A control method and device for testing a vehicle intelligent restraint system

Abstract

This invention relates to the field of automotive crash testing technology, specifically to a control method and equipment for testing a vehicle intelligent restraint system. The method includes: acquiring the airbag controller acceleration curve and the front impact sensor acceleration curve from a real vehicle crash test; filtering, identifying the collision initiation time, and performing cubic spline interpolation on the airbag controller acceleration curve to convert it into theoretical acceleration; using the theoretical acceleration as a command to control the movement of the impact slide; acquiring the actual acceleration and calculating the tracking error; selecting PID coefficients from multiple sets of parameters based on the waveform stage and dynamically correcting them according to the error; using the corrected coefficients to calculate the thrust compensation and correct the motor output; and sending a trigger signal to the impact slide to synchronously initiate the signal injection of the front impact sensor acceleration curve with the impact movement. This invention can suppress instantaneous acceleration deviation and long-term displacement drift, improve waveform reproduction capability, and achieve synchronous reproduction of physical impact and collision sensor signals.

Description

Technical Field

[0001] This invention generally relates to the field of automotive crash testing technology, and specifically to a control method and equipment for testing a vehicle intelligent restraint system. Background Technology

[0002] In the development of intelligent vehicle safety, bench impact tests are necessary to verify the reliability of intelligent restraint systems. Traditional impact test benches often employ hydraulic or pneumatic drives, resulting in delayed drive response. They do not consider the effects of guide rail friction, viscous damping, and buffer reaction forces on the actual movement of the slide, easily leading to peak deviations, timing deviations, and displacement drift in the later stages between the target acceleration curve and the measured acceleration curve. Furthermore, when resampling the collision curve using linear interpolation or low-order discretization, the slopes of adjacent sampling intervals may be discontinuous, potentially causing abrupt changes in the driving force command.

[0003] Furthermore, the collision judgment of a vehicle's intelligent restraint system does not solely rely on the physical impact received by the airbag controller itself; it also typically depends on signals from external sensors such as front and side impact sensors. If the physical loading of the impact platform is not synchronized with the injection of analog signals from the sensors, or if the communication time slots between the analog signals and the airbag controller are mismatched, the relative timing sequence between the signals during a real-vehicle collision cannot be accurately reproduced, thus affecting the test results. Existing test benches may also employ only a single acceleration closed-loop or a single displacement closed-loop, making it difficult to simultaneously achieve both instantaneous acceleration tracking accuracy and full-time-domain displacement consistency.

[0004] Therefore, a control method and equipment for testing vehicle intelligent restraint systems are needed to solve the above problems. Summary of the Invention

[0005] The purpose of this invention is to provide a control method and equipment for testing intelligent restraint systems of vehicles to solve the above-mentioned problems.

[0006] In a first aspect, the present invention provides a control method for testing a vehicle intelligent restraint system, comprising: Obtain the airbag controller acceleration curve and the front collision sensor acceleration curve collected from the actual vehicle crash test; The acceleration curve of the airbag controller is converted into theoretical acceleration with a preset sampling period; The theoretical acceleration is used as the theoretical acceleration command to cause the airbag controller of the vehicle under test, which is carried by the impact slide, to start the impact movement. At the same time, the signal of the acceleration curve of the front collision sensor is injected into the collision sensor interface of the airbag controller of the vehicle under test. During the impact motion of the impact slide, the actual acceleration of the impact slide is collected in real time, and the tracking error between the theoretical acceleration command and the actual acceleration is calculated. The waveform is determined based on the theoretical acceleration value at the current moment, and the stage includes at least two of the following: rising edge stage, peak segment stage, falling edge stage, and braking segment stage. Based on the stage of the waveform, select the corresponding proportional coefficient, integral coefficient, and derivative coefficient from a set of preset control parameters; The selected proportional coefficient and differential coefficient are dynamically corrected based on the tracking error. Using the corrected proportional coefficient, integral coefficient, and differential coefficient, the tracking error is proportionally, integrally, and differentially calculated to obtain the thrust compensation amount. The motor output of the impact slide is then corrected based on the thrust compensation amount.

[0007] According to the technical solution provided by the present invention, converting the acceleration curve of the airbag controller into theoretical acceleration with a preset sampling period specifically includes: The acceleration curve of the airbag controller is filtered. The moment of collision start is identified based on the filtered acceleration, and the acceleration curve of the airbag controller is extracted with that moment as the zero point of time. The truncated airbag controller acceleration curve is converted into theoretical acceleration with a preset sampling period using cubic spline interpolation.

[0008] According to the technical solution provided by the present invention, the collision start time is identified based on the filtered acceleration, specifically including: when the absolute value of the acceleration difference between adjacent filtered times is higher than a preset starting threshold, the time is identified as the collision start time.

[0009] According to the technical solution provided by the present invention, the filtered airbag controller acceleration curve is divided into multiple intervals according to the original sampling time, and a cubic polynomial is constructed in each interval; A tridiagonal linear equation system is constructed using the following constraints: amplitude matching at interval endpoints, continuity of first and second derivatives at internal nodes, and zero second derivatives at the beginning and end of the waveform. The tridiagonal linear equations are solved using the chasing method to obtain the undetermined coefficients of each piecewise polynomial, and the theoretical acceleration corresponding to each time step is obtained.

[0010] According to the technical solution provided by the present invention, determining the stage of the waveform based on the theoretical acceleration value at the current moment specifically includes: When the theoretical acceleration value at the current moment is greater than the theoretical acceleration value at the previous moment, it is determined to be the rising edge phase; When the absolute value of the difference between the theoretical acceleration value at the current moment and the theoretical acceleration value at the previous moment is less than or equal to the preset peak slope threshold, it is determined to be the peak segment stage; When the theoretical acceleration value at the current moment is less than the theoretical acceleration value at the previous moment, it is determined to be the falling edge phase; When the theoretical acceleration value at the current moment is less than the preset braking acceleration threshold, it is determined to be the braking phase.

[0011] According to the technical solution provided by the present invention, the selected proportional coefficient and differential coefficient are dynamically corrected based on the tracking error, specifically including: The selected scaling factor is multiplied by 1 and the error correction factor is summed, where the error correction factor is equal to the first preset correction factor multiplied by the tracking error; The selected differential coefficient is multiplied by 1 and summed with the error correction coefficient, wherein the error correction coefficient is equal to the second preset correction coefficient multiplied by the tracking error; Furthermore, upper limits are set for the dynamically corrected proportional coefficient and differential coefficient respectively. When the dynamically corrected proportional coefficient or differential coefficient exceeds the corresponding upper limit, the upper limit is used as the corrected coefficient.

[0012] According to the technical solution provided by the present invention, the thrust compensation amount further includes a Coulomb friction compensation term, a viscous damping compensation term, and a theoretical missing thrust term. The Coulomb friction compensation term is preset based on the guide rail friction of the impact slide. The viscous damping compensation term is determined based on the product of the equivalent viscous damping coefficient of the impact slide and the target velocity at the current moment. The theoretical missing thrust term is determined based on the product of the total mass of the impact slide and the tracking error.

[0013] According to the technical solution provided by the present invention, a displacement loop correction step is also included: The actual displacement of the impact slide is collected in real time, and the theoretical velocity and theoretical displacement are recursively derived based on the theoretical acceleration. Calculate the displacement error between the theoretical displacement and the actual displacement, and generate an additional acceleration correction value based on the displacement error; The theoretical acceleration command for the next moment is corrected using the additional acceleration correction value.

[0014] According to the technical solution provided by the present invention, a trigger signal is sent to the impact slide to synchronize the signal injection of the acceleration curve of the front impact sensor with the impact motion of the impact slide, specifically as follows: The synchronization pulse emitted by the airbag controller of the vehicle under test is captured, and the communication period and time slot are determined based on the synchronization pulse. The acceleration curve of the front collision sensor is converted into a current modulation signal adapted to the PSI5 protocol, and the current modulation signal is output in the corresponding time slot. Simultaneously, a voltage jump trigger signal is sent to the impact slide.

[0015] Secondly, the present invention provides a vehicle intelligent restraint system testing device for implementing the control method for testing the vehicle intelligent restraint system as described above, comprising: An impact slide is used to support the airbag controller of a vehicle under test and apply impact motion. The impact slide includes a linear motor, a grating encoder, and an acceleration sensor. The grating encoder is used to obtain the actual displacement of the impact slide, and the acceleration sensor is used to obtain the actual acceleration of the impact slide. The control unit consists of a data preprocessing module, a signal injection module, and a motion control module. The control unit is connected to the collision sensor interface of the impact slide and the airbag controller of the vehicle under test, respectively. The data preprocessing module is used to convert the airbag controller acceleration curve into theoretical acceleration with a preset sampling period; The motion control module is used to execute the method described above; The signal injection module is used to inject a collision simulation signal corresponding to the acceleration curve of the front collision sensor into the airbag controller of the vehicle under test.

[0016] The beneficial effects of the embodiments of the present invention are as follows: 1. This invention generates a theoretical acceleration sequence that meets a fixed sampling period by digitally filtering, identifying the collision start time, and resampling segmented cubic splines on the acceleration curve of the airbag controller collected from a real vehicle collision. This ensures that the acceleration and rate of change of acceleration remain continuous between adjacent interpolation intervals, avoids motor thrust spikes caused by abrupt changes in discrete commands, and effectively preserves key dynamic characteristics such as collision peak value, slope, and trigger threshold, providing a foundation for high-fidelity waveform reproduction.

[0017] 2. This invention adopts a coordinated control strategy of acceleration closed loop and displacement closed loop: the acceleration loop corrects the driving force compensation amount of the current control cycle in real time based on the real-time acceleration tracking error, which helps to eliminate instantaneous acceleration deviation; the displacement loop generates additional acceleration correction value according to the deviation between actual displacement and theoretical displacement, and pre-corrects the theoretical acceleration, theoretical velocity and theoretical displacement commands of subsequent control cycles, simultaneously suppressing the accumulation of velocity error and displacement error, and solving the problems of timing drift and steady-state deviation in traditional single-loop control.

[0018] 3. Based on the different stages of the collision waveform, such as the rising edge, peak segment, falling edge, and braking segment, as well as the amplitude of the real-time tracking error, this invention adaptively selects and dynamically corrects the PID control parameters. At the same time, it incorporates the inertial force of the total mass of the platform, the Coulomb friction of the guide rail, and the equivalent viscous damping of the system into the driving force compensation, thus avoiding the distortion of local key waveforms by using only a single overall index. Attached Figure Description

[0019] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings.

[0020] Figure 1 This is a schematic diagram of the test equipment provided in the embodiments of this application.

[0021] Figure 2 This is a schematic diagram of the signal injection module structure provided in an embodiment of this application.

[0022] The diagram is labeled as follows: 100, vehicle under test; 200, impact slide; 300, control unit; 301, module controller; 302, FPGA; 303, output channel; 304, parameter extraction and feedback module. Detailed Implementation

[0023] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0024] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0025] Example 1 This invention provides a control method for testing a vehicle intelligent restraint system, comprising the following steps: First, the airbag controller acceleration curve and the front impact sensor acceleration curve collected from the actual vehicle collision test are acquired. The airbag controller acceleration curve is converted into theoretical acceleration with a preset sampling period. The theoretical acceleration is used as the theoretical acceleration command to start the impact movement of the airbag controller of the vehicle under test on the impact slide. At the same time, the signal of the front impact sensor acceleration curve is injected into the collision sensor interface of the airbag controller of the vehicle under test. That is, the start time and end time of the acquisition of the above acceleration signals should be the same, so as to preserve the relative timing between the physical impact input and the external sensor input during the actual vehicle collision.

[0026] The collision curves acquired from the actual vehicle need to be filtered and discretized by the control unit. The acceleration curve discretization uses a uniform discretization method, discretizing the curves according to the original sampling rate, and then filtering the discretized curves. A CFC600 filter is used for filtering, and the acceleration is calculated after filtering as follows:

[0027] In the formula, The filtered acceleration is at time m. Let i be the original acceleration shifted forward by i cycles from the current point. The i-th order tap coefficient of the filter, where N is the order of the filter.

[0028] For the filtered curve, when Automatic identification when the preset starting threshold is exceeded. The corresponding time is the collision start time t0, and t0=0 is set. The preset starting threshold can be set according to the vehicle type, collision speed and target waveform noise level. In this embodiment, the preset starting threshold is 3g.

[0029] Since the sampling rate can be set to different values ​​in real vehicle collisions, to ensure the consistency of data entering and exiting the impact slide, data with a sampling rate lower than 10kHz is output as data with a fixed sampling rate of 10kHz. Using cubic spline interpolation, the filtered airbag controller acceleration curve is divided into multiple intervals according to the original sampling time. Within each small interval (t... m With t m+1 Construct a cubic polynomial within ) :

[0030] In the formula, m is the interval label, m=0, 1, 2, ..., M-1, with a total of M time nodes, divided into M-1 intervals; Let be the acceleration function obtained by interpolation in the m-th interval; t is the current time, within the interval t m ≤t≤t m+1 Inside; c m0 c m1 c m2 c m3 These are coefficients to be determined.

[0031] The constraints are: amplitude matching at the interval endpoints, continuity of the first derivative at the internal nodes, continuity of the second derivative at the internal nodes, and zero second derivatives at the beginning and end of the waveform, i.e.:

[0032] Specifically: After interpolating the m-th interval, the left endpoint of this interval Corresponding acceleration Equal to time node Measured acceleration in the original curve ; After interpolating the m-th interval, the right endpoint of this interval Corresponding acceleration Equal to time node Measured acceleration in the original curve ; Public Time Nodes The first derivative at the end of the m-th interval is equal to the first derivative at the beginning of the (m+1)-th interval; Public Time Nodes The second derivative at the end of the m-th interval is equal to the second derivative at the beginning of the (m+1)-th interval. Collision start moment The second derivative is 0; Collision End Time The second derivative is 0.

[0033] Wherein, to solve for the undetermined coefficient c m0， c m1 c m2 c m3 Simplify the equation, let , , Taking the derivative of the constructed cubic polynomial, we get:

[0034]

[0035] At this point, when we substitute the two ends of the m-th interval, the left endpoint... That is, when x=0, t= That is, x= We can obtain: ,

[0036] Similarly, by substituting the left endpoint of the m-th interval into the acceleration value, we can obtain:

[0037] By substituting the right endpoint of the m-th interval into the acceleration value, we can obtain:

[0038] The above , , Substituting the expression yields

[0039] At this time, c m1 c m2 c m3 In the solution process, only the solution needs to be found. , In this embodiment, a tridiagonal system of equations can be constructed based on the continuity condition of the first derivative at the time node. The coefficient matrix of this system of equations is in sparse tridiagonal form, so the chasing method is used to solve it. , Furthermore, the coefficients of the cubic polynomial in each interval were determined to obtain the theoretical acceleration corresponding to each time point.

[0040] After the impact slide begins to move, the actual acceleration of the impact slide is collected in real time, and the tracking error between the theoretical acceleration and the actual acceleration is calculated:

[0041] In the formula, for Theoretical acceleration at time t, for The actual acceleration at any given moment, for Acceleration tracking error at any given moment.

[0042] To achieve piecewise adaptive control, the stage of the waveform is determined based on the theoretical acceleration value at the current moment. The determination method is as follows: when When this occurs, it is determined to be the rising edge phase; when At that time, it was determined to be the peak period ( (The peak slope threshold). when When this occurs, it is determined to be the falling edge phase; when At that time, it is determined to be the braking phase ( (This refers to the acceleration threshold during the braking phase).

[0043] in, This value, close to 0, can be selected based on the actual situation, indicating that the theoretical acceleration has reached its peak and is in a stable phase. It is also a value close to 0, which can be selected according to the actual situation, indicating that the collision is nearing its end and entering the final braking phase.

[0044] Based on the stage of the waveform, select the corresponding basic proportional coefficient from a set of preset control parameters. Basic integral coefficient and fundamental differential coefficients In one specific implementation, the basic control parameters for each stage are shown in the table below:

[0045] After selecting the basic parameters, adjust the basic scaling factor according to the tracking error. and fundamental differential coefficients Perform dynamic correction:

[0046]

[0047] In the formula , These are the first preset correction factor and the second preset correction factor (in this embodiment, they are 0.02 and 0.03 respectively). Furthermore, to prevent the time-varying parameters from being infinitely amplified, the following settings are provided: , The upper limits are 30 and 3 respectively. When the dynamically corrected proportional coefficient or differential coefficient exceeds the corresponding upper limit value, the upper limit value is used as the corrected coefficient.

[0048] Using the corrected proportional, integral, and derivative coefficients, proportional, integral, and derivative operations are performed on the tracking error to obtain the controller thrust compensation:

[0049] In the formula, for Thrust compensation for the controller at any time, Set a fixed sampling period for the controller.

[0050] At this point, the theoretical missing driving force is:

[0051] In the formula, for The theory of time lacks a driving force. The linear motor driving the impact slide table determines the total mass of the platform (including the vehicle controller and other structures). The thrust compensation is: =

[0052] In the formula, The Coulomb friction force can be preset based on the friction force of the guide rail of the impact slide. The equivalent viscous damping coefficient of the system is... For the first Target speed at all times.

[0053] Furthermore, acceleration errors accumulate over time, resulting in both velocity and displacement errors. Correcting only the acceleration thrust can suppress instantaneous deviations; a closed-loop displacement loop is necessary to synchronously correct both velocity and displacement commands to achieve global synchronous adjustment. In the uncorrected state, the theoretical velocity and displacement can be obtained from the following formula:

[0054]

[0055] for The theoretical velocity at time t, for The theoretical displacement at time t, The period is the discrete sampling period.

[0056] The actual velocity and displacement of the impact slide can be obtained by the following formula:

[0057]

[0058] for The actual speed at that moment for The actual displacement at any given moment.

[0059] Therefore, the single-step speed error increment is:

[0060] The single-step displacement error increment is:

[0061] The displacement loop is adjusted using the following formula for PID adaptive additional acceleration correction:

[0062] In the formula, , , These are the displacement proportionality coefficient, displacement integral coefficient, and displacement differential coefficient, respectively. Taking a specific actual loading acceleration waveform as an example, the displacement proportionality coefficient... Directly generate additional acceleration correction values; the initial value range can be set to 1-3; displacement integral coefficient. To eliminate long-term displacement drift, the initial value range can be set to 0.001-0.01; displacement differential coefficient The initial value range can be set to 0.001-0.003. The above parameter values ​​are for reference only; in actual applications, adjustments need to be made based on the actual loading curve. , , The parameters should be adjusted in a trial-and-error order, and all the above parameters should be adjusted from small to large. Finally, the parameter with the highest equipment stability and curve reproducibility should be selected as the setting parameter for this test.

[0063] The acceleration correction value at the next moment can be calculated by the following formula:

[0064] Therefore, the acceleration loop is responsible for correcting the instantaneous acceleration deviation in the current control cycle, while the displacement loop is responsible for pre-correcting the theoretical acceleration command for the next control cycle based on the accumulated displacement deviation. Together, they form an acceleration-displacement dual closed-loop control.

[0065] The operating mode and parameters of the signal injection module are configured through the control unit. In this embodiment, the injection characteristic parameters of the signal injection module include: baud rate set to 125 kbit / s, low-level current magnitude of the current waveform during communication set to 5.45 mA, high-level current magnitude set to 32.73 mA, verification mode set to CRC16, synchronization voltage threshold set to 3V, time slot set to 500 μs, and sensor initialization data retransmission count set to 5 times.

[0066] During the test, the control unit sends a start command to the signal injection module. The module controller receives the command, converts it to PSI5 protocol format, and sends a configuration command to the FPGA, controlling the FPGA to enter synchronization mode. Once in synchronization mode, the FPGA captures the synchronization pulse emitted by the airbag controller of the vehicle under test. Based on the synchronization pulse, it determines the communication period and time slot, dynamically calculates the phase offset of the front impact sensor analog signal relative to the SYNC frame, ensuring that the communication timing of the analog signal matches that of the airbag controller. The FPGA converts the stored front impact sensor analog signal into a current modulation signal under the PSI5 protocol (using Manchester encoding, where a decrease in current represents logic "0" and an increase in current represents logic "1") according to preset parameters and PSI5 protocol requirements, and injects it into the airbag controller through the output channel. Simultaneously, the signal injection module sends a 3V-10V voltage jump trigger signal to the impact slide, activating the impact slide to physically impact the airbag controller, achieving synchronous execution of signal injection and slide impact. During the test, the parameter extraction feedback module judges whether there are any abnormal phenomena such as injection interruption in the front impact sensor analog signal. If an abnormal phenomenon occurs, the test is stopped and restarted.

[0067] After one impact loading cycle, the actual acceleration curve of the impact slide and the acceleration curve of the airbag controller are divided into multiple corresponding segments according to peak values, troughs, or inflection points. The average relative error of each segment, the global average relative error of the entire curve, and the normalized root mean square error are calculated respectively.

[0068] Pass thresholds can be set for the errors of each segment, the global average relative error, and the normalized root mean square error. In one specific embodiment, each threshold can be set to 5%. When the error in a local critical segment exceeds the limit, even if the global error does not exceed the limit, the collision condition reproduction can be deemed unqualified, and the control parameters of the corresponding segment can be readjusted according to the segment where the error is located.

[0069] Example 2 like Figure 1-2 As shown, the present invention provides a vehicle intelligent restraint system testing device including a vehicle under test 100, an impact slide 200, and a control unit 300. The airbag controller of the vehicle under test 100 is placed on the impact slide 200. The control unit 300 is connected to both the impact slide 200 and the collision sensor interface of the airbag controller of the vehicle under test 100. Specifically, the front collision sensor interface of the airbag controller is connected to the control unit 300, and the control unit 300 injects the front collision sensor acceleration curve signal into the airbag controller through this interface to reproduce the collision condition.

[0070] The impact slide 200 includes a linear motor, a grating encoder, and a buffer mechanism. The linear motor drives the carrier platform to move. The grating encoder acquires the actual displacement of the carrier platform in real time. An acceleration sensor is fixed to the carrier platform or a component rigidly connected to the carrier platform to acquire the actual acceleration. The buffer mechanism is set at the end of the slide's stroke to absorb the remaining kinetic energy and limit the stroke of the carrier platform.

[0071] The control unit 300 includes a data preprocessing module, a signal injection module, and a motion control module. The data preprocessing module is used to filter and discretize the acceleration curves acquired from the real vehicle crash test; the signal injection module is used to inject a crash simulation signal into the airbag controller of the vehicle under test; and the motion control module is used to execute the control method as described in Embodiment 1.

[0072] The signal injection module includes a module controller 301, an FPGA 302, an output channel 303, and a parameter extraction and feedback module 304. The module controller 301 communicates with the control unit 300 to receive configuration commands, configure input / output channels, baud rate, reference voltage or current level, verification mode, time slot, and signals to be transmitted. The FPGA 302 executes protocol timing control and signal encoding according to the configuration parameters, and outputs the modulated signal through the output channel 303. The parameter extraction and feedback module 304 feeds back the actual output synchronization pulse, modulated waveform, time slot, phase offset, and abnormal status to the control unit 300.

[0073] The communication interface types provided in this embodiment include PSI5-CAN, CPCIe-PSI5, and PCIe-PSI5. The above interfaces are used to realize the transmission of commands from the control unit 300 and collision sensor signals.

[0074] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A control method for testing a vehicle intelligent restraint system, characterized in that, include: Obtain the airbag controller acceleration curve and the front collision sensor acceleration curve collected from the actual vehicle crash test; The acceleration curve of the airbag controller is converted into theoretical acceleration with a preset sampling period; The theoretical acceleration is used as the theoretical acceleration command to cause the airbag controller of the vehicle under test, which is carried by the impact slide, to start the impact movement. At the same time, the signal of the acceleration curve of the front collision sensor is injected into the collision sensor interface of the airbag controller of the vehicle under test. During the impact motion of the impact slide, the actual acceleration of the impact slide is collected in real time, and the tracking error between the theoretical acceleration command and the actual acceleration is calculated. The waveform is determined based on the theoretical acceleration value at the current moment, and the stage includes at least two of the following: rising edge stage, peak segment stage, falling edge stage, and braking segment stage. Based on the stage of the waveform, select the corresponding proportional coefficient, integral coefficient, and derivative coefficient from a set of preset control parameters; The selected proportional coefficient and differential coefficient are dynamically corrected based on the tracking error. Using the corrected proportional coefficient, integral coefficient, and differential coefficient, the tracking error is proportionally, integrally, and differentially calculated to obtain the thrust compensation amount. The motor output of the impact slide is then corrected based on the thrust compensation amount.

2. The control method for testing a vehicle intelligent restraint system according to claim 1, characterized in that, Converting the airbag controller acceleration curve into theoretical acceleration with a preset sampling period specifically includes: The acceleration curve of the airbag controller is filtered. The moment of collision start is identified based on the filtered acceleration, and the acceleration curve of the airbag controller is extracted with that moment as the zero point of time. The truncated airbag controller acceleration curve is converted into theoretical acceleration with a preset sampling period using cubic spline interpolation.

3. The control method for testing a vehicle intelligent restraint system according to claim 2, characterized in that, The collision start time is identified based on the filtered acceleration. Specifically, when the absolute value of the difference between the accelerations of adjacent moments after filtering is higher than a preset starting threshold, that moment is identified as the collision start time.

4. The control method for testing a vehicle intelligent restraint system according to claim 2, characterized in that, The filtered airbag controller acceleration curve is divided into multiple intervals according to the original sampling time, and a cubic polynomial is constructed in each interval. A tridiagonal linear equation system is constructed using the following constraints: amplitude matching at interval endpoints, continuity of first and second derivatives at internal nodes, and zero second derivatives at the beginning and end of the waveform. The tridiagonal linear equations are solved using the chasing method to obtain the undetermined coefficients of each piecewise polynomial, and the theoretical acceleration corresponding to each time step is obtained.

5. The control method for testing a vehicle intelligent restraint system according to claim 1, characterized in that, Determining the stage of the waveform based on the theoretical acceleration value at the current moment specifically includes: When the theoretical acceleration value at the current moment is greater than the theoretical acceleration value at the previous moment, it is determined to be the rising edge phase; When the absolute value of the difference between the theoretical acceleration value at the current moment and the theoretical acceleration value at the previous moment is less than or equal to the preset peak segment slope threshold, it is determined to be a peak segment stage. When the theoretical acceleration value at the current moment is less than the theoretical acceleration value at the previous moment, it is determined to be the falling edge phase; When the theoretical acceleration value at the current moment is less than the preset braking acceleration threshold, it is determined to be the braking phase.

6. The control method for testing a vehicle intelligent restraint system according to claim 1, characterized in that, The selected proportional coefficient and derivative coefficient are dynamically corrected based on the tracking error, specifically including: The selected scaling factor is multiplied by 1 and the error correction factor is summed, where the error correction factor is equal to the first preset correction factor multiplied by the tracking error; The selected differential coefficient is multiplied by 1 and summed with the error correction coefficient, wherein the error correction coefficient is equal to the second preset correction coefficient multiplied by the tracking error; Furthermore, upper limits are set for the dynamically corrected proportional coefficient and differential coefficient respectively. When the dynamically corrected proportional coefficient or differential coefficient exceeds the corresponding upper limit, the upper limit is used as the corrected coefficient.

7. The control method for testing a vehicle intelligent restraint system according to claim 1, characterized in that, The thrust compensation amount also includes a Coulomb friction compensation term, a viscous damping compensation term, and a theoretical missing thrust term. The Coulomb friction compensation term is preset based on the guide rail friction of the impact slide. The viscous damping compensation term is determined based on the product of the equivalent viscous damping coefficient of the impact slide and the target velocity at the current moment. The theoretical missing thrust term is determined based on the product of the total mass of the impact slide and the tracking error.

8. The control method for testing a vehicle intelligent restraint system according to claim 1, characterized in that, It also includes a displacement loop correction step: The actual displacement of the impact slide is collected in real time, and the theoretical velocity and theoretical displacement are recursively derived based on the theoretical acceleration. Calculate the displacement error between the theoretical displacement and the actual displacement, and generate an additional acceleration correction value based on the displacement error; The theoretical acceleration command for the next moment is corrected using the additional acceleration correction value.

9. The control method for testing a vehicle intelligent restraint system according to claim 1, characterized in that, A trigger signal is sent to the impact slide to synchronize the injection of the acceleration curve signal from the front impact sensor with the impact motion of the impact slide, specifically: The synchronization pulse emitted by the airbag controller of the vehicle under test is captured, and the communication period and time slot are determined based on the synchronization pulse. The acceleration curve of the front collision sensor is converted into a current modulation signal adapted to the PSI5 protocol, and the current modulation signal is output in the corresponding time slot. Simultaneously, a voltage jump trigger signal is sent to the impact slide.

10. A vehicle intelligent restraint system testing device, used to implement the control method for testing the vehicle intelligent restraint system as described in any one of claims 1 to 9, characterized in that, An impact slide (200) is used to carry the airbag controller of the vehicle under test (100) and apply impact motion. The impact slide (200) includes a linear motor, a grating encoder and an acceleration sensor. The grating encoder is used to obtain the actual displacement of the impact slide (200) and the acceleration sensor is used to obtain the actual acceleration of the impact slide. The control unit (300) consists of a data preprocessing module, a signal injection module and a motion control module. The control unit (300) is connected to the collision sensor interface of the impact slide (200) and the airbag controller of the vehicle under test (100), respectively. The data preprocessing module is used to convert the airbag controller acceleration curve into theoretical acceleration with a preset sampling period; The motion control module is used to perform the method as described in any one of claims 1 to 9; The signal injection module is used to inject a collision simulation signal corresponding to the acceleration curve of the front collision sensor into the airbag controller of the vehicle under test.

Images