency check again, the vehicle state quantity obtained from the reconstruction will be stopped, and the corresponding reconstruction flag will be cleared. The restored signal will then participate in the cross-consistency check, signal reconstruction, and generation of cooperative control commands based on the measured values.

[0069] When the marker is maintained and the corresponding vehicle state quantity is reconstructed by the signal reconstruction module, the signal reconstruction module performs the first reconstruction based on the encrypted data collected during the re-verification period; after the re-verification ends and the collection cycle is restored, the signal reconstruction module continues to maintain the reconstruction based on the restored collection data until the marker is removed.

[0070] As an optional embodiment, the signal mapping relationship is related to the vehicle's operating conditions. The same conversion coefficient may not be applicable under different operating conditions. For example, the transmission ratio of a shifting vehicle changes with the gear. Continuing to use the original signal mapping relationship after the operating conditions are changed will lead to distortion of the reconstruction results. Moreover, when the operating conditions are just switched, the reliability of the alternative mapping relationship has not been verified, and there is a risk in directly switching.

[0071] Therefore, this application establishes signal mapping relationships according to operating condition partitions, and calls alternative mapping relationships in a dual-mapping parallel mutual verification manner during operating condition migration, as detailed below: Predetermine the representative conversion factors for each operating condition zone; when the conversion factor corresponding to a certain operating condition... Representative conversion factor with corresponding working condition zone satisfy: When this operating condition is specified, it is assigned to that operating condition zone; among which, To allow for fluctuations, its dimensions are the same as those of the conversion factor, and it is determined according to the maximum deviation of the measured conversion factor of each operating condition within the same operating condition zone during the calibration phase relative to the representative conversion factor.

[0072] Signal mapping relationships are established according to operating condition zones. Operating conditions within the same operating condition zone share the corresponding signal mapping relationship. That is, each operating condition within the same zone uses the same conversion factor and conversion method. With this setting, the mapping relationship is established on a zone-by-zone basis, avoiding parameter redundancy caused by establishing mapping relationships for each operating condition separately.

[0073] In some implementations, when the operating condition is migrated to another operating condition partition, the conversion factor corresponding to the migrated operating condition is... The difference between the conversion factor and the original operating condition zone exceeds the allowable fluctuation. The conversion coefficients of the original signal mapping relationship are no longer applicable to the migrated operating conditions. Therefore, it is determined that the signal mapping relationship being called does not match the operating conditions, and the alternative mapping relationship corresponding to the operating condition partition to which the migrated operating conditions belong is called.

[0074] Since the reliability of the alternative mapping relationships under the new operating conditions has not yet been verified when the operating conditions have just changed, during the transition period, the signal reconstruction module uses both the original signal mapping relationships and the alternative mapping relationships in parallel to reconstruct the same vehicle state variable from the same set of unlabeled measured associated signals. Let the vehicle state variable reconstructed from the original signal mapping relationships be denoted as... The vehicle state variables obtained by reconstructing the alternative mapping relationships are: Calculate according to the aforementioned method for calculating deviation. and The deviation between them.

[0075] When the deviation between the vehicle state quantities obtained from the two reconstructions is within the allowable deviation range determined according to the measurement error, it is determined that the alternative mapping relationship is suitable for the migrated operating conditions, the alternative mapping relationship is enabled, and the original signal mapping relationship is stopped from being used for reconstruction.

[0076] When the migration transition period ends, if the deviation between the vehicle state quantities obtained from the two reconstructions still exceeds the allowable deviation range, the alternative mapping relationship will not be enabled, the corresponding reconstructed vehicle state quantities will be marked as invalid, and the identification results of the operating conditions will be reviewed.

[0077] When the verification result is the operating condition before migration, the current operating condition migration is determined to be a misidentification. The signal reconstruction module calls the signal mapping relationship corresponding to the operating condition before migration and clears the invalid flag of the reconstructed vehicle state quantity. When the verification result is still the operating condition after migration, the invalid flag of the reconstructed vehicle state quantity is maintained, and the original signal mapping relationship and the alternative mapping relationship are used in parallel for reconstruction verification in the subsequent time window until the deviation between the vehicle state quantities obtained by the two reconstructions is within the allowable deviation range, and then the alternative mapping relationship is enabled.

[0078] When the reconstructed vehicle state quantity is marked as invalid, the cooperative control module suspends the use of the reconstructed vehicle state quantity and uses the vehicle state quantity verified in the previous time window to maintain the current control; when there is no vehicle state quantity verified in the previous time window, the cooperative control module calls the pre-stored safety target state quantity to generate safety control instructions until the signal reconstruction module re-outputs a valid reconstructed vehicle state quantity.

[0079] As an optional embodiment, the collaborative control command acts on multiple vehicle execution units simultaneously. If the signal on which the command is based is not verified, the abnormal signal will directly mislead the execution unit's actions. Moreover, the control objects of each execution unit are different, and the driving goal of the whole vehicle needs to be decomposed into the target state quantities of each execution unit. In addition, after the command is issued, it is also necessary to verify whether the feedback of the execution unit is consistent with the actual state of the vehicle: if the execution unit reports that it has acted, but the vehicle state does not change accordingly, it indicates that there is an abnormality in the link from the command to the execution, which cannot be found by simply checking the feedback signal itself.

[0080] Therefore, this application is executed in three steps: instruction generation, response signal verification, and execution status determination, as follows: In S901, the target driving state of the vehicle is determined based on the unmarked signals in each signal and the reconstructed vehicle state quantity. The unmarked signals are those that have been confirmed to be normal by cross-consistency check and fluctuation check. When the signal required to generate the instruction is marked, the reconstructed vehicle state quantity is used to replace the signal value of the signal in the calculation to ensure that the instruction generation is not interrupted due to the abnormality of a single signal.

[0081] Specifically, the target driving state of the vehicle is determined according to the accelerator pedal opening signal, brake pedal opening signal, and steering angle signal from the multiple signals. The target driving state is then allocated according to the control object of each vehicle execution unit, and corresponding coordinated control commands are generated for each vehicle execution unit. ; ; ; The target driving torque of the corresponding drive motor is expressed in N·m; The target braking pressure of the brake actuator, in Pa; The target steering angle for the corresponding steering actuator, in degrees; This is the normalized accelerator pedal opening. The normalized brake pedal opening is represented by both values, which range from 0 to 1, and both are dimensionless percentage opening values. This is the signal value of the steering angle signal, in degrees. , , These are the conversion coefficients for drive, braking, and steering, respectively, written during vehicle factory calibration. Each coordinated control command carries the corresponding vehicle execution unit's... , or As the target state quantity, that is, the target quantity value that the corresponding vehicle execution unit should achieve, it serves as the comparison benchmark for subsequent determination of the execution state. The vehicle execution unit includes a drive motor, a brake actuator, and a steering actuator.

[0082] In S902, the response signal and signals representing the same vehicle state quantity from multiple signals are used as correlated signals for cross-consistency verification. The response signal is the feedback signal of the actual action of the vehicle actuator, including the actual torque signal of the drive motor, the braking pressure signal of the brake actuator, and the actual steering angle signal of the steering actuator. It is acquired through the corresponding acquisition channel and shares the same clock source and clock reference with the multiple signals.

[0083] The verification mechanism is as follows: the response signal and the corresponding multi-channel signal represent the same physical process. For example, the actual steering angle signal and the steering angle signal both represent the vehicle's steering state, and the actual torque signal and the motor speed signal both represent the state of the drive system. Under normal circumstances, the response signal changes in tandem with the corresponding multi-channel signal according to the signal mapping relationship.

[0084] When the response signal deviates from the other related signals, it indicates that the feedback of the execution unit is inconsistent with the actual state of the vehicle. The execution unit has acted according to the instruction but the vehicle state has not changed accordingly, or the feedback signal itself is distorted. Both of these are abnormalities in the link from instruction to execution. Therefore, the execution link of the corresponding vehicle execution unit is determined to be abnormal and recorded as the basis for the maintenance of the vehicle execution unit.

[0085] In S903, a corresponding response time period is preset according to the type of vehicle execution unit. The response time period is determined based on the response duration calibrated under normal conditions for each type of execution unit, with a margin, so that the confirmation wait is adapted to the actual response capability of each execution unit. Response signals are collected within the response time period initiated by the cooperative control command, and the response signals are compared with the target state variables carried by the cooperative control command. Where y is the magnitude of the response signal during the response period. The target state quantity carried by the coordinated control command, i.e., in S901 , or The units for both correspond to the type of vehicle actuator, and are N·m, Pa, or °. The permissible error of the response is determined based on the measurement error of the corresponding response signal.

[0086] The reason for using the target value carried by the instruction itself as the judgment criterion is that each cooperative control instruction has different requirements for the execution unit, and there is no fixed judgment criterion applicable to all instructions. The target state quantity carried by the instruction is the specific requirement of that instruction for the execution unit. When the response signal satisfies the requirements within the response time period... If the execution is normal, it is determined that the execution is normal; otherwise, it is determined that the execution is abnormal, and the abnormal vehicle execution unit is recorded.

[0087] As an optional embodiment, the vehicle execution unit can still complete the actions required by the command in the early stage of aging, but the response speed decreases, and it cannot be detected by simply determining "whether to execute". The performance degradation is the state in which the vehicle execution unit can still complete the actions required by the command, but the response speed decreases. This is different from the execution abnormality: the execution abnormality is that the command action is not completed, while the performance degradation is that the command action has been completed, but the completion process has slowed down.

[0088] The mechanism is as follows: factors such as aging of the drive motor windings, wear of the brake actuator friction components, and increased transmission clearance of the steering actuator reduce the drive response capability of the actuator. This manifests as the response signal still reaching the target value, but the rise time becoming longer. Therefore, this application further determines performance degradation under the premise of normal operation, as follows: When the response signal reaches the target state quantity, the reference rise time stored according to the type of cooperative control instruction is retrieved; when the rise time of the response signal exceeds the allowable time range determined according to the reference rise time and normal response fluctuation, the performance of the vehicle execution unit is determined to be degraded and the number of degradations is accumulated. The response time period corresponding to the subsequent cooperative control instruction is extended according to the number of degradations by a preset step size. The extended response time period does not exceed the maximum response time period of the corresponding vehicle execution unit.

[0089] Let the rise time of the response signal be . ; ,in, The time elapsed from the issuance of the coordinated control command to the entry of the response signal into the response range determined based on the target state quantity and the allowable error of the response; The baseline rise time is the rise time obtained by calibration under normal execution conditions. This represents the allowable margin for normal response fluctuations, and the unit is milliseconds (ms).

[0090] The distribution range of normal response fluctuations is determined according to the statistics of the calibration phase. It is used to exclude normal random fluctuations caused by ambient temperature and load fluctuations. Only when the increase in rise time exceeds the normal fluctuation range can it be attributed to the performance degradation of the execution unit itself.

[0091] When the above formula is true, it is determined that the performance of the vehicle execution unit has degraded and the number of degradations N is accumulated. The number of degradations N is also used as the basis for maintenance prompts for the execution unit. After performance degradation is determined, the response time period corresponding to subsequent collaborative control commands is extended by a preset step size according to the number of degradations: ; For the extended response period, The response time before the extension is defined as follows: N is the number of degradations, and s is the preset step size, both in milliseconds. s is calibrated according to the magnitude of the increase in response time caused by the performance degradation of the execution unit.

[0092] The reason for the extension is that performance degradation is a gradual process; the more severe the degradation, the slower the response. If the original response time is still used for judgment, the response signal of the degraded execution unit may not have reached the target value and would be misjudged as an execution anomaly in the execution status determination. Extending the response time by multiplying the number of degradations by a preset step size allows the judgment benchmark to be gradually adjusted with the degree of degradation, correctly distinguishing between "slower execution" and "no execution." The extended response time is used for the execution status determination of subsequent collaborative control instructions.

[0093] It should be noted that the number of degradations is accumulated separately for each vehicle execution unit and stored in the system. The maximum response time period is the upper limit of the response time period written when the vehicle is manufactured; the extended response time period shall not exceed the maximum response time period.

[0094] After the vehicle actuator is repaired or replaced, a maintenance reset signal is sent to the collaborative control module through the vehicle diagnostic interface. After receiving the maintenance reset signal, the collaborative control module clears the degradation count of the corresponding vehicle actuator to zero and restores the response time period corresponding to subsequent collaborative control commands to the response time period before the extension.

[0095] In some implementations, the transmission delay of each signal is not constant: changes in bus load, temperature fluctuations, and device aging can all cause transmission delay drift. If the factory-calibrated initial value is used for a long time, the alignment of the actual generation time will be misaligned, leading to misjudgment in cross-compliance checking. At the same time, the actual measured delay is affected by bus scheduling, clock synchronization errors, and sampling fluctuations, and direct replacement will cause a jump in transmission delay. Therefore, this application updates the transmission delay online by weighted fusion of the measured delay of the response signal and the current transmission delay while determining the execution status, as follows: The vehicle actuator writes the time of response signal generation when generating the response signal, and then the acquisition channel calculates it: , To measure the actual transmission delay of the corresponding acquisition channel in response to the signal, To respond to the time when the signal arrives at the acquisition channel, The generation time written when generating a response signal for the vehicle execution unit. and All signals are based on the same clock source and are measured in milliseconds (ms); the response signals are transmitted through the corresponding acquisition channels. and The difference represents the time elapsed from the generation of the response signal by the vehicle actuator to its arrival at the corresponding acquisition channel. Its variation reflects the drift in the transmission delay of the corresponding acquisition channel, and therefore can be used to correct the transmission delay. Subsequently, the transmission delay of the corresponding acquisition channel is updated according to weighted fusion: :in The transmission latency before the update is the initial value set at the factory during the first update; This is the updated transmission latency; The fusion weight is the measured latency, which is a dimensionless value ranging from 0 to 0.5.

[0096] The basis for setting ω is that the measured latency in a single instance is affected not only by the transmission latency but also by the fluctuation of the execution unit's action duration. Its reliability is lower than the current transmission latency obtained through long-term statistics, so it is set to be less than 1- The value is set so that the update is based primarily on the current transmission delay and is slowly corrected using single measured values. The transmission delay can gradually track real-world drifts such as bus load and device aging, without jumping due to single fluctuations. The updated transmission delay is used by the acquisition scheduling module to determine the acquisition trigger time and by the signal verification module to calculate the actual occurrence time.

[0097] Furthermore, the fusion weights The reliability is determined based on the reliability of the corresponding acquisition channel. The reliability is calculated based on the number of transmission delay corrections for the corresponding acquisition channel within a preset statistical period and the number of times the signal is marked and maintained after verification. ; in The value range is 0, corresponding to the reliability of the acquisition channel. <C≤1; This represents the number of times the transmission delay is corrected for the corresponding acquisition channel within the preset statistical period. The number of times the signal of the corresponding acquisition channel is marked and maintained after verification within the preset statistical period; wherein the preset statistical period is a fixed statistical duration traced back from the current transmission delay update time, and is written when the vehicle is calibrated at the factory. Fusion weights Determined according to the following relationship: ,in , These are the lower and upper bounds of the fusion weights, respectively, and satisfy the following conditions: ; Therefore, the fusion weight Always less than 1- This ensures that the update result of the transmission delay is based primarily on the current transmission delay; when the number of transmission delay corrections or the number of times the marker is maintained increases within the preset statistical period for the corresponding acquisition channel, the reliability C decreases, and the fusion weights are adjusted. This reduces the impact of a single measured delay of the abnormal acquisition channel on the transmission delay update result.

[0098] The mechanism is that channels with frequent transmission delay drift or frequent signal anomalies are more likely to be affected by their own anomalies in a single measured delay. By using a smaller weight for fusion, the measured values ​​of abnormal channels can be avoided from polluting the transmission delay update results, ensuring that the delay update is dominated by long-term statistical values.

[0099] In summary, this invention determines the acquisition trigger time of associated signals according to the transmission delay of each signal, so that signals from different transmission paths correspond to the same physical time of the vehicle, reducing misjudgments caused by time misalignment; in cross-consistency verification, the entire group of same-direction offsets are identified as transmission delay drift, and single-path offsets are identified as signal anomalies, avoiding the misreconstruction of normal signals. The abnormal signal is reconstructed according to the signal mapping relationship based on the operating condition, and the reconstructed value is restricted from participating in the verification again by the reconstruction identifier to prevent the propagation of the calculation error. At the same time, the response signal of the coordinated control command is used to verify the execution link and update the transmission delay online, so that the acquisition scheduling, signal verification and execution feedback form a closed loop, ensuring that the vehicle status acquisition and control command generation remain continuous, accurate and reliable under the conditions of bus load changes, device aging and single-channel signal abnormality.

[0100] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art 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 appended claims and their equivalents.

Claims

1. A vehicle multi-channel signal intelligent acquisition and collaborative control system, characterized in that, include: The multi-channel signal acquisition module is used to acquire multiple signals from the vehicle according to the acquisition cycle of each signal and add timestamps; The operating condition identification module is used to identify the vehicle's operating condition based on multiple signals; The acquisition and scheduling module is used to determine the acquisition trigger time of each associated signal according to the transmission delay of each signal, so that the actual generation time of each associated signal falls into the same time window; the associated signal is a signal that represents the state of the same vehicle or the state of the same vehicle component. The signal verification module is used to subtract the transmission delay of the corresponding acquisition channel from the timestamp to obtain the actual generation time, group the associated signals within the same time window according to the acquisition channel and perform cross-consistency verification; if the associated signal group of the same acquisition channel is offset in the same direction relative to other associated signal groups, it is determined that the transmission delay drift of the acquisition channel is determined and corrected according to the time offset of the corresponding change point; if the deviation of the vehicle status quantity of one signal after unified characterization exceeds the allowable deviation range determined by the measurement error among three or more associated signals, it is marked as an abnormal deviation signal. The signal reconstruction module is used to call the pre-established signal mapping relationship according to the operating conditions, and reconstruct the corresponding vehicle state quantity based on the unmarked measured associated signal corresponding to the deviation abnormal signal, and replace the deviation abnormal signal. The collaborative control module is used to generate collaborative control commands carrying target state quantities based on unlabeled signals and reconstructed vehicle state quantities, and send them to the corresponding vehicle execution unit; during the response period after the command is issued, the response signal is included in the corresponding time window for verification according to its actual generation time, and the transmission delay of the corresponding acquisition channel is updated according to the difference between the generation time and the arrival time carried by the response signal. The corrected or updated transmission delay is used to determine the acquisition trigger time and calculate the actual generation time.

2. The intelligent acquisition and collaborative control system for multi-channel vehicle signals according to claim 1, characterized in that: The multi-channel signal acquisition module includes multiple acquisition channels, and each signal is acquired through its corresponding acquisition channel; the multiple signals include vehicle sensor signals, bus signals, and response signals from the vehicle's actuators; Each acquisition channel adds a timestamp to the acquired signal based on the same clock source. The timestamp is the moment when the signal arrives at the acquisition channel. Each vehicle actuator writes the generation time to the generated response signal based on the same clock source.

3. The intelligent acquisition and collaborative control system for multi-channel vehicle signals according to claim 1, characterized in that: The operating conditions include starting, acceleration, cruising, deceleration, and idling. The operating condition identification module identifies the operating condition based on the vehicle speed signal, accelerator pedal opening signal, and brake pedal opening signal; After the system is powered on, the operating condition identification module continuously identifies the operating conditions; Operating condition migration will not be triggered until the initial operating condition is confirmed. When two consecutive identical identification results appear for the first time, the operating condition corresponding to the identification result is confirmed as the initial operating condition. If the initial operating condition has been confirmed, and two consecutive identification results are consistent, the operating condition corresponding to the identification result shall be confirmed as the current operating condition of the vehicle. If two consecutive identification results are inconsistent, the original confirmed operating conditions shall be kept unchanged until the two consecutive identification results are consistent.

4. The intelligent acquisition and collaborative control system for multi-channel vehicle signals according to claim 2, characterized in that: The method for determining the acquisition trigger time of the acquisition scheduling module is as follows: For each associated signal, determine its acquisition time within the same time window, and subtract the transmission delay of the corresponding acquisition channel from each acquisition time to obtain the acquisition trigger time; the width of the same time window is determined according to the maximum value in the acquisition period of each associated signal.

5. The intelligent acquisition and collaborative control system for multi-channel vehicle signals according to claim 4, characterized in that: The signal verification module performs verification according to the following steps: S501. Subtract the transmission delay of the corresponding acquisition channel from the timestamp to obtain the actual generation time, and select the associated signals whose actual generation time falls within the same time window. S502. When each associated signal represents the same vehicle state quantity, the vehicle state quantity represented by each associated signal is directly compared; when each associated signal represents the state of the same vehicle component, one signal is mapped to the vehicle state quantity represented by another signal according to the signal mapping relationship before comparison. S503. When the deviation of the vehicle state quantity represented by the two signals being compared is greater than the sum of the measurement errors of the two signals, and the signals within one group of associated signals are consistent with each other, and the corresponding change points of the associated signals in this group are offset in the same direction in time relative to the corresponding change points of other associated signal groups, it is determined that the transmission delay drift of the acquisition channel corresponding to the associated signals in this group is corrected according to the time offset of the corresponding change points. S504. Recalculate the actual generation time based on the corrected transmission delay; if the deviation is not greater than the sum of measurement errors, confirm that the corresponding signal is normal. When the verification deviation is greater than the sum of the measurement errors and there are more than three associated signals, the one with deviations exceeding the allowable deviation range from all other associated signals will be marked as an abnormal deviation signal. When there are two associated signals or when the abnormal signal cannot be determined based on the consistency of the associated signals, the expected vehicle state quantity, which is converted from the target state quantity carried by the reference coordinated control command through the signal mapping relationship, is used as the benchmark, and the one with the deviation exceeding the allowable deviation range is marked as a deviation abnormal signal.

6. The intelligent acquisition and collaborative control system for multi-channel vehicle signals according to claim 2, characterized in that: The signal verification module is also used to divide the fluctuation components after high-pass filtering of each signal into m frequency bands, calculate the power of each frequency band and the total power, where m is the preset number of frequency bands; when the total power is equal to zero, the signal is marked as a fluctuation abnormal signal; when the total power is not equal to zero, the power distribution is calculated and compared with the power distribution of the normal noise characteristics corresponding to the operating condition; when the deviation exceeds the allowable deviation range determined by the calibration result of the corresponding operating condition, the signal is marked as a fluctuation abnormal signal. Once any signal is marked, the acquisition scheduling module adjusts the acquisition period of that signal and its associated signals to a preset re-verification acquisition period, which is shorter than the acquisition period before the signal was marked. If the verification results for a preset number of consecutive time windows are all abnormal, the abnormal state is confirmed to continue and the corresponding mark is retained. The signal reconstruction module then reconstructs the corresponding vehicle state quantity. If the verification results for a preset number of consecutive time windows are all normal, it is determined that the previous marking was caused by a momentary disturbance and the corresponding mark is cleared. After the re-verification is completed, the acquisition period of that signal and its associated signals is restored to the acquisition period before the signal was marked.

7. A vehicle multi-channel signal intelligent acquisition and collaborative control system according to claim 6, characterized in that: The signal reconstruction module calculates the unmarked measured associated signals according to the invoked signal mapping relationship to obtain the vehicle state quantity corresponding to the deviation abnormal signal or fluctuation abnormal signal that maintains the marking. When the reconstructed vehicle state quantity replaces the corresponding marked signal, a reconstruction identifier is added to the reconstructed vehicle state quantity. The vehicle state quantity with the reconstruction identifier does not participate in cross-consistency verification, nor is it used as the input for the reconstruction of other marked signals.

8. The intelligent acquisition and collaborative control system for multi-channel vehicle signals according to claim 7, characterized in that: The signal reconstruction module establishes and invokes the signal mapping relationship in the following way: The representative conversion coefficients for each operating condition zone are predetermined. Operating conditions whose differences between the corresponding conversion coefficients and the representative conversion coefficients are within the allowable fluctuation range are classified into the same operating condition zone. Operating conditions within the same operating condition zone share the corresponding signal mapping relationship. When the operating condition is migrated to another operating condition partition, the alternative mapping relationship corresponding to the operating condition partition to which the migrated operating condition belongs is invoked; During the migration transition period, the original signal mapping relationship and the alternative mapping relationship are used in parallel to reconstruct the same vehicle state quantity; when the deviation between the two reconstruction results is within the allowable deviation range determined according to the measurement error, the alternative mapping relationship is activated. If the deviation still exceeds the allowable range when the migration transition period ends, the alternative mapping relationship will not be enabled, the reconstructed vehicle state quantity will be marked as invalid, and the operating condition will be reviewed; the migration transition period is a preset period calculated from the start of the operating condition migration.

9. The intelligent acquisition and collaborative control system for multi-channel vehicle signals according to claim 1, characterized in that: The collaborative control module generates collaborative control instructions and determines the execution status according to the following steps: S901. Determine the target driving state based on the unlabeled signal and the reconstructed vehicle state quantity, and generate cooperative control instructions corresponding to each vehicle execution unit. The cooperative control instructions carry the target state quantity. The vehicle execution unit includes a drive motor, a brake actuator, and a steering actuator. S902. The response signal and the signal representing the same vehicle state quantity or the same vehicle component state among the multiple signals are used as associated signals. After unified representation, cross-consistency verification is performed. When the deviation between the response signal and the other associated signals exceeds the allowable deviation range, the execution link of the corresponding vehicle execution unit is determined to be abnormal. S903. When the response signal enters the response range determined by the measurement error of the corresponding response signal based on the target state quantity during the response period, the execution is deemed normal; otherwise, the execution is deemed abnormal.

10. A vehicle multi-channel signal intelligent acquisition and collaborative control system according to claim 9, characterized in that: The collaborative control module determines performance degradation and updates transmission latency in the following way: When the response signal enters the response range, the reference rise time pre-stored according to the type of cooperative control instruction is obtained; When the rise time of the response signal exceeds the allowable time range determined by the baseline rise time and normal response fluctuation, the corresponding vehicle execution unit is judged to have degraded performance and the number of degradations is accumulated. The response time period corresponding to the subsequent coordinated control command is extended according to the number of degradations by a preset step size, and does not exceed the maximum response time period of the corresponding vehicle execution unit. The response signal carries the generation time written by the vehicle execution unit based on the same clock source; the difference between the time when the response signal arrives at the corresponding acquisition channel and the generation time is used as the measured delay of the corresponding acquisition channel, and the transmission delay of the corresponding acquisition channel is updated after being weighted and fused with the current transmission delay; the fusion weight of the measured delay is less than the fusion weight of the current transmission delay, and is determined according to the reliability of the corresponding acquisition channel. The reliability is generated according to the number of transmission delay corrections within a preset statistical period and the number of times the mark is retained after verification.

Images