Vehicle state detection method and device, computer device, and storage medium

Abstract

The present disclosure provides a vehicle state detection method and device, computer equipment and a storage medium. The vehicle state detection method of the present disclosure comprises: obtaining target data of a vehicle in a first time period; the target data comprises at least one of speed, multi-axis angular velocity, multi-axis acceleration, altitude, Euler angle, speed average, speed change, multi-axis angular velocity change, multi-axis acceleration change, altitude change and Euler angle change; performing sliding window processing on at least part of the target data to obtain N groups of sliding window data; matching M groups of sliding window data with identification information of at least one detection target, and according to the matching results corresponding to each detection target respectively, determining the number of groups of sliding window data that meet a first preset condition in the matching results corresponding to each detection target respectively, and taking the number as a first quantity; M is a positive integer less than or equal to N; and determining the actual state of the vehicle according to the first quantity corresponding to each detection target.

Description

Technical Field

[0001] This disclosure belongs to the field of intelligent driving technology, specifically relating to a method and apparatus for detecting vehicle status, computer equipment, and storage medium. Background Technology

[0002] Intelligent driving relies on various sensors to perceive the surrounding environment, but the performance of these sensors is limited under different external conditions. For example, in adverse weather conditions such as rain and fog, the penetration performance of radar signals is significantly weakened; when the positioning signal of the positioning system is obstructed by obstacles (such as buildings, vegetation, or terrain barriers), its positioning accuracy decreases significantly, leading to increased errors. Therefore, how to accurately perceive the surrounding driving environment and determine the vehicle's current driving status is a technical problem that urgently needs to be solved in the field of intelligent driving. Summary of the Invention

[0003] This disclosure aims to at least solve one of the technical problems existing in the prior art, and to provide a method and apparatus for detecting vehicle status, a computer device and a storage medium.

[0004] Firstly, the technical solution adopted to solve the technical problem of this disclosure is a method for detecting vehicle status, comprising: acquiring target data of the vehicle within a first time period; the target data includes at least one of speed, multi-axis angular velocity, multi-axis acceleration, elevation, Euler angle, average speed, speed change, multi-axis angular velocity change, multi-axis acceleration change, elevation change, and Euler angle change;

[0005] At least a portion of the target data is subjected to sliding window processing to obtain N sets of sliding window data; N is a positive integer greater than or equal to 1;

[0006] The M groups of sliding window data are matched with the identification information of at least one detection target, and the number of sliding window data groups that satisfy the first preset condition in the matching results corresponding to each detection target is determined according to the matching results corresponding to each detection target, and is used as the first quantity; M is a positive integer less than or equal to N;

[0007] The actual state of the vehicle is determined based on the first quantity corresponding to each of the detection targets.

[0008] Secondly, this disclosure also provides a vehicle status detection device, including a data acquisition module, a first detection module, and a second detection module;

[0009] The data acquisition module is configured to acquire target data of the vehicle within a first time period; the target data includes at least one of speed, multi-axis angular velocity, multi-axis acceleration, elevation, Euler angle, mean speed, speed change, multi-axis angular velocity change, multi-axis acceleration change, elevation change, and Euler angle change.

[0010] The sliding window processing module is configured to perform sliding window processing on at least a portion of the target data to obtain N sets of sliding window data;

[0011] The first detection module is configured to match M sets of sliding window data with the identification information of at least one detection target, and determine the number of sets of sliding window data that satisfy the first preset condition in the matching results corresponding to each detection target according to the matching results corresponding to each detection target, and use this as the first quantity; M is less than or equal to N;

[0012] The second detection module is configured to determine the actual state of the vehicle based on the first quantity corresponding to each of the detection targets.

[0013] Thirdly, embodiments of this disclosure also provide a computer device, including: a processor, a memory, and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the computer device is running, the processor communicates with the memory via the bus, and when the machine-readable instructions are executed by the processor, the steps of the vehicle state detection method as described in any one of the first aspects are performed.

[0014] Fourthly, embodiments of this disclosure also provide a computer non-transient readable storage medium, wherein a computer program is stored on the computer non-transient readable storage medium, and the computer program is executed by a processor to perform the steps of the vehicle state detection method as described in any one of the first aspects.

[0015] The vehicle state detection method, apparatus, computer equipment, and storage medium provided in this disclosure include: The vehicle state detection method involves acquiring target data of the vehicle within a first time period and performing sliding window processing on the target data to obtain N sets of sliding window data, each set including multiple target data. This disclosure utilizes sliding window data in sets, which improves detection accuracy compared to using only single sensor data. Furthermore, the algorithm is simple, and the sliding window processing is efficient and convenient. Additionally, the number of sliding window data sets that satisfy a first preset condition is determined; a higher number of sets indicates a greater likelihood that the vehicle meets the candidate state of the first preset condition. Finally, the actual state of the vehicle is determined based on the number of sliding window data sets satisfying the first preset condition for each detected target. This disclosure analyzes the complementarity and redundancy of sliding window data among the detected targets and integrates different sensor data (different types of target data) to perform comprehensive and multi-angle identification of vehicle driving conditions. This improves the accuracy and reliability of road condition identification, resulting in a more accurate understanding of the vehicle's actual state and helping the vehicle make more accurate judgments and decisions in the face of complex and changing road conditions. Attached Figure Description

[0016] Figure 1 A flowchart of a vehicle status detection method provided in an embodiment of this disclosure.

[0017] Figure 2 This is a three-dimensional spatial coordinate distribution diagram of the vehicle condition detection system.

[0018] Figure 3a and Figure 3b These are schematic diagrams illustrating different sliding window methods provided in the embodiments of this disclosure.

[0019] Figure 4 This is a schematic diagram illustrating sliding window processing of different types of target data within a first time period, provided as an embodiment of this disclosure.

[0020] Figure 5 A flowchart illustrating the accumulation of a first quantity of sliding window data under a single type, as provided in this embodiment of the disclosure.

[0021] Figure 6 This is a flowchart illustrating a vehicle status detection method provided in an embodiment of the present disclosure.

[0022] Figure 7 This is a schematic diagram of a vehicle status detection device provided in an embodiment of this disclosure.

[0023] Figure 8 This is a schematic diagram of the structure of a computer device provided in an embodiment of this disclosure. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of the embodiments of this disclosure clearer, the technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this disclosure, and not all of them. The components of the embodiments of this disclosure described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this disclosure provided in the accompanying drawings is not intended to limit the scope of the claimed disclosure, but merely represents selected embodiments of this disclosure. All other embodiments obtained by those skilled in the art based on the embodiments of this disclosure without inventive effort are within the scope of protection of this disclosure.

[0025] Unless otherwise defined, the technical or scientific terms used in this disclosure shall have the ordinary meaning understood by one of ordinary skill in the art to which this disclosure pertains. The terms “first,” “second,” and similar terms used in this disclosure do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Similarly, the terms “an,” “a,” or “the,” and similar terms do not indicate a quantity limitation, but rather indicate the presence of at least one. The terms “including,” “comprising,” or “containing,” and similar terms mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. The terms “connected,” “linked,” or similar terms are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. The terms “upper,” “lower,” “left,” and “right,” etc., are used only to indicate relative positional relationships, and these relative positional relationships may change accordingly when the absolute position of the described objects changes.

[0026] In this disclosure, "multiple or several" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0027] Vehicle driving status detection technology faces a series of challenges. For example, when using data from sensors such as LiDAR, millimeter-wave radar, and cameras in combination, the differences in the output frequencies of each sensor (e.g., a typical camera exposure frequency is 30Hz, while the scanning frequency of LiDAR is between 10 and 30Hz) can lead to difficulties in aligning data triggering times. Furthermore, LiDAR outputs massive amounts of point cloud data; to ensure sufficient coverage and high resolution in different scenarios, multiple camera devices are often required. This not only increases the complexity of the hardware system but also presents significant technical obstacles to building and integrating such a multi-source sensor fusion system in the short term, hindering its rapid application in vehicle systems. In addition, the efficient operation of this system places stringent requirements on the processor's computing power and performance standards. For example, in the process of constructing and optimizing filters using vehicle dynamics models to obtain real-time vehicle driving status and slope information, the following problems exist: First, the vehicle dynamics model is essentially a complex nonlinear system. When using filtering techniques such as Kalman filtering (KF), extended Kalman filtering (EFK), and unscented Kalman filtering (UKF) for model optimization, there is a risk that the filter may converge to a local optimum, which deviates from the global optimum of the actual driving scenario. Second, there is an unavoidable time delay in the process of transmitting torque from the engine to each transmission module and in the process of the braking system transmitting braking torque to each braking module. This will further exacerbate the mismatch between the detected vehicle driving status and the actual driving scenario, resulting in significant errors. Furthermore, using autonomous driving scene recognition algorithms presents problems such as complex algorithm structure, long algorithm processing time, and inability to quickly respond to the vehicle's current driving status. Moreover, at the current stage of development of intelligent driving vehicles, the confidence level of the autonomous driving algorithm output is not high in extreme situations or abnormal case scenarios, and there is a possibility of misjudgment.

[0028] Furthermore, vehicle driving status detection typically relies on deep learning for obstacle detection and localization, while reinforcement learning is used for decision-making. However, this method is highly dependent on sensor data and has stringent requirements for computing power (especially CPUs and GPUs). When vehicles are in harsh environments, some sensors may fail, affecting the accuracy of the final detection results. Moreover, the high standards and cost of such hardware systems supporting vehicle driving status detection technology are often not conducive to practical applications.

[0029] Therefore, this disclosure provides a method for detecting vehicle status, wherein the executing entity of the method is a device with certain computing capabilities. For example, the executing entity of the method may be a vehicle controller, and the hardware type of the controller is not limited to various types of chips.

[0030] Figure 1 A flowchart of the vehicle status detection method provided in the embodiments of this disclosure is shown below. Figure 1 As shown, it includes S11 to S14.

[0031] S11. Obtain the target data of the vehicle within the first time period.

[0032] The target data is sensor data (e.g., IMU data and P-BOX data) or data determined based on sensor data. Sensor data can be, for example, IMU data (e.g., multi-axis angular velocity and multi-axis acceleration), P-BOX data (e.g., position, velocity, attitude (Euler angles), elevation, and confidence level), and wheel speedometer data (e.g., wheel speed). Specifically, the target data can include at least one of the following: velocity, multi-axis angular velocity, multi-axis acceleration, elevation, Euler angles, mean velocity, velocity variation, multi-axis angular velocity variation, multi-axis acceleration variation, elevation variation, and Euler angle variation. The variations in parameters such as velocity variation, multi-axis angular velocity variation, multi-axis acceleration variation, elevation variation, and Euler angle variation are calculated based on the sensor data.

[0033] The vehicle condition detection system determines the vehicle's attitude in three-dimensional space using parameters in the X, Y, and Z directions. For example... Figure 2 The diagram shows the three-dimensional spatial coordinate distribution of the vehicle condition detection system. In this system, the X-axis represents the lateral direction (from the driver's seat to the passenger's seat); the Y-axis represents the longitudinal direction (from the rear of the vehicle to the front); and the Z-axis represents the vertical direction (from the vehicle chassis to the roof).

[0034] Speed ​​can be obtained through direct measurement, such as using a P-BOX positioning device. Alternatively, speed can be obtained through acceleration integration. Taking acceleration integration as an example, the acceleration corresponding to each of the three axes (X-axis, Y-axis, and Z-axis) can be integrated separately to obtain the three-axis speed. The speeds of the three axes (X-axis, Y-axis, and Z-axis), that is, the speeds in the three directions, are combined into a single speed, which is recorded as the speed in the target data. Alternatively, the integrated X-axis speed, Y-axis speed, and Z-axis speed can all be recorded as the speed in the target data.

[0035] Multi-axis angular velocities include X-axis angular velocity, Y-axis angular velocity, and Z-axis angular velocity. The X-axis angular velocity, Y-axis angular velocity, and Z-axis angular velocity can be measured by three single-axis gyroscopes in an Inertial Measurement Unit (IMU).

[0036] Multi-axis acceleration includes X-axis acceleration, Y-axis acceleration, and Z-axis acceleration. X-axis acceleration, Y-axis acceleration, and Z-axis acceleration can be measured by three single-axis accelerometers in an Inertial Measurement Unit (IMU).

[0037] Elevation refers to the change in a vehicle's height under stress. Elevation can be calculated by combining IMU and wheel speedometer data, or it can be directly measured using a P-BOX positioning device.

[0038] Euler angles include pitch, roll, and yaw. Pitch can be obtained by integrating the angular velocity along the X-axis. Roll can be obtained by integrating the angular velocity along the Y-axis. Yaw can be obtained by integrating the angular velocity along the Z-axis.

[0039] The average speed is the average speed over a fixed time period.

[0040] Changes in velocity, multi-axis angular velocity, multi-axis acceleration, elevation, and Euler angles are collectively referred to as parameter changes. Parameter changes can be understood as the variation of each parameter within a sliding window over the entire window, characterizing the fluctuation of parameters over a period of time. A sliding window can be understood as a pre-set fixed time period, or as a selection of a fixed number of parameters.

[0041] In one possible implementation, the target data includes multi-axis angular velocity, multi-axis angular velocity variation, Euler angles, Euler angle variation, and elevation variation. For example, the target data includes X-axis angular velocity, X-axis angular velocity variation, pitch angle, pitch angle variation, and elevation variation. Using the above target data, the vehicle's uphill and downhill status can be detected.

[0042] In one possible implementation, the target data includes multi-axis angular velocities, multi-axis accelerations, and Euler angle changes. For example, the target data includes Z-axis angular velocity, X-axis acceleration, Y-axis acceleration, and yaw angle changes. This target data can be used to detect the vehicle's turning state.

[0043] In one possible implementation, the target data includes multi-axis acceleration, Euler angle changes, and velocity changes. For example, the target data includes Y-axis acceleration, pitch angle changes, and velocity changes. Using this target data, the vehicle's acceleration and deceleration states can be detected.

[0044] In one possible implementation, the target data includes multi-axis angular velocities, multi-axis accelerations, and Euler angle changes. For example, the target data includes X-axis angular velocity, Y-axis angular velocity, Z-axis angular velocity, X-axis acceleration, Y-axis acceleration, Z-axis acceleration, pitch angle change, and roll angle change. Using the above target data, vehicle bump conditions can be detected.

[0045] In one possible implementation, the target data includes multi-axis angular velocity, multi-axis acceleration, Euler angle variation, and average velocity. For example, the target data includes Z-axis angular velocity, Z-axis acceleration, yaw angle variation, and a first average velocity. Here, the first average velocity refers to the average value of the X-axis velocity over a fixed time period. Using the above target data, the straight-line state of the vehicle can be detected.

[0046] In one possible implementation, the target data includes multi-axis angular velocities, changes in multi-axis angular velocities, average velocity, and changes in velocity. For example, the target data includes X-axis angular velocity, Y-axis angular velocity, Z-axis angular velocity, X-axis acceleration, Y-axis acceleration, Z-axis acceleration, average second velocity, and changes in velocity. Here, the average second velocity refers to the average value of the combined velocity of the three axes (X, Y, and Z) in one direction over a fixed time period. Using the above target data, the detection of a stationary vehicle state can be achieved.

[0047] S12. Perform sliding window processing on at least a portion of the target data to obtain N sets of sliding window data.

[0048] Where N is a positive integer greater than or equal to 1. At least some of the target data are, for example, multiple target data collected continuously within a first time period. In one implementation, the second time period corresponding to each set of sliding window data is determined based on the frequency of the sensor collecting the sensor data; the target data is sensor data (e.g., IMU data and P-BOX data, etc.) or determined based on sensor data; the magnitude of the sensor frequency is inversely proportional to the length of the second time period. Figure 3a and Figure 3b As shown, with the number of target data selected by the sliding window remaining constant, a higher sensor frequency results in a shorter second time period for the same amount of data. The second time period is shorter than the first time period.

[0049] Figure 3a and Figure 3b These are schematic diagrams illustrating different sliding window methods provided in embodiments of this disclosure. Sliding window processing: In one case, such as... Figure 3a As shown, when the second time period is short, or when it can be understood that the amount of target data within the sliding window is small, the sliding window method can be that the target data within adjacent sliding windows does not overlap. In another case, such as... Figure 3bAs shown, when the second time period is long, or when the amount of target data in the sliding window is large, the sliding window method can be that the target data in adjacent sliding windows has some overlap.

[0050] For example, a sliding window process is performed on all target data within the first time period. One sliding window yields a set of sliding window data, and so on, resulting in N sets of sliding window data corresponding to N sliding windows. Figure 3a and Figure 3b As shown, the sub-data within the same sliding window are of the same type, and a sub-data is also a target data within the first time period.

[0051] The target data includes various types listed above, such as X-axis velocity, Y-axis velocity, Z-axis velocity, composite velocity (the velocity resulting from the X, Y, and Z axes), X-axis angular velocity, Y-axis angular velocity, Z-axis angular velocity, X-axis acceleration, Y-axis acceleration, Z-axis acceleration, elevation, pitch angle, roll angle, and yaw angle, first average velocity, second average velocity, velocity change, X-axis angular velocity change, Y-axis angular velocity change, Z-axis angular velocity change, X-axis acceleration change, Y-axis acceleration change, Z-axis acceleration change, elevation change, pitch angle change, roll angle change, and yaw angle change.

[0052] Figure 4 This is a schematic diagram illustrating sliding window processing of different types of target data within a first time period, provided by an embodiment of this disclosure. Figure 4 As shown, for target data of the same type, such as type A, sliding window processing within the first time period yields v1 sets of sliding window data. Sliding window processing for type B within the first time period yields v2 sets of sliding window data. Both v1 and v2 are positive integers greater than or equal to 1 and less than N. Sliding window processing is performed simultaneously on multiple target data of different types, yielding N sets of sliding window data, where N = v1 + v2 + ... The terms "type A" and "type B" refer to one of the types listed above; A and B are of different types.

[0053] S13. Match the M groups of sliding window data with the identification information of at least one detection target, and determine the number of sliding window data groups that meet the first preset condition in the matching results corresponding to each detection target according to the matching results corresponding to each detection target, and use it as the first quantity.

[0054] Where M is a positive integer less than or equal to N. The detection target corresponds to the type of target data, or in other words, the detection target corresponds to the type of sliding window data. For example, the detection target includes detecting any one of the following: X-axis velocity, Y-axis velocity, Z-axis velocity, composite velocity (the velocity synthesized from the X, Y, and Z axes), X-axis angular velocity, Y-axis angular velocity, Z-axis angular velocity, X-axis acceleration, Y-axis acceleration, Z-axis acceleration, elevation, pitch angle, roll angle, yaw angle, first average velocity, second average velocity, velocity change, X-axis angular velocity change, Y-axis angular velocity change, Z-axis angular velocity change, X-axis acceleration change, Y-axis acceleration change, Z-axis acceleration change, elevation change, pitch angle change, roll angle change, and yaw angle change.

[0055] The identification information of the detected target is a defining value, such as a pre-set detection threshold used to measure the current detected target.

[0056] The number of groups in group M can be the sum of the number of sliding window data groups under the detection type corresponding to each detection target, such as... Figure 4 As shown, if the detection targets are detection type A and detection type B, then M = v1 + v2.

[0057] The matching result of the detection target is either the result where the sub-data in the sliding window data is less than the detection threshold, or the result where the sub-data in the sliding window data is greater than or equal to the detection threshold.

[0058] Different detection targets may correspond to different first preset conditions. For a single detection target, based on the matching result corresponding to the detection target, it is necessary to count the total number of sub-data points less than the detection threshold in a single sliding window, or the total number of sub-data points greater than the detection threshold in a single sliding window. In one case, the total number of sub-data points less than the detection threshold in a single sliding window is greater than the preset threshold, which satisfies the first preset condition corresponding to the current detection target, and also means that the vehicle corresponds to one or more possible candidate states of the detection target, such as being stationary. In another case, the total number of sub-data points greater than or equal to the detection threshold in a single sliding window is greater than the preset threshold, which satisfies the first preset condition corresponding to the current detection target, and also means that the vehicle corresponds to one or more possible candidate states of the detection target, such as uphill / downhill, turning, acceleration / deceleration, bumpy, or straight-line states when not stationary.

[0059] The number of sliding window data sets that meet the first preset condition out of the cumulative M sets of sliding window data is taken as the first quantity.

[0060] S14. Determine the actual state of the vehicle based on the first quantity corresponding to each detection target.

[0061] Based on the candidate states of the vehicle initially predicted in S13 that meet the first preset condition, the magnitude of the first quantity directly affects the actual state of the vehicle (which can also be understood as the current actual driving state). In one possible implementation, if the first quantities corresponding to multiple detection targets for the target state are all greater than or equal to the number of times a threshold is reached, then the actual state of the vehicle corresponding to the first time period is determined to be the target state. The target state includes at least one of the following: stationary state, uphill / downhill state, turning state, acceleration / deceleration state, bumpy state, and straight-line state. The so-called "target state" refers to a hypothetical candidate state of the vehicle, which is further verified by the first quantities corresponding to each detection target under this hypothetical state. If the verification is successful, it proves that the currently detected target state is indeed the actual state of the vehicle.

[0062] Each target state corresponds to multiple detection targets, which improves the accuracy of vehicle state detection. The first number corresponding to each of the multiple detection targets is greater than or equal to a threshold number, meaning that if the vehicle is in the target state multiple times within a first time period, the vehicle's actual state can be determined as the target state.

[0063] The vehicle state detection method provided in this disclosure collects and calculates target data using sensors already present in the vehicle (such as IMU, GPS, and wheel speedometers). This method of collecting information using existing vehicle sensors is real-time and reliable. Furthermore, by performing sliding window processing on at least a portion of the target data within a first time period, N sets of sliding window data can be obtained; each set of sliding window data includes multiple target data. This disclosure detects targets using sliding window data in sets, which improves detection accuracy compared to using only single sensor data. Simultaneously, the algorithm is simple, and the sliding window processing is efficient and convenient. Additionally, the number of sliding window data sets that satisfy a first preset condition is determined. The more sets there are, the greater the probability that the vehicle meets the candidate state of the first preset condition. Finally, the actual state of the vehicle is determined based on the number of sliding window data sets that satisfy the first preset condition for each detected target. This disclosure analyzes the complementarity and redundancy of sliding window data among various detection targets, and integrates data from different sensors (different types of target data) to identify vehicle driving conditions from all angles and perspectives. This can improve the accuracy and reliability of road condition identification, thereby obtaining a more accurate understanding of the actual state of the vehicle. In turn, it can help the vehicle make more accurate judgments and decisions in a timely manner when facing complex and ever-changing road conditions.

[0064] In some embodiments, for step S13 described above, such as Figure 5As shown, it is a flowchart of the first quantity of sliding window data accumulation under a single type provided in the embodiments of this disclosure, specifically including S131 to S134.

[0065] S131. For any set of sliding window data, determine the detection threshold of the target based on the type of sliding window data.

[0066] The types of sliding window data, that is, the types of sub-data (or target data) within the sliding window data, include any one of the following: X-axis velocity, Y-axis velocity, Z-axis velocity, composite velocity (the velocity synthesized from the X, Y, and Z axes), X-axis angular velocity, Y-axis angular velocity, Z-axis angular velocity, X-axis acceleration, Y-axis acceleration, Z-axis acceleration, elevation, pitch angle, roll angle, yaw angle, first speed mean, second speed mean, velocity change, X-axis angular velocity change, Y-axis angular velocity change, Z-axis angular velocity change, X-axis acceleration change, Y-axis acceleration change, Z-axis acceleration change, elevation change, pitch angle change, roll angle change, and yaw angle change. It should be noted that sliding window data of the same type may include multiple groups, for example, v groups, where v is a positive integer less than or equal to M. Figure 4 As shown, the sliding window data under type A is group v1, and the sliding window data under type B is group v2. If M groups of sliding window data correspond to multiple detection targets, then the total number of sliding window data groups corresponding to each detection target type is M.

[0067] For example, if the sliding window data consists of multiple X-axis angular velocities, then the detection target is to detect the X-axis angular velocity. In this way, the pre-set detection threshold of the X-axis angular velocity can be directly obtained, which is the first threshold mentioned below.

[0068] Specifically, the detection thresholds may include a first threshold and / or an eleventh threshold compared to the X-axis angular velocity, a second threshold compared to the pitch angle, a third threshold compared to the change in X-axis angular velocity, a fourth threshold and / or a nineteenth threshold compared to the change in pitch angle, a fifth threshold compared to the change in elevation, a sixth threshold and / or a fourteenth threshold compared to the Z-axis angular velocity, a seventh threshold and / or a fifteenth threshold compared to the X-axis acceleration, an eighth threshold and / or a sixteenth threshold compared to the Y-axis acceleration, a ninth threshold compared to the change in yaw angle, a tenth threshold compared to the change in velocity (Y-axis velocity change), a twelfth threshold and / or a thirteenth threshold compared to the Y-axis angular velocity, a seventeenth threshold and / or an eighteenth threshold compared to the Z-axis acceleration, a twentieth threshold and / or a twenty-first threshold compared to the change in roll angle, a twenty-second threshold compared to the first average velocity, a twenty-third threshold compared to the second average velocity, and a twenty-fourth threshold compared to the change in velocity (composite velocity change).

[0069] S132. Compare the absolute value of each sub-data in the sliding window data with the detection threshold to obtain the second number of sub-data that meet the comparison conditions.

[0070] like Figure 4 As shown, taking X-axis angular velocity as the type of sliding window data as an example, a set of sliding window data includes multiple sub-data, such as multiple X-axis angular velocities. The absolute value of each X-axis angular velocity is compared with a first threshold. If the absolute value of the X-axis angular velocity is greater than the first threshold, it meets the comparison condition of the current detection target (X-axis angular velocity). Finally, the second number of sub-data that meet the comparison condition in a set of sliding window data is counted.

[0071] This step only uses X-axis angular velocity as an example. The same applies to other types of sliding window data, and repeated parts will not be repeated.

[0072] S133. If the second quantity is greater than or equal to the preset threshold, then determine the first sliding window state of the sliding window data.

[0073] The preset threshold is set based on the detection target and practical experience, and this embodiment does not limit the specific value. The first sliding window state only represents the state of a single sliding window data point, which can characterize the trend that the vehicle may generate this state, but cannot characterize the candidate states of the vehicle. The first sliding window state is characterized by a second quantity being greater than or equal to the preset threshold, which also means that there are more sub-data points in the sliding window data that meet the comparison conditions.

[0074] S134. The number of the first sliding window states of the v group of sliding window data corresponding to the type of cumulative sliding window data is used as the first quantity.

[0075] Taking X-axis angular velocity as the type of sliding window data as an example, iterate through each group of v sliding window data. If the absolute value of the X-axis angular velocity in the sliding window data is greater than the first threshold, and the second number is greater than or equal to the preset threshold, then the sliding window data is incremented by 1. The number of sliding window data groups accumulated by iterating through all v sliding window data is used as the first number.

[0076] In one possible implementation, for detecting uphill / downhill states, the detection targets include detecting X-axis angular velocity, detecting the change in X-axis angular velocity, detecting pitch angle, detecting the change in pitch angle, and detecting the change in elevation. If the first quantity corresponding to each of the above detection targets is greater than or equal to a threshold number, then the actual state of the vehicle within the first time period is determined to be uphill / downhill.

[0077] In one possible implementation, the detection targets for turning include detecting the Z-axis angular velocity, X-axis acceleration, Y-axis acceleration, and the change in yaw angle. If the first quantity corresponding to each of the above detection targets is greater than or equal to a threshold number, then the actual state of the vehicle within the first time period is determined to be a turning state.

[0078] In one possible implementation, for detecting acceleration / deceleration, the detection targets include detecting Y-axis acceleration, detecting pitch angle change, and detecting speed change. If the first quantity corresponding to each of the above detection targets is greater than or equal to a threshold number, then the actual state of the vehicle within the first time period is determined to be an acceleration / deceleration state.

[0079] In one possible implementation, the detection targets for the bumpy state include detecting the X-axis angular velocity, Y-axis angular velocity, Z-axis angular velocity, X-axis acceleration, Y-axis acceleration, Z-axis acceleration, pitch angle change, and roll angle change. If the first quantity corresponding to each of the above detection targets is greater than or equal to a threshold number, then the actual state of the vehicle during the first time period is determined to be a bumpy state.

[0080] In one possible implementation, for detecting a straight-line state, the detection targets include detecting the Z-axis angular velocity, detecting the Z-axis acceleration, detecting the change in yaw angle, and detecting the average value of a first velocity. If the first quantity corresponding to each of the above detection targets is greater than or equal to a threshold number, then the actual state of the vehicle within the first time period is determined to be a straight-line state.

[0081] In one possible implementation, for detecting a stationary state, the detection targets include detecting the X-axis angular velocity, the Y-axis angular velocity, the Z-axis angular velocity, the X-axis acceleration, the Y-axis acceleration, the Z-axis acceleration, the average second velocity, and the velocity change. If the first quantity corresponding to each of the above detection targets is greater than or equal to a threshold number, then the actual state of the vehicle within the first time period is determined to be a stationary state.

[0082] For ease of understanding, the following explanation uses a single driving status detection as an example; however, this does not constitute a limitation for simultaneous detection of multiple driving statuses in actual applications.

[0083] In some embodiments, the detection of uphill / downhill conditions specifically includes the following process.

[0084] The detection target is the X-axis angular velocity. There is a set of sliding window data of type v1. The number of X-axis angular velocities whose absolute values ​​exceed a first threshold is counted in each set of sliding window data. The total number of sets in group v1 where the second value exceeds a1 is accumulated, and this number is used as the first value corresponding to the X-axis angular velocity. Here, a1 is a preset threshold used when detecting the X-axis angular velocity.

[0085] The detection target is the pitch angle. There are two groups of pitch angles corresponding to type v2. The absolute value of multiple pitch angles in each group of sliding window data is counted to be greater than a second threshold. The total number of groups in group v2 where the second threshold is greater than a2 is then accumulated, and this number is used as the first threshold for the pitch angle. Here, a2 is a preset threshold used for pitch angle detection.

[0086] The detection target is the change in X-axis angular velocity. There are X-axis angular velocity changes under type v3. For each group of sliding window data, the absolute value of multiple X-axis angular velocity changes is counted to be greater than a third threshold (the second number). The number of groups in group v3 whose second number is greater than a3 is then accumulated and used as the first number corresponding to the X-axis angular velocity change. Here, a3 is a preset threshold used when detecting the change in X-axis angular velocity.

[0087] The detection target is the pitch angle change. There are pitch angle changes under type v4. For each group of sliding window data, the absolute value of multiple pitch angle changes exceeding the fourth threshold is counted as the second number. The number of groups in v4 where the second number exceeds a4 is accumulated, and this is taken as the first number corresponding to the pitch angle change. Here, a4 is the preset threshold used for detecting pitch angle changes.

[0088] The detection target is to detect elevation changes. In this case, there are elevation changes under the corresponding type in group v5. The number of groups in each sliding window data set whose absolute values ​​of multiple elevation changes are greater than the fifth threshold is counted. The total number of groups in group v5 whose second value is greater than a5 is then accumulated, and this number is used as the first value corresponding to the elevation change. Here, a5 is a preset threshold used when detecting pitch angle changes.

[0089] Here, the sum of v1, v2, v3, v4, and v5 is M. v1, v2, v3, v4, and v5 may be the same or different, depending on the size of the sliding window corresponding to each type.

[0090] If the first quantity corresponding to the X-axis angular velocity is greater than the number of times threshold b1, the first quantity corresponding to the pitch angle is greater than the number of times threshold b2, the first quantity corresponding to the change in X-axis angular velocity is greater than the number of times threshold b3, the first quantity corresponding to the change in pitch angle is greater than the number of times threshold b4, and the first quantity corresponding to the change in elevation is greater than the number of times threshold b5, then the actual state of the vehicle in the first time period is determined to be an uphill or downhill state.

[0091] For ease of understanding, the above method of determining uphill and downhill states is expressed in the form of Expression 1.

[0092] Expression 1:

[0093]

[0094] For example, if the X-axis angular velocity is greater than a positive first threshold, the pitch angle is greater than a positive second threshold, the change in X-axis angular velocity is greater than a positive third threshold, the change in pitch angle is greater than a positive fourth threshold, and the change in elevation is greater than a positive fifth threshold, and the other two quantitative conditions in Expression 1 are satisfied, then the vehicle is determined to be in an uphill state.

[0095] For example, if the X-axis angular velocity is less than or equal to a negative first threshold, the pitch angle is less than or equal to a negative second threshold, the change in X-axis angular velocity is less than or equal to a negative third threshold, the change in pitch angle is less than or equal to a negative fourth threshold, and the change in elevation is less than or equal to a negative fifth threshold, and the other two quantitative conditions in Expression 1 are satisfied, then the vehicle is determined to be in a downhill state.

[0096] In this embodiment, for detecting uphill and downhill conditions, when the intelligent vehicle is traveling on sloping surfaces such as ramps, the system can accurately identify turning and other rotational movements of the vehicle by measuring the X-axis angular velocity and its change; it can accurately identify the slope of the vehicle on the road by measuring the pitch angle and its change; and it can effectively determine the elevation changes of the vehicle on the road surface. Combining these data, it can accurately determine whether the vehicle is currently in an uphill or downhill scenario.

[0097] In some embodiments, the detection of a turning state specifically includes the following process.

[0098] The detection target is the Z-axis angular velocity. There are 6 groups of sliding window data corresponding to this type. For each group of sliding window data, the number of groups where the absolute value of multiple Z-axis angular velocities is greater than a sixth threshold is counted. The total number of groups of Z-axis angular velocities whose second value is greater than a6 is accumulated and used as the first value corresponding to the Z-axis angular velocity. Here, a6 is a preset threshold used when detecting the Z-axis angular velocity.

[0099] The detection target is X-axis acceleration. There are sliding window data sets under type v7. For each group of sliding window data, the number of absolute values ​​of multiple X-axis accelerations exceeding the seventh threshold is counted. The total number of groups in group v7 where the second number of X-axis accelerations exceeds a7 is then accumulated, and this number is used as the first quantity corresponding to the X-axis acceleration. Here, a7 is a preset threshold used when detecting X-axis acceleration.

[0100] The detection target is Y-axis acceleration. There are 8 groups of sliding window data corresponding to this type. For each group of sliding window data, the number of absolute values ​​of multiple Y-axis accelerations exceeding the eighth threshold is counted. The total number of groups in group V8 where the second number of Y-axis accelerations exceeds a8 is then accumulated, and this number is used as the first quantity corresponding to the Y-axis acceleration. Here, a8 is a preset threshold used when detecting Y-axis acceleration.

[0101] The detection target is the change in yaw angle. There are sliding window data sets under type v9. For each group of sliding window data, the absolute value of multiple yaw angle changes exceeding the ninth threshold is counted as the second number. The number of groups in group v9 where the second number exceeds a9 is accumulated, and this is taken as the first number corresponding to the yaw angle change. Here, a9 is a preset threshold used when detecting the change in yaw angle.

[0102] Here, the sum of v6, v7, v8, and v9 is M. v6, v7, v8, and v9 may be the same or different, depending on the size of the sliding window corresponding to each type.

[0103] If the first quantity corresponding to the Z-axis angular velocity is greater than the number of times threshold b6, the first quantity corresponding to the X-axis acceleration is greater than the number of times threshold b7, the first quantity corresponding to the Y-axis acceleration is greater than the number of times threshold b8, and the first quantity corresponding to the yaw angle change is greater than the number of times threshold b9, then the actual state of the vehicle in the first time period is determined to be a turning state.

[0104] For ease of understanding, the above method of determining the turning state is represented by expression two.

[0105] Expression 2:

[0106] For example, if the Z-axis angular velocity is greater than the positive sixth threshold, the X-axis acceleration is greater than the positive seventh threshold, the Y-axis acceleration is greater than the positive eighth threshold, and the yaw change is greater than the positive ninth threshold, and the other two quantitative conditions in Expression 2 are satisfied, then the vehicle is determined to be in a left-turn state.

[0107] For example, if the Z-axis angular velocity is less than or equal to the negative sixth threshold, the X-axis acceleration is less than or equal to the negative seventh threshold, the Y-axis acceleration is less than or equal to the negative eighth threshold, and the yaw change is less than or equal to the negative ninth threshold, and the other two quantitative conditions in Expression 2 are satisfied, then the vehicle is determined to be in a right-turn state.

[0108] In this embodiment, for the detection of the aforementioned turning state, when the intelligent vehicle is driving on a road surface with a turning scenario, the vehicle's rotational movements such as turning can be accurately identified through the vehicle's Euler angles, especially the yaw angle, and the Z-axis angular velocity; simultaneously, the vehicle's acceleration and deceleration movements on the road surface can be identified using the X-axis and Y-axis acceleration. Combining the above data, it is possible to accurately determine whether the vehicle is currently in a turning scenario (wide turn, tight turn, left turn, and right turn, etc.).

[0109] In some embodiments, the detection of acceleration / deceleration states specifically includes the following process.

[0110] The detection target is to detect Y-axis acceleration. At this time, there are sliding window data under the corresponding type in group v8. The second number of multiple Y-axis accelerations with absolute values ​​greater than the eighth threshold in each group of sliding window data is counted, and the number of groups with the second number of Y-axis accelerations greater than a8 in group v8 is accumulated as the first number corresponding to Y-axis acceleration.

[0111] The detection target is to detect the pitch angle change. At this time, there is a pitch angle change under the corresponding type of group v4. The second number of absolute values ​​of multiple pitch angle changes in each group of sliding window data that are greater than the fourth threshold is counted. The number of groups in group v4 where the second number is greater than a4 is accumulated and used as the first number corresponding to the pitch angle change.

[0112] The detection target is the change in velocity. Here, the change in velocity can be the change in Y-axis velocity or the composite velocity change. There are 10 groups of velocity changes corresponding to this type. For each group of sliding window data, the absolute value of multiple velocity changes exceeding the tenth threshold is counted as the second number. The number of groups in group v10 whose second number exceeds a10 is then accumulated, and this number is used as the first quantity corresponding to the velocity change. Here, a10 is a preset threshold used when detecting velocity changes.

[0113] Here, the sum of v8, v4, and v10 is M. v8, v4, and v10 may be the same or different, depending on the size of the sliding window corresponding to each type.

[0114] If the first quantity corresponding to the Y-axis acceleration is greater than the number of times threshold b8, the first quantity corresponding to the pitch angle change is greater than the number of times threshold b4, and the first quantity corresponding to the speed change is greater than the number of times threshold b10, then the actual state of the vehicle in the first time period is determined to be an acceleration / deceleration state.

[0115] For ease of understanding, the above method for determining acceleration and deceleration states is expressed in the form of Expression 3.

[0116] Expression 3:

[0117] For example, if the Y-axis acceleration is greater than the positive eighth threshold, the pitch angle change is greater than the positive fourth threshold, and the velocity change is greater than the positive tenth threshold, and the other two quantitative conditions in Expression 3 are satisfied, then the vehicle is determined to be in an acceleration state.

[0118] For example, if the Y-axis acceleration is less than or equal to the negative eighth threshold, the pitch angle change is less than or equal to the negative fourth threshold, and the velocity change is less than or equal to the negative tenth threshold, and the other two quantitative conditions of expression three are satisfied, then the vehicle is determined to be in a deceleration state.

[0119] Regarding the detection of acceleration and deceleration states, when an intelligent vehicle is traveling on a road surface with acceleration and deceleration scenarios, changes in the vehicle's forward speed can be identified through changes in Y-axis velocity, acceleration can be identified through Y-axis acceleration, and turning and other rotational movements can be identified through changes in pitch angle. Combining these data, it is possible to accurately determine whether the vehicle is currently in an acceleration or deceleration scenario.

[0120] In some embodiments, the detection of bumpy conditions specifically includes the following process.

[0121] The detection target is to detect the X-axis angular velocity. At this time, there is a sliding window data of type v1. The second number of multiple X-axis angular velocities in each sliding window data group that have an absolute value greater than the first threshold and less than the eleventh threshold is counted. The number of groups in group v1 where the second number is greater than a1 is accumulated and used as the first number corresponding to the X-axis angular velocity.

[0122] The detection target is the Y-axis angular velocity. There are 11 groups of sliding window data corresponding to this type. For each group, the absolute value of multiple Y-axis angular velocities is counted as a second number greater than the twelfth threshold and less than the thirteenth threshold. The total number of groups in group v11 whose second number is greater than a11 is then counted as the first number corresponding to the Y-axis angular velocity. Here, a11 is a preset threshold used when detecting the Y-axis angular velocity.

[0123] The detection target is to detect the Z-axis angular velocity. At this time, there are sliding window data under the corresponding type of v6 groups. The second number of multiple Z-axis angular velocities in each sliding window data group that have an absolute value greater than the sixth threshold and less than the fourteenth threshold is counted. The number of groups in v6 whose second number is greater than a6 is accumulated and used as the first number corresponding to the Z-axis angular velocity.

[0124] The detection target is to detect X-axis acceleration. At this time, there is sliding window data under the corresponding type of v7 group. The second number of multiple X-axis accelerations in each sliding window data group that have an absolute value greater than the seventh threshold and less than the fifteenth threshold is counted. The number of groups in v7 group whose second number is greater than a7 is accumulated and used as the first number corresponding to the X-axis acceleration.

[0125] The detection target is to detect Y-axis acceleration. At this time, there are sliding window data under the corresponding type in group v8. The second number of multiple Y-axis accelerations in each sliding window data group that have an absolute value greater than the eighth threshold and less than the sixteenth threshold is counted. The number of groups in group v8 whose second number of Y-axis accelerations is greater than a8 is accumulated and used as the first number corresponding to Y-axis acceleration.

[0126] The detection target is Z-axis acceleration. There are 12 groups of sliding window data corresponding to this type. For each group, the absolute value of multiple Z-axis accelerations is counted as a second number greater than the seventeenth threshold and less than the eighteenth threshold. The total number of groups in the 12 groups whose second number is greater than a12 is then counted as the first number corresponding to the Z-axis acceleration. Here, a12 is a preset threshold used for detecting Z-axis acceleration.

[0127] The detection target is to detect the pitch angle change. At this time, there are sliding window data under the corresponding type of v4 group. The second number of absolute values ​​of multiple pitch angle changes in each group of sliding window data that are greater than the fourth threshold and less than the nineteenth threshold is counted. The number of groups in v4 group whose second number is greater than a4 is accumulated and used as the first number corresponding to the pitch angle change.

[0128] The detection target is the change in roll angle. There are sliding window data sets of type v13. For each group of sliding window data, the absolute value of multiple roll angle changes is counted as a second number greater than the twentieth threshold and less than the twenty-first threshold. The number of groups in v13 whose second number is greater than a13 is then accumulated and used as the first number corresponding to the roll angle change. Here, a13 is a preset threshold used when detecting Z-axis acceleration.

[0129] Here, the sum of v1, v11, v6, v7, v8, v12, v14, and v13 is M. v1, v11, v6, v7, v8, v12, v14, and v13 may be the same or different, depending on the size of the sliding window corresponding to the different types.

[0130] If the first quantity corresponding to the X-axis angular velocity is greater than the number of times threshold b1, the first quantity corresponding to the Y-axis angular velocity is greater than the number of times threshold b11, the first quantity corresponding to the Z-axis angular velocity is greater than the number of times threshold b6, the first quantity corresponding to the X-axis acceleration is greater than the number of times threshold b7, the first quantity corresponding to the Y-axis acceleration is greater than the number of times threshold b8, the first quantity corresponding to the Z-axis acceleration is greater than the number of times threshold b12, the first quantity corresponding to the pitch angle change is greater than the number of times threshold b4, and the first quantity corresponding to the roll angle change is greater than the number of times threshold b13, then the actual state of the vehicle in the first time period is determined to be a bumpy state.

[0131] For ease of understanding, the above method for judging the bumpy state is represented by expression four.

[0132] Expression 4:

[0133]

[0134] In this embodiment, for detecting bumpy conditions, when the intelligent vehicle is driving on a bumpy road, the three-axis angular velocity output by the gyroscope is used to accurately identify whether the vehicle's angular velocity changes frequently, and the three-axis acceleration output by the accelerometer is used to accurately identify whether the vehicle's acceleration changes frequently. The slope of the vehicle on the road can be accurately identified by the pitch angle change and roll angle change. By combining the above data, it is possible to accurately determine whether the vehicle is currently in a bumpy scenario.

[0135] In some embodiments, the detection of a straight line state specifically includes the following process.

[0136] The detection target is to detect the Z-axis angular velocity. At this time, there are sliding window data under the corresponding type of v6 groups. The second number of multiple Z-axis angular velocities in each sliding window data group whose absolute value is less than the sixth threshold is counted. The number of groups in v6 whose second number is greater than a6 is accumulated and used as the first number corresponding to the Z-axis angular velocity.

[0137] The detection target is to detect Z-axis acceleration. At this time, there are 12 groups of sliding window data of the corresponding type. The number of absolute values ​​of multiple Z-axis accelerations in each group of sliding window data that are less than the seventeenth threshold is counted. The number of groups of Z-axis accelerations in the 12 groups that have the second number greater than a12 is accumulated and used as the first number corresponding to Z-axis acceleration.

[0138] The detection target is to detect the change in yaw angle. At this time, there is sliding window data under the corresponding type of v9 group. The second number of multiple yaw angle changes in each group of sliding window data with absolute values ​​less than the ninth threshold is counted. The number of groups with the second number of yaw angle changes in v9 group that is greater than a9 is accumulated and used as the first number corresponding to the change in yaw angle.

[0139] The detection target is the average first velocity. This average first velocity can be the average X-axis velocity, i.e., the average lateral velocity. There are 14 groups of sliding window data corresponding to this type. For each group, the absolute value of multiple average first velocities is counted as a second number less than the twenty-second threshold. The total number of groups in group v14 where this second number is greater than a14 is then counted as the first quantity corresponding to the average first velocity. Here, a14 is the preset threshold used when detecting the average first velocity.

[0140] Here, the sum of v6, v12, v9, and v14 is M. v6, v12, v9, and v14 may be the same or different, depending on the size of the sliding window corresponding to the different types.

[0141] If the first quantity corresponding to the Z-axis acceleration is greater than the number threshold b6, the first quantity corresponding to the Z-axis acceleration is greater than the number threshold b12, the first quantity corresponding to the yaw angle change is greater than the number threshold b9, and the first quantity corresponding to the first speed average is greater than the number threshold b14, then the actual state of the vehicle in the first time period is determined to be a straight line state.

[0142] For ease of understanding, the method for determining the state of the straight line is expressed in the form of expression five.

[0143] Expression 5:

[0144]

[0145] For example, if the Z-axis acceleration, Z-axis acceleration, yaw change, and the average value of the first velocity are all positive, then the vehicle is moving forward.

[0146] For example, if the Z-axis acceleration, Z-axis acceleration, yaw change, and the average value of the first velocity are all negative, then the vehicle is reversing.

[0147] In this embodiment, for the detection of straight-line states, when the intelligent vehicle is traveling on a straight road segment, the Z-axis acceleration is detected to accurately determine whether the vehicle is turning. If not, it is considered to be in a straight-line state. Further, the vehicle's attitude information, such as yaw angle change and lateral velocity (mean of the first velocity), is detected to verify whether the vehicle is turning. Combining the above data, it is possible to accurately determine whether the vehicle is currently moving forward or backward.

[0148] In some embodiments, the detection of a stationary state specifically includes the following process.

[0149] The detection targets are X-axis angular velocity, Y-axis angular velocity, Z-axis angular velocity, X-axis acceleration, Y-axis acceleration, Z-axis acceleration, average second velocity, and velocity change.

[0150] The detection target is to detect the X-axis angular velocity. At this time, there is a sliding window data of type v1. The second number of multiple X-axis angular velocities in each sliding window data group whose absolute value is less than the first threshold is counted. The number of groups in group v1 whose second number is greater than a1 is accumulated and used as the first number corresponding to the X-axis angular velocity.

[0151] The detection target is to detect the Y-axis angular velocity. At this time, there are sliding window data under the corresponding type of v11 group. The second number of multiple Y-axis angular velocities in each group of sliding window data whose absolute value is less than the twelfth threshold is counted. The number of groups in v11 whose second number is greater than a11 is accumulated and used as the first number corresponding to the Y-axis angular velocity.

[0152] The detection target is to detect the Z-axis angular velocity. At this time, there are sliding window data under the corresponding type of v6 groups. The second number of multiple Z-axis angular velocities in each sliding window data group whose absolute value is less than the sixth threshold is counted. The number of groups in v6 whose second number is greater than a6 is accumulated and used as the first number corresponding to the Z-axis angular velocity.

[0153] The detection target is to detect X-axis acceleration. At this time, there is sliding window data under the corresponding type of group v7. The second number of multiple X-axis accelerations with absolute values ​​less than the seventh threshold in each group of sliding window data is counted. The number of groups with the second number of X-axis accelerations in group v7 that is greater than a7 is accumulated and used as the first number corresponding to the X-axis acceleration.

[0154] The detection target is to detect Y-axis acceleration. At this time, there are sliding window data under the corresponding type in group v8. The second number of multiple Y-axis accelerations with absolute values ​​less than the eighth threshold in each group of sliding window data is counted. The number of groups with the second number of Y-axis accelerations in group v8 that is greater than a8 is accumulated and used as the first number corresponding to Y-axis acceleration.

[0155] The detection target is to detect Z-axis acceleration. At this time, there are 12 groups of sliding window data of the corresponding type. The number of absolute values ​​of multiple Z-axis accelerations in each group of sliding window data that are less than the seventeenth threshold is counted. The number of groups of Z-axis accelerations in the 12 groups that have the second number greater than a12 is accumulated and used as the first number corresponding to Z-axis acceleration.

[0156] The detection target is the average of the second velocity. This average of the second velocity can be the average of the synthesized velocities. There are 15 groups of sliding window data corresponding to this type. The number of groups in each group whose absolute value of the average of the second velocity is less than the twenty-third threshold is counted. The total number of groups in the v15 groups whose average of the second velocity is greater than a15 is then recorded as the first quantity corresponding to the average of the second velocity. Here, a15 is a preset threshold used when detecting the average of the second velocity.

[0157] The detection target is the change in velocity. This change in velocity can be the change in synthetic velocity. There are 16 groups of sliding window data corresponding to this type. The number of groups in each group whose absolute value of multiple velocity changes is less than the 24th threshold is counted. The number of groups in the 16 groups whose second number is greater than a16 is accumulated and used as the first number corresponding to the velocity change.

[0158] Here, the sum of v1, v11, v6, v7, v8, v12, v15, and v16 is M. v1, v11, v6, v7, v8, v12, v15, and v16 may be the same or different, depending on the size of the sliding window corresponding to the different types.

[0159] If the first quantity corresponding to the X-axis angular velocity is greater than the number of times threshold b1, the first quantity corresponding to the Y-axis angular velocity is greater than the number of times threshold b11, the first quantity corresponding to the Z-axis angular velocity is greater than the number of times threshold b6, the first quantity corresponding to the X-axis acceleration is greater than the number of times threshold b7, the first quantity corresponding to the Y-axis acceleration is greater than the number of times threshold b8, the first quantity corresponding to the Z-axis acceleration is greater than the number of times threshold b12, the first quantity corresponding to the second speed average is greater than the number of times threshold b15, and the first quantity corresponding to the speed change is greater than the number of times threshold b16, then the actual state of the vehicle in the first time period is determined to be a stationary state.

[0160] For ease of understanding, the above method of determining the static state is represented by expression six.

[0161] Expression Six:

[0162]

[0163] In this embodiment, for detecting stationary states, the data output by the sensors of the intelligent vehicle is relatively stable when it is stationary. Taking the IMU sensor as an example, although its data contains some noise signals, its output value will tend to "0" under static conditions. At the same time, the vehicle's speed value will also remain near "0". Therefore, by using the target data in Expression Six above that are less than the threshold, such as a threshold close to 0, the stationary state of the vehicle can be effectively identified.

[0164] The above embodiments illustrate methods for detecting six vehicle driving states: uphill / downhill, turning, acceleration / deceleration, bumpy conditions, straight-line driving, and stationary conditions. In practical applications, all six states can be detected simultaneously, and multiple states may coexist, such as uphill / downhill and acceleration / deceleration, uphill / downhill and bumpy conditions, acceleration / deceleration and bumpy conditions, turning and bumpy conditions, etc. However, a stationary state cannot coexist with the other five states.

[0165] This disclosure analyzes the complementarity and redundancy of sliding window data among various detection targets, and fuses data from different sensors (different types of target data) to perform comprehensive and multi-angle identification of vehicle driving conditions. This improves the accuracy and reliability of road condition identification, resulting in a more accurate understanding of the vehicle's actual state. This, in turn, helps the vehicle make more precise judgments and decisions in the face of complex and changing road conditions. Currently, intelligent vehicles are equipped with positioning devices such as PBOX and IMU. Based on the above algorithm, the driving state of intelligent vehicles can be detected and identified at high frequency and in real time. This can be applied to calculate the vehicle's posture (e.g., angular offset) during driving, achieving real-time image compensation. Alternatively, when the vehicle is turning, blind spot detection can be triggered based on the detection results to promptly identify potential risks and issue warnings to the driver. Furthermore, based on the detected current vehicle motion state, combined with kinematic models, vehicle dynamics models, and filtering algorithms, the vehicle's pose can be predicted, further ensuring the positioning accuracy during vehicle operation. In addition, accurate state detection can further enhance the auxiliary functions of intelligent driving, such as automatic parking and adaptive cruise control, making the driving process smoother and more comfortable, ultimately greatly improving the overall travel experience for drivers and passengers.

[0166] In some embodiments, the calculation methods for the aforementioned velocity changes, multi-axis angular velocity changes, multi-axis acceleration changes, elevation changes, and Euler angle changes are all the same. This embodiment denotes any one of the velocity changes, multi-axis angular velocity changes, multi-axis acceleration changes, elevation changes, and Euler angle changes as parameter changes, and only describes the steps for determining the parameter changes to represent the steps for determining the changes corresponding to any of the aforementioned target parameters.

[0167] The target parameters are velocity, multi-axis acceleration, elevation, and Euler angles.

[0168] The steps for determining the parameter change corresponding to the target parameter include: For any set of sliding window data, the difference between the maximum and minimum values ​​of the n1 target parameters in the sliding window data can be used as the parameter change for one sliding window output; n1 is an integer greater than or equal to 2, where n1 represents that the sliding window data includes n1 target parameters. Alternatively, if n1 is greater than 2, the fluctuation values ​​of two temporally adjacent target parameters in the sliding window data can be determined, where the fluctuation value equals the difference between the absolute values ​​of adjacent target parameters. Then, any one of the maximum fluctuation value, minimum fluctuation value, and average fluctuation value in the sliding window data can be used as the parameter change for one sliding window output. Within the first time period, according to... Figure 3a and Figure 3b This is done by sequentially sliding the window to obtain multiple parameter changes from multiple sliding window outputs. For example, ... Figure 3a and Figure 3b As shown, during the first time period, the sliding window is executed v times, outputting v parameter changes. These v parameter changes constitute a set of sliding window data consisting of parameter changes.

[0169] Taking X-axis angular velocity as an example, the difference between the maximum and minimum values ​​of n1 X-axis angular velocities in a sliding window dataset containing X-axis angular velocities is taken as the change in X-axis angular velocity output in one sliding window operation. Alternatively, when n1 is greater than 2, the difference between the absolute values ​​of two temporally adjacent X-axis angular velocities in the sliding window dataset is determined as the fluctuation value; then, any one of the maximum fluctuation value, minimum fluctuation value, and average fluctuation value is taken as the change in X-axis angular velocity output in one sliding window operation.

[0170] In some embodiments, the average first velocity is taken as the average X-axis velocity. The X-axis acceleration can be directly measured by the IMU. The X-axis accelerations within the first time period are integrated to obtain multiple consecutive X-axis velocities within the first time period, which are denoted as the first integration parameter.

[0171] The steps to determine the average first velocity include:

[0172] For any set of sliding window data, the average of the n² first integral parameters in the sliding window data is denoted as the first velocity mean, where n² is an integer greater than or equal to 2, and n² indicates that the sliding window data includes n² first integral parameters. Within the first time interval, according to... Figure 3a and Figure 3b The process involves sequential sliding window processing to obtain multiple average values ​​of the first velocity from multiple sliding window outputs. For example, as... Figure 3a and Figure 3bAs shown, during the first time period, v sliding windows are executed, outputting v average first speed values. These v average first speed values ​​constitute a set of sliding window data consisting of average first speed values.

[0173] In some embodiments, the average second velocity is taken as the average composite velocity. The IMU can directly measure the X-axis acceleration, Y-axis acceleration, and Z-axis acceleration. Each X-axis acceleration within a first time period is integrated to obtain multiple consecutive X-axis velocities within the first time period, denoted as the first integration parameter; each Y-axis acceleration within the first time period is integrated to obtain multiple consecutive Y-axis velocities within the first time period, denoted as the second integration parameter; and each Z-axis acceleration within the first time period is integrated to obtain multiple consecutive Z-axis velocities within the first time period, denoted as the third integration parameter.

[0174] The steps to determine the average second velocity include:

[0175] For three sets of sliding window data at the same time sequence (one set containing X-axis acceleration, one set containing Y-axis acceleration, and one set containing Z-axis acceleration), the i-th fused data in the new sliding window data is determined based on the i-th first integral parameter, second integral parameter, and third integral parameter in the three sets of sliding window data; i is less than or equal to n3, where n3 is the number of sub-data in the sliding window data. The i-th first integral parameter, second integral parameter, and third integral parameter represent velocities in different directions (X, Y, Z), and can be obtained by vector addition to obtain the fused data.

[0176] For any new set of sliding window data, the average of the n4 fused data points in the new sliding window data is denoted as the second velocity mean, where n4 is an integer greater than or equal to 2, and n4 indicates that the new sliding window data includes n4 fused data points. Within the first time period, according to... Figure 3a and Figure 3b The process involves sequential sliding window processing to obtain multiple average values ​​of the second velocity from the multiple sliding window outputs. For example, as... Figure 3a and Figure 3b As shown, during the first time period, v sliding windows are executed, outputting v average values ​​of the second velocity. These v average values ​​of the second velocity constitute a set of sliding window data consisting of the average values ​​of the second velocity.

[0177] The speed mentioned above can be calculated by integrating acceleration, or it can be directly measured using data from the P-BOX positioning device.

[0178] In some embodiments, for S11, acquiring target data of the vehicle over a period of time may specifically include: first, acquiring raw data; raw data refers to data directly collected by sensors, such as IMU data (e.g., multi-axis acceleration and multi-axis acceleration) collected by the inertial measurement unit (IMU), P-BOX data (e.g., position, velocity, attitude (Euler angles), elevation, and confidence level) collected by the positioning device (P-BOX), wheel speed data (e.g., wheel speed) collected by the wheel speed meter, etc.

[0179] Next, it is determined whether the confidence level of the P-BOX data is greater than a set value. If the confidence level of the P-BOX data is greater than or equal to the set value, it means that the accuracy of the data collected by the positioning device's P-BOX meets the user's requirements, and the IMU data (multi-axis acceleration and multi-axis accelerometer data) and P-BOX data (velocity, elevation, and Euler angles) can be directly used as target data. If the confidence level of the P-BOX data is less than the set value, it means that the accuracy of the data collected by the positioning device's P-BOX is low and does not meet the user's requirements. In this case, the speed, elevation, and Euler angles can be calculated using the IMU data and wheel speedometer data, and used together with the IMU data (multi-axis acceleration and multi-axis accelerometer data) as target data.

[0180] For ease of understanding, the vehicle status detection method of this disclosure will be described below as a complete process. The details of the processing will not be repeated, but can be found in the description of the corresponding embodiments above. Figure 6 A detailed flowchart of a vehicle status detection method provided in this disclosure embodiment is shown below. Figure 6 As shown, it includes S101 to S108.

[0181] S101. Obtain the target data of the vehicle within the first time period. The target data includes speed, multi-axis angular velocity, multi-axis acceleration, elevation, Euler angles, mean speed, speed change, multi-axis angular velocity change, multi-axis acceleration change, elevation change, and Euler angle change.

[0182] S102. For target data of the same type, perform sliding windowing on all target data of that type within the first time period to obtain v groups of sliding window data.

[0183] S103. Various types of target data are slid-windowed separately to obtain v sets of sliding window data corresponding to various types of target data, for a total of N sets of sliding window data.

[0184] S104. For any set of sliding window data, determine the detection threshold of the target based on the type of sliding window data.

[0185] S105. Compare the absolute value of each sub-data in the sliding window data with the detection threshold to obtain the second number of sub-data that meet the comparison conditions.

[0186] S106. The first number in the cumulative v group sliding window data is greater than or equal to the preset threshold.

[0187] S107. Obtain the first number corresponding to each of the multiple detection targets corresponding to the target state.

[0188] S108. If the first number corresponding to the multiple detection targets corresponding to the target state is greater than or equal to the number of times threshold, then the actual state of the vehicle corresponding to the first time period is determined to be the target state.

[0189] Those skilled in the art will understand that, in the above-described method of the specific implementation, the order in which each step is written does not imply a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of each step should be determined by its function and possible internal logic.

[0190] In addition, this disclosure also provides a vehicle state detection device corresponding to the vehicle state detection method. Since the principle of the vehicle state detection device in this disclosure is similar to the vehicle state detection method described above in this disclosure, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be described again.

[0191] Figure 7 This is a schematic diagram of a vehicle status detection device provided in an embodiment of the present disclosure, as shown below. Figure 7 As shown, the vehicle status detection device includes a data acquisition module 71, a first detection module 72, and a second detection module 73.

[0192] The data acquisition module 71 is configured to acquire target data of the vehicle within a first time period; the target data includes at least one of speed, multi-axis angular velocity, multi-axis acceleration, elevation, Euler angle, mean speed, speed change, multi-axis angular velocity change, multi-axis acceleration change, elevation change, and Euler angle change.

[0193] It should be noted that the data acquisition module 71 in this embodiment is configured to perform step S11 in the above-described vehicle state detection method, and repeated parts will not be described again.

[0194] The sliding window processing module is configured to perform sliding window processing on at least a portion of the target data to obtain N sets of sliding window data.

[0195] It should be noted that the data acquisition module 71 in this embodiment is configured to perform step S12 in the above-described vehicle state detection method, and repeated parts will not be described again.

[0196] The first detection module 72 is configured to match M sets of sliding window data with the identification information of at least one detection target, and determine the number of sets of sliding window data that meet the first preset condition in the matching results corresponding to each detection target according to the matching results corresponding to each detection target, and use it as the first quantity; M is less than or equal to N.

[0197] It should be noted that the data acquisition module 71 in this embodiment is configured to perform step S13 in the above-described vehicle state detection method, and repeated parts will not be described again.

[0198] The second detection module 73 is configured to determine the actual state of the vehicle based on the first quantity corresponding to each detection target.

[0199] It should be noted that the data acquisition module 71 in this embodiment is configured to perform step S14 in the above-described vehicle state detection method, and repeated parts will not be described again.

[0200] In some embodiments, the second detection module 73 is specifically configured to determine the actual state of the vehicle in the first time period as the target state if the first number corresponding to the multiple detection targets corresponding to the target state is greater than or equal to the number of times threshold; the target state includes at least one of the following: stationary state, uphill / downhill state, turning state, acceleration / deceleration state, bumpy state, and straight state.

[0201] In some embodiments, the first detection module 72 is specifically configured to, for any set of sliding window data, determine the detection threshold of the detection target according to the type of the sliding window data; compare the absolute value of each sub-data in the sliding window data with the detection threshold to obtain a second number of sub-data that meet the comparison conditions; if the second number is greater than or equal to a preset threshold, determine the first sliding window state of the sliding window data; accumulate the number of the first sliding window states in v sets of sliding window data corresponding to the type of sliding window data, and use it as the first number; v is a positive integer less than or equal to M.

[0202] In some embodiments, Euler angles include pitch angles; Euler angle changes include pitch angle changes; multiple detection targets corresponding to uphill and downhill states include X-axis angular velocity, pitch angle, X-axis angular velocity changes, pitch angle changes, and elevation changes; detection thresholds include a first threshold compared to X-axis angular velocity, a second threshold compared to pitch angle, a third threshold compared to X-axis angular velocity changes, a fourth threshold compared to pitch angle changes, and a fifth threshold compared to elevation changes.

[0203] Taking the target state as an uphill or downhill state as an example, the first detection module 72 is specifically configured to: count the second number of absolute values ​​of multiple X-axis angular velocities in each group of sliding window data that are greater than a first threshold, and accumulate the number of groups in group v1 where the second number of X-axis angular velocities is greater than a1, as the first number corresponding to the X-axis angular velocity; count the second number of absolute values ​​of multiple pitch angles in each group of sliding window data that are greater than a second threshold, and accumulate the number of groups in group v2 where the second number of pitch angles is greater than a2, as the first number corresponding to the pitch angle; and count the second number of absolute values ​​of multiple X-axis angular velocity changes in each group of sliding window data that are greater than a third threshold, and accumulate v... The number of groups with the second largest quantity greater than a3 among the three groups of X-axis angular velocity changes is taken as the first quantity corresponding to the X-axis angular velocity change. The number of groups with the second largest quantity greater than the fourth threshold for the absolute value of multiple pitch angle changes in each group of sliding window data is counted, and the number of groups with the second largest quantity greater than a4 in the pitch angle change of group v4 is accumulated as the first quantity corresponding to the pitch angle change. The number of groups with the second largest quantity greater than the fifth threshold for the absolute value of multiple elevation changes in each group of sliding window data is counted, and the number of groups with the second largest quantity greater than a5 in the elevation change of group v5 is accumulated as the first quantity corresponding to the elevation change. The sum of v1, v2, v3, v4 and v5 is M.

[0204] In some embodiments, Euler angles include yaw angles; Euler angle changes include yaw angle changes; the multiple detection targets corresponding to the turning state include Z-axis angular velocity, X-axis acceleration, Y-axis acceleration, and yaw angle changes; the detection thresholds include a sixth threshold compared with Z-axis angular velocity, a seventh threshold compared with X-axis acceleration, an eighth threshold compared with Y-axis acceleration, and a ninth threshold compared with yaw angle changes.

[0205] Taking the target state as a turning state as an example, the first detection module 72 is specifically configured to: count the number of absolute values ​​of multiple Z-axis angular velocities in each group of sliding window data that are greater than the sixth threshold, and accumulate the number of groups in group v6 where the second number of Z-axis angular velocities is greater than a6, as the first number corresponding to the Z-axis angular velocity; count the number of absolute values ​​of multiple X-axis accelerations in each group of sliding window data that are greater than the seventh threshold, and accumulate the number of groups in group v7 where the second number of X-axis accelerations is greater than a7, as the first number corresponding to the X-axis acceleration; count the number of absolute values ​​of multiple Y-axis accelerations in each group of sliding window data that are greater than the eighth threshold, and accumulate the number of groups in group v8 where the second number of Y-axis accelerations is greater than a8, as the first number corresponding to the Y-axis acceleration; count the number of absolute values ​​of multiple yaw angle changes in each group of sliding window data that are greater than the ninth threshold, and accumulate the number of groups in group v9 where the second number of yaw angle changes is greater than a9, as the first number corresponding to the yaw angle change; the sum of v6, v7, v8, and v9 is M.

[0206] In some embodiments, Euler angles include pitch angles; Euler angle changes include pitch angle changes; multiple detection targets corresponding to acceleration / deceleration states include Y-axis acceleration, pitch angle changes, and velocity changes; detection thresholds include an eighth threshold compared to Y-axis acceleration, a fourth threshold compared to pitch angle changes, and a tenth threshold compared to velocity changes.

[0207] Taking the target state as an acceleration / deceleration state as an example, the first detection module 72 is specifically configured to: count the number of absolute values ​​of multiple Y-axis accelerations in each group of sliding window data that are greater than the eighth threshold, and accumulate the number of groups in group v8 where the second number of Y-axis accelerations is greater than a8, as the first number corresponding to the Y-axis acceleration; count the number of absolute values ​​of multiple pitch angle changes in each group of sliding window data that are greater than the fourth threshold, and accumulate the number of groups in group v4 where the second number of pitch angle changes is greater than a4, as the first number corresponding to the pitch angle changes; count the number of absolute values ​​of multiple velocity changes in each group of sliding window data that are greater than the tenth threshold, and accumulate the number of groups in group v10 where the second number of velocity changes is greater than a10, as the first number corresponding to the velocity changes; the sum of v8, v4, and v10 is M.

[0208] In some embodiments, Euler angles include pitch angle and roll angle; Euler angle changes include pitch angle changes and roll angle changes; multiple detection targets corresponding to the turbulence state include X-axis angular velocity, Y-axis angular velocity, Z-axis angular velocity, X-axis acceleration, Y-axis acceleration, Z-axis acceleration, pitch angle changes, and roll angle changes; detection thresholds include a first threshold and an eleventh threshold compared to the X-axis angular velocity, a twelfth threshold and a thirteenth threshold compared to the Y-axis angular velocity, a sixth threshold and a fourteenth threshold compared to the Z-axis angular velocity, a seventh threshold and a fifteenth threshold compared to the X-axis acceleration, an eighth threshold and a sixteenth threshold compared to the Y-axis acceleration, a seventeenth threshold and an eighteenth threshold compared to the Z-axis acceleration, a fourth threshold and a nineteenth threshold compared to the pitch angle changes, and a twentieth threshold and a twenty-first threshold compared to the roll angle changes.

[0209] Taking the target state as a bumpy state as an example, the first detection module 72 is specifically configured to: count the number of absolute values ​​of multiple X-axis angular velocities in each group of sliding window data that are greater than the first threshold and less than the eleventh threshold, and accumulate the number of groups in group v1 where the second number is greater than a1, as the first number corresponding to the X-axis angular velocity; count the number of absolute values ​​of multiple Y-axis angular velocities in each group of sliding window data that are greater than the twelfth threshold and less than the thirteenth threshold, and accumulate the number of groups in group v11 where the second number is greater than a11, as the first number corresponding to the Y-axis angular velocity; The first quantity corresponding to angular velocity; For each group of sliding window data, the second quantity of multiple Z-axis angular velocities whose absolute values ​​are greater than the sixth threshold and less than the fourteenth threshold is counted, and the number of groups in group v6 where the second quantity of Z-axis angular velocities is greater than a6 is accumulated, which is used as the first quantity corresponding to Z-axis angular velocity; For each group of sliding window data, the second quantity of multiple X-axis accelerations whose absolute values ​​are greater than the seventh threshold and less than the fifteenth threshold is counted, and the number of groups in group v7 where the second quantity of X-axis acceleration is greater than a7 is accumulated, which is used as the first quantity corresponding to X-axis acceleration ... sixth threshold and less than the fourteenth threshold is counted, and the number of groups in group v7 where the second quantity of X-axis acceleration is greater than a7 is accumulated, which is used as the first quantity corresponding to X-axis acceleration; For each group of sliding window data, the second quantity of multiple X-axis accelerations whose absolute The data includes the second number of Y-axis accelerations whose absolute values ​​are greater than the eighth threshold and less than the sixteenth threshold, and the cumulative number of groups in group v8 where the second number is greater than a8, which is used as the first number corresponding to the Y-axis acceleration; the data also includes the second number of Z-axis accelerations whose absolute values ​​are greater than the seventeenth threshold and less than the eighteenth threshold, and the cumulative number of groups in group v12 where the second number is greater than a12, which is used as the first number corresponding to the Z-axis acceleration; and the data also includes the second number of pitch angle changes whose absolute values ​​are greater than a8 in each group of sliding window data. The second number of pitch angle changes is greater than a4 in the fourth threshold and less than the nineteenth threshold, and the number of groups with the second number greater than a4 in the v4 group pitch angle change is accumulated as the first number corresponding to the pitch angle change; the second number of roll angle changes in each group of sliding window data whose absolute value is greater than the twentieth threshold and less than the twenty-first threshold is counted, and the number of groups with the second number greater than a13 in the v13 group roll angle change is accumulated as the first number corresponding to the roll angle change; the sum of v1, v11, v6, v7, v8, v12, v14 and v13 is M.

[0210] In some embodiments, Euler angles include yaw angles; Euler angle changes include yaw angle changes; the average velocity includes a first average velocity; the multiple detection targets corresponding to the straight-line state include Z-axis angular velocity, Z-axis acceleration, yaw angle changes, and the first average velocity; the detection thresholds include a sixth threshold compared with Z-axis angular velocity, a seventeenth threshold compared with Z-axis acceleration, a ninth threshold compared with yaw angle changes, and a twenty-second threshold compared with the first average velocity.

[0211] Taking a straight line as the target state as an example, the first detection module 72 is specifically configured to: count the number of absolute values ​​of multiple Z-axis angular velocities in each group of sliding window data that are less than the sixth threshold, and accumulate the number of groups in group v6 where the second number is greater than a6, as the first number corresponding to the Z-axis angular velocity; count the number of absolute values ​​of multiple Z-axis accelerations in each group of sliding window data that are less than the seventeenth threshold, and accumulate the number of groups in group v12 where the second number is greater than a12, as the first number corresponding to the Z-axis acceleration; count the number of absolute values ​​of multiple yaw angle changes in each group of sliding window data that are less than the ninth threshold, and accumulate the number of groups in group v9 where the second number is greater than a9, as the first number corresponding to the yaw angle change; count the number of absolute values ​​of multiple first speed averages in each group of sliding window data that are less than the twenty-second threshold, and accumulate the number of groups in group v14 where the second number is greater than a14, as the first number corresponding to the first speed average; the sum of v6, v12, v9, and v14 is M.

[0212] In some embodiments, the average velocity includes a second average velocity; the multiple detection targets corresponding to the stationary state include X-axis angular velocity, Y-axis angular velocity, Z-axis angular velocity, X-axis acceleration, Y-axis acceleration, Z-axis acceleration, the second average velocity, and velocity change; the detection thresholds include a first threshold compared with the X-axis angular velocity, a twelfth threshold compared with the Y-axis angular velocity, a sixth threshold compared with the Z-axis angular velocity, a seventh threshold compared with the X-axis acceleration, an eighth threshold compared with the Y-axis acceleration, a seventeenth threshold compared with the Z-axis acceleration, a twenty-third threshold compared with the second average velocity, and a twenty-fourth threshold compared with the velocity change.

[0213] Taking a stationary target state as an example, the first detection module 72 is specifically configured to: count the number of times the absolute value of multiple X-axis angular velocities in each group of sliding window data is less than a first threshold, and accumulate the number of groups in group v1 where the second number is greater than a1, as the first number corresponding to the X-axis angular velocity; count the number of times the absolute value of multiple Y-axis angular velocities in each group of sliding window data is less than a twelfth threshold, and accumulate the number of groups in group v11 where the second number is greater than a11, as the first number corresponding to the Y-axis angular velocity; count the number of times the absolute value of multiple Z-axis angular velocities in each group of sliding window data is less than a sixth threshold, and accumulate the number of groups in group v6 where the second number is greater than a6, as the first number corresponding to the Z-axis angular velocity; count the number of times the absolute value of multiple X-axis accelerations in each group of sliding window data is less than a seventh threshold, and accumulate the number of groups in group v7 where the second number is greater than a7, as the first number corresponding to the X-axis acceleration; and count the number of times the absolute value of multiple X-axis accelerations in each group of sliding window data is less than a seventh threshold, and accumulate the number of groups in group v7 where the second number is greater than a7, as the first number corresponding to the X-axis acceleration; and count the number of times the absolute value of multiple X-axis angular velocities in each group of sliding window data is less than a first ... The data includes the second number of absolute values ​​of multiple Y-axis accelerations that are less than the eighth threshold, and the cumulative number of groups in group v8 where the second number is greater than a8, which is used as the first number corresponding to the Y-axis acceleration; the data also includes the second number of absolute values ​​of multiple Z-axis accelerations that are less than the seventeenth threshold, and the cumulative number of groups in group v12 where the second number is greater than a12, which is used as the first number corresponding to the Z-axis acceleration; the data also includes the second number of absolute values ​​of multiple average second velocities that are less than the twenty-third threshold, and the cumulative number of groups in group v15 where the second number is greater than a15, which is used as the first number corresponding to the average second velocity; the data also includes the second number of absolute values ​​of multiple velocity changes that are less than the twenty-fourth threshold, and the cumulative number of groups in group v16 where the second number is greater than a16, which is used as the first number corresponding to the velocity change; the sum of v1, v11, v6, v7, v8, v12, v15, and v16 is M.

[0214] In some embodiments, the second time period corresponding to each group of sliding window data is determined based on the frequency of the sensor that collects the sensor data; the target data is sensor data or determined based on sensor data; the magnitude of the sensor frequency is inversely proportional to the length of the second time period.

[0215] In some embodiments, any one of the velocity change, multi-axis angular velocity change, multi-axis acceleration change, elevation change, and Euler angle change is denoted as a parameter change; each parameter change has a target parameter of the same type. The vehicle state detection device also includes a parameter calculation module 74. The parameter calculation module 74 is configured to, for any set of sliding window data, take the difference between the maximum and minimum values ​​of the n1 target parameters in the sliding window data as the parameter change of one sliding window output, so as to obtain multiple parameter changes of multiple sliding window outputs within a first time period; n1 is an integer greater than or equal to 2; or, for any set of sliding window data, if n1 is greater than 2, determine the fluctuation values ​​of two temporally adjacent target parameters in the sliding window data, and take any one of the maximum fluctuation value, minimum fluctuation value, and average fluctuation value as the parameter change of one sliding window output, so as to obtain multiple parameter changes of multiple sliding window outputs within a first time period.

[0216] In some embodiments, velocity includes a first integral parameter of X-axis acceleration; the velocity mean includes a first velocity mean. The parameter calculation module 74 is also configured to, for any set of sliding window data, record the average of n2 first integral parameters in the sliding window data as the first velocity mean, so as to obtain multiple first velocity mean values ​​of multiple sliding window outputs within a first time period; n2 is an integer greater than or equal to 2.

[0217] In some embodiments, the velocity includes a first integral parameter of the X-axis acceleration, a second integral parameter of the Y-axis acceleration, and a second integral parameter of the Z-axis acceleration; the velocity mean includes a second velocity mean. The parameter calculation module 74 is also configured to, for three sets of sliding window data in the same time sequence, determine the i-th fused data in the new sliding window data based on the i-th first integral parameter, second integral parameter, and third integral parameter in the three sets of sliding window data; i is less than or equal to n3, where n3 is the number of sub-data in the sliding window data; for any new set of sliding window data, the average of the n4 fused data in the new sliding window data is recorded as the second velocity mean, so as to obtain multiple second velocity mean values ​​of multiple sliding window outputs within the first time period; n4 is an integer greater than or equal to 2.

[0218] In some embodiments, the data acquisition module 71 is specifically configured to acquire raw data, including inertial measurement unit (IMU) data, positioning device (P-BOX) data, and wheel speedometer data; the IMU data includes multi-axis angular velocity and multi-axis acceleration; the P-BOX data includes velocity, elevation, and Euler angles; if the confidence level of the P-BOX data is greater than or equal to a set value, the IMU data and P-BOX data are used as target data; if the confidence level of the P-BOX data is less than the set value, the velocity, elevation, and Euler angles are calculated using the IMU data and wheel speedometer data, and used together with the IMU data as target data.

[0219] Figure 8 This is a schematic diagram of the structure of a computer device provided in an embodiment of this disclosure. Figure 8 As shown, this disclosure provides a computer device including: one or more processors 801, a memory 802, and one or more I / O interfaces 803. The memory 802 stores one or more programs, which, when executed by the one or more processors, cause the one or more processors to implement a vehicle state detection method as described in any of the above embodiments; the one or more I / O interfaces 803 are connected between the processor and the memory, configured to enable information interaction between the processor and the memory.

[0220] Among them, processor 801 is a device with data processing capabilities, including but not limited to central processing unit (CPU); memory 802 is a device with data storage capabilities, including but not limited to random access memory (RAM, more specifically SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory (FLASH); I / O interface (read-write interface) 803 is connected between processor 801 and memory 802, and can realize information interaction between processor 801 and memory 802, including but not limited to data bus (Bus).

[0221] In some embodiments, the processor 801, memory 802, and I / O interface 803 are interconnected via bus 804, and thus connected to other components of the computing device.

[0222] According to embodiments of this disclosure, a computer non-transient readable storage medium is also provided. This computer non-transient readable storage medium stores a computer program, wherein, when executed by a processor, the program implements the steps in the vehicle state detection method as described in any of the above embodiments.

[0223] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a machine-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication component, and / or installed from a removable medium. When the computer program is executed by a central processing unit (CPU), it performs the functions defined above in the system of this disclosure.

[0224] It should be noted that the computer-readable non-transient readable medium disclosed herein may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. Computer-readable storage media may be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. The transmitted data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The computer-readable signal medium can also be any non-transient readable computer storage medium other than a computer-readable storage medium, which can transmit, propagate, or transfer a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the non-transient readable computer storage medium can be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0225] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the drawings. For example, two adjacent blocks may actually represent substantially parallel execution, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0226] It is understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of this disclosure, and this disclosure is not limited thereto. For those skilled in the art, various modifications and improvements can be made without departing from the spirit and substance of this disclosure, and these modifications and improvements are also considered to be within the scope of protection of this disclosure.

Claims

1. A method for detecting the condition of a vehicle, characterized in that, include: Acquire target data of the vehicle within a first time period; the target data includes at least one of the following: speed, multi-axis angular velocity, multi-axis acceleration, elevation, Euler angles, mean speed, speed change, multi-axis angular velocity change, multi-axis acceleration change, elevation change, and Euler angle change; At least a portion of the target data is subjected to sliding window processing to obtain N sets of sliding window data; N is a positive integer greater than or equal to 1; The M groups of sliding window data are matched with the identification information of at least one detection target, and the number of sliding window data groups that satisfy the first preset condition in the matching results corresponding to each detection target is determined according to the matching results corresponding to each detection target, and is used as the first quantity; M is a positive integer less than or equal to N; The actual state of the vehicle is determined based on the first quantity corresponding to each of the detection targets.

2. The vehicle status detection method according to claim 1, characterized in that, Determining the actual state of the vehicle based on the first quantity corresponding to each of the detection targets includes: If the first number corresponding to each of the multiple detection targets corresponding to the target state is greater than or equal to the number of times threshold, then the actual state of the vehicle in the first time period is determined to be the target state; the target state includes at least one of the following: stationary state, uphill / downhill state, turning state, acceleration / deceleration state, bumpy state, and straight state.

3. The vehicle status detection method according to claim 1, characterized in that, The step of matching the M groups of sliding window data with the identification information of at least one detected target, and determining the number of groups of sliding window data that satisfy the first preset condition in the matching results corresponding to each detected target, and using this as the first quantity, includes: For any set of sliding window data, the detection threshold of the target is determined according to the type of the sliding window data; The absolute value of each sub-data in the sliding window data is compared with the detection threshold to obtain a second number of sub-data that meet the comparison conditions. If the second quantity is greater than or equal to a preset threshold, then the first sliding window state of the sliding window data is determined; The number of the first sliding window states in v groups of sliding window data corresponding to the type of the sliding window data is accumulated as the first quantity; v is a positive integer less than or equal to M.

4. The method for detecting the vehicle condition according to any one of claims 1 to 3, characterized in that, The Euler angle includes the pitch angle; the change in Euler angle includes the change in pitch angle; the multiple detection targets corresponding to the uphill and downhill states include the X-axis angular velocity, the pitch angle, the change in the X-axis angular velocity, the change in pitch angle, and the change in elevation; the detection thresholds include a first threshold compared with the X-axis angular velocity, a second threshold compared with the pitch angle, a third threshold compared with the change in X-axis angular velocity, a fourth threshold compared with the change in pitch angle, and a fifth threshold compared with the change in elevation. Based on the matching results corresponding to each of the detection targets, determine the number of groups of sliding window data that satisfy the first preset condition in the matching results corresponding to each of the detection targets, and use this as the first quantity, including: The number of absolute values ​​of multiple X-axis angular velocities in each group of sliding window data that are greater than the first threshold is counted separately, and the number of groups in group v1 where the second number is greater than a1 is accumulated, which is taken as the first number corresponding to the X-axis angular velocity; The number of absolute values ​​of multiple pitch angles in each group of sliding window data that are greater than the second threshold is counted, and the number of groups in group v2 where the second number is greater than a2 is accumulated, which is taken as the first number corresponding to the pitch angle. The number of absolute values ​​of multiple X-axis angular velocity changes in each group of sliding window data that are greater than the third threshold is counted separately, and the number of groups in group v3 where the second number is greater than a3 is accumulated, which is taken as the first number corresponding to the X-axis angular velocity change. The number of absolute values ​​of multiple pitch angle changes in each group of sliding window data that are greater than the fourth threshold is counted separately, and the number of groups in group v4 where the second number is greater than a4 is accumulated, which is taken as the first number corresponding to the pitch angle change. The number of absolute values ​​of multiple elevation changes in each group of sliding window data that are greater than the fifth threshold is counted separately, and the number of groups in group v5 whose second number is greater than a5 is accumulated as the first number corresponding to the elevation change; the sum of v1, v2, v3, v4 and v5 is M.

5. The method for detecting the vehicle condition according to any one of claims 1 to 3, characterized in that, The Euler angles include the yaw angle; the change in Euler angles includes the change in yaw angle; the multiple detection targets corresponding to the turning state include Z-axis angular velocity, X-axis acceleration, Y-axis acceleration, and change in yaw angle; the detection thresholds include a sixth threshold compared with the Z-axis angular velocity, a seventh threshold compared with the X-axis acceleration, an eighth threshold compared with the Y-axis acceleration, and a ninth threshold compared with the change in yaw angle. Based on the matching results corresponding to each of the detection targets, determine the number of groups of sliding window data that satisfy the first preset condition in the matching results corresponding to each of the detection targets, and use this as the first quantity, including: The number of absolute values ​​of multiple Z-axis angular velocities in each group of sliding window data that are greater than the sixth threshold is counted separately, and the number of groups in group v6 where the second number is greater than a6 is accumulated, which is taken as the first number corresponding to the Z-axis angular velocity; The number of absolute values ​​of multiple X-axis accelerations in each group of sliding window data that are greater than the seventh threshold is counted separately, and the number of groups in group v7 where the second number is greater than a7 is accumulated, which is taken as the first number corresponding to the X-axis acceleration; The number of absolute values ​​of multiple Y-axis accelerations in each group of sliding window data that are greater than the eighth threshold is counted separately, and the number of groups in group v8 where the second number is greater than a8 is accumulated, which is taken as the first number corresponding to the Y-axis acceleration; The number of absolute values ​​of multiple yaw angle changes in each group of sliding window data that are greater than the ninth threshold is counted separately, and the number of groups in group v9 whose second number is greater than a9 is accumulated as the first number corresponding to the yaw angle change; the sum of v6, v7, v8 and v9 is M.

6. The method for detecting the vehicle condition according to any one of claims 1 to 3, characterized in that, The Euler angles include pitch angles; the Euler angle change includes pitch angle change; the multiple detection targets corresponding to the acceleration / deceleration states include Y-axis acceleration, pitch angle change, and velocity change; the detection thresholds include an eighth threshold compared with the Y-axis acceleration, a fourth threshold compared with the pitch angle change, and a tenth threshold compared with the velocity change. Based on the matching results corresponding to each of the detection targets, determine the number of groups of sliding window data that satisfy the first preset condition in the matching results corresponding to each of the detection targets, and use this as the first quantity, including: The number of absolute values ​​of multiple Y-axis accelerations in each group of sliding window data that are greater than the eighth threshold is counted separately, and the number of groups in group v8 where the second number is greater than a8 is accumulated, which is taken as the first number corresponding to the Y-axis acceleration; The number of absolute values ​​of multiple pitch angle changes in each group of sliding window data that are greater than the fourth threshold is counted separately, and the number of groups in group v4 where the second number is greater than a4 is accumulated, which is taken as the first number corresponding to the pitch angle change. The number of absolute values ​​of multiple speed changes in each group of sliding window data that are greater than the tenth threshold is counted separately, and the number of groups in v10 whose second number is greater than a10 is accumulated as the first number corresponding to the speed change; the sum of v8, v4 and v10 is M.

7. The method for detecting the vehicle condition according to any one of claims 1 to 3, characterized in that, The Euler angles include pitch angle and roll angle; the Euler angle changes include pitch angle changes and roll angle changes; the multiple detection targets corresponding to the bumpy state include X-axis angular velocity, Y-axis angular velocity, Z-axis angular velocity, X-axis acceleration, Y-axis acceleration, Z-axis acceleration, pitch angle changes, and roll angle changes; the detection thresholds include a first threshold and an eleventh threshold compared to the X-axis angular velocity, a twelfth threshold and a thirteenth threshold compared to the Y-axis angular velocity, a sixth threshold and a fourteenth threshold compared to the Z-axis angular velocity, a seventh threshold and a fifteenth threshold compared to the X-axis acceleration, an eighth threshold and a sixteenth threshold compared to the Y-axis acceleration, a seventeenth threshold and an eighteenth threshold compared to the Z-axis acceleration, a fourth threshold and a nineteenth threshold compared to the pitch angle changes, and a twentieth threshold and a twenty-first threshold compared to the roll angle changes. Based on the matching results corresponding to each of the detection targets, determine the number of groups of sliding window data that satisfy the first preset condition in the matching results corresponding to each of the detection targets, and use this as the first quantity, including: The number of absolute values ​​of multiple X-axis angular velocities in each group of sliding window data that are greater than the first threshold and less than the eleventh threshold is counted separately, and the number of groups in group v1 where the second number is greater than a1 is accumulated, which is taken as the first number corresponding to the X-axis angular velocity; The number of absolute values ​​of multiple Y-axis angular velocities in each group of sliding window data that are greater than the twelfth threshold and less than the thirteenth threshold is counted separately, and the number of groups in group v11 where the second number is greater than a11 is accumulated, which is taken as the first number corresponding to the Y-axis angular velocity; The number of absolute values ​​of multiple Z-axis angular velocities in each group of sliding window data that are greater than the sixth threshold and less than the fourteenth threshold is counted separately, and the number of groups in group v6 where the second number is greater than a6 is accumulated, which is taken as the first number corresponding to the Z-axis angular velocity; The number of absolute values ​​of multiple X-axis accelerations in each group of sliding window data that are greater than the seventh threshold and less than the fifteenth threshold is counted separately, and the number of groups in group v7 where the second number is greater than a7 is accumulated, which is taken as the first number corresponding to the X-axis acceleration; The number of absolute values ​​of multiple Y-axis accelerations in each group of sliding window data that are greater than the eighth threshold and less than the sixteenth threshold is counted separately, and the number of groups in group v8 where the second number is greater than a8 is accumulated, which is taken as the first number corresponding to the Y-axis acceleration; The number of absolute values ​​of multiple Z-axis accelerations in each group of sliding window data that are greater than the seventeenth threshold and less than the eighteenth threshold is counted separately, and the number of groups of Z-axis accelerations in group v12 that have the second number greater than a12 is accumulated as the first number corresponding to the Z-axis acceleration; The number of absolute values ​​of multiple pitch angle changes in each group of sliding window data that are greater than the fourth threshold and less than the nineteenth threshold is counted separately, and the number of groups in group v4 whose second number is greater than a4 is accumulated, which is taken as the first number corresponding to the pitch angle change. The number of absolute values ​​of multiple roll angle changes in each group of sliding window data that are greater than the twentieth threshold and less than the twenty-first threshold is counted separately, and the number of groups in group v13 whose second number is greater than a13 is accumulated as the first number corresponding to the roll angle change; the sum of v1, v11, v6, v7, v8, v12, v14 and v13 is M.

8. The method for detecting the vehicle condition according to any one of claims 1 to 3, characterized in that, The Euler angles include the yaw angle; the Euler angle change includes the yaw angle change; the average speed includes the first average speed; the multiple detection targets corresponding to the straight-line state include Z-axis angular velocity, Z-axis acceleration, the yaw angle change, and the first average speed; the detection thresholds include a sixth threshold compared with the Z-axis angular velocity, a seventeenth threshold compared with the Z-axis acceleration, a ninth threshold compared with the yaw angle change, and a twenty-second threshold compared with the first average speed. Based on the matching results corresponding to each of the detection targets, determine the number of groups of sliding window data that satisfy the first preset condition in the matching results corresponding to each of the detection targets, and use this as the first quantity, including: The number of absolute values ​​of multiple Z-axis angular velocities in each group of sliding window data that are less than the sixth threshold is counted separately, and the number of groups in group v6 where the second number is greater than a6 is accumulated, which is taken as the first number corresponding to the Z-axis angular velocity; The number of absolute values ​​of multiple Z-axis accelerations in each group of sliding window data that are less than the seventeenth threshold is counted separately, and the number of groups of Z-axis accelerations in group v12 that have the second number greater than a12 is accumulated as the first number corresponding to the Z-axis acceleration; The number of absolute values ​​of multiple yaw angle changes in each group of sliding window data that are less than the ninth threshold is counted separately, and the number of groups in group v9 where the second number is greater than a9 is accumulated, which is taken as the first number corresponding to the yaw angle change. The number of absolute values ​​of multiple first speed averages in each group of sliding window data that are less than the twenty-second threshold is counted separately, and the number of groups in group v14 where the second number is greater than a14 is accumulated as the first number corresponding to the first speed average; the sum of v6, v12, v9 and v14 is M.

9. The method for detecting the vehicle condition according to any one of claims 1 to 3, characterized in that, The average velocity includes a second average velocity; the multiple detection targets corresponding to the stationary state include X-axis angular velocity, Y-axis angular velocity, Z-axis angular velocity, X-axis acceleration, Y-axis acceleration, Z-axis acceleration, the second average velocity, and velocity change; the detection thresholds include a first threshold compared with the X-axis angular velocity, a twelfth threshold compared with the Y-axis angular velocity, a sixth threshold compared with the Z-axis angular velocity, a seventh threshold compared with the X-axis acceleration, an eighth threshold compared with the Y-axis acceleration, a seventeenth threshold compared with the Z-axis acceleration, a twenty-third threshold compared with the second average velocity, and a twenty-fourth threshold compared with the velocity change. Based on the matching results corresponding to each of the detection targets, determine the number of groups of sliding window data that satisfy the first preset condition in the matching results corresponding to each of the detection targets, and use this as the first quantity, including: The number of absolute values ​​of multiple X-axis angular velocities in each group of sliding window data that are less than a first threshold is counted, and the number of groups in group v1 where the second number is greater than a1 is accumulated, which is taken as the first number corresponding to the X-axis angular velocity; The number of absolute values ​​of multiple Y-axis angular velocities in each group of sliding window data that are less than the twelfth threshold is counted separately, and the number of groups in group v11 where the second number is greater than a11 is accumulated, which is taken as the first number corresponding to the Y-axis angular velocity; The number of absolute values ​​of multiple Z-axis angular velocities in each group of sliding window data that are less than the sixth threshold is counted separately, and the number of groups in group v6 where the second number is greater than a6 is accumulated, which is taken as the first number corresponding to the Z-axis angular velocity; The number of absolute values ​​of multiple X-axis accelerations in each group of sliding window data that are less than the seventh threshold is counted separately, and the number of groups in group v7 where the second number is greater than a7 is accumulated, which is taken as the first number corresponding to the X-axis acceleration; The number of absolute values ​​of multiple Y-axis accelerations in each group of sliding window data that are less than the eighth threshold is counted separately, and the number of groups in group v8 where the second number is greater than a8 is accumulated, which is taken as the first number corresponding to the Y-axis acceleration; The number of absolute values ​​of multiple Z-axis accelerations in each group of sliding window data that are less than the seventeenth threshold is counted separately, and the number of groups of Z-axis accelerations in group v12 that have the second number greater than a12 is accumulated as the first number corresponding to the Z-axis acceleration; The number of absolute values ​​of multiple second speed averages in each group of sliding window data that are less than the twenty-third threshold is counted, and the number of groups in group v15 where the second number is greater than a15 is accumulated, which is taken as the first number corresponding to the second speed average. The number of absolute values ​​of multiple speed changes in each group of sliding window data that are less than the twenty-fourth threshold is counted separately, and the number of groups in group v16 whose second number is greater than a16 is accumulated as the first number corresponding to the speed change; the sum of v1, v11, v6, v7, v8, v12, v15 and v16 is M.

10. The vehicle status detection method according to claim 1, characterized in that, The second time period corresponding to each group of sliding window data is determined based on the frequency of the sensor that collects the sensor data; the target data is the sensor data or determined based on the sensor data; the magnitude of the sensor frequency is inversely proportional to the length of the second time period.

11. The vehicle status detection method according to claim 1, characterized in that, Any one of the velocity change, the angular velocity change, the acceleration change, the elevation change, and the Euler angle change is denoted as the parameter change. Each of the aforementioned parameter changes has a target parameter of the same type; The steps for determining the parameter change corresponding to the target parameter include: For any set of sliding window data, the difference between the maximum and minimum values ​​of the n1 target parameters in the sliding window data is taken as the parameter change amount of one sliding window output, so as to obtain multiple parameter changes of multiple sliding window outputs within the first time period; n1 is an integer greater than or equal to 2; or, For any set of sliding window data, if n1 is greater than 2, determine the fluctuation values ​​of two time-adjacent target parameters in the sliding window data, and take any one of the maximum fluctuation value, minimum fluctuation value and average fluctuation value as the parameter change amount of one sliding window output, so as to obtain multiple parameter changes of multiple sliding window outputs within the first time period.

12. The vehicle status detection method according to claim 1, characterized in that, The velocity includes the first integral parameter of the X-axis acceleration; The average speed includes the first average speed; The steps for determining the first average speed include: For any set of sliding window data, the average value of the n2 first integral parameters in the sliding window data is recorded as the first speed average value, so as to obtain multiple first speed average values ​​of multiple sliding window outputs within the first time period; n2 is an integer greater than or equal to 2.

13. The vehicle status detection method according to claim 1, characterized in that, The velocity includes a first integral parameter of the X-axis acceleration, a second integral parameter of the Y-axis acceleration, and a second integral parameter of the Z-axis acceleration; the average velocity includes the average second velocity. The steps for determining the second average speed include: For three sets of sliding window data in the same time sequence, the i-th fused data in the new sliding window data is determined based on the i-th first integral parameter, the second integral parameter and the third integral parameter in the three sets of sliding window data; i is less than or equal to n3, where n3 is the number of sub-data in the sliding window data; For any set of new sliding window data, the average value of the n4 fused data in the new sliding window data is recorded as the second speed average value, so as to obtain multiple second speed average values ​​of multiple sliding window outputs within the first time period; n4 is an integer greater than or equal to 2.

14. The vehicle status detection method according to claim 1, characterized in that, Obtain target data for the vehicle within the first time period, including: Acquire raw data, which includes inertial measurement unit (IMU) data, positioning device (P-BOX) data, and wheel speedometer data; the IMU data includes multi-axis angular velocity and multi-axis acceleration; the P-BOX data includes the velocity, elevation, and Euler angles. If the confidence level of the P-BOX data is greater than or equal to a set value, the IMU data and the P-BOX data are used as the target data. If the confidence level of the P-BOX data is less than a set value, the speed, elevation, and Euler angle are calculated using the IMU data and the wheel speedometer data, and are used together with the IMU data as the target data.

15. A vehicle condition detection device, characterized in that, It includes a data acquisition module, a first detection module, and a second detection module; The data acquisition module is configured to acquire target data of the vehicle within a first time period; the target data includes at least one of speed, multi-axis angular velocity, multi-axis acceleration, elevation, Euler angle, mean speed, speed change, multi-axis angular velocity change, multi-axis acceleration change, elevation change, and Euler angle change. The sliding window processing module is configured to perform sliding window processing on at least a portion of the target data to obtain N sets of sliding window data; The first detection module is configured to match the M groups of sliding window data with the identification information of at least one detection target, and determine the number of groups of sliding window data that satisfy the first preset condition in the matching results corresponding to each detection target according to the matching results corresponding to each detection target, and use this as the first quantity; M is less than or equal to N; The second detection module is configured to determine the actual state of the vehicle based on the first quantity corresponding to each of the detection targets.

16. A computer device, characterized in that, include: The computer device includes a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the computer device is running, the processor communicates with the memory via the bus. When the machine-readable instructions are executed by the processor, they perform the steps of the vehicle state detection method as described in any one of claims 1 to 14.

17. A computer non-transient readable storage medium, characterized in that, The computer non-transient readable storage medium stores a computer program that, when executed by a processor, performs the steps of the vehicle state detection method as described in any one of claims 1 to 14.

Images