A method and system for multi-point sampling and synchronous archiving of a motion sickness test of an electric passenger vehicle

Abstract

The present application relates to the technical field of data processing, in particular to a kind of electric passenger car motion sickness test multi-point sampling synchronous archiving method and system.The method comprises: defining multiple key spatial sampling points on the preset cycle condition route;Real-time monitoring vehicle position information, when determining that the vehicle reaches the key spatial sampling point, read the time stamp captured by the frame start bit hardware of the vehicle CAN controller as a unified time anchor point, and broadcast the trigger message containing the unified time anchor point;Vehicle acquisition terminal responds to trigger message, cuts data snapshot from ring buffer with unified time anchor point as reference, and writes metadata tag containing relative time stamp in data snapshot;After completing a cycle of cycle condition, the data snapshot of each key spatial sampling point is aggregated and packaged, and uploaded to archive.The technical scheme can simultaneously meet the needs of accurate anchoring of condition structure, multi-source data hardware level synchronization and data semantic indexing in motion sickness test.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, specifically to a method and system for synchronous archiving of multi-point sampling in motion sickness tests of electric passenger vehicles. Background Technology

[0002] In road tests of motion sickness comfort in electric passenger vehicles, the test vehicle typically needs to drive multiple laps along a pre-set cyclical route (approximately 5 km to 10 km in total length), while simultaneously collecting multi-source heterogeneous data for subsequent analysis. These data sources usually include: multi-axis motion perception data from dummy dummy (such as the three-axis acceleration and angular velocity of a six-freedom dummy, with sampling frequencies typically in the range of 100 Hz to 1000 Hz), three-axis motion data of the vehicle (vehicle speed, acceleration, pedal opening, etc., acquired via the OBD-II / CAN bus, with timestamps mostly captured by the CAN controller at the frame start bit), in-vehicle environmental parameter data (such as CO2 concentration, noise level, temperature, etc.), and operational log data from the condition alert system (recording indicators of driving conditions such as flat roads, acceleration, deceleration, curves, and bumps). These four types of data sources are characterized by large differences in sampling frequencies and physically independent acquisition links, but when analyzing the causes of motion sickness, they must be precisely aligned on the timeline and strongly correlated with specific driving condition stages.

[0003] The existing experimental data archiving and management technologies mainly include the following processing methods: A common approach is to perform coarse archiving by whole train runs or fixed-duration slices. This method sets the smallest unit of archiving as the entire test task or a fixed time block, with files identified only by "task number + start timestamp". However, this design primarily aims to ensure the secure storage of raw data, with the archiving granularity designed as a management unit rather than an analysis unit. In motion sickness testing scenarios, key operating condition boundaries such as straight sections, acceleration, deceleration, curves, and bumps only exist in the real-time state machine of the operating condition prompting system and are not permanently embedded in the data storage structure. This leads to subsequent data analysis relying solely on manual segmentation or rough time intervals for guesswork, resulting in significant subjective errors and difficulty in reproducing the data. Furthermore, the sheer volume of whole train run files makes batch statistical analysis across operating conditions and locations costly.

[0004] Another approach is to extract data segments N seconds before and after an event based on a Mark trigger signal. This method is typically used in fault or special event recording scenarios, and its triggering mechanism relies on event-driven mechanisms. However, motion sickness assessment requires coverage of all typical operating conditions, including starting, straightaways, acceleration, deceleration, curves, bumps, and stopping. Most of these stages do not traditionally generate "event" signals, meaning this method can only capture sporadic anchor points and cannot form a complete coverage of the operating condition structure. Furthermore, the extracted segments usually only have single labels such as "Mark=1," lacking semantic information such as point numbers and operating condition stage identifiers. Also, the extraction window is often of fixed duration, making it difficult to accurately match the dynamic temporal characteristics of actual operating conditions.

[0005] Furthermore, location-triggered technology based on GNSS (Global Navigation Satellite System) geofencing has also been used for data archiving. This method triggers data recording by determining whether a vehicle has reached a specific geographical location. However, in actual field test environments, GNSS signals are susceptible to multipath effects and obstruction in areas such as tunnels, tree-lined roads, or urban canyons, resulting in a drift of 5 m to 15 m. This leads to false or missed geofencing triggers, resulting in missing data for critical operating points. Simultaneously, GNSS outputs position estimates, and its time base is not synchronized with the hardware clock of the vehicle-mounted acquisition system, resulting in implicit alignment errors between the captured location points and the CAN message timeline.

[0006] Finally, regarding the time alignment problem of multi-source data, existing technologies generally rely excessively on post-processing algorithm compensation. Since the dummy sensor link and the vehicle CAN link are physically independent, the buffer queuing delay during data transmission is non-constant, especially during abrupt acceleration and deceleration. Post-processing linear correction or interpolation padding is essentially mathematical processing based on estimation, which cannot recover the original high-frequency data that has been emptied and covered by the buffer. Furthermore, the region with the largest correction residual is often precisely the most critical dynamic peak segment. More importantly, simple time alignment only solves the problem of data occurring at the same moment, failing to address the semantic structure problem of "what operating condition stage this data is from," making it difficult to meet the needs of large-scale data reuse. Summary of the Invention

[0007] The purpose of this invention is to propose a multi-point sampling and synchronous archiving method and system for motion sickness testing of electric passenger vehicles, which can simultaneously meet the requirements of accurate anchoring of working condition structure, hardware-level synchronization of multi-source data, and semantic indexing of data in motion sickness testing.

[0008] To achieve the above objectives, in a first aspect, the present invention proposes a method for simultaneous archiving of multi-point sampling in motion sickness tests of electric passenger vehicles, comprising: Define multiple key spatial sampling points on the preset cyclic operating route; The vehicle location information is monitored in real time. When it is determined that the vehicle has arrived at the key spatial sampling point, the timestamp captured by the on-board CAN controller at the start of the frame is read as a unified time anchor point, and a trigger message containing the unified time anchor point is broadcast. In response to the trigger message, the vehicle-mounted data acquisition terminal cuts a data snapshot from the circular buffer based on the unified time anchor point, and writes a metadata tag containing a relative timestamp into the data snapshot; After completing one cycle of the test, the data snapshots of each key spatial sampling point are aggregated, packaged, and uploaded to the test data management platform for archiving.

[0009] Beneficial effects of the basic solution: This technical solution differs from traditional archiving methods based on fuzzy timelines or isolated events. By directly using key spatial sampling points in a pre-defined cyclical operating condition route as the trigger source for data extraction, it makes the structure of dizziness-inducing operating conditions explicit. Existing technologies generally only record the trigger at a certain moment, and the operating condition information carried by the data is implicit, requiring manual comparison or decoding later. This technical solution writes metadata tags containing relative timestamps at the same time as the data snapshot is generated, enabling the original collected data to have self-descriptive operating conditions at the moment of generation. For downstream data governance, dizziness map construction, or AI training, the input data is no longer a timeline lacking operating condition annotations and requiring computational power for identification, but rather structured data with built-in operating condition key values, lowering the threshold for data cleaning and feature extraction.

[0010] This technical solution uses the timestamp captured by the hardware of the Start of Frame (SOF) of the vehicle CAN controller as a unified time anchor point, and performs circular buffer clipping based on this anchor point, which solves the clock drift and asynchronous problem between vehicle sensors, dummy sensors and environmental data.

[0011] Traditional solutions often rely on software clock calibration or post-processing interpolation alignment between different terminals, which not only accumulates errors but also fails to recover data where signal peaks are truncated or distorted due to clock drift. This technical solution uses the timestamp captured by the onboard CAN controller at the Start of Frame (SOF) as a unified and unique time anchor point, and uses this as the basis for cropping data snapshots from the circular buffer. This method eliminates the errors caused by post-processing linear correction or interpolation due to differences in clock sources from different devices, ensuring that data from dummies, vehicles, and the environment fall on the same timeline at the source of acquisition. It solves the problem of peak data being masked or misaligned due to time shifts, improving the consistency of multi-point sampling data.

[0012] As a feasible and preferred solution, real-time monitoring of vehicle location information is implemented. When it is determined that the vehicle has arrived at the key spatial sampling point, the timestamp captured by the on-board CAN controller at the start-of-frame hardware is read as a unified time anchor point. Specifically, this includes: Triggering determination is performed using GNSS geofencing determination mode or DMI cumulative mileage determination mode. In the GNSS geofence determination mode, when the GNSS positioning reliability meets the preset conditions and the vehicle enters the circular determination area with the sampling point coordinates as the center and the preset radius as the radius, the determination is triggered; In the DMI cumulative mileage determination mode, when the GNSS positioning reliability does not meet the preset conditions, the cumulative driving distance is calculated using the wheel speed signal on the vehicle CAN bus. When the difference between the cumulative mileage and the preset sampling point mileage calibration value is less than the preset threshold, a determination is triggered. Regardless of the determination mode used, the timestamp captured by the CAN controller at the SOF sampling point of the current frame is immediately read as the unified time anchor point when the determination is triggered.

[0013] As a feasible preferred solution, the vehicle-mounted data acquisition terminal responds to the trigger message by cropping a data snapshot from the circular buffer based on the unified time anchor point, specifically including: Parse the trigger message to obtain the unified time anchor point and operating condition stage identifier; The cutting window is dynamically set according to the working condition stage identifier, wherein a first cutting window is set for working condition stages involving sudden changes in longitudinal acceleration, and a second cutting window is set for working condition stages with stable motion. The data snapshot is generated by extracting data frames for the corresponding time period from the circular buffer of the dummy sensor, vehicle CAN message, and environmental sensor channels according to the clipping window.

[0014] As a feasible and preferred approach, metadata tags containing relative timestamps are written into the data snapshot, specifically including writing three-level tags: The first-level label is used to identify the spatial anchor point identity of the data source; The second-level label is used to identify the semantics of the working condition stage to which the data belongs; The third-level label is a relative timestamp, which records the offset of the timestamp of each frame of data relative to the unified time anchor point.

[0015] As a feasible and preferred solution, the data snapshots of each key spatial sampling point after completing one cycle of operation are aggregated and encapsulated, specifically including: Once the cycle is determined to be complete and all preset key spatial sampling point data are collected, all snapshot data of this cycle will be aggregated into an archive package according to a preset format. The archive package contains the test task number, lap number, a collection of snapshot data of all points in that lap, and test element information.

[0016] As a feasible and preferred option, the data is uploaded to the experimental data management platform for archiving, specifically including: After receiving the archive package, the test data management platform establishes a five-level index directory structure, which is in the following order: task level, circle level, point level, working condition level, and data type level. Perform pre-verification on the archived packages entering the database. The pre-verification includes timescale consistency verification and data integrity verification. If the verification finds time alignment abnormalities or the proportion of valid frames is lower than the lower limit, the data for that cycle will be marked as a downgraded use flag.

[0017] As a feasible and preferred solution, timescale consistency verification specifically includes: For peak-dominated operating conditions, extract the peak time of vehicle motion data and the peak time of dummy response data, calculate the time deviation between the two, and determine that the time alignment is abnormal if it exceeds a preset threshold. For non-peak-dominated operating conditions, the continuity of CAN frame sequence numbers is the primary verification method.

[0018] As a feasible preferred solution, the key spatial sampling points are located according to the dizzying working condition structure, covering seven typical working condition stages: stationary start, straight and constant speed, acceleration and overtaking, deceleration and braking, driving on a curve, bumpy road surface and end of parking; the key spatial sampling points are not evenly distributed, and each sampling point corresponds to an independent trigger radius.

[0019] As a feasible preferred solution, the unified time anchor point originates from a free-running counter inside the CAN controller, whose current value is automatically captured by hardware at the frame start bit sampling point of each CAN message and latched as a timestamp field.

[0020] Secondly, the present invention also proposes a multi-point sampling and synchronous archiving system for motion sickness testing of electric passenger vehicles, which is applied to the above-mentioned multi-point sampling and synchronous archiving method for motion sickness testing of electric passenger vehicles. Attached Figure Description

[0021] Figure 1 This is a logical diagram of a multi-point sampling and synchronous archiving method for motion sickness testing of electric passenger vehicles.

[0022] Figure 2 This is a schematic diagram showing the locations of P1 to P7 on the cyclic operating route.

[0023] Figure 3 This is a schematic diagram of the trigger / broadcast / trimming timing.

[0024] Figure 4 This is a schematic diagram of a five-level index directory hierarchy.

[0025] Figure 5 This is a schematic diagram of the adaptive timescale consistency verification process.

[0026] Figure 6 This is a schematic diagram of the system architecture. Detailed Implementation

[0027] To make the technical solution and advantages of this application clearer, the technical solution of the present invention will be further described in detail below with reference to the accompanying drawings. It is understood that the specific embodiments described herein are only some embodiments of the present invention, and are only used to explain this application, not to limit it. It should be noted that the technical features or combinations of technical features described in the following embodiments should not be considered isolated; they can be combined with each other to achieve better technical effects. The same reference numerals appearing in the accompanying drawings of the following embodiments represent the same features or components, and can be applied to different embodiments.

[0028] Furthermore, unless otherwise defined, the technical or scientific terms used in this invention description shall have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains.

[0029] The present invention will now be described in further detail with reference to the accompanying drawings.

[0030] Example 1 Reference Figure 1 This invention provides a method for simultaneous archiving of multi-point sampling in motion sickness tests of electric passenger vehicles, comprising: Step S100: Hardware setup and connection of the test system. The test system includes a working condition prompting system, an on-board data acquisition terminal (DAQ), and a test data management platform, including: Step S101, Installation and configuration of the operating condition notification system, including: The operating condition alert system serves as the trigger and time base for this solution, and its core hardware is an in-vehicle embedded host. This host is installed below the dashboard in front of the passenger side of the vehicle under test or near the center console, and secured with a shock-absorbing bracket to prevent equipment vibration caused by vehicle bumps.

[0031] In terms of electrical connectivity, the system connects to the vehicle's CAN bus (powertrain / chassis CAN) via the OBD-II (On-Board Diagnostics II) diagnostic interface for real-time reading of vehicle motion data such as vehicle speed, accelerator pedal opening, brake pedal opening, longitudinal / lateral / vertical acceleration (a_x / a_y / a_z), and yaw rate. A GNSS antenna, mounted inside the windshield or externally on the roof, connects to the GNSS module inside the main unit to acquire high-precision latitude and longitude coordinates and positioning accuracy. For power supply, it draws power from the vehicle's 12V power interface and is equipped with a UPS capacitor or a small-capacity battery as backup power to prevent system restarts due to vehicle shutdown or voltage fluctuations.

[0032] The key hardware principle of this system lies in utilizing the 16-bit free-running counter within the CAN controller (such as bxCAN / MCAN / FDCAN). During vehicle CAN bus communication, when each CAN message frame is transmitted, the CAN controller automatically captures the current value of this counter at the SOF (Start of Frame) sampling point of each CAN message frame and latches it as the timestamp field of that frame message, denoted as t_CAN. This unified time anchor point is hardware-latched by the free-running counter within the CAN controller at the SOF sampling point, providing a unified hardware time base for all data sources.

[0033] Step S102, installation and configuration of the vehicle-mounted data acquisition terminal (DAQ), including: The vehicle-mounted data acquisition terminal (DAQ), serving as the snapshot cropping and tagging actuator, is installed in the rear compartment, luggage compartment, or rear seat rail of the vehicle. It is equipped with a shock absorber and is placed as close as possible to the dummy amplifier chassis to shorten the sensor wiring length.

[0034] For signal connectivity, the dummy sensor (six-axis free-roaming dummy) is installed in the driver's or passenger's position. Its multi-axis motion sensing signal output is sent to an amplifier, whose analog output is connected to the DAQ's analog input channel. In-vehicle environmental parameter sensors (such as CO2 concentration sensors, noise sound pressure level sensors, and temperature sensors) are connected to the DAQ's analog or digital input channels, respectively. A branch of the vehicle's CAN bus is split off via an OBD-Y splitter and connected to the DAQ's CAN interface, ensuring that the DAQ and the operating condition alert system are connected to the same CAN physical bus, thus guaranteeing that they share the same clock source.

[0035] The causal chain relationship between motion state and data flow is as follows: The test vehicle's power source is a drive motor. When the driver depresses the accelerator pedal, the drive motor, as the prime mover, outputs torque. This torque is transmitted to the wheels through the reducer and half-shafts, causing the wheels to rotate as driven components. The interaction between the wheels and the ground generates longitudinal thrust, which in turn propels the vehicle body into longitudinal translational motion (forward during starting or acceleration, and backward during deceleration and braking). When the driver depresses the brake pedal, the brake calipers clamp the brake discs, generating braking torque that reduces wheel speed. When the driver turns the steering wheel, the steering column and tie rod change the wheel angle, causing the vehicle to yaw about its vertical axis, manifested as cornering motion. When the vehicle travels over bumpy roads, the vertical profile of the road surface causes elastic deformation of the tires, which in turn causes suspension compression or extension, resulting in vertical translation and pitch rotation of the vehicle body. All the aforementioned vehicle body motion states are characterized by vehicle CAN messages (including signals such as a_x / a_y / a_z and yaw rate), and are captured by the CAN controller SOF sampling point hardware and solidified into data frames with t_CAN timestamps. At the same time, the dummy's inertial response relative to the seat is sampled by an amplifier and inherits the same CAN clock source within the DAQ, forming a precisely aligned frame sequence.

[0036] Step S103, deployment of the experimental data management platform, includes: The test data management platform is deployed in the test center's computer room and consists of rack servers and storage arrays. This platform does not directly participate in vehicle movement but communicates with the onboard devices via a network. During implementation, the DAQ connects to the test site's local area network via onboard Ethernet or WLAN to upload processed data packets to the management platform server.

[0037] Step S200, defining and deploying key spatial sampling points, including: On a pre-set cyclical route, based on the mechanism of motion sickness, multiple key spatial sampling points are defined. In this embodiment, the total length is approximately 6.8 km, and seven key spatial sampling points, P1 to P7, are defined, referring to... Figure 2 The key spatial sampling points are not evenly distributed, but are located according to the dizzying working conditions, covering seven typical working conditions: starting from a standstill, driving at a constant speed on a straight road, accelerating and overtaking, decelerating and braking, driving on a curve, driving on a bumpy road, and finishing a stop, as shown in Table 1.

[0038] Table 1

[0039] Step S300, sampling point triggering mechanism, includes: Step S301, parameter entry and initialization: Enter the geographic coordinates (lat_i, lon_i), trigger radius R_i, and corresponding mileage calibration value S_i of P1 to P7 in the working condition prompt system. The system default startup mode is Mode A (GNSS geofence).

[0040] Step S302, GNSS geofence determination – Mode A (normal): During vehicle operation, the status alert system obtains the current vehicle position in real time via the GNSS antenna. The system constructs a circular determination area with the coordinates (lat_i, lon_i) of each sampling point as the center and R_i (e.g., 15m~20m) as the radius.

[0041] The determination formula is:

[0042] in, and This refers to the latitude and longitude output by the current GNSS. This is the conversion factor between latitude and longitude and meters.

[0043] Simultaneously, the system monitors the positioning confidence parameters (such as HDOP or confidence flags) output by the GNSS module in real time. When the positioning confidence is greater than or equal to 0.8, the system maintains mode A operation. Once it is determined that the vehicle has entered the circular region of a certain sampling point Pi, the system immediately enters the trigger state.

[0044] Step S303, DMI (Distance Measurement Interface) cumulative mileage determination - Mode B (backoff): When the vehicle enters a GNSS signal obstruction area (such as a tunnel), and the GNSS positioning reliability is detected to be less than 0.8 for N consecutive cycles (N=3 in this embodiment), the system automatically and smoothly switches to Mode B.

[0045] In Mode B, the system uses the wheel speed signals on the vehicle's CAN bus to calculate the cumulative driving distance using the DMI algorithm. The determination formula has been changed to:

[0046] in, The mileage trigger threshold is set to 8m to 12m in this embodiment. Triggering is determined when the difference between the accumulated mileage and the preset sampling point mileage calibration value is less than the threshold.

[0047] This scheme constructs a dual-mode triggering mechanism for GNSS and DMI (Distance Measurement Interface). When the GNSS confidence level decreases, the system can smoothly retreat to DMI accumulated mileage for triggering. Regardless of the triggering path used, the same CAN hardware time anchor point is broadcast, avoiding the introduction of new time base splits due to triggering path switching and ensuring the stability of position triggering under various road conditions.

[0048] Step S304: Unified Time Anchor Point Broadcast. Regardless of whether Mode A or Mode B is used for triggering, when the judgment condition is met, the operating condition prompt system does not rely on GNSS time, but immediately reads the timestamp captured by the CAN controller at the SOF sampling point of the current frame. The system broadcasts the trigger message to the entire network via the CAN bus. This message has a triplet structure: For example, when the vehicle reaches P3 (the acceleration and overtaking section), the broadcast message is: This step ensures that the time base of the trigger signal always originates from the CAN hardware clock, avoiding time axis splitting caused by trigger source switching.

[0049] Step S400: Snapshot cropping and tagging by the vehicle-mounted data acquisition terminal (DAQ). After receiving the trigger message broadcast by the operating condition alert system, the DAQ performs snapshot cropping and tagging operations. Taking the vehicle reaching P3 (acceleration overtaking section) as an example, the specific steps are as follows: Step S401: Receive the trigger message; the DAQ's CAN interface module parses the triplet message. The current trigger position P3 is identified, the operating condition is acceleration (ACC), and the time anchor point is... .

[0050] Step S402, circular buffer trimming: The DAQ internally maintains a circular buffer for all acquisition channels. The system uses... Using this as a reference point, the cutting window is dynamically set according to the characteristics of different working conditions, with reference to... Figure 2 .

[0051] For P3 (acceleration overtaking segment) and P4 (deceleration braking segment), due to the drastic change in longitudinal acceleration, the peak characteristics are obvious and the duration is short. It is necessary to accurately capture the complete transient process before and after the peak. The clipping window is set as follows:

[0052] For the remaining sampling points (such as P1 / P2 / P5 / P7), the motion state is relatively stable, and the clipping window is set as follows:

[0053] Based on the aforementioned window parameters, DAQ accurately extracts data frames for the corresponding time periods from the circular buffers of channels such as dummy triaxial acceleration, vehicle CAN messages, and environmental sensors (CO2, noise, temperature).

[0054] Step S403: Write a level 3 tag. In the header of the snapshot data packet generated by cropping, DAQ writes a level 3 metadata tag to explicitly solidify the working condition structure into the data structure.

[0055] First-level tag tag_point = P3: Identifies the spatial anchor point identity of the data source.

[0056] The second-level tag tag_stage = ACC: identifies the semantics of the working condition stage to which the data belongs.

[0057] Third-level tag rel_ts = t_frame t_CAN calculates the timestamp of each frame of data relative to the trigger anchor point. The relative offset.

[0058] By using rel_ts relative timestamps, the internal timing sequence of the snapshot is decoupled from the system's absolute clock. Even if the system time changes during subsequent data playback, the time series inside the snapshot still maintains relative correctness.

[0059] Step S500, aggregating data by cycle and archiving it into the database, includes: Step S501, Loop Aggregation and Encapsulation: After the vehicle completes one lap cycle, passing through P1 to P7 sequentially and triggering data collection, the DAQ's internal logic monitoring module determines "Loop completed / P1...P7 collected". At this time, the DAQ aggregates all snapshot data of this lap according to a preset format.

[0060] The archive package structure is defined as follows:

[0061] in, This experiment task is numbered. For the circle number, Includes snapshot data of all seven points in the circle. It includes metadata such as the tester, time, and vehicle model.

[0062] After packaging, DAQ uploads the archived package to the test data management platform via vehicle Ethernet / WLAN.

[0063] Step S502: The five-level index is stored in the database. After receiving the archive package, the experimental data management platform parses the data structure and establishes a five-level index directory on the server storage array, referring to... Figure 4 The index hierarchy is as follows: TaskID (task level) / LapID (lap level) / PointID (point level) / StageID (stage level) / DataType (data type level) For example, the Task_001 / Lap_002 / P3 / ACCEL / fake_3axis / directory stores the dummy's three-axis acceleration data under the acceleration condition at point P3 in the second lap of Task 1.

[0064] This index structure allows complex retrieval needs (such as "finding all train deceleration segment data" or "finding the vertical trend of the same physical location") to be directly located using a combination key, avoiding the inefficient overhead of scanning large files of the entire train.

[0065] In step S503, before the data is formally entered into the analysis model library, the management platform performs a pre-verification on each archive package to identify and mark abnormal data.

[0066] Step S503-1, timescale consistency verification, refer to Figure 5 .

[0067] Since this solution uses the CAN hardware clock as the sole anchor point, theoretically, vehicle motion data and dummy response data should be precisely aligned on the time axis. This characteristic is verified using a peak matching algorithm.

[0068] For P3 (acceleration segment), extract the longitudinal resultant acceleration amplitude |a| or a x peak time t v And the peak time of the axial response of the dummy. t f .

[0069] For P4 (deceleration phase), extract the peak moment of vehicle braking deceleration and the peak moment of dummy response.

[0070] Calculate time deviation:

[0071] Set threshold (Typical value 50ms), if > If the time alignment is abnormal, the system will automatically set the flag flag_tmisalign=1.

[0072] For non-peak-dominated segments such as P1 / P2 / P5 / P7, the continuity of the CAN frame sequence number is prioritized, and the peak alignment weight is reduced.

[0073] Verification 2: Data integrity verification. The platform parses the archive package and calculates the percentage of valid frames for each channel, including dummy sensor data, vehicle CAN data, environmental data, and log data.

[0074] Calculation formula:

[0075] Set a lower limit for the percentage of valid frames (e.g., 80%), if any type of data If the value is below the lower limit, the flag_incomplete is set to 1.

[0076] If an archive package is marked with `flag_tmisalign` or `flag_incomplete`, the data for that cycle is set to be used in a downgraded manner. This data can be used for trend observation or fault location analysis, but it is excluded from the standard input set for subsequent motion modeling and knowledge graphs, thus ensuring the data quality for higher-level analysis.

[0077] This solution moves data quality control to the pre-archiving stage. By monitoring the deviation Δt between the peak times of vehicles and dummies within the same snapshot, it can promptly detect and flag hidden data slippage issues caused by link latency configuration errors (e.g., Δt > 50ms). This changes the passive situation of traditional solutions where data mismatches are only discovered during post-event compensation or modeling, effectively intercepting invalid data from entering the database and ensuring the availability of experimental data from the source.

[0078] Step S600, retrieval and playback, can be performed through the API structure provided by the management platform, including: Full cycle playback: Input TaskID, and the system will automatically stitch together snapshots of each point in the order of P1 to P7 to generate continuous time series data of the full cycle working condition for playback.

[0079] Cross-train operation condition comparison: Input StageID (e.g., "deceleration and braking"), and the system will retrieve the data segments corresponding to StageID from all TaskIDs to form a cross-train stage comparison dataset.

[0080] Same-point evolution analysis: Input PointID (e.g., P6), and the system will extract data from the same physical location on different dates and in different test tasks to analyze the vertical excitation time series evolution of that location.

[0081] Anomaly Reverse Lookup: System output includes The list helps to quickly locate abnormal aspects of the experiment.

[0082] This disclosure also provides a multi-point sampling and synchronous archiving system for motion sickness testing of electric passenger vehicles, applied to the aforementioned multi-point sampling and synchronous archiving method for motion sickness testing of electric passenger vehicles, with reference to... Figure 6 It includes a working condition alert system, an on-board data acquisition terminal, and a test data management platform.

[0083] The above content is merely an embodiment of the present invention. Commonly known structures and characteristics are not described in detail here. Those skilled in the art are aware of all common technical knowledge in the field prior to the application date or priority date, are aware of all existing technologies in that field, and have the ability to apply conventional experimental methods prior to that date. Those skilled in the art can improve and implement the present invention based on the guidance provided in this application and their own capabilities. Typical well-known structures or operating methods should not be obstacles for those skilled in the art to implement this application. It should be noted that those skilled in the art can make several modifications and improvements without departing from the structure of the present invention, and these should also be considered within the scope of protection of the present invention. These modifications and improvements will not affect the effectiveness of the present invention or the practicality of the patent. The scope of protection claimed in this application should be determined by the content of its claims, and the specific embodiments described in the specification can be used to interpret the content of the claims.

Claims

1. A method for simultaneous archiving of multi-point sampling in motion sickness tests of electric passenger vehicles, characterized in that, include: Define multiple key spatial sampling points on the preset cyclic operating route; The vehicle location information is monitored in real time. When it is determined that the vehicle has arrived at the key spatial sampling point, the timestamp captured by the on-board CAN controller at the start of the frame is read as a unified time anchor point, and a trigger message containing the unified time anchor point is broadcast. In response to the trigger message, the vehicle-mounted data acquisition terminal cuts a data snapshot from the circular buffer based on the unified time anchor point, and writes a metadata tag containing a relative timestamp into the data snapshot; After completing one cycle of the test, the data snapshots of each key spatial sampling point are aggregated, packaged, and uploaded to the test data management platform for archiving.

2. The method for simultaneous archiving of multi-point sampling in motion sickness testing of electric passenger vehicles according to claim 1, characterized in that, Real-time monitoring of vehicle location information; when it is determined that the vehicle has arrived at the key spatial sampling point, the timestamp captured by the on-board CAN controller at the start-of-frame hardware is read as a unified time anchor point, specifically including: Triggering determination is performed using GNSS geofencing determination mode or DMI cumulative mileage determination mode. In the GNSS geofence determination mode, when the GNSS positioning reliability meets the preset conditions and the vehicle enters the circular determination area with the sampling point coordinates as the center and the preset radius as the radius, the determination is triggered; In the DMI cumulative mileage determination mode, when the GNSS positioning reliability does not meet the preset conditions, the cumulative driving distance is calculated using the wheel speed signal on the vehicle CAN bus. When the difference between the cumulative mileage and the preset sampling point mileage calibration value is less than the preset threshold, a determination is triggered. Regardless of the determination mode used, the timestamp captured by the CAN controller at the SOF sampling point of the current frame is immediately read as the unified time anchor point when the determination is triggered.

3. The method for simultaneous archiving of multi-point sampling in motion sickness testing of electric passenger vehicles according to claim 2, characterized in that, In response to the trigger message, the vehicle-mounted data acquisition terminal cuts a data snapshot from the circular buffer based on the unified time anchor point, specifically including: Parse the trigger message to obtain the unified time anchor point and operating condition stage identifier; The cutting window is dynamically set according to the working condition stage identifier, wherein a first cutting window is set for working condition stages involving sudden changes in longitudinal acceleration, and a second cutting window is set for working condition stages with stable motion. The data snapshot is generated by extracting data frames for the corresponding time period from the circular buffer of the dummy sensor, vehicle CAN message, and environmental sensor channels according to the clipping window.

4. The method for simultaneous archiving of multi-point sampling in motion sickness testing of electric passenger vehicles according to claim 1, characterized in that, Write metadata tags containing relative timestamps into the data snapshot, specifically including writing three-level tags: The first-level label is used to identify the spatial anchor point identity of the data source; The second-level label is used to identify the semantics of the working condition stage to which the data belongs; The third-level label is a relative timestamp, which records the offset of the timestamp of each frame of data relative to the unified time anchor point.

5. The method for simultaneous archiving of multi-point sampling in motion sickness testing of electric passenger vehicles according to claim 1, characterized in that, The data snapshots of each key spatial sampling point after completing one cycle of operation are aggregated and encapsulated, specifically including: Once the cycle is determined to be complete and all preset key spatial sampling point data are collected, all snapshot data of this cycle will be aggregated into an archive package according to a preset format. The archive package contains the test task number, lap number, a collection of snapshot data of all points in that lap, and test element information.

6. The method for simultaneous archiving of multi-point sampling in motion sickness testing of electric passenger vehicles according to claim 1, characterized in that, Upload the data to the experimental data management platform for archiving, specifically including: After receiving the archive package, the test data management platform establishes a five-level index directory structure, which is in the following order: task level, circle level, point level, working condition level, and data type level. Perform pre-verification on the archived packages entering the database. The pre-verification includes timescale consistency verification and data integrity verification. If the verification finds time alignment abnormalities or the proportion of valid frames is lower than the lower limit, the data for that cycle will be marked as a downgraded use flag.

7. The method for simultaneous archiving of multi-point sampling in motion sickness testing of electric passenger vehicles according to claim 6, characterized in that, Timescale consistency verification specifically includes: For peak-dominated operating conditions, extract the peak time of vehicle motion data and the peak time of dummy response data, calculate the time deviation between the two, and determine that the time alignment is abnormal if it exceeds a preset threshold. For non-peak-dominated operating conditions, the continuity of CAN frame sequence numbers is the primary verification method.

8. The method for simultaneous archiving of multi-point sampling in motion sickness testing of electric passenger vehicles according to claim 1, characterized in that, The key spatial sampling points are located according to the dizzying working condition structure, covering seven typical working condition stages: stationary start, straight and constant speed, acceleration and overtaking, deceleration and braking, cornering, bumpy road surface and end of parking; the key spatial sampling points are not evenly distributed, and each sampling point corresponds to an independent trigger radius.

9. The method for simultaneous archiving of multi-point sampling in motion sickness testing of electric passenger vehicles according to claim 1, characterized in that, The unified time anchor point originates from the free-running counter inside the CAN controller. The counter automatically captures the current value at the frame start bit sampling point of each CAN message and latches it as a timestamp field.

10. A multi-point sampling and synchronous archiving system for motion sickness testing of electric passenger vehicles, characterized in that, This method is applied to a multi-point sampling and synchronous archiving method for motion sickness testing of electric passenger vehicles as described in any one of claims 1-9.

Images