Roll-over risk determination method and vehicle

Abstract

The present disclosure provides a roll-out-of-control risk determination method and a vehicle, and relates to the technical field of data processing. The method comprises: acquiring motion posture information and vehicle speed information of a target vehicle, environmental weather information and curve curvature information of a location where the target vehicle is located; in a case where it is determined based on the environmental weather information, the vehicle speed information and the motion posture information that the target vehicle is in a rainy environment and is in a curve passing state, determining a roll-out-of-control risk level based on at least one of a lateral adhesion margin parameter, a speed-curvature matching parameter and a roll-impact rate parameter, so as to identify a roll-out-of-control risk based on the roll-out-of-control risk level. The present disclosure can solve the problem of inaccurate identification of the roll-out-of-control risk possibly caused by the curve compression behavior of the target vehicle in the rainy environment.

Description

Technical Field

[0001] This disclosure relates to the field of data processing technology, specifically to a method for determining rollover risk and a vehicle. Background Technology

[0002] In the field of safety testing for two-wheeled vehicles (such as motorcycles), accurately identifying the risk of loss of control when cornering on wet, slippery roads is of significant practical importance for preventing traffic accidents such as skidding and crashes. Especially with the development of vehicle-to-everything (V2X) technology, how to accurately assess the risks of these specific dangerous driving scenarios using data collected by onboard sensors has become a key technological objective. Summary of the Invention

[0003] In view of this, the present disclosure aims to provide a method and vehicle for determining roll loss of control risk, so as to solve the problem of inaccurate identification of roll loss of control risk that may be caused by cornering behavior in rainy weather in the prior art.

[0004] In a first aspect, one embodiment of this disclosure provides a method for determining roll loss of control risk, comprising: acquiring motion posture information and vehicle speed information of a target vehicle, environmental weather information and curve curvature information of the target vehicle's location; and, when it is determined based on the environmental weather information, vehicle speed information, and motion posture information that the target vehicle is in a rainy environment and is in a curve, determining a roll loss of control risk level based on at least one of a lateral adhesion margin parameter, a velocity curvature matching parameter, and a roll impact rate parameter, thereby identifying roll loss of control risk based on the roll loss of control risk level, wherein the lateral adhesion margin parameter is determined based on the environmental weather information and motion posture information, the velocity curvature matching parameter is determined based on the vehicle speed information, environmental weather information, and curve curvature information, and the roll impact rate parameter is determined based on the roll angular velocity and roll acceleration included in the motion posture information.

[0005] In conjunction with the first aspect, in some implementations of the first aspect, the roll impact rate parameter is determined based on the roll angular velocity and roll acceleration, including: within a preset sliding time window, sorting the absolute values ​​of the roll angular velocity in ascending order and taking the first preset percentile value as the representative value of the roll rate; within a preset sliding time window, sorting the absolute values ​​of the roll acceleration in ascending order and taking the second preset percentile value as the representative value of the roll acceleration; and determining the roll impact rate parameter based on the representative values ​​of the roll rate and roll acceleration.

[0006] In conjunction with the first aspect, in certain implementations of the first aspect, the roll impact rate parameter is determined based on the representative values ​​of roll rate and roll acceleration, including: determining a first normalized value based on the representative value of roll rate and a preset normalized reference value of roll rate; determining a second normalized value based on the representative value of roll acceleration and a preset normalized reference value of roll acceleration; weighting and fusing the first normalized value and the second normalized value to obtain a target normalized value; and determining the roll impact rate parameter based on the target normalized value.

[0007] In conjunction with the first aspect, in certain implementations of the first aspect, the roll loss-of-control risk level is determined based on at least one of the lateral adhesion margin parameter, velocity curvature matching parameter, and roll impact rate parameter, including: determining a lateral adhesion sub-score based on the lateral adhesion margin parameter; determining a velocity curvature matching sub-score based on the velocity curvature matching parameter; determining a roll impact rate sub-score based on the roll impact rate parameter; weightedly fusing the lateral adhesion sub-score, velocity curvature matching sub-score, and roll impact rate sub-score to determine a roll loss-of-control risk score; and determining the roll loss-of-control risk level based on the roll loss-of-control risk score.

[0008] In conjunction with the first aspect, in some implementations of the first aspect, the motion attitude information of the target vehicle includes lateral acceleration, wherein the lateral adhesion margin parameter is determined based on environmental weather information and motion attitude information, including: determining the road surface adhesion coefficient based on environmental weather information; sorting the lateral accelerations within a preset sliding time window and taking the maximum value as the target lateral acceleration; determining the lateral adhesion limit value based on the road surface adhesion coefficient; and determining the lateral adhesion margin parameter based on the lateral adhesion limit value and the target lateral acceleration.

[0009] In conjunction with the first aspect, in some implementations of the first aspect, the lateral adhesion margin parameter is determined based on the lateral adhesion limit value and the target lateral acceleration, including: determining the lateral adhesion utilization ratio based on the lateral adhesion limit value and the target lateral acceleration; determining the remaining lateral adhesion ratio based on the lateral adhesion utilization ratio; and performing a boundary-truncation linear normalization process on the remaining lateral adhesion ratio to obtain the lateral adhesion margin parameter.

[0010] In conjunction with the first aspect, in some implementations of the first aspect, the speed curvature matching parameters are determined based on vehicle speed information, environmental weather information, and curve curvature information, including: determining the road surface adhesion coefficient based on environmental weather information; determining a safe speed limit based on the road surface adhesion coefficient, a preset safe roll angle threshold, and curve curvature when the curve curvature indicated by the curve curvature information is greater than a preset curvature threshold; sorting the vehicle speed information within a preset sliding time window and taking the speed value corresponding to the maximum vehicle speed information as the target vehicle speed; determining the speed over-limit ratio based on the target vehicle speed and the safe speed limit; and performing linear normalization processing based on boundary truncation on the speed over-limit ratio to obtain the speed curvature matching parameters.

[0011] In conjunction with the first aspect, in some implementations of the first aspect, the motion attitude information includes the heading angle and the roll angle. Determining that the target vehicle is in a cornering state includes: acquiring the target vehicle's system power mode and engine speed; determining the target vehicle's riding state based on the system power mode, vehicle speed information, and engine speed; if the target vehicle's riding state indicates that the target vehicle is in a true riding state, determining the radius of curvature based on the vehicle speed information and heading angle, and determining the roll angle index based on the roll angle; and determining that the target vehicle is in a cornering state if the radius of curvature is less than a preset radius of curvature threshold, the vehicle speed information is greater than a preset lower vehicle speed threshold, and the roll angle index is greater than a preset index threshold, and this condition persists for a first preset duration.

[0012] In conjunction with the first aspect, in some implementations of the first aspect, obtaining environmental weather information of the target vehicle's location includes: obtaining the target vehicle's GPS location data; retrieving the rainfall intensity level from a network weather service interface based on the GPS location data, and using the rainfall intensity level as environmental weather information.

[0013] Secondly, one embodiment of this disclosure provides a roll loss of control risk determination device, comprising: an acquisition module, configured to acquire motion posture information and vehicle speed information of a target vehicle, environmental weather information and curve curvature information of the target vehicle's location; and a determination module, configured to, when determining that the target vehicle is in a rainy environment and is in a curve based on the environmental weather information, vehicle speed information, and motion posture information, determine a roll loss of control risk level based on at least one of a lateral adhesion margin parameter, a velocity curvature matching parameter, and a roll impact rate parameter, so as to identify roll loss of control risk based on the roll loss of control risk level, wherein the lateral adhesion margin parameter is determined based on the environmental weather information and motion posture information, the velocity curvature matching parameter is determined based on the vehicle speed information, environmental weather information, and curve curvature information, and the roll impact rate parameter is determined based on the roll angular velocity and roll acceleration included in the motion posture information.

[0014] Thirdly, one embodiment of this disclosure provides a computer-readable storage medium storing a computer program for performing the roll-off risk determination method described in the first aspect.

[0015] Fourthly, one embodiment of this disclosure provides a vehicle comprising: a processor; a memory for storing processor-executable instructions; the processor being configured to execute the roll-off risk determination method described in the first aspect.

[0016] Fifthly, one embodiment of this disclosure provides a computer program product including instructions that, when executed on a vehicle, cause the vehicle to implement the roll loss risk determination method described in the first aspect.

[0017] In the embodiments of this disclosure, by acquiring the target vehicle's motion posture information, vehicle speed information, environmental weather information and curve curvature information, and based on this, the subsequent roll loss of control risk level determination process is triggered only when the target vehicle is simultaneously in a rainy environment and in a curve, thus avoiding invalid operations in irrelevant scenarios (such as driving on a straight road in sunny weather) and reducing computing power consumption. Furthermore, three risk parameters with clear physical meanings are defined. The lateral adhesion margin parameter, determined based on environmental weather information and motion posture information, characterizes the remaining adhesion between the tire and the road surface. The speed-curvature matching parameter, determined based on vehicle speed information, environmental weather information, and curve curvature information, reflects the degree to which the actual vehicle speed exceeds the safe cornering speed. The roll impact rate parameter, determined based on roll angular velocity and roll angular acceleration, quantifies the dynamic severity of the cornering action. The roll loss of control risk level is determined by at least one of these three risk parameters, and roll loss of control risk identification is performed based on the roll loss of control risk level. This allows for both flexible selection of a single risk parameter for rapid risk assessment based on sensor configuration and operating conditions, and more accurate and comprehensive risk judgment through the combination of multiple risk parameters. It effectively solves the problem in related technologies of difficulty in scenario-adaptive and multi-dimensional quantitative assessment of roll loss of control risk in rainy curves based on onboard sensor data. It achieves efficient, flexible, and accurate comprehensive judgment of roll loss of control risk, which helps to provide timely warning of skidding accidents and improve driving safety. Attached Figure Description

[0018] The above and other objects, features, and advantages of this disclosure will become more apparent from the more detailed description of the embodiments thereof in conjunction with the accompanying drawings. The drawings are provided to further illustrate the embodiments of this disclosure and form part of the specification. They are used together with the embodiments of this disclosure to explain the disclosure and do not constitute a limitation thereof. In the drawings, the same reference numerals generally represent the same components or steps.

[0019] Figure 1 The diagram shown is a flowchart illustrating a method for determining rollover risk according to an embodiment of this disclosure.

[0020] Figure 2 The diagram shown is a structural schematic of a rollover risk determination device provided in an embodiment of this disclosure.

[0021] Figure 3 The diagram shown is a structural schematic of a vehicle provided in an embodiment of this disclosure. Detailed Implementation

[0022] 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 some embodiments of this disclosure, and not all embodiments. Based on the embodiments of this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0023] Furthermore, to better illustrate this disclosure, numerous specific details are set forth in the following detailed description. Those skilled in the art will understand that this disclosure can be practiced without certain specific details. In some instances, methods and means well-known to those skilled in the art have not been described in detail in order to highlight the main points of this disclosure.

[0024] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0025] Furthermore, the terms “first,” “second,” “third,” and “fourth” are used only for distinguishing descriptions and should not be interpreted as indicating or implying relative importance.

[0026] To assess and warn of potential roll and loss of control risks associated with cornering in rainy weather, common technologies rely on driver experience or offline analysis of driving videos. Some vehicles with basic internet connectivity attempt to directly utilize lateral acceleration or roll angle signals from the onboard inertial measurement unit, triggering alarms by setting fixed thresholds. This approach compares real-time acceleration with preset limits to make decisions, aiming to achieve basic cornering warning functionality with low computational overhead and simple logic.

[0027] However, this fixed threshold-based solution performs poorly when vehicles are in wet, slippery cornering conditions. This is mainly because, in order to maintain low complexity and hardware cost, the judgment logic in related technologies relies on a fixed threshold value, making it difficult to dynamically adapt to road surface friction characteristics, corner geometry, and actual vehicle speed. This leads to false alarms or missed alarms in rainy conditions where accurate detection of roll and loss of control risk is crucial. Specifically, the road surface adhesion coefficient decreases significantly in rainy conditions; the same lateral acceleration approaches the tire adhesion limit, but the fixed threshold cannot reflect this change in physical limits. Furthermore, the relationship between corner radius and vehicle speed is not taken into account, resulting in a lack of awareness of the danger of excessive speed and forced cornering.

[0028] To address the aforementioned technical problems, this disclosure provides a method and vehicle for determining roll loss of control risk. The method includes: acquiring motion attitude information and vehicle speed information of a target vehicle, environmental weather information and curve curvature information of the target vehicle's location; and, if the target vehicle is determined to be in a rainy environment and cornering based on the environmental weather information, vehicle speed information, and motion attitude information, determining a roll loss of control risk level based on at least one of a lateral adhesion margin parameter, a velocity curvature matching parameter, and a roll impact rate parameter, thereby identifying roll loss of control risk based on the roll loss of control risk level. The lateral adhesion margin parameter is determined based on the environmental weather information and motion attitude information; the velocity curvature matching parameter is determined based on the vehicle speed information, environmental weather information, and curve curvature information; and the roll impact rate parameter is determined based on the roll angular velocity and roll acceleration included in the motion attitude information. In this way, by acquiring the target vehicle's motion attitude information, speed information, environmental weather information, and curve curvature information, and based on this, the subsequent process of determining the risk level of roll loss of control is only triggered when the target vehicle is simultaneously in a rainy environment and in a curve. This avoids invalid operations in irrelevant scenarios (such as driving on a straight road in sunny weather) and reduces computing power consumption. Furthermore, three risk parameters with clear physical meanings are defined. The lateral adhesion margin parameter, determined based on environmental weather information and motion posture information, characterizes the remaining adhesion between the tire and the road surface. The speed-curvature matching parameter, determined based on vehicle speed information, environmental weather information, and curve curvature information, reflects the degree to which the actual vehicle speed exceeds the safe cornering speed. The roll impact rate parameter, determined based on roll angular velocity and roll angular acceleration, quantifies the dynamic severity of the cornering action. The roll loss of control risk level is determined by at least one of these three risk parameters, and roll loss of control risk identification is performed based on the roll loss of control risk level. This allows for both flexible selection of a single risk parameter for rapid risk assessment based on sensor configuration and operating conditions, and more accurate and comprehensive risk judgment through the combination of multiple risk parameters. It effectively solves the problem in related technologies of difficulty in scenario-adaptive and multi-dimensional quantitative assessment of roll loss of control risk in rainy curves based on onboard sensor data. It achieves efficient, flexible, and accurate comprehensive judgment of roll loss of control risk, which helps to provide timely warning of skidding accidents and improve driving safety.

[0029] The following is based on Figure 1 The method for determining the rollover and loss of control risk provided in this disclosure is described in detail.

[0030] Figure 1 The diagram shown is a flowchart illustrating a method for determining rollover and loss of control risk according to an embodiment of this disclosure. This method can be applied to vehicles. Figure 1 As shown in the embodiments of this disclosure, the method for determining the risk of rollover loss of control may include the following steps.

[0031] Step S110: Obtain the target vehicle's motion attitude information, vehicle speed information, environmental weather information and curve curvature information of the target vehicle's location.

[0032] For example, the target vehicle can represent a two-wheeled vehicle or similar motor vehicle capable of collecting its own motion attitude information, speed information, and location information during driving and communicating with vehicle-to-everything (V2X) services. For example, it may include, but is not limited to, motorcycles, electric motorcycles, three-wheeled motor vehicles with a tilt driving mode, or combinations thereof. Motion attitude information can represent data reflecting the target vehicle's current motion state and body posture. For example, it may include, but is not limited to, lateral acceleration, roll rate, roll acceleration, heading angle, and roll angle obtained using an inertial measurement unit, or equivalent information obtained using other attitude sensors, or combinations thereof.

[0033] In some implementations, motion attitude information can be acquired, for example, through an onboard inertial measurement unit (IMU), including but not limited to data such as three-axis acceleration and three-axis angular velocity. In other implementations, motion attitude information can also be acquired individually or in combination through onboard gyroscopes, accelerometers, and other inertial sensors. In some implementations, acquiring motion attitude information through onboard inertial sensors allows for real-time perception of dynamic attitude changes in the target vehicle.

[0034] For example, environmental weather information can characterize real-time information that reflects the location of the target vehicle, such as rainfall or slippery road conditions.

[0035] For example, curve curvature information can characterize information reflecting the degree of curvature of the road where the target vehicle is currently located. This may include, but is not limited to, curve curvature values ​​preset from high-precision maps, road curvature extracted based on historical driving trajectories, or the reciprocal of the radius of curvature estimated in real time by onboard sensors, or combinations thereof. In some implementations, a greater curve curvature means a sharper curve, and a greater lateral force is required for the target vehicle to navigate the curve.

[0036] For example, vehicle speed information can characterize data that reflects the current speed of the target vehicle. This may include, but is not limited to, ground speed output from wheel speed sensors, ground speed from Global Positioning System (GPS) positioning data, speed calculated from engine speed and gear ratio, or combinations thereof. In some implementations, vehicle speed information can be obtained from ground speed measured by GPS or from vehicle speed measured by wheel speed sensors.

[0037] Step S120: If it is determined that the target vehicle is in a rainy environment and is in a curve based on environmental weather information, vehicle speed information and motion attitude information, the roll loss control risk level is determined based on at least one of the lateral adhesion margin parameter, speed curvature matching parameter and roll impact rate parameter, so as to identify the roll loss control risk based on the roll loss control risk level.

[0038] For example, the lateral adhesion margin parameter is determined based on environmental weather information and motion attitude information, the speed curvature matching parameter is determined based on vehicle speed information, environmental weather information and cornering curvature information, and the roll impact rate parameter is determined based on the roll angular velocity and roll acceleration included in the motion attitude information.

[0039] In some implementations, the roll loss risk level determination process only proceeds after determining that the target vehicle is in a rainy environment and cornering, based on environmental weather information, vehicle speed information, and motion attitude information. If the target vehicle does not meet the above prerequisites (e.g., in a sunny environment or driving on a straight road), the roll loss risk level determination process is skipped, and the default state is output directly. By determining these prerequisites, the roll loss risk level determination process is limited to valid scenarios of rain and cornering, avoiding invalid operations in irrelevant scenarios (such as driving on a straight road in a sunny environment), thus reducing computational power consumption.

[0040] For example, the lateral adhesion margin parameter can be a quantitative indicator characterizing the remaining lateral adhesion capability of a target vehicle's tires in a curve. In some implementations, the lateral adhesion margin parameter can be determined based on environmental weather information and motion attitude information. The environmental weather information is used to determine the adhesion capability under current road conditions, and the motion attitude information is used to determine the actual lateral force level experienced by the target vehicle. The lateral adhesion margin parameter quantifies the roll depth of the target vehicle under current road conditions from the perspective of adhesion. A smaller lateral adhesion margin parameter indicates that the actual lateral force is closer to the road adhesion limit, and the risk of sideslip and loss of control is higher, thus providing a quantitative basis for determining the risk level of sideslip and loss of control from the adhesion dimension.

[0041] For example, the speed curvature matching parameter can characterize the degree of matching between the target vehicle's current actual speed and the theoretical speed limit for safely navigating a curve in rainy weather. In some implementations, the speed curvature matching parameter is determined based on vehicle speed information, ambient weather information, and curve curvature information. When the target vehicle's actual speed exceeds the safe cornering speed (theoretical speed limit) under the current curve and road surface conditions, the target vehicle is forced to increase its roll angle to balance centrifugal force. The speed curvature matching parameter quantifies this degree of speeding, providing a quantitative basis for determining the speed-curvature matching dimension of the roll loss risk level.

[0042] In some embodiments, the motion attitude information of the target vehicle may include roll rate and roll acceleration. For example, roll rate can characterize the roll rotation rate of the target vehicle about its longitudinal axis, and roll acceleration can characterize the rate of change of roll rate. Roll rate can be obtained, for example, by taking the absolute value of the angular velocity component along the longitudinal axis of the vehicle body from a gyroscope in an IMU, and roll acceleration can be obtained, for example, by taking the time derivative of roll rate. Roll rate reflects the speed of the target vehicle's roll action, and roll acceleration reflects the degree of abrupt change in roll action; together, they characterize the severity of the roll dynamics. Excessively fast roll rate or sudden changes in roll acceleration can easily exceed the adhesion limit of a wet road surface; therefore, incorporating these two parameters into the motion attitude information provides a data basis for calculating the roll impact rate parameter.

[0043] In some implementations, the roll impact rate parameter can be determined based on roll angular velocity and roll acceleration. The roll impact rate parameter characterizes the severity of the target vehicle's roll within a preset time window. It quantifies the intensity of the cornering action from the perspective of dynamic roll changes; aggressive cornering actions (i.e., large roll angular velocity and drastic changes in roll acceleration) are more likely to cause the tires to exceed the road surface adhesion limit and skid on wet surfaces.

[0044] In some implementations, the roll impact rate parameter can quantify the risk of roll loss of control from the perspective of the severity of the dynamic changes in roll. For example, the roll loss of control risk level can characterize a classification result reflecting the likelihood of the target vehicle experiencing sideslip loss of control, based on a comprehensive assessment of one or more risk parameters. This classification result can serve as the basis for identifying roll loss of control risks.

[0045] In some implementations, the roll loss-of-control risk level can be determined based on at least one of the following parameters: lateral adhesion margin parameter, speed curvature matching parameter, and roll impact rate parameter. That is, the roll loss-of-control risk can be assessed from at least one of multiple dimensions, including adhesion margin, speed curvature matching, and roll impact. For example, the roll loss-of-control risk level can be determined based on the lateral adhesion margin parameter, speed curvature matching parameter, and roll impact rate parameter. Compared to assessment methods that rely on only a single dimension, this approach more comprehensively and accurately reflects the actual risk level of the target vehicle in rainy curve scenarios, improving the comprehensiveness and robustness of risk identification. The output classification results facilitate differentiated intervention measures by drivers or fleet management systems based on different risk levels.

[0046] By acquiring the target vehicle's motion attitude information, speed information, environmental weather information, and curve curvature information, and based on this, the determination of the subsequent roll loss risk level is only triggered when the target vehicle is simultaneously in a rainy environment and in a curve, thus avoiding invalid operations in irrelevant scenarios (such as driving on a straight road in sunny weather) and reducing computing power consumption. Furthermore, three risk parameters with clear physical meanings are defined. The lateral adhesion margin parameter, determined based on environmental weather information and motion posture information, characterizes the remaining adhesion between the tire and the road surface. The speed-curvature matching parameter, determined based on vehicle speed information, environmental weather information, and curve curvature information, reflects the degree to which the actual vehicle speed exceeds the safe cornering speed. The roll impact rate parameter, determined based on roll angular velocity and roll angular acceleration, quantifies the dynamic severity of the cornering action. The roll loss of control risk level is determined by at least one of these three risk parameters, and roll loss of control risk identification is performed based on the roll loss of control risk level. This allows for both flexible selection of a single risk parameter for rapid risk assessment based on sensor configuration and operating conditions, and more accurate and comprehensive risk judgment through the combination of multiple risk parameters. It effectively solves the problem in related technologies of difficulty in scenario-adaptive and multi-dimensional quantitative assessment of roll loss of control risk in rainy curves based on onboard sensor data. It achieves efficient, flexible, and accurate comprehensive judgment of roll loss of control risk, which helps to provide timely warning of skidding accidents and improve driving safety.

[0047] In some embodiments, obtaining environmental weather information of the target vehicle's location may specifically include: obtaining the target vehicle's GPS location data; retrieving the rainfall intensity level from a network weather service interface based on the GPS location data, and using the rainfall intensity level as environmental weather information.

[0048] In some implementations, GPS location data of the target vehicle can be obtained. This GPS location data may include, for example, the vehicle's current longitude and latitude. GPS location data provides precise geographical location information for the target vehicle and is a necessary input for subsequently retrieving corresponding location weather data from a network weather service interface. Based on the GPS location data, rainfall intensity levels can be retrieved from the network weather service interface and used as environmental weather information. This network weather service interface may, for example, be a publicly available weather application programming interface (API) service. Rainfall intensity levels may include multiple levels, such as no rain, light rain, moderate rain, heavy rain, etc. Using rainfall intensity levels as environmental weather information provides a discrete representation of weather conditions.

[0049] In some embodiments, environmental weather information is used as the basis for determining whether a target vehicle is in a rainy environment. If the rainfall intensity level indicated by the environmental weather information is greater than a preset level threshold, it is determined that the target vehicle is in a rainy environment.

[0050] By utilizing the vehicle's existing GPS location data, the rainfall intensity level can be retrieved in real time from the network weather service interface. This eliminates the need for additional hardware such as rain sensors on the vehicle, making full use of the current mature vehicle-to-everything (V2X) and meteorological service infrastructure. While reducing hardware costs and deployment complexity, it can dynamically update weather information according to the vehicle's location, ensuring the real-time and spatial accuracy of rainy weather assessment.

[0051] In some embodiments, motion posture information may include heading angle and roll angle. Accordingly, determining that the target vehicle is in a cornering state may specifically include: acquiring the target vehicle's system power mode and engine speed; determining the target vehicle's riding state based on the system power mode, vehicle speed information, and engine speed; if the target vehicle's riding state indicates that the target vehicle is in a true riding state, determining the radius of curvature based on the vehicle speed information and heading angle, and determining the roll angle index based on the roll angle; and determining that the target vehicle is in a cornering state if the radius of curvature is less than a preset radius of curvature threshold, the vehicle speed information is greater than a preset lower vehicle speed threshold, the roll angle index is greater than a preset index threshold, and this condition persists for a first preset duration.

[0052] For example, the heading angle can be obtained via GPS to characterize the target vehicle's direction of travel; the roll angle can be obtained via IMU to characterize the target vehicle's tilt around its longitudinal axis. The system power mode can characterize the target vehicle's electrical system operating status, and the engine speed can characterize the engine's operating status. The system power mode and engine speed are auxiliary criteria for determining whether the target vehicle is in a true riding state, and can eliminate misjudgments when the vehicle is off, stationary, or in an abnormal driving state.

[0053] In some implementations, the riding status of a target vehicle can be determined based on system power mode, vehicle speed information, and engine speed. This riding status indicates whether the target vehicle is in a real riding state. By comprehensively considering three conditions—system power mode, vehicle speed information, and engine speed—to determine the riding status, compared to relying on only a single condition (such as looking only at vehicle speed), it can more accurately distinguish between real riding and non-riding scenarios (such as the vehicle being towed or the engine idling).

[0054] For example, the radius of curvature can characterize the curvature of the target vehicle's current trajectory. A smaller radius of curvature indicates a sharper curve, and it is one of the key indicators for determining whether the target vehicle is cornering. The yaw rate can be obtained through an IMU. In some implementations, when the target vehicle is in a real riding state, the radius of curvature can be determined based on the vehicle speed information and the yaw rate, satisfying the following formula (1).

[0055] (1); in, Indicates the radius of curvature. This indicates the speed value corresponding to the vehicle speed information. This indicates the angular velocity of the oscillation.

[0056] For example, the roll angle index can characterize the degree of matching between the roll angle of the target vehicle and the theoretical roll angle calculated by vehicle kinematics. In some implementations, determining the roll angle index based on vehicle speed information, yaw rate, and roll angle may specifically include: determining the theoretical roll angle based on vehicle speed information and yaw rate, satisfying the following formula (2). Determining the roll angle deviation based on the theoretical roll angle and roll angle, satisfying the following formula (3). When the absolute value of the roll angle is less than or equal to a first preset angle threshold (e.g., 5°), the roll angle index is determined to be a first preset value (e.g., 0); when the absolute value of the roll angle is greater than the first preset angle threshold, the roll angle index is determined based on the roll angle deviation, satisfying the following formula (4).

[0057] (2); (3); (4); in, Indicates the theoretical roll angle. This represents the acceleration due to gravity (e.g., 9.8 m / s²). Indicates the roll angle. Indicates the roll angle deviation. This indicates the roll angle index.

[0058] In some implementations, a preset curvature radius threshold is used to distinguish between driving on a curve and driving on a near-straight road. A curvature radius smaller than this preset threshold indicates that the target vehicle is navigating a curve with significant curvature. A preset lower speed limit threshold is used to exclude low-speed scenarios (such as parking, idling, etc.). Even if a certain curvature radius and roll angle exist in low-speed scenarios, they usually do not pose a risk of roll loss of control. By setting a lower speed limit threshold to filter low-speed scenarios, false alarms for low-risk conditions are avoided. A first preset duration (e.g., 0.5 seconds) is used to ensure the continuity of the curve-taking state. The curve-taking state is only confirmed when all conditions are met for a continuous period of time, effectively suppressing transient false triggers caused by sensor noise or brief road surface disturbances.

[0059] By simultaneously requiring the radius of curvature, vehicle speed information, and roll angle index to all meet their respective thresholds, and requiring this state to continue for a first preset duration, it can effectively eliminate misjudgments caused by non-cornering scenarios such as brief lane changes, road bumps, or transient sensor fluctuations. It ensures that subsequent roll loss of control risk index calculations are only triggered when the target vehicle is indeed in a continuous cornering state, thus improving the accuracy and robustness of cornering state determination.

[0060] By comprehensively judging the system's power mode, engine speed, and vehicle speed information, the true riding status of the vehicle is confirmed, effectively eliminating interference from non-riding scenarios such as the vehicle being stationary, turned off, or being towed. Based on this, the radius of curvature is calculated using the heading angle and vehicle speed, requiring that the radius of curvature, vehicle speed, and roll angle exponent all meet their respective thresholds, and that this state be maintained for a first preset duration. This effectively suppresses false triggers caused by instantaneous sensor noise or brief lane changes by the driver or road bumps, improving the accuracy and reliability of curve state recognition.

[0061] In some embodiments, the motion attitude information of the target vehicle may include lateral acceleration. Accordingly, the lateral adhesion margin parameter is determined based on environmental weather information and motion attitude information. Specifically, it may include: determining the road surface adhesion coefficient based on environmental weather information; sorting the lateral accelerations within a preset sliding time window and taking the maximum value as the target lateral acceleration; determining the lateral adhesion limit value based on the road surface adhesion coefficient; and determining the lateral adhesion margin parameter based on the lateral adhesion limit value and the target lateral acceleration.

[0062] For example, lateral acceleration can be the acceleration component along the lateral direction in the vehicle coordinate system, which can be acquired by an IMU. Lateral acceleration directly reflects the level of lateral force experienced by the target vehicle when driving in a curve and is a key physical quantity for measuring the roll depth.

[0063] In some implementations, the road surface adhesion coefficient can be determined based on environmental weather information. For example, the road surface adhesion coefficient can be determined based on the rainfall intensity level indicated by the environmental weather information; the greater the rainfall intensity, the lower the road surface adhesion coefficient. The decrease in road surface adhesion coefficient with increasing rainfall intensity reflects the weakening effect of wet road surface slippage on tire adhesion in rainy weather.

[0064] In some implementations, the lateral accelerations are sorted within a preset sliding time window, and the maximum value is taken as the target lateral acceleration. Before sorting, the lateral accelerations can be low-pass filtered to suppress high-frequency noise. Taking the maximum lateral acceleration within the preset sliding time window as the target lateral acceleration captures the maximum lateral force level experienced by the target vehicle within that window, reflecting the lateral force state at the most dangerous moment. Low-pass filtering helps remove high-frequency noise from the sensor, improving the reliability of the target lateral acceleration.

[0065] For example, the lateral adhesion limit value represents the upper limit of lateral acceleration corresponding to the maximum lateral adhesion force that the tire can provide under the current road surface conditions, and is a reference benchmark for judging whether the target vehicle is approaching the critical state of sideslip. In some implementations, the lateral adhesion limit value can be determined based on the road surface adhesion coefficient. The lateral adhesion limit value can be the product of the road surface adhesion coefficient and the gravitational acceleration, that is, it satisfies the following formula (5). The road surface adhesion coefficient can be determined based on the rainfall intensity level, satisfying the following formula (6).

[0066] (5); (6); in, Indicates the lateral adhesion limit value. Indicates the road surface adhesion coefficient. This indicates the rainfall intensity level.

[0067] In some embodiments, determining the lateral adhesion margin parameter based on the lateral adhesion limit and the target lateral acceleration may specifically include: determining the lateral adhesion utilization ratio based on the lateral adhesion limit and the target lateral acceleration; determining the remaining lateral adhesion ratio based on the lateral adhesion utilization ratio; and performing a boundary-truncation linear normalization process on the remaining lateral adhesion ratio to obtain the lateral adhesion margin parameter.

[0068] In some implementations, the lateral adhesion utilization ratio is determined based on the lateral adhesion limit and the target lateral acceleration, satisfying the following formula (7). The remaining lateral adhesion ratio is determined based on the lateral adhesion utilization ratio, satisfying the following formula (8). The remaining lateral adhesion ratio is then subjected to boundary-truncation linear normalization to obtain the lateral adhesion margin parameter, satisfying the following formula (9).

[0069] (7); (8); (9); in, This indicates the ratio of lateral adhesion utilization. Indicates the target's lateral acceleration. This indicates the remaining ratio of lateral adhesion. This represents the lateral adhesion margin parameter.

[0070] The lateral adhesion margin parameter characterizes the magnitude of the remaining adhesion margin. When the target lateral acceleration approaches the lateral adhesion limit, the lateral adhesion margin parameter approaches 0, indicating an extremely high risk of sideslip.

[0071] By dynamically determining the road surface adhesion coefficient using environmental weather information, the lateral adhesion limit value can adaptively change with rainfall intensity, truly reflecting the current road surface friction characteristics. The maximum lateral acceleration within a preset sliding time window is used as the target lateral acceleration. A calculation logic based on the ratio of lateral adhesion utilization and the ratio of lateral adhesion residue is introduced, and the results are linearly normalized with boundary truncation. This allows the determination of the lateral adhesion margin parameter to adaptively reflect the real-time changes in road surface friction characteristics under different rainfall intensities, effectively filtering out the interference of sensor transient noise on peak sampling, and improving the adaptability and robustness of roll risk assessment to complex and variable road environments and dynamic driving conditions.

[0072] The determination method for the lateral adhesion margin parameter has been explained above. The determination method for the velocity curvature matching parameter will be further explained below.

[0073] In some embodiments, the speed curvature matching parameters are determined based on vehicle speed information, ambient weather information, and curve curvature information. Specifically, this may include: determining the road surface adhesion coefficient based on ambient weather information; determining a safe speed limit based on the road surface adhesion coefficient, a preset safe roll angle threshold, and curve curvature when the curve curvature indicated by the curve curvature information is greater than a preset curvature threshold; sorting the vehicle speed information within a preset sliding time window and taking the speed value corresponding to the maximum vehicle speed information as the target vehicle speed; determining the speed over-limit ratio based on the target vehicle speed and the safe speed limit; and performing linear normalization processing based on boundary truncation on the speed over-limit ratio to obtain the speed curvature matching parameters.

[0074] In some implementations, the road surface adhesion coefficient is determined based on environmental weather information. The method for determining this road surface adhesion coefficient is the same as or similar to the method used to determine the road surface adhesion coefficient in the section on lateral adhesion margin parameters mentioned earlier.

[0075] In some implementations, the safe speed limit determination process is only executed when the curve curvature indicated by the curve curvature information is greater than a preset curvature threshold. This preset curvature threshold can be, for example, 0.001 m⁻¹. When the curve curvature is less than or equal to the preset curvature threshold, it indicates that the road is close to a straight section, and the speed curvature matching parameter can be set to 0. By setting the preset curvature threshold for filtering, abnormal safe speed limit calculation results (such as approaching infinity) are avoided due to extremely small curvature in near-straight road scenarios, thus improving the robustness of the speed curvature matching parameter.

[0076] For example, the preset safe roll angle threshold can be 25° (i.e., 0.436 rad). In some implementations, the safe speed limit can be determined based on the road surface adhesion coefficient, the preset safe roll angle threshold, and the curve curvature, satisfying the following formula (10).

[0077] (10); in, Indicates the safe speed limit. Indicates the road surface adhesion coefficient. Indicates the curvature of the curve. This indicates the preset safe roll angle threshold.

[0078] In some implementations, the safe speed limit takes into account both road surface adhesion conditions and safe roll angle constraints, taking the smaller of the corresponding upper speed limits to ensure that the target vehicle will not exceed either the road surface adhesion limit or the safe roll angle limit at that speed.

[0079] In some implementations, the maximum vehicle speed within a preset sliding time window is taken as the target vehicle speed. This can capture the speed level at the most dangerous moment within the preset sliding time window, ensuring that the speed curvature matching parameter reflects the degree of speeding under the most unfavorable operating conditions.

[0080] In some implementations, the speed over-limit ratio can reflect the degree to which the target vehicle speed exceeds the safe speed limit. When the ratio is greater than 1, it indicates that the target vehicle speed has exceeded the safe cornering speed. The larger the ratio, the more serious the speeding, and the greater the degree to which the target vehicle is forced to increase its roll angle. The speed over-limit ratio can be, for example, the ratio of the target vehicle speed to the safe speed limit, satisfying the following formula (11).

[0081] (11); in, Indicates the speed over-limit ratio. Indicates the target vehicle speed.

[0082] In some implementations, the velocity over-limit ratio is linearly normalized based on boundary truncation to obtain the velocity curvature matching parameters. This linear normalization process can, for example, map the velocity over-limit ratio to the target interval, satisfying the following formula (12).

[0083] (12); in, This represents the velocity curvature matching parameter.

[0084] The speed overspeed ratio is mapped to the target range using linear normalization, and then truncation is applied to ensure that the speed curvature matching parameter has a consistent numerical scale with other parameters. Setting the lower bound of the truncation boundary (e.g., 0.8) means that the parameter is 0 when the speed overspeed ratio is below this value (i.e., no overspeed risk is considered), and setting the upper bound of the truncation boundary (e.g., 1.5) means that the parameter saturates to 1 when the speed overspeed ratio reaches or exceeds this value (i.e., the overspeed risk is considered to be at its maximum). This boundary truncation design makes the speed curvature matching parameter linearly sensitive to overspeed changes in the intermediate region, while also being robust to extreme values.

[0085] The safe speed limit is determined by a dual constraint based on the road adhesion coefficient and the vehicle roll safety boundary (preset safe roll angle), ensuring that the judgment of speeding is based on the actual physical safety boundary. Furthermore, a preset sliding time window is used to capture the maximum vehicle speed, ensuring that the parameters reflect the most unfavorable operating conditions. Linear normalization processing based on boundary truncation maps the overspeed ratio to the target interval, making the speed curvature matching parameter values ​​smooth and bounded. This objectively quantifies the degree of danger caused by excessive vehicle speed forcing increased cornering, facilitating integration with other parameters.

[0086] For example, even if the adhesion margin and vehicle speed are well matched, an extremely aggressive cornering maneuver can still lead to loss of control. By introducing the roll impact rate parameter, this blind spot in risk assessment is compensated for, significantly improving the comprehensiveness and robustness of risk identification.

[0087] In some implementations, the roll impact rate parameter is determined based on the roll angular velocity and roll acceleration information included in the motion attitude information. Specifically, it may include: within a preset sliding time window, sorting the absolute values ​​of the roll angular velocity in ascending order and taking the first preset percentile value as the representative value of the roll rate; within a preset sliding time window, sorting the absolute values ​​of the roll acceleration in ascending order and taking the second preset percentile value as the representative value of the roll acceleration; and determining the roll impact rate parameter based on the representative values ​​of the roll rate and roll acceleration.

[0088] For example, the preset sliding time window can be a fixed-length time interval (e.g., 2 seconds) used to segment continuously acquired sensor data. The preset percentile value can represent the data value at a specified percentage position in an ascending sequence of a dataset.

[0089] In some implementations, the preset percentile can be adjusted within the range of 85% to 99% based on the actual sensor noise level and anti-interference requirements. For example, the first preset percentile value can be set to the 95th percentile, which is the roll rate value that will not be exceeded for 95% of the time within the preset sliding time window. Selecting the 95th percentile value instead of the maximum value as the representative value of the roll rate can eliminate the top 5% of extreme noise spikes while retaining the roll rate information close to the peak level within the preset sliding time window. This achieves a balance between noise immunity and sensitivity, ensuring that the representative value of the roll rate reflects the roll rate level for the vast majority of the time within the preset sliding time window, thus improving the noise immunity of the roll impact rate parameter.

[0090] In some implementations, similar to the method for determining the representative value of the roll rate, the absolute values ​​of the roll acceleration are sorted in ascending order within a preset sliding time window, and the second preset percentile value is taken as the representative value of the roll acceleration. Taking the percentile value of the absolute value of the roll acceleration can suppress noise spikes in the acceleration signal, making the representative value of the roll acceleration robustly reflect the overall level of dynamic changes in roll within the preset sliding time window. The second preset percentile value can be, for example, the 90th percentile value, that is, the roll acceleration value that will not be exceeded for 90% of the time within the preset sliding time window. The roll acceleration signal is usually noisier than the roll angular velocity signal, so selecting a relatively lower 90th percentile value (compared to the 95th percentile value of the representative value of the roll rate) can more effectively suppress noise interference in the acceleration signal while still retaining sufficient dynamic change information.

[0091] In some implementations, the roll impact rate parameter can be determined based on representative values ​​of roll rate and roll acceleration. By using both representative values ​​of roll rate and roll acceleration as inputs to determine the roll impact rate parameter, this parameter simultaneously reflects both the sustained rapidity and the sudden, violent nature of the roll maneuver, characterizing the danger of the dynamic roll impact from two complementary perspectives.

[0092] By sorting the absolute values ​​of roll angular velocity and roll acceleration within a preset sliding time window, and using the first and second preset percentile values ​​as representative values ​​instead of directly using the maximum value, this processing method effectively suppresses the interference of instantaneous spike noise generated by the sensor under vibration or impact conditions on the calculation of roll impact rate parameters. This allows the calculated representative values ​​of roll rate and acceleration to more robustly and realistically reflect the continuous bending impact level within the preset sliding time window, thereby improving the reliability of the roll impact rate parameters and the accuracy of rain risk assessment.

[0093] In some embodiments, the roll impact rate parameter is determined based on the representative values ​​of roll rate and roll acceleration. Specifically, this may include: determining a first normalized value based on the representative value of roll rate and a preset normalized reference value of roll rate; determining a second normalized value based on the representative value of roll acceleration and a preset normalized reference value of roll acceleration; performing a weighted fusion of the first normalized value and the second normalized value to obtain a target normalized value; and determining the roll impact rate parameter based on the target normalized value.

[0094] In some implementations, the preset normalized reference value for the roll rate can be, for example, 1.2 rad / s, representing the maximum roll rate reference level in a typical scenario. The first normalized value can be, for example, the ratio of the representative roll rate value to the preset normalized reference value for the roll rate. By normalizing the representative roll rate value by dividing it by the preset normalized reference value for the roll rate, the roll rate, which has physical dimensions, is converted into a dimensionless relative value.

[0095] In some implementations, the preset normalized reference value for roll acceleration can be, for example, 3.0 rad / s², representing the maximum roll acceleration reference level in a typical scenario. The second normalized value can be, for example, the ratio of the representative roll acceleration value to the preset normalized reference value. Similar to the first normalized value, the representative roll acceleration value is normalized to a dimensionless relative value, placing it on a comparable numerical scale with the first normalized value, facilitating subsequent weighted fusion.

[0096] In some implementations, the first normalized value and the second normalized value can be weighted and fused based on the following formula (13) to obtain the target normalized value. Based on the target normalized value, the roll impact rate parameter is determined, satisfying the following formula (14). Through weighted fusion, differential weighting can be applied according to the different contributions of roll rate and roll acceleration to the roll impact hazard. For example, when the weight of roll rate is greater than the weight of roll acceleration, it indicates that the contribution of sustained high roll rate to the risk of runaway is considered greater than that of transient roll acceleration abrupt change. This differential weighting allows the roll impact rate parameter to more reasonably reflect the degree of danger of different roll dynamic characteristics.

[0097] (13); (14); in, This represents the roll impact rate parameter. Represents the target normalized value. This represents the roll rate as a representative value. This represents the preset normalized reference value for roll rate. This represents the roll acceleration value. This represents the preset normalized reference value for roll acceleration.

[0098] By comparing representative values ​​of roll rate and roll acceleration with different physical dimensions with preset normalized reference values, dimensionless first and second normalized values ​​are obtained. These two normalized values ​​are then weighted and fused. Based on the target normalized value, a bounded roll impact rate parameter with uniform dimensions is finally determined. This solves the problem of different physical quantities being unable to be directly fused and avoids fusion distortion caused by inconsistent scales of various parameters.

[0099] The determination methods for the lateral adhesion margin parameter, velocity curvature matching parameter, and roll impact rate parameter have been explained above. The following section explains how to determine the roll loss of control risk level based on at least one of these parameters.

[0100] In some embodiments, determining the roll loss-of-control risk level based on at least one of the lateral adhesion margin parameter, velocity curvature matching parameter, and roll impact rate parameter includes: determining a lateral adhesion sub-score based on the lateral adhesion margin parameter; determining a velocity curvature matching sub-score based on the velocity curvature matching parameter; determining a roll impact rate sub-score based on the roll impact rate parameter; weighting and fusing the lateral adhesion sub-score, velocity curvature matching sub-score, and roll impact rate sub-score to determine a roll loss-of-control risk score; and determining the roll loss-of-control risk level based on the roll loss-of-control risk score.

[0101] For example, the lateral adhesion margin parameter represents the remaining lateral adhesion capacity of the tire (the larger the value, the more sufficient the remaining capacity and the safer it is), and the lateral adhesion sub-score can represent the degree of danger (the larger the value, the more dangerous it is). Therefore, the lateral adhesion sub-score can be obtained, for example, based on the lateral adhesion margin parameter by the following formula (15).

[0102] (15); in, Indicates the score for lateral attachment. This represents the lateral adhesion margin parameter.

[0103] By converting the lateral adhesion margin parameter into a lateral adhesion sub-score, the semantics of tire lateral adhesion remaining capacity are transformed into semantics of hazard level.

[0104] In some implementations, the speed curvature matching parameter itself already characterizes the degree of speeding danger. For example, the speed curvature matching sub-score can be directly equal to the speed curvature matching parameter, satisfying the following formula (16). The speed curvature matching parameter has quantified the degree of exceeding the limit of the actual vehicle speed relative to the safe cornering speed. Its numerical direction is consistent with the degree of danger, so it can be directly used as a sub-score to participate in the fusion, maintaining the simplicity of the calculation.

[0105] (16); in, This represents the velocity curvature matching sub-score. Represents velocity curvature matching parameters In some implementations, the roll impact rate parameter itself characterizes the severity of the roll dynamic impact. For example, the roll impact rate sub-score can be directly equal to the roll impact rate parameter, satisfying the following formula (17). The roll impact rate parameter has quantified the severity of the roll action, and its numerical direction is consistent with the degree of danger, so it can be directly used as a sub-score in the fusion.

[0106] (17); in, This represents the sub-score of the roll impact rate. This represents the roll impact rate parameter.

[0107] In some implementations, the lateral attachment sub-score, velocity curvature matching sub-score, and roll impact rate sub-score are weighted and fused using the following formula (18) to determine the roll loss of control risk score.

[0108] (18); in, This indicates the risk score for rollover and loss of control.

[0109] In some implementations, the weighting reflects the engineering judgment of the relative importance of different runaway factors. For example, the weights of lateral adhesion margin and velocity curvature matching are the same and slightly higher than the weight of roll impact rate, indicating that insufficient adhesion and overspeed cornering are considered equally important major runaway factors.

[0110] In some implementations, the risk level of roll loss of control is determined based on a roll loss of control risk score. Multiple score ranges can be preset, each corresponding to a risk level. For example, the risk level can include four levels: Risk Level 0, Risk Level 1, Risk Level 2, and Risk Level 3. At Risk Level 0, the roll angle is small, the speed is below the safety limit, and it is considered normal cornering; at Risk Level 1, there is a certain roll posture, with a relatively low risk; at Risk Level 2, there is significant roll, approaching the road adhesion limit; at Risk Level 3, there is extreme roll, making it extremely easy to lose control. Different risk levels can correspond to different levels of danger, allowing drivers or fleet management systems to take differentiated warning and intervention measures based on the level, such as issuing a warning for Level 2 and triggering emergency intervention for Level 3.

[0111] In another optional embodiment, the above-mentioned method for determining roll loss of control risk may further include: selecting interpretable text information corresponding to the roll loss of control risk level from a preset template library; and outputting the interpretable text information and the risk level together to a human-machine interface or a vehicle network service platform.

[0112] For example, interpretable text information may include: “Normal cornering, small roll angle, speed below safety limit”, “Slight cornering, please be aware of slippery road surface”, “Moderate cornering, approaching the adhesion limit, please slow down”, “Dangerous cornering, extremely easy to lose control, please slow down immediately and reduce roll”.

[0113] For example, the in-vehicle human-machine interface can display the current roll and loss of control risk level to the driver in real time, and issue tiered warnings at risk levels 2 and 3, prompting the driver to reduce speed and minimize roll. Furthermore, the risk level can be uploaded to the fleet management system via the vehicle network, allowing managers to remotely monitor the real-time safety status of each motorcycle and intervene proactively when necessary.

[0114] By weighted fusion of the lateral attachment sub-score, velocity-curvature matching sub-score, and roll impact rate sub-score, differentiated weighting can be applied based on the contribution of each dimension to the roll runaway risk, resulting in a comprehensive roll runaway risk score that reflects multiple risk factors. Discretizing the continuous risk score into multiple risk levels makes the risk assessment results more intuitive, easier to understand, and easier to communicate.

[0115] It should be noted that any parts not fully described in the above examples can be referred to the aforementioned embodiments, and will not be repeated here.

[0116] The above is based on Figure 1 The present disclosure describes in detail an embodiment of the method for determining rollover risk. The following is in conjunction with... Figure 2This document describes in detail embodiments of the roll runaway risk determination device. It should be understood that the descriptions of the roll runaway risk determination method embodiments correspond to the descriptions of the roll runaway risk determination device embodiments; therefore, any parts not described in detail can be found in the preceding method embodiments.

[0117] Figure 2 The diagram shown is a structural schematic of a rollover risk determination device provided in an embodiment of this disclosure.

[0118] like Figure 2 As shown, the roll loss risk determination device 200 provided in this embodiment includes: an acquisition module 210, used to acquire motion posture information and vehicle speed information of the target vehicle, environmental weather information and curve curvature information of the target vehicle's location; and a determination module 220, used to determine the roll loss risk level based on at least one of a lateral adhesion margin parameter, a speed curvature matching parameter, and a roll impact rate parameter when the target vehicle is determined to be in a rainy environment and in a curve state based on the environmental weather information, vehicle speed information, and motion posture information. The lateral adhesion margin parameter is determined based on the environmental weather information and motion posture information, the speed curvature matching parameter is determined based on the vehicle speed information, environmental weather information, and curve curvature information, and the roll impact rate parameter is determined based on the roll angular velocity and roll acceleration included in the motion posture information.

[0119] In some embodiments, the determining module 220 is further configured to: sort the absolute values ​​of the roll angular velocity in ascending order within a preset sliding time window, and take the first preset percentile value as the representative value of the roll rate; sort the absolute values ​​of the roll acceleration in ascending order within the preset sliding time window, and take the second preset percentile value as the representative value of the roll acceleration; and determine the roll impact rate parameter based on the representative value of the roll rate and the representative value of the roll acceleration.

[0120] In some embodiments, the determining module 220 is further configured to: determine a first normalized value based on a representative value of roll rate and a preset normalized reference value of roll rate; determine a second normalized value based on a representative value of roll acceleration and a preset normalized reference value of roll acceleration; perform weighted fusion of the first normalized value and the second normalized value to obtain a target normalized value; and determine a roll impact rate parameter based on the target normalized value.

[0121] In some embodiments, the determining module 220 is further configured to: determine a lateral adhesion sub-score based on a lateral adhesion margin parameter; determine a velocity curvature matching sub-score based on a velocity curvature matching parameter; determine a roll impact rate sub-score based on a roll impact rate parameter; weight and fuse the lateral adhesion sub-score, velocity curvature matching sub-score, and roll impact rate sub-score to determine a roll loss of control risk score; and determine a roll loss of control risk level based on the roll loss of control risk score.

[0122] In some embodiments, the motion attitude information of the target vehicle includes lateral acceleration, wherein the determining module 220 is further configured to: determine the road surface adhesion coefficient based on environmental weather information; sort the lateral accelerations within a preset sliding time window and take the maximum value as the target lateral acceleration; determine the lateral adhesion limit value based on the road surface adhesion coefficient; and determine the lateral adhesion margin parameter based on the lateral adhesion limit value and the target lateral acceleration.

[0123] In some embodiments, the determining module 220 is further configured to: determine the lateral adhesion utilization ratio based on the lateral adhesion limit value and the target lateral acceleration; determine the remaining lateral adhesion ratio based on the lateral adhesion utilization ratio; and perform boundary-truncation linear normalization processing on the remaining lateral adhesion ratio to obtain the lateral adhesion margin parameter.

[0124] In some embodiments, the determining module 220 is further configured to: determine the road surface adhesion coefficient based on environmental weather information; determine a safe speed limit based on the road surface adhesion coefficient, a preset safe roll angle threshold, and the curve curvature when the curve curvature indicated by the curve curvature information is greater than a preset curvature threshold; sort the vehicle speed information within a preset sliding time window and take the speed value corresponding to the maximum vehicle speed information as the target vehicle speed; determine the speed over-limit ratio based on the target vehicle speed and the safe speed limit; and perform linear normalization processing based on boundary truncation on the speed over-limit ratio to obtain the speed curvature matching parameters.

[0125] In some embodiments, the motion posture information includes yaw rate and roll angle, wherein the determining module 220 is further configured to: acquire the system power mode and engine speed of the target vehicle; determine the riding state of the target vehicle based on the system power mode, vehicle speed information and engine speed; when the riding state of the target vehicle indicates that the target vehicle is in a real riding state, determine the radius of curvature based on the vehicle speed information and yaw rate, and determine the roll angle index based on the vehicle speed information, yaw rate and roll angle; and determine that the target vehicle is in a cornering state when the radius of curvature is less than a preset radius of curvature threshold, the vehicle speed information is greater than a preset lower limit threshold, the roll angle index is greater than a preset index threshold, and this condition persists for a first preset duration.

[0126] In some embodiments, the acquisition module 210 is further configured to: acquire the GPS location data of the target vehicle; retrieve the rainfall intensity level from the network weather service interface based on the GPS location data, and use the rainfall intensity level as environmental weather information.

[0127] Below, for reference Figure 3 To describe a vehicle according to an embodiment of this disclosure. Figure 3 The diagram shown is a structural schematic of a vehicle provided in an embodiment of this disclosure.

[0128] like Figure 3 As shown, vehicle 300 includes one or more processors 301 and memory 302.

[0129] The processor 301 may be a central processing unit (CPU) or other form of processing unit with data processing capabilities and / or instruction execution capabilities, and may control other components in the vehicle 300 to perform desired functions.

[0130] The memory 302 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 301 may execute the program instructions to implement the roll loss risk determination method of the various embodiments of this disclosure described above and / or other desired functions. The computer-readable storage medium may also store various information such as the target vehicle's motion attitude information and speed information, environmental weather information and curve curvature information of the target vehicle's location, lateral adhesion margin parameters, speed curvature matching parameters, etc.

[0131] In one example, vehicle 300 may also include input device 303 and output device 304, which are interconnected via a bus system and / or other forms of connection mechanism (not shown).

[0132] The input device 303 may include, for example, a keyboard, a mouse, etc.

[0133] The output device 304 can output various information to the outside, including the target vehicle's motion attitude and speed information, the target vehicle's location, environmental weather information and curve curvature information, lateral adhesion margin parameters, speed curvature matching parameters, etc. The output device 304 may include, for example, a display, a speaker, a printer, and a communication network and its connected remote output devices, etc.

[0134] Of course, for the sake of simplicity, Figure 3 Only some of the components of the vehicle 300 relevant to this disclosure are shown in this illustration; components such as buses, input / output interfaces, etc., are omitted. In addition, the vehicle 300 may include any other suitable components depending on the specific application.

[0135] In addition to the methods and apparatus described above, embodiments of this disclosure may also be computer program products comprising computer program instructions that, when executed by a processor, cause the processor to perform the steps in the roll loss-of-control risk determination methods according to various embodiments of this disclosure described above.

[0136] The computer program product can be written in any combination of one or more programming languages ​​to perform the operations of the embodiments of this disclosure. The programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on a user's computing device, partially on a user's computing device, as a standalone software package, partially on a user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0137] Furthermore, embodiments of this disclosure may also be computer-readable storage media storing computer program instructions that, when executed by a processor, cause the processor to perform the steps in the roll loss-of-control risk determination method according to various embodiments of this disclosure described above.

[0138] The computer-readable storage medium may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof.

[0139] The basic principles of this disclosure have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this disclosure are merely examples and not limitations, and should not be considered as essential features of each embodiment of this disclosure. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the scope of this disclosure to the necessity of employing the aforementioned specific details for implementation.

[0140] The block diagrams of devices, apparatuses, devices, and systems disclosed herein are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0141] It should also be noted that in the apparatus, devices, and methods of this disclosure, the components or steps can be disassembled and / or recombined. These disassemblies and / or recombinations should be considered as equivalent solutions to this disclosure.

[0142] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of this disclosure. Therefore, this disclosure is not intended to be limited to the aspects shown herein, but rather to be carried out within the widest scope consistent with the principles and novel features disclosed herein.

[0143] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of this disclosure to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations therein.

Claims

1. A method for determining the risk of rollover loss of control, characterized in that, include: Acquire the target vehicle's motion attitude and speed information, as well as the environmental weather information and curve curvature information of the target vehicle's location; When the target vehicle is determined to be in a rainy environment and cornering based on the environmental weather information, vehicle speed information, and motion posture information, a roll loss of control risk level is determined based on at least one of the following: lateral adhesion margin parameter, velocity curvature matching parameter, and roll impact rate parameter. Roll loss of control risk identification is then performed based on this risk level. The lateral adhesion margin parameter is determined based on the environmental weather information and motion posture information; the velocity curvature matching parameter is determined based on the vehicle speed information, environmental weather information, and corner curvature information; and the roll impact rate parameter is determined based on the roll angular velocity and roll acceleration included in the motion posture information.

2. The method according to claim 1, characterized in that, The roll impact rate parameter is determined based on the roll angular velocity and the roll acceleration, and includes: Within a preset sliding time window, the absolute values ​​of the roll angular velocity are sorted in ascending order, and the first preset percentile value is taken as the representative value of the roll rate. Within a preset sliding time window, the absolute values ​​of the roll acceleration are sorted in ascending order, and the second preset percentile value is taken as the representative value of the roll acceleration. The roll impact rate parameter is determined based on the representative values ​​of the roll rate and the roll acceleration.

3. The method according to claim 2, characterized in that, The determination of the roll impact rate parameter based on the representative value of the roll rate and the representative value of the roll acceleration includes: Based on the representative value of the roll rate and the preset normalized reference value of the roll rate, a first normalized value is determined; Based on the representative value of the roll acceleration and the preset normalized reference value of the roll acceleration, a second normalized value is determined; The first normalized value and the second normalized value are weighted and fused to obtain the target normalized value; The roll impact rate parameter is determined based on the target normalized value.

4. The method according to claim 1, characterized in that, The determination of the roll-off risk level based on at least one of the lateral adhesion margin parameter, velocity curvature matching parameter, and roll impact rate parameter includes: Based on the lateral adhesion margin parameter, the lateral adhesion sub-score is determined; Based on the velocity curvature matching parameters, determine the velocity curvature matching sub-score; Based on the roll impact rate parameter, determine the roll impact rate sub-score; The lateral attachment sub-score, the velocity curvature matching sub-score, and the roll impact rate sub-score are weighted and fused to determine the roll loss control risk score. The roll loss of control risk level is determined based on the roll loss of control risk score.

5. The method according to any one of claims 1 to 4, characterized in that, The motion attitude information of the target vehicle includes lateral acceleration. The lateral adhesion margin parameter is determined based on the environmental weather information and the motion posture information, and includes: Based on the aforementioned environmental weather information, the road surface adhesion coefficient is determined; Within a preset sliding time window, the lateral accelerations are sorted, and the maximum value is taken as the target lateral acceleration. Based on the road surface adhesion coefficient, determine the lateral adhesion limit value; The lateral adhesion margin parameter is determined based on the lateral adhesion limit value and the target lateral acceleration.

6. The method according to claim 5, characterized in that, The determination of the lateral adhesion margin parameter based on the lateral adhesion limit value and the target lateral acceleration includes: Based on the lateral adhesion limit value and the target lateral acceleration, the lateral adhesion utilization ratio is determined; Based on the lateral adhesion utilization ratio, determine the remaining lateral adhesion ratio; The remaining ratio of lateral adhesion is subjected to boundary-truncation linear normalization to obtain the lateral adhesion margin parameter.

7. The method according to any one of claims 1 to 4, characterized in that, The speed curvature matching parameters are determined based on the vehicle speed information, the environmental weather information, and the curve curvature information, and include: Based on the aforementioned environmental weather information, the road surface adhesion coefficient is determined; If the curvature of the curve indicated by the curve curvature information is greater than a preset curvature threshold, a safe speed limit is determined based on the road surface adhesion coefficient, the preset safe roll angle threshold, and the curve curvature. Within a preset sliding time window, the vehicle speed information is sorted, and the speed value corresponding to the maximum vehicle speed information is taken as the target vehicle speed. Based on the target vehicle speed and the safe speed limit, the speed over-limit ratio is determined; The velocity over-limit ratio is subjected to linear normalization based on boundary truncation to obtain the velocity curvature matching parameters.

8. The method according to claim 1, characterized in that, The motion posture information includes yaw rate and roll angle. Determining that the target vehicle is in a curve includes: Obtain the system power mode and engine speed of the target vehicle; The riding status of the target vehicle is determined based on the system power mode, the vehicle speed information, and the engine speed. When the target vehicle riding status indicates that the target vehicle is in a real riding state, the radius of curvature is determined based on the vehicle speed information and the yaw rate, and the roll angle index is determined based on the vehicle speed information, the yaw rate, and the roll angle. If the radius of curvature is less than a preset radius of curvature threshold, the vehicle speed is greater than a preset lower limit threshold, the roll angle index is greater than a preset index threshold, and this condition persists for a first preset duration, then the target vehicle is determined to be in a curve.

9. The method according to claim 1, characterized in that, The acquisition of environmental weather information at the location of the target vehicle includes: Obtain the GPS location data of the target vehicle; Based on the location data from the Global Positioning System, the rainfall intensity level is retrieved from the network weather service interface, and the rainfall intensity level is used as the environmental weather information.

10. A vehicle, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is used to execute the roll loss risk determination method according to any one of claims 1 to 9.

Images