Vehicle detection dynamic rollover warning method, system and electronic device

Abstract

The application relates to the field of automobile safety technology and provides a vehicle detection dynamic rollover early warning method, a system and electronic equipment, which comprises the following steps: based on a target vehicle, receiving multi-source data of a real-time monitoring unit integrated with an accelerometer and a gyroscope, setting a complementary filtering framework, combining vehicle posture information and EKF principles to establish a dynamic model; predicting the vehicle posture and calculating dynamic parameters, comparing the dynamic parameters with preset threshold values, judging whether a rollover early warning signal is triggered or not, solving the technical problem that the accuracy of rollover early warning is insufficient under the influence of various interference factors in a complex and constantly changing driving environment, setting the complementary filtering framework, effectively filtering out noise, improving the accuracy of vehicle posture estimation by using the EKF principle, simultaneously, deeply analyzing the multidimensional correlation among the vehicle posture, road conditions and dynamic running parameters, quickly judging the rollover risk and triggering the early warning signal, and reducing the probability of rollover accidents.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automobile safety, in particular to a vehicle detection dynamic rollover warning method, system and electronic device. BACKGROUND

[0002] Vehicle rollover accidents, as one of the most serious types of traffic accidents, not only threaten the safety of passengers, therefore, researching and developing a rollover warning system with active safety protection function has become an important direction to improve the safety performance of vehicles, in recent years, with the progress of sensing technology, data fusion algorithm and vehicle dynamic control technology, vehicle safety systems gradually move towards intelligentization, providing technical support for rollover risk assessment and warning.

[0003] Existing rollover warning technology usually relies on single sensor data (such as a gyroscope), which is easily affected by data noise, sensor drift and other factors in complex driving environments, resulting in a decrease in the reliability of rollover risk assessment results, in addition, traditional rollover detection methods, such as physical testing and simulation models, although to some extent, provide a basis for vehicle rollover prevention, but still have limitations such as limited test site, unable to cover all actual scenarios, at the same time, traditional methods mostly use simple threshold judgment or fixed rule models, which cannot fully consider the multi-dimensional correlation between vehicle posture, road conditions and dynamic running parameters, and also face problems such as precision, response speed and false positives / misses.

[0004] In summary, the existing technology has the technical problem of insufficient accuracy of rollover warning in complex and constantly changing driving environments affected by various interference factors. SUMMARY

[0005] The present application provides a vehicle detection dynamic rollover warning method, system and electronic device, which aims to solve the technical problem of insufficient accuracy of rollover warning in complex and constantly changing driving environments affected by various interference factors in the prior art, by combining the data of accelerometers and gyroscopes, setting up a complementary filtering architecture, effectively filtering out noise, further utilizing the EKF principle to improve the accuracy of vehicle posture estimation, at the same time, deeply analyzing the multi-dimensional correlation between vehicle posture, road conditions and dynamic running parameters, quickly judging the rollover risk and triggering the warning signal, and reducing the probability of rollover accidents.

[0006] In view of the above problems, the present application provides a vehicle detection dynamic rollover warning method, system and electronic device.

[0007] The first aspect disclosed in the present application provides a vehicle detection dynamic rollover warning method, wherein the method includes: based on the target vehicle, receiving multi-source monitoring data corresponding to the real-time monitoring unit, the real-time monitoring unit is integrated with an accelerometer and a gyroscope; setting a complementary filtering architecture through the multi-source monitoring data; introducing vehicle posture information, utilizing the EKF principle, and establishing a vehicle dynamics model; connecting the complementary filtering architecture and the vehicle dynamics model, determining the predicted vehicle posture information, and calculating the obtained dynamic parameter set; based on the dynamic parameter set, comparing with a preset rollover threshold to determine whether a rollover warning signal is triggered.

[0008] The second aspect disclosed in the present application provides a vehicle detection dynamic rollover warning system, wherein the system includes: a data receiving module for receiving multi-source monitoring data corresponding to a real-time monitoring unit based on a target vehicle, wherein the real-time monitoring unit is integrated with an accelerometer and a gyroscope; an architecture setting module for setting a complementary filtering architecture through the multi-source monitoring data; a model building module for introducing vehicle posture information, utilizing the EKF principle, and establishing a vehicle dynamics model; a posture determination module for connecting the complementary filtering architecture and the vehicle dynamics model, determining the predicted vehicle posture information, and calculating the obtained dynamic parameter set; a comparison module for comparing the dynamic parameter set with a preset rollover threshold to determine whether a rollover warning signal is triggered.

[0009] The third aspect disclosed in the present application provides an electronic device, including a memory and a processor, wherein the memory stores executable instructions, and when the processor executes the executable instructions stored in the memory, any step of the first aspect disclosed in the present application is implemented.

[0010] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0011] Due to the use of multi-source data from a real-time monitoring unit based on a target vehicle, which receives accelerometers and gyroscopes, a complementary filtering architecture is set up, and a dynamic model is established by combining vehicle posture information and the EKF principle; the vehicle posture is predicted and dynamic parameters are calculated, and compared with preset thresholds to determine whether a rollover warning signal is triggered. The present application provides a vehicle detection dynamic rollover warning method, system and electronic equipment. By combining data from accelerometer and gyroscope sensors, a complementary filtering architecture is set up to effectively filter out noise, and the EKF principle is further utilized to improve the accuracy of vehicle posture estimation. At the same time, the multi-dimensional correlation between vehicle posture, road conditions and dynamic operating parameters is deeply analyzed to quickly determine the rollover risk and trigger a warning signal, thereby reducing the probability of rollover accidents.

[0012] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 A flowchart of a vehicle detection dynamic rollover warning method is provided for an embodiment of the present application;

[0014] Figure 2 A structural diagram of a vehicle detection dynamic rollover warning system is provided for an embodiment of the present application;

[0015] Figure 3 This is a diagram of the internal structure of an electronic device provided in an embodiment of the present application.

[0016] Explanation of the accompanying drawings: data receiving module 11, architecture setting module 12, model building module 13, posture determination module 14, comparison module 15, bus 300, receiver 301, processor 302, transmitter 303, memory 304, bus interface 305. DETAILED DESCRIPTION

[0017] The following description of exemplary embodiments of the present invention is provided in conjunction with the accompanying drawings, in which various details of the embodiments of the present invention are included to facilitate understanding. These details should be considered as merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0018] Example 1

[0019] like Figure 1 As shown, an embodiment of the present application provides a vehicle detection dynamic rollover warning method, wherein the method includes:

[0020] S100: Based on the target vehicle, receiving multi-source monitoring data corresponding to the real-time monitoring unit, the real-time monitoring unit is integrated with an accelerometer and a gyroscope; S200: setting a complementary filtering architecture based on the multi-source monitoring data; S300: introducing vehicle posture information, using the EKF principle, and establishing a vehicle dynamics model.

[0021] Specifically, multi-source monitoring data refers to data from multiple sensors. The real-time monitoring unit is integrated with an accelerometer and a gyroscope. The accelerometer is used to measure the linear acceleration of the vehicle, while the gyroscope is used to measure the angular velocity of the vehicle. The data from these two sensors are combined to more comprehensively reflect the dynamic state of the vehicle. The complementary filtering architecture optimizes the algorithm model for collecting data by combining the advantages of low-pass filters and high-pass filters. The low-pass filter can smooth the data and reduce noise, while the high-pass filter can capture rapidly changing signals.

[0022] The EKF (extended Kalman filter) principle is an extended form of Kalman filtering and is suitable for state estimation problems in nonlinear systems. Vehicle dynamics models usually have nonlinear characteristics. EKF can effectively estimate the dynamic state of the vehicle, such as posture and speed, through linearization processing, thereby improving the estimation accuracy of vehicle posture information; the vehicle dynamics model is used to describe the motion characteristics of the vehicle under various working conditions. The model takes into account multiple factors such as the vehicle's mass, inertia, friction, driving force, etc., and can accurately simulate the dynamic behavior of the vehicle. By establishing a vehicle dynamics model, we can better understand the vehicle's motion state and provide theoretical support for rollover warning.

[0023] The accelerometer and gyroscope integrated in the real-time monitoring unit will continuously collect dynamic data of the target vehicle. The accelerometer measures the linear acceleration of the vehicle in three directions, and the gyroscope measures the angular velocity of the vehicle in three directions. These data reflect the acceleration changes and rotational motion of the vehicle during driving, providing a basis for subsequent data processing and posture estimation; the collected multi-source monitoring data is input into the complementary filtering architecture for processing. The complementary filtering architecture optimizes the data through the synergistic effect of low-pass filters and high-pass filters. The low-pass filter is used to smooth the data, reduce the impact of noise, and make the data more stable; the high-pass filter is used to capture the rapidly changing parts of the data, such as the vehicle's rapid turning or acceleration. Through complementary filtering, the accuracy and reliability of the data can be effectively improved, providing higher-quality data support for subsequent vehicle posture estimation and rollover risk assessment.

[0024] After obtaining the optimized data, the vehicle posture information is introduced, and the dynamic state of the vehicle is estimated using the EKF principle. The vehicle dynamics model is used as the state equation of the EKF. The vehicle dynamics model describes the motion characteristics of the vehicle under various working conditions. The data of the accelerometer and gyroscope are used as observation data. The observation equation of the EKF is combined with the state equation to perform real-time estimation of the vehicle's posture, speed and other state variables, eliminate the state estimation offset of the nonlinear system, and accurately track the dynamic state changes of the vehicle by continuously updating the state estimate. Even under complex nonlinear conditions, reliable estimation accuracy can be maintained.

[0025] S400: Connecting the complementary filtering architecture and the vehicle dynamics model to determine the predicted vehicle posture information and calculate the obtained dynamic parameter set; S500: Based on the dynamic parameter set, comparing with the preset rollover threshold to determine whether a rollover warning signal is triggered.

[0026] Specifically, predicting vehicle posture information refers to estimating the posture of the vehicle at any future moment through the combination of the vehicle dynamics model and the complementary filtering architecture. It is based on the current vehicle state and known physical laws to infer the posture that the vehicle may reach at the next moment, such as the tilt angle, roll angle, etc.; the dynamic parameter set is a set of parameters that describe the dynamic state of the vehicle, including roll angle elements, roll speed elements, roll acceleration elements, etc. These parameters are obtained through calculations of the vehicle dynamics model and the complementary filtering architecture, and can comprehensively reflect the dynamic behavior of the vehicle and possible rollover risks; the rollover threshold is used to determine whether the vehicle is at risk of rollover. When one or some parameters in the dynamic parameter set exceed this threshold, the vehicle is considered to be in danger of rollover, thereby triggering a rollover warning signal.

[0027] Multi-source monitoring data is processed using a complementary filtering architecture to obtain stable and accurate vehicle status data. This data is input into a vehicle dynamics model, and the model's calculations predict the posture that the vehicle may reach in the next moment. This prediction takes into account the vehicle's physical characteristics and current motion state, and can therefore provide more accurate vehicle posture information. After obtaining the predicted vehicle posture information, a set of dynamic parameters is calculated, including roll angle, roll speed, roll acceleration, etc. These parameters can comprehensively describe the dynamic behavior of the vehicle and are an important basis for assessing rollover risks. The calculated dynamic parameter set is compared with a preset rollover threshold. If one or more parameters exceed the rollover threshold, it indicates that the vehicle may be at risk of rollover, and a rollover warning signal is immediately triggered to provide a timely warning to the driver, so that appropriate measures can be taken to avoid rollover accidents, thereby improving vehicle safety.

[0028] Furthermore, the present application method includes:

[0029] The dynamic parameter set includes a roll angle element, a roll speed element, and a roll acceleration element; a first risk item, a second risk item, and a third risk item are set respectively through the roll angle element, the roll speed element, and the roll acceleration element in the dynamic parameter set; based on the first risk item, the second risk item, and the third risk item, a risk assessment model is established, and the risk assessment model evaluates the rollover risk index of the target vehicle using a deviation value corresponding to a preset rollover threshold.

[0030] Specifically, the roll angle element refers to the tilt angle of the vehicle relative to the horizontal plane at a certain moment, reflecting the degree of tilt of the vehicle during the rollover process, and is one of the important indicators for assessing rollover risk; the roll speed element represents the angular velocity of the vehicle's roll, that is, the rate of change of the vehicle's roll angle per unit time. The faster the roll speed, the higher the risk of the vehicle rollover; the roll acceleration element refers to the angular acceleration of the vehicle's roll, that is, the rate of change of the roll speed, reflecting the change in acceleration of the vehicle's rollover motion; the risk item is a quantitative indicator of the vehicle's rollover risk, corresponding to different roll angles, roll speeds, and roll accelerations. Dynamic parameters of each risk item are converted into risk values ​​through a specific calculation formula or method, which are used to evaluate the rollover risk of the vehicle; the risk assessment model is used to combine multiple risk items and derive a comprehensive risk index through a certain calculation method, comprehensively considering the rollover risk of the vehicle under different dynamic parameters, and providing a more accurate basis for rollover warning; the deviation value corresponding to the preset rollover threshold refers to a reference value in the risk assessment model used to judge whether the rollover risk meets the warning standard. When the risk index exceeds the deviation value, the vehicle is considered to have a rollover risk, thereby triggering a rollover warning signal.

[0031] Through calculations using a vehicle dynamics model and a complementary filtering architecture, dynamic parameters such as the vehicle's roll angle, roll velocity, and roll acceleration are obtained. These parameters reflect the changes in the vehicle's tilt, roll velocity, and roll acceleration during the roll process, respectively, and serve as the basis for assessing rollover risk. Based on the roll angle, roll velocity, and roll acceleration elements, a first risk item (such as the roll angle), a second risk item (such as the roll velocity), and a third risk item (such as the roll acceleration) are set, and the corresponding dynamic parameters of the first, second, and third risk items are converted into risk values. Through the analysis steps described above, the first, second, and third risk items are combined to establish a risk assessment model. The values ​​of each risk item are combined to derive a rollover risk index, which comprehensively reflects the degree of rollover risk of the vehicle under the current dynamic parameters. By comparing the risk index with the deviation value corresponding to the preset rollover threshold, it is determined whether the vehicle is at risk of rollover, thereby achieving real-time assessment and early warning of rollover risk.

[0032] Furthermore, by using the multi-source monitoring data, a complementary filtering architecture is set up, and the method of the present application includes:

[0033] The first processed data is extracted by combining the multi-source monitoring data with the low-pass filter network layer; the second processed data is extracted by combining the multi-source monitoring data with the high-pass filter network layer; and a complementary filtering architecture is set based on the low-pass filter network layer and the first processed data, and the high-pass filter network layer and the second processed data.

[0034] Specifically, the low-pass filter network layer is used to extract the low-frequency components in the signal. The low-pass filter can effectively smooth the signal, remove high-frequency noise, and make the signal more stable. In the structure of the network layer, the low-pass filter can process the input multi-source monitoring data and extract the data components that reflect the stable state of the vehicle; the first processed data refers to the data obtained after processing by the low-pass filter network layer. These data mainly contain low-frequency information in vehicle motion, such as the vehicle's stable acceleration changes, slow steering movements, etc., which can reflect the vehicle's long-term motion trend and stability; the high-pass filter network layer is used to extract the high-frequency components in the signal. The high-pass filter can effectively capture the rapidly changing parts of the signal and remove low-frequency Interference. In the structure of the network layer, the high-pass filter can process the input multi-source monitoring data and extract the data components that reflect the rapid change of the vehicle state; the second processed data refers to the data obtained after processing by the high-pass filter network layer. These data mainly contain high-frequency information in vehicle motion, such as the vehicle's rapid acceleration changes, sudden steering actions, etc., which can reflect the vehicle's short-term motion changes and dynamic responses; the complementary filtering architecture refers to the algorithm model corresponding to the comprehensive processing of the signal by combining the advantages of low-pass filtering and high-pass filtering. The complementary filtering architecture can make full use of the smoothing characteristics of the low-pass filter and the fast response characteristics of the high-pass filter to extract more accurate and reliable vehicle dynamic information.

[0035] The multi-source monitoring data is input into the low-pass filter network layer for processing. The low-pass filter will smooth the data, remove high-frequency noise and interference, and extract low-frequency information reflecting the stable state of the vehicle. These low-frequency information include the vehicle's stable acceleration changes, slow steering movements, etc., which can reflect the vehicle's long-term motion trend and stability. The first processed data provides a stable foundation for subsequent vehicle posture estimation and rollover risk assessment; the multi-source monitoring data is input into the high-pass filter network layer for processing. The high-pass filter will quickly respond to the data, capture the high-frequency changes, and extract high-frequency information reflecting the vehicle's rapidly changing state. These high-frequency information include the vehicle's rapid acceleration changes, sudden steering movements, etc., which can reflect the vehicle's short-term motion changes and dynamic responses. The second processed data provides important dynamic information for subsequent vehicle dynamic analysis and rollover risk warning.

[0036] Based on the first processed data obtained by the low-pass filtering network layer and the second processed data obtained by the high-pass filtering network layer, a complementary filtering architecture is set up. The complementary filtering architecture comprehensively processes the first processed data and the second processed data by reasonably allocating the functions of low-pass filtering and high-pass filtering. In this way, the smoothing characteristics of the low-pass filter and the fast response characteristics of the high-pass filter are fully utilized to extract more accurate and reliable vehicle dynamic information. The complementary filtering architecture can effectively improve the accuracy of vehicle posture estimation and rollover risk assessment, and provide more reliable data support for the rollover warning system.

[0037] Furthermore, based on the low-pass filter network layer and the first processed data, and the high-pass filter network layer and the second processed data, a complementary filtering architecture is set. The method of the present application includes:

[0038] Determine low-pass filtering parameters based on the low-pass filtering network layer and the first processed data; determine high-pass filtering parameters based on the high-pass filtering network layer and the second processed data; and configure a complementary filtering architecture based on the low-pass filtering network layer and the high-pass filtering network layer through the low-pass filtering parameters and the high-pass filtering parameters.

[0039] Specifically, the low-pass filter parameters are used to define the characteristics of the low-pass filter, mainly including the cutoff frequency, etc. The cutoff frequency determines the maximum frequency that the low-pass filter can pass. Signals below this frequency will be retained, while signals above this frequency will be attenuated or filtered out. By adjusting the low-pass filter parameters, the filter's processing effect on low-frequency signals can be optimized; the high-pass filter parameters are used to define the characteristics of the high-pass filter, mainly including the cutoff frequency, etc. The cutoff frequency determines the minimum frequency that the high-pass filter can pass. Signals above this frequency will be retained, while signals below this frequency will be attenuated or filtered out. By adjusting the high-pass filter parameters, the filter's processing effect on high-frequency signals can be optimized; configuring a complementary filtering architecture refers to setting and adjusting the complementary filtering architecture according to the determined low-pass filtering parameters and high-pass filtering parameters. The complementary filtering architecture realizes comprehensive processing of signals by reasonably configuring the parameters of the low-pass filter and the high-pass filter to achieve better filtering effects and data processing performance.

[0040] According to the first processed data obtained by the low-pass filter network layer, the low-frequency signal characteristics therein are analyzed, and by observing the frequency distribution and stability of the data, the appropriate low-pass filter parameters, such as the cutoff frequency, are determined. This cutoff frequency should be selected within the frequency range that can retain the vehicle's smooth motion state information while effectively filtering out high-frequency noise. For example, if the vehicle's smooth acceleration changes are mainly concentrated in the lower frequency range, the cutoff frequency of the low-pass filter can be set within this range to retain these important low-frequency information; according to the second processed data obtained by the high-pass filter network layer, the high-frequency signal characteristics therein are analyzed, and by observing the rapid changes and dynamic responses of the data, the appropriate high-pass filter parameters, such as the cutoff frequency, are determined. This cutoff frequency should be selected within the frequency range that can capture the vehicle's rapid motion state information. While capturing dynamic change information, it effectively filters out low-frequency interference within the frequency range. For example, if the rapid acceleration changes of the vehicle are mainly concentrated in the higher frequency range, the cutoff frequency of the high-pass filter can be set within this range to capture these important high-frequency information; based on the determined low-pass filter parameters and high-pass filter parameters, the complementary filtering architecture is configured. By reasonably setting the parameters of the low-pass filter and the high-pass filter, the complementary filtering architecture can make full use of the smoothing characteristics of the low-pass filter and the fast response characteristics of the high-pass filter to comprehensively process the vehicle dynamic signals. In this way, the stable motion information and fast-changing information of the vehicle can be effectively extracted, the accuracy of vehicle posture estimation and the accuracy of rollover risk assessment can be improved, and more reliable data support can be provided for the rollover warning system.

[0041] Furthermore, according to the low-pass filter network layer and the first processed data, a low-pass filter parameter is determined. The method of the present application includes:

[0042] A frequent start-stop feature is introduced; based on the frequent start-stop feature, a first stability coefficient is output in a first driving state associated with the longitudinal acceleration; and the low-pass filter parameter is adjusted once using the first stability coefficient.

[0043] Specifically, the frequent start-stop feature refers to the frequent starts and stops of a vehicle during driving. This feature is often associated with urban traffic conditions, such as frequent stops at traffic lights and stop-and-go traffic on congested roads. The frequent start-stop feature reflects the frequent changes in the vehicle's longitudinal acceleration, which has a significant impact on the vehicle's stability and dynamic response. The first driving state associated with longitudinal acceleration refers to the specific driving state exhibited by the vehicle in terms of longitudinal acceleration. In the case of frequent starts and stops, the vehicle's longitudinal acceleration will increase rapidly when starting and decrease rapidly when stopping. This state can be regarded as the first driving state and reflects the acceleration changes of the vehicle during the start-stop process. The first stability coefficient is a coefficient used to describe the degree of stability of the vehicle under the frequent start-stop feature. It is calculated based on the vehicle's performance in the first driving state associated with longitudinal acceleration and reflects the vehicle's ability to maintain stability during the start-stop process. A higher first stability coefficient indicates a more stable vehicle during the start-stop process. Primary adjustment refers to optimizing the parameters of the low-pass filter based on the results of the first stability coefficient to better adapt to the dynamic characteristics of the vehicle under frequent starts and stops and improve the low-pass filter's processing of vehicle stable motion state information.

[0044] The frequent start-stop feature of the vehicle is introduced. Further, when the vehicle experiences rapid increases and decreases in acceleration multiple times in a short period of time, the system can determine that the vehicle is in a state of frequent start-stop. The introduction of this feature helps to more accurately evaluate the stability of the vehicle in complex traffic environments. After identifying the frequent start-stop feature, a first stability coefficient is calculated based on the vehicle's performance in a first driving state associated with the longitudinal acceleration. Specifically, the first stability coefficient is derived by analyzing the changes in the vehicle's acceleration during the start-stop process, as well as the vehicle's stability performance under these changes. For example, if the vehicle can maintain stability well during the start-stop process without obvious shaking or tilting, then the first stability coefficient will be higher.

[0045] According to the calculated first stability coefficient, the parameters of the low-pass filter are adjusted once. If the first stability coefficient is high, it means that the vehicle is relatively stable during the start-stop process, then the cutoff frequency of the low-pass filter can be appropriately increased to better retain the information of the vehicle's stable motion state; if the first stability coefficient is low, it means that the vehicle is less stable during the start-stop process, then the cutoff frequency of the low-pass filter can be appropriately lowered to enhance the smoothing processing capability of the vehicle's stable motion state information. Through this adjustment, the low-pass filter can better adapt to the dynamic characteristics of the vehicle under frequent start-stop conditions, and improve the accuracy of vehicle posture estimation and rollover risk assessment.

[0046] Furthermore, according to the low-pass filter network layer and the first processed data, a low-pass filter parameter is determined. The method of the present application includes:

[0047] A steering sensitivity feature is introduced; based on the steering sensitivity feature, a second stability coefficient is output in a second driving state associated with the lateral acceleration; and the low-pass filter parameter is adjusted secondary by the second stability coefficient.

[0048] Specifically, the steering sensitivity characteristic refers to the vehicle's response speed and sensitivity to driver input during steering. This characteristic reflects the vehicle's dynamic characteristics in terms of lateral acceleration and is typically related to the vehicle's suspension system, steering system, and other factors. The steering sensitivity characteristic can affect the vehicle's stability and handling during cornering. The second driving state associated with lateral acceleration refers to the specific driving state exhibited by the vehicle in terms of lateral acceleration. During steering, the vehicle's lateral acceleration varies with the steering angle. This state can be considered a second driving state and reflects the changes in the vehicle's acceleration during steering. The second stability coefficient is a coefficient used to describe the vehicle's stability under the steering sensitivity characteristic. It is calculated based on the vehicle's performance in the second driving state associated with lateral acceleration and reflects the vehicle's ability to maintain stability during steering. A higher second stability coefficient indicates a more stable vehicle during steering. Secondary adjustment refers to optimizing the low-pass filter parameters based on the results of the second stability coefficient, based on the first adjustment, to better adapt to the vehicle's dynamic characteristics under steering sensitivity and further improve the low-pass filter's processing of vehicle stable motion state information.

[0049] By introducing the vehicle's steering sensitivity feature, the system can further determine that the vehicle has high steering sensitivity when the vehicle can quickly and accurately respond to the driver's operations during the steering process, and the change in lateral acceleration matches the change in steering angle. The introduction of this feature helps to more accurately evaluate the stability and handling of the vehicle when turning; after identifying the steering sensitivity feature, the system will calculate a second stability coefficient based on the vehicle's performance in a second driving state associated with the lateral acceleration. Specifically, the second stability coefficient is derived by analyzing the changes in the vehicle's lateral acceleration during the steering process, and the vehicle's stability performance under these changes. For example, if the vehicle can maintain stability well during the steering process without obvious skidding or tilting, then the second stability coefficient will be higher.

[0050] According to the calculated second stability coefficient, the parameters of the low-pass filter are adjusted secondary. If the second stability coefficient is high, it means that the vehicle is relatively stable during the steering process, then the cutoff frequency of the low-pass filter can be appropriately increased to better retain the information of the vehicle's stable motion state; if the second stability coefficient is low, it means that the vehicle is less stable during the steering process, then the cutoff frequency of the low-pass filter can be appropriately lowered to enhance the smoothing processing capability of the vehicle's stable motion state information. Through this adjustment, the low-pass filter can better adapt to the dynamic characteristics of the vehicle under sensitive steering conditions, further improving the accuracy of vehicle posture estimation and rollover risk assessment.

[0051] Furthermore, the complementary filtering architecture and the vehicle dynamics model are connected to determine the predicted vehicle posture information. The method of the present application includes:

[0052] A random noise term corresponding to the unpaved road surface is added to the vehicle dynamics model; and a vibration compensation mechanism is set based on the random noise term, wherein the vibration compensation mechanism is used to correct the vehicle posture deviation caused by the unpaved road surface.

[0053] Specifically, unpaved roads refer to roads that are not paved or have poor paving quality, such as dirt roads and gravel roads. These roads are usually rough and easily cause vehicle vibration and bumps, affecting the vehicle's driving stability and posture control. In the vehicle dynamics model, the random noise term is used to represent the uncertainty and randomness in vehicle motion caused by external environmental factors (such as unpaved roads). Including the random noise term can make the model more realistically reflect the dynamic behavior of the vehicle under complex road conditions. The vibration compensation mechanism refers to detecting and analyzing the vibration of the vehicle on the unpaved road, calculating the corresponding compensation amount, and adjusting and correcting the vehicle's posture to reduce the posture deviation caused by the uneven road surface and improve the vehicle's driving stability and safety.

[0054] A random noise term is introduced into the vehicle dynamics model to simulate the random disturbances in vehicle motion caused by unpaved roads. This noise term can be modeled based on the characteristics of the unpaved road. For example, the probability distribution and related parameters of the noise term can be determined by statistically analyzing the vibration data of the vehicle on the unpaved road. Including the random noise term in the model can enable the model to more accurately describe the dynamic behavior of the vehicle on the unpaved road. Based on the added random noise term, a vibration compensation mechanism is designed and implemented. This mechanism monitors the vibration of the vehicle on the unpaved road in real time and uses sensor data (such as accelerometers and gyroscopes) to detect the vehicle's posture changes and vibration amplitude. Based on the detected vibration information, the corresponding compensation amount is calculated to adjust and correct the vehicle's posture. For example, when the vehicle tilts due to bumps on the unpaved road, the vibration compensation mechanism can calculate a reverse compensation torque to restore the vehicle to a more level state, thereby reducing the posture deviation caused by the unpaved road. In the above steps, the vibration compensation mechanism is used to effectively correct the vehicle's posture deviation when driving on unpaved roads, thereby improving the vehicle's driving stability and enabling the vehicle to maintain good driving performance and safety under complex road conditions. At the same time, the corrected vehicle posture information also provides more accurate data support for the rollover warning system, helping to improve the accuracy of rollover risk assessment and reduce the probability of rollover accidents.

[0055] Furthermore, to determine whether a rollover warning signal is triggered, the present application method includes:

[0056] Connecting to an ADAS unit; when a rollover warning signal is triggered, the ADAS unit immediately responds to an emergency protection operation, which includes any one or more of an automatic braking command, a lane keeping command, and a vehicle stabilization command.

[0057] Specifically, ADAS (Advanced Driver Assistance Systems) is a system that integrates multiple driving assistance functions to improve vehicle safety and driving comfort. ADAS units usually include functional modules such as automatic braking, lane keeping, and vehicle stability control. They can monitor and analyze the vehicle's driving status in real time through sensors and algorithms, and automatically perform corresponding operations when necessary; emergency protection operation refers to a series of protective measures automatically performed by the ADAS unit when it detects that the vehicle is at risk of rollover or other emergency situations, in order to prevent or mitigate the occurrence of accidents. Emergency protection operations include automatic braking commands, lane keeping commands, and vehicle stability commands; automatic braking commands are the first type of emergency protection operation, which automatically controls the vehicle's braking The automatic braking system slows down or stops the vehicle. When the risk of rollover is detected, the automatic braking command can quickly reduce the speed of the vehicle and reduce the possibility of rollover. The lane keeping command is the second emergency protection operation. It automatically controls the vehicle's steering system to keep the vehicle in the current lane. When it is detected that the vehicle may deviate from the lane, the lane keeping command can automatically adjust the vehicle's driving direction to prevent the vehicle from rollover or collision with other vehicles. The vehicle stability command is the third emergency protection operation. It automatically controls the vehicle's braking and power output to keep the vehicle in a stable driving state. When it is detected that the vehicle loses stability, the vehicle stability command can automatically adjust the vehicle's braking force and power output to prevent the vehicle from rollover or loss of control.

[0058] Connect the rollover warning system to the vehicle's ADAS unit. This connection can be achieved through the vehicle's electronic control unit (ECU). The rollover warning system transmits the detected rollover risk signal to the ADAS unit. After connecting to the ADAS unit, the rollover warning system can utilize the various driving assistance functions of the ADAS unit to achieve more effective rollover risk protection; when the rollover warning system detects that the vehicle is at risk of rollover and triggers the rollover warning signal, the ADAS unit will respond immediately and perform emergency protection operations. Specific operations include: the ADAS unit will send an automatic braking command to the vehicle's braking system to quickly decelerate or stop the vehicle, effectively reducing the vehicle's speed , reducing the possibility of the vehicle rolling over and buying more reaction time for the driver; if the vehicle tends to deviate from the lane in a rollover risk state, the ADAS unit will send a lane keeping command to the vehicle's steering system, automatically adjusting the vehicle's driving direction to keep the vehicle in the current lane. This can prevent the vehicle from rolling over or colliding with other vehicles, thereby improving vehicle safety; the ADAS unit will also send a vehicle stability command to the vehicle's braking and power output systems based on the vehicle's dynamic state, automatically adjusting the vehicle's braking force and power output to keep the vehicle in a stable driving state, preventing the vehicle from losing stability in a rollover risk state and avoiding the vehicle from rolling over or losing control. In the above steps, by connecting to the ADAS unit and performing emergency protection operations, the rollover warning system can take effective protective measures in a timely manner when a rollover risk is detected, thereby improving vehicle safety. This integrated protection mechanism can not only effectively reduce the probability of rollover accidents, but also mitigate the severity of accidents to a certain extent.

[0059] In summary, the vehicle detection dynamic rollover warning method, system, and electronic device provided by the embodiments of the present application have the following technical effects:

[0060] Due to the use of a target vehicle-based, multi-source monitoring data corresponding to the real-time monitoring unit, the real-time monitoring unit is integrated with an accelerometer and a gyroscope; a complementary filtering architecture is set up through the multi-source monitoring data; vehicle posture information is introduced, the EKF principle is used, and a vehicle dynamics model is established; the complementary filtering architecture and the vehicle dynamics model are connected to determine the predicted vehicle posture information, and a dynamic parameter set is calculated; based on the dynamic parameter set, it is compared with the preset rollover threshold to determine whether a rollover warning signal is triggered. The present application provides a vehicle detection dynamic rollover warning method, system and electronic equipment, which sets a complementary filtering architecture by combining the data of the accelerometer and gyroscope sensors to effectively filter out noise, and further uses the EKF principle to improve the accuracy of vehicle posture estimation. At the same time, it deeply analyzes the multi-dimensional correlation between vehicle posture, road conditions and dynamic operating parameters, quickly determines the rollover risk and triggers a warning signal, and reduces the probability of rollover accidents.

[0061] Example 2

[0062] Based on the same inventive concept as the vehicle detection dynamic rollover warning method in the above embodiment, Figure 2 As shown, an embodiment of the present application provides a vehicle detection dynamic rollover warning system, wherein the system includes:

[0063] The data receiving module 11 is used to receive multi-source monitoring data corresponding to the real-time monitoring unit based on the target vehicle, and the real-time monitoring unit is integrated with an accelerometer and a gyroscope;

[0064] An architecture setting module 12 is used to set a complementary filtering architecture based on the multi-source monitoring data;

[0065] The model building module 13 is used to introduce vehicle posture information, use the EKF principle, and build a vehicle dynamics model;

[0066] a posture determination module 14 for connecting the complementary filtering architecture and the vehicle dynamics model to determine and predict vehicle posture information and calculate a set of dynamic parameters;

[0067] The comparison module 15 is configured to compare the dynamic parameter set with a preset rollover threshold value to determine whether a rollover warning signal is triggered.

[0068] Furthermore, the system is also used to perform the following method:

[0069] Setting a first risk item, a second risk item, and a third risk item respectively according to the roll angle element, the roll speed element, and the roll acceleration element in the dynamic parameter set;

[0070] A risk assessment model is established based on the first risk item, the second risk item, and the third risk item. The risk assessment model assesses a rollover risk index of a target vehicle using a deviation value corresponding to a preset rollover threshold.

[0071] Furthermore, the system is also used to perform the following method:

[0072] Extracting first processed data through the multi-source monitoring data in combination with a low-pass filter network layer;

[0073] Extracting second processed data through the multi-source monitoring data in combination with a high-pass filter network layer;

[0074] A complementary filtering architecture is set based on the low-pass filtering network layer and the first processed data, and the high-pass filtering network layer and the second processed data.

[0075] Furthermore, the system is also used to perform the following method:

[0076] Determining low-pass filtering parameters according to the low-pass filtering network layer and the first processed data;

[0077] Determining high-pass filtering parameters according to the high-pass filtering network layer and the second processed data;

[0078] Based on the low-pass filtering network layer and the high-pass filtering network layer, a complementary filtering architecture is configured through the low-pass filtering parameters and the high-pass filtering parameters.

[0079] Furthermore, the system is also used to perform the following method:

[0080] Introducing the frequent start-stop feature;

[0081] Based on the frequent start-stop feature, outputting a first stability coefficient in a first driving state associated with longitudinal acceleration;

[0082] The low-pass filter parameters are adjusted once using the first stability coefficient.

[0083] Furthermore, the system is also used to perform the following method:

[0084] Introducing steering-sensitive features;

[0085] outputting a second stability coefficient in a second driving state associated with lateral acceleration based on the steering sensitivity characteristic;

[0086] The low-pass filter parameters are adjusted secondary using the second stability coefficient.

[0087] Furthermore, the system is also used to perform the following method:

[0088] Adding a random noise term corresponding to an unpaved road surface to the vehicle dynamics model;

[0089] Based on the random noise term, a vibration compensation mechanism is set up, which is used to correct the vehicle posture deviation caused by the unpaved road.

[0090] Furthermore, the system is also used to perform the following method:

[0091] Connect to ADAS unit;

[0092] When the rollover warning signal is triggered, the ADAS unit immediately responds with an emergency protection operation, which includes any one or more of an automatic braking command, a lane keeping command, and a vehicle stabilization command.

[0093] Example 3

[0094] like Figure 3 As shown, it is a schematic diagram of the structure of an exemplary electronic device of the present application. Figure 3In particular embodiments, bus architecture 300 is represented by bus 300, which can include any number of interconnecting buses and bridges needed to support various components of the system. Bus 300 can include a bus to connect various circuits of the system, including one or more processors, represented by processor 302, and memory, represented by memory 304. Bus 300 can also connect various other circuits, such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art, and therefore, will not be described any further. Bus interface 305 provides an interface between bus 300 and receiver 301 and transmitter 303. Receiver 301 and transmitter 303 can be the same component, i.e., a transceiver, providing a means for communicating with various other apparatus over a transmission medium. Processor 302 is responsible for managing bus 300 and general processing, while memory 304 can be used for storing data used by processor 302 in executing operational processes.

[0095] The technical features of the above embodiments can be combined in any manner, and for the sake of brevity, not all possible combinations are described in the above description. However, it is to be understood that any combination of the technical features is possible and within the scope of the present disclosure.

[0096] The above description of disclosed embodiments provides enabling concepts for those skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and generic principles defined herein can be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A vehicle detection dynamic rollover warning method, characterized in that: The method comprises: Based on the target vehicle, receiving multi-source monitoring data corresponding to the real-time monitoring unit, wherein the real-time monitoring unit is integrated with an accelerometer and a gyroscope; Setting a complementary filtering architecture based on the multi-source monitoring data; Introducing vehicle posture information, using EKF principle, and establishing vehicle dynamics model; Connecting the complementary filtering architecture and the vehicle dynamics model to determine predicted vehicle posture information and calculate a set of dynamic parameters; Based on the dynamic parameter set, the system compares the dynamic parameter set with a preset rollover threshold value to determine whether a rollover warning signal is triggered; The method of setting up a complementary filtering architecture based on the multi-source monitoring data includes: Extracting first processed data through the multi-source monitoring data in combination with a low-pass filter network layer; Extracting second processed data through the multi-source monitoring data in combination with a high-pass filter network layer; Setting a complementary filtering architecture based on the low-pass filtering network layer and the first processed data, and the high-pass filtering network layer and the second processed data; Wherein, based on the low-pass filter network layer and the first processed data, and the high-pass filter network layer and the second processed data, a complementary filter architecture is set, including: Determining low-pass filtering parameters according to the low-pass filtering network layer and the first processed data; Determining high-pass filtering parameters according to the high-pass filtering network layer and the second processed data; Based on the low-pass filtering network layer and the high-pass filtering network layer, a complementary filtering architecture is configured by using the low-pass filtering parameters and the high-pass filtering parameters; Wherein, determining low-pass filter parameters according to the low-pass filter network layer and the first processed data further includes: Introducing the frequent start-stop feature; Based on the frequent start-stop feature, outputting a first stability coefficient in a first driving state associated with longitudinal acceleration; The low-pass filter parameters are adjusted once using the first stability coefficient.

2. The method according to claim 1, wherein The dynamic parameter set includes a roll angle element, a roll speed element, and a roll acceleration element; Setting a first risk item, a second risk item, and a third risk item respectively according to the roll angle element, the roll speed element, and the roll acceleration element in the dynamic parameter set; A risk assessment model is established based on the first risk item, the second risk item, and the third risk item. The risk assessment model assesses a rollover risk index of a target vehicle using a deviation value corresponding to a preset rollover threshold.

3. The method according to claim 1, wherein Determining low-pass filtering parameters according to the low-pass filtering network layer and the first processed data, the method further comprising: Introducing steering-sensitive features; outputting a second stability coefficient in a second driving state associated with lateral acceleration based on the steering sensitivity characteristic; The low-pass filter parameters are adjusted secondary using the second stability coefficient.

4. The method according to claim 1, wherein Connecting the complementary filtering architecture and the vehicle dynamics model to determine predicted vehicle posture information, the method includes: Adding a random noise term corresponding to an unpaved road surface to the vehicle dynamics model; Based on the random noise term, a vibration compensation mechanism is set up, which is used to correct the vehicle posture deviation caused by the unpaved road.

5. The method according to claim 1, wherein Determining whether a rollover warning signal is triggered, the method includes: Connect to ADAS unit; When the rollover warning signal is triggered, the ADAS unit immediately responds with an emergency protection operation, which includes any one or more of an automatic braking command, a lane keeping command, and a vehicle stabilization command.

6. A vehicle detection dynamic rollover warning system, characterized in that: A vehicle detection dynamic rollover warning method for implementing any one of claims 1 to 5, comprising: A data receiving module is used to receive multi-source monitoring data corresponding to a real-time monitoring unit based on a target vehicle, wherein the real-time monitoring unit is integrated with an accelerometer and a gyroscope; An architecture setting module, configured to set a complementary filtering architecture based on the multi-source monitoring data; Model building module, used to introduce vehicle posture information, use EKF principle, and build vehicle dynamics model; a posture determination module, configured to connect the complementary filtering architecture and the vehicle dynamics model, determine and predict vehicle posture information, and calculate a set of dynamic parameters; a comparison module, configured to compare the dynamic parameter set with a preset rollover threshold value to determine whether a rollover warning signal is triggered; The method of setting up a complementary filtering architecture based on the multi-source monitoring data includes: Extracting first processed data through the multi-source monitoring data in combination with a low-pass filter network layer; Extracting second processed data through the multi-source monitoring data in combination with a high-pass filter network layer; Setting a complementary filtering architecture based on the low-pass filtering network layer and the first processed data, and the high-pass filtering network layer and the second processed data; Wherein, based on the low-pass filter network layer and the first processed data, and the high-pass filter network layer and the second processed data, a complementary filter architecture is set, including: Determining low-pass filtering parameters according to the low-pass filtering network layer and the first processed data; Determining high-pass filtering parameters according to the high-pass filtering network layer and the second processed data; Based on the low-pass filtering network layer and the high-pass filtering network layer, a complementary filtering architecture is configured by using the low-pass filtering parameters and the high-pass filtering parameters; Wherein, according to the low-pass filter network layer and the first processed data, a low-pass filter parameter is determined, and the system is further used to perform the following method: Introducing the frequent start-stop feature; Based on the frequent start-stop feature, outputting a first stability coefficient in a first driving state associated with longitudinal acceleration; The low-pass filter parameters are adjusted once using the first stability coefficient.

7. An electronic device, characterized in that: The electronic device comprises: a memory for storing executable instructions; The processor is configured to implement a vehicle detection dynamic rollover warning method according to any one of claims 1 to 5 when executing the executable instructions stored in the memory.

Images