Vehicle rollover warning system and method based on state prediction and working condition clustering

Abstract

The application provides a vehicle rollover early warning system and method based on state prediction and working condition clustering, wherein the system comprises: a state prediction module for predicting the motion state of an electric vehicle through a neural network; a state prediction module for determining the current working condition of the electric vehicle through clustering; a model updating module for updating the vehicle roll dynamics model of the electric vehicle in real time based on the predicted motion state and the determined current working condition; a rollover early warning module for analyzing and warning the vehicle rollover risk based on the real-time updated vehicle roll dynamics model. Through the four steps of state prediction, working condition clustering, model updating and risk warning, the real-time monitoring and early warning of the electric vehicle rollover risk are realized. The model parameters can be adjusted according to the real-time state and working condition, and the early warning accuracy is improved. The rollover risk can be predicted and the driver can be reminded in time, thereby improving the driving safety. An innovative solution is provided for the safety of intelligent electric vehicles.

Description

Technical Field

[0001] This invention relates to the field of electric vehicle control technology, and in particular to a vehicle rollover early warning system and method based on state prediction and operating condition clustering. Background Technology

[0002] Currently, vehicle rollover accidents are a significant issue in traffic safety, especially in complex driving scenarios such as high-speed turns and emergency avoidance, where the risk of rollover increases significantly. Therefore, developing an efficient and accurate vehicle rollover warning system has become an urgent need to improve driving safety.

[0003] Existing vehicle rollover warning systems are mostly based on traditional dynamic models, combining sensors to monitor parameters such as lateral acceleration and roll angle in real time to determine rollover risk. However, these systems have significant limitations: on the one hand, fixed model parameters are difficult to adapt to changing operating conditions and vehicle states, resulting in insufficient warning accuracy; on the other hand, in dynamic and complex driving environments, the real-time performance of the system is difficult to guarantee, limiting its effectiveness in practical applications.

[0004] To address the above problems, this invention proposes a vehicle rollover early warning system and method based on state prediction and working condition clustering. Summary of the Invention

[0005] One of the objectives of this invention is to provide a vehicle rollover early warning system and method based on state prediction and working condition clustering to solve the problems mentioned in the background art.

[0006] In a first aspect, the vehicle rollover warning system based on state prediction and working condition clustering provided in the embodiments of the present invention includes:

[0007] The state prediction module is used to predict the motion state of an electric vehicle through a neural network.

[0008] The operating condition clustering module is used to determine the current operating condition of electric vehicles through clustering.

[0009] The model update module is used to update the vehicle roll dynamics model of electric vehicles in real time based on the predicted motion state and the determined current operating conditions.

[0010] The rollover warning module is used to analyze vehicle rollover risk and provide early warning based on the real-time updated vehicle rollover dynamics model.

[0011] Optionally, the state prediction module predicts the motion state of the electric vehicle through a neural network, including:

[0012] A pre-trained recurrent neural network is used as the neural network;

[0013] The neural network is fed with vehicle parameters of an electric vehicle, including at least vehicle wheel speed, steering wheel angle, yaw rate, longitudinal acceleration, lateral acceleration, body roll angle, suspension travel sensor data, and motor torque / speed.

[0014] The predicted motion state of an electric vehicle is based on at least the predicted vehicle yaw rate, predicted vehicle lateral acceleration, predicted vehicle roll angle, and predicted tire vertical load, as output by a neural network.

[0015] Optionally, the operating condition clustering module determines the current operating condition of the electric vehicle through clustering, including:

[0016] Based on electric vehicles, a feature vector set for clustering is constructed, including at least the current vehicle speed, steering wheel angle, yaw rate, longitudinal acceleration, lateral acceleration, body roll angle, tire slip ratio, and motor output torque.

[0017] Based on the K-means algorithm, the working conditions are clustered according to the feature vector set used for clustering to obtain the current working conditions.

[0018] The current operating conditions include at least the following categories: straight driving, steady-state steering, transient steering, acceleration, braking, driving conditions under different road surface adhesion coefficients, and different road types.

[0019] Optionally, the model update module updates the vehicle roll dynamics model of the electric vehicle in real time based on the predicted motion state and the determined current operating conditions, including:

[0020] Based on the predicted motion state and the determined current operating conditions, adjust the corresponding parameters or coefficients in the vehicle roll dynamics model;

[0021] Among them, the parameters or coefficients include at least: roll stiffness, roll damping, roll moment of inertia, and tire lateral stiffness.

[0022] Optionally, the rollover warning module analyzes the vehicle rollover risk based on a real-time updated vehicle roll dynamics model, including:

[0023] For each preset vehicle rollover risk index, the index value of the real-time updated vehicle rollover dynamics model under that vehicle rollover risk index is calculated and compared with the corresponding preset index threshold. The vehicle rollover risk is determined based on the comparison result.

[0024] Among them, vehicle rollover risk indicators include at least: roll angle, load transfer rate, and rollover coefficient.

[0025] Optionally, the method of providing a vehicle rollover risk warning includes:

[0026] Step a: Divide the critical time window into at least two consecutive decision nodes, and repeat steps b to f within each consecutive decision node;

[0027] Step b: Real-time collection of cognitive load assessment data for electric vehicle drivers;

[0028] Step c: Based on the cognitive load assessment, determine the driver's current cognitive load and match the corresponding solution reception sensitivity level; wherein, the solution reception sensitivity level is negatively correlated with the maximum intensity of auxiliary intervention that the driver can accept;

[0029] Step d: Based on the driver's current sensitivity level to the scheme reception, dynamically select the corresponding level of the auxiliary execution scheme from the preset anti-rollover assist strategy library; wherein, the anti-rollover assist strategy library contains multi-level auxiliary execution schemes from the active control layer to the passive prompting layer, the level of each level of the auxiliary execution scheme is positively correlated with its corresponding scheme reception sensitivity level, and the auxiliary intervention intensity of each level of the auxiliary execution scheme decreases as the level increases;

[0030] Step e: Implement a dynamic selection-assisted execution plan for the driver;

[0031] Step f: When the similarity between the active operation behavior vector of the driver operating the electric vehicle and the auxiliary target vector of the auxiliary execution scheme executed on the driver exceeds a threshold, stop executing the corresponding auxiliary execution scheme on the driver.

[0032] Optionally, the step of obtaining the safety critical time window includes:

[0033] The time window from when the risk of vehicle rollover is determined to the future target time is taken as the safety critical time window.

[0034] Among them, the future target time is the smaller value between the physical rollover prevention limit time and the driver's cognitive buffer time;

[0035] Among them, the physical rollover prevention limit time includes: the conservative time for an electric vehicle to roll over irreversibly in the future, determined based on the real-time updated vehicle roll dynamics model;

[0036] Cognitive buffer time includes: the maximum permissible time for a driver to complete the cognitive perception and operational response to vehicle rollover risk, determined based on the driver's cognitive load during the time spent analyzing and determining the vehicle rollover risk.

[0037] Optionally, the cognitive load assessment may be based on: the driver's eye gaze trajectory, changes in steering wheel grip pressure, and the history of operation response delays within a previously preset time period.

[0038] Secondly, the vehicle rollover early warning method based on state prediction and working condition clustering provided in the embodiments of the present invention includes:

[0039] Predicting the motion state of electric vehicles using neural networks;

[0040] The current operating condition of electric vehicles is determined by clustering;

[0041] The vehicle roll dynamics model of the electric vehicle is updated in real time based on the predicted motion state and the determined current operating conditions.

[0042] Based on the real-time updated vehicle rollover dynamics model, the risk of vehicle rollover is analyzed and a vehicle rollover risk warning is issued.

[0043] Optionally, the method of providing a vehicle rollover risk warning includes:

[0044] Step a: Divide the critical time window into at least two consecutive decision nodes, and repeat steps b to f within each consecutive decision node;

[0045] Step b: Real-time collection of cognitive load assessment data for electric vehicle drivers;

[0046] Step c: Based on the cognitive load assessment, determine the driver's current cognitive load and match the corresponding solution reception sensitivity level; wherein, the solution reception sensitivity level is negatively correlated with the maximum intensity of auxiliary intervention that the driver can accept;

[0047] Step d: Based on the driver's current sensitivity level to the scheme reception, dynamically select the corresponding level of the auxiliary execution scheme from the preset anti-rollover assist strategy library; wherein, the anti-rollover assist strategy library contains multi-level auxiliary execution schemes from the active control layer to the passive prompting layer, the level of each level of the auxiliary execution scheme is positively correlated with its corresponding scheme reception sensitivity level, and the auxiliary intervention intensity of each level of the auxiliary execution scheme decreases as the level increases;

[0048] Step e: Implement a dynamic selection-assisted execution plan for the driver;

[0049] Step f: When the similarity between the active operation behavior vector of the driver operating the electric vehicle and the auxiliary target vector of the auxiliary execution scheme executed on the driver exceeds a threshold, stop executing the corresponding auxiliary execution scheme on the driver.

[0050] The present invention has achieved the following beneficial effects:

[0051] This system predicts the motion state of electric vehicles through neural networks and uses clustering technology to identify current operating conditions, thereby updating the vehicle's roll dynamics model in real time. This dynamic adaptive design significantly improves the accuracy and real-time performance of warnings, providing an innovative solution for the safety of intelligent electric vehicles. In practical applications, such as high-speed cornering scenarios, the system can predict rollover risks in advance and promptly alert the driver, effectively reducing the probability of accidents.

[0052] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.

[0053] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0054] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0055] Figure 1 This is a schematic diagram of a vehicle rollover warning system based on state prediction and working condition clustering in an embodiment of the present invention;

[0056] Figure 2 This is a schematic diagram illustrating the execution steps of each module in the vehicle rollover early warning system based on state prediction and working condition clustering in an embodiment of the present invention. Detailed Implementation

[0057] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0058] The research and development approach of this application is to combine neural network state prediction and operating condition clustering technology to update the vehicle roll dynamics model in real time, thereby more accurately assessing and warning of the rollover risk of electric vehicles. This method can dynamically adapt to different driving conditions and vehicle states, improving the accuracy and timeliness of warnings and providing technical support for intelligent driving safety.

[0059] Figure 1 This is a schematic diagram of a vehicle rollover warning system based on state prediction and working condition clustering provided in an embodiment of this application, as shown below. Figures 1 to 2 As shown, the system includes the following four main modules:

[0060] The state prediction module 100 is used to perform step 101, predicting the motion state of the electric vehicle through a neural network. Step 101 includes:

[0061] 201. Use a pre-trained recurrent neural network as the neural network.

[0062] In this step, a pre-trained recurrent neural network (RNN) is used as the core prediction model. RNNs can process time-series data and capture the time dependence of vehicle parameters through their recurrent structure, making them suitable for predicting future dynamic states based on historical and current vehicle parameters. Historical data collected by the electric vehicle's sensors is used as training data, and the RNN is pre-trained with the goal of minimizing the error between the predicted and actual states.

[0063] 202. Input the vehicle parameters of the electric vehicle, including at least the vehicle wheel speed, steering wheel angle, yaw rate, longitudinal acceleration, lateral acceleration, body roll angle, suspension travel sensor data, and motor torque / speed, into the neural network.

[0064] In this step, the input data is collected in real time by sensors on the vehicle and input into the RNN in a time series.

[0065] 203. The predicted motion state of an electric vehicle, as output by a neural network, includes at least the predicted vehicle yaw rate, the predicted vehicle lateral acceleration, the predicted vehicle roll angle, and the predicted tire vertical load.

[0066] Specific implementation example: Taking an electric SUV traveling at high speed as an example, sensors collect data such as the current wheel speed of 120 km / h, steering wheel angle of 15°, and yaw rate of 0.2 rad / s. After receiving these inputs, the RNN predicts the motion state within the next second: yaw rate of 0.25 rad / s, lateral acceleration of 0.4g, body roll angle of 5°, and tire vertical load distribution change rate of 10%. This predicted data provides real-time status information for subsequent modules.

[0067] The operating condition clustering module 200 is used to execute step 102, determining the current operating condition of the electric vehicle through clustering. Step 102 includes:

[0068] 301. Based on electric vehicles, construct a feature vector set for clustering using vehicle parameters including at least the current vehicle speed, steering wheel angle, yaw rate, longitudinal acceleration, lateral acceleration, body roll angle, tire slip ratio, and motor output torque.

[0069] In this step, the above parameters are normalized and combined into multi-dimensional feature vectors (such as [vehicle speed, steering wheel angle, yaw rate, longitudinal acceleration, lateral acceleration, body roll angle, tire slip ratio, motor output torque]), forming a feature vector set for clustering.

[0070] 302. Based on the K-means algorithm, cluster the driving conditions according to the feature vector set used for clustering to obtain the current driving conditions. The categories of the current driving conditions include at least: straight driving, steady-state steering, transient steering, acceleration, braking, driving conditions under different road surface adhesion coefficients, and different road types.

[0071] In this step, the K-means algorithm is used to cluster the feature vector set. The algorithm iteratively optimizes and classifies similar feature vectors, and finally outputs the current working condition category.

[0072] Continuing with the specific implementation example above: the feature vector is [120km / h, 15°, 0.2rad / s, 0.1g, 0.3g, 4°, 0.05, 500Nm]. After analysis using the K-means algorithm, it is classified as a "high-speed steady-state steering" condition. This result reflects the dynamic characteristics of the vehicle during high-speed cornering, providing a basis for model updates.

[0073] The model update module 300 is used to execute step 103, updating the vehicle roll dynamics model of the electric vehicle in real time based on the predicted motion state and the determined current operating conditions. The model update module dynamically adjusts the parameters of the vehicle roll dynamics model according to the predicted motion state and the current operating conditions to reflect real-time vehicle behavior. The roll dynamics model is a mathematical model used to describe the dynamic response of the vehicle under lateral forces, and can be pre-built based on relevant simulation software. Step 103 includes:

[0074] 401. Based on the predicted motion state and the determined current operating conditions, adjust the corresponding parameters or coefficients in the vehicle roll dynamics model. These parameters or coefficients shall include at least: roll stiffness, roll damping, roll moment of inertia, and tire lateral stiffness.

[0075] In this step, a mapping relationship between different predicted motion states or determined current operating conditions and model parameters or coefficients can be established in advance. During adjustment, the corresponding parameters or coefficients are found based on this mapping relationship and updated accordingly. A specific implementation example: Under high-speed steady-state steering conditions, with a predicted lateral acceleration of 0.4g, the system increases roll stiffness (from 5000 Nm / rad to 5500 Nm / rad) and roll damping (from 1000 Ns / m to 1200 Ns / m) to simulate the increased rigidity of the vehicle during cornering. The updated model more accurately reflects the current state.

[0076] The rollover warning module 400 is used to execute step 104, analyzing the vehicle rollover risk based on the real-time updated vehicle rollover dynamics model, and issuing a vehicle rollover risk warning. Step 104 includes:

[0077] 501. For each preset vehicle rollover risk index, calculate the index value of the real-time updated vehicle rollover dynamics model under that index, and compare it with the corresponding preset index threshold. Determine the vehicle rollover risk based on the comparison result. The vehicle rollover risk index includes at least: roll angle, load transfer rate, and rollover coefficient.

[0078] In this step, different vehicle rollover risk indicators and their corresponding calculation formulas and threshold values ​​can be preset. The calculated indicator values ​​are compared with the corresponding threshold values; if the values ​​exceed the thresholds, the vehicle rollover risk is determined. The roll angle reflects the degree of vehicle tilt, the load transfer rate reflects the tire load distribution during cornering, and the rollover coefficient reflects the comprehensive dynamic stability parameters.

[0079] Specific implementation example: The calculated roll angle is 5° (less than 10°), load transfer rate is 0.6 (less than 0.8), and rollover coefficient is 0.7 (less than 1.0). The system determines that there is currently no risk of rollover. However, if the turn intensifies and the roll angle rises to 12°, a warning is triggered, prompting the driver to slow down.

[0080] This system achieves real-time monitoring and early warning of rollover risks in electric vehicles through four steps: state prediction, operating condition clustering, model updating, and risk warning. Its core advantage lies in its dynamic adaptability, enabling it to adjust model parameters based on real-time states and operating conditions, thereby improving warning accuracy. In practical applications, such as high-speed turning scenarios, the system can predict rollover risks and promptly alert the driver, thus enhancing driving safety. This technology provides an innovative solution for the safety of intelligent electric vehicles.

[0081] In some embodiments, the method of providing vehicle rollover risk warning includes:

[0082] Step a: Divide the critical time window into at least two consecutive decision nodes, and repeat steps b to f within each consecutive decision node.

[0083] In this step, the critical safety time window is the key time range within which the vehicle evolves from its current state to an irreversible rollover state. By dividing this time window into at least two consecutive decision nodes, the system can achieve dynamic monitoring and phased intervention over time. This division ensures that the system can continuously adjust its assistance strategies based on real-time changes in driver status and vehicle dynamics, avoiding the blindness or lag of a one-time intervention.

[0084] In practice, the length of the safety critical time window is first determined through the acquisition steps described later (e.g., assumed to be 3 seconds). The system divides this into two decision nodes, each lasting 1.5 seconds. Within each 1.5-second node, the system repeats steps b through f to achieve continuous assessment and response to the driver and vehicle status. For example, when an electric vehicle is traveling at 80 km / h on a curve, if the system detects an increase in the roll angle, it immediately initiates the time window division and processes the data in 1.5-second increments.

[0085] Specific implementation example: Suppose an electric vehicle is driving on a mountain road, and the sensors detect that the vehicle's tilt angle reaches 5 degrees, triggering a rollover risk warning. The system divides the safety critical time window (e.g., 3 seconds) into two nodes: Node 1 (0-1.5 seconds) and Node 2 (1.5-3 seconds). Within each node, the system collects data and dynamically adjusts the strategy to ensure that intervention measures are synchronized with the evolution of the risk.

[0086] Step b: Real-time acquisition of cognitive load assessment data for electric vehicle drivers. This cognitive load assessment data includes: the driver's eye gaze trajectory, changes in steering wheel grip pressure, and the historical operational response delays over a previously preset time period.

[0087] In this step, the driver's cognitive load reflects their current ability to process information and cope with risks, and is a key basis for determining assistance strategies. This step comprehensively assesses the driver's psychological and behavioral state through multi-dimensional data collection, including eye gaze trajectory, changes in steering wheel grip pressure, and operational response delay history, providing data support for subsequent decision-making.

[0088] The eye gaze trajectory is recorded by an in-vehicle camera and an eye-tracking algorithm, which records the position and movement path of the driver's gaze focus (the acquisition of the driver's eye gaze trajectory is an existing technology, such as the automatic steering and lane changing function that BMW has already applied to its vehicles based on the eye gaze trajectory, which will not be elaborated here).

[0089] Steering wheel grip pressure changes are detected using pressure sensors mounted on the steering wheel. A sudden increase in pressure may indicate tension or an emergency response, while a decrease in pressure may suggest relaxation or fatigue.

[0090] Operation response delay history analyzes the driver's response time to commands such as braking, accelerator, or steering within a preset time period (e.g., the first 10 seconds). A relatively increased average delay may indicate excessive cognitive load.

[0091] Specific implementation example: When an electric vehicle is traveling at 60 km / h, the system detects that the driver's gaze shifts from the road to the central control screen three times within one second, the steering wheel grip pressure increases from 20N to 30N, and the brake response delay increases from 0.3 seconds to 0.5 seconds within the first 10 seconds. These data are recorded in real time and used as the basis for cognitive load assessment.

[0092] Step c: Based on the cognitive load assessment, determine the driver's current cognitive load and match it with the corresponding scheme reception sensitivity level. The scheme reception sensitivity level is negatively correlated with the driver's maximum acceptable level of auxiliary intervention.

[0093] In this step, the level of cognitive load directly affects the driver's ability to accept external assistance. By quantifying assessment data, the system determines the driver's current cognitive load level and maps it to a preset sensitivity level for receiving the intervention. The sensitivity level is negatively correlated with the maximum acceptable intensity of assistance intervention for the driver; that is, the higher the sensitivity level, the more likely the driver is to accept low-intensity assistance.

[0094] The system inputs the collected data into a cognitive load assessment model (e.g., a weighted average algorithm that calculates the load index by weighting the number of eye gaze deviations from the road, changes in grip pressure, and the increase in average response delay time). For example, if the percentage of eye gaze deviations is 40%, the change in grip pressure is 30%, and the increase in response delay is 30%, the load index is 0.75 (out of 1). Based on the load index, a sensitivity level is matched: a load index > 0.7 indicates high sensitivity (receiving weak intervention), 0.3 ≤ load index ≤ 0.7 indicates medium sensitivity (receiving moderate intervention), and a load index < 0.3 indicates low sensitivity (receiving strong intervention). In this example, a load index of 0.75 corresponds to a high sensitivity level.

[0095] Continuing with the specific implementation example above: A driver load index of 0.75 indicates highly distracted attention. The system determines that the driver's sensitivity to receiving the plan is high, meaning that subsequent assistance strategies should avoid overly complex interventions to prevent increasing the driver's burden. Step d: Based on the driver's current sensitivity level to receiving the plan, dynamically select the corresponding level of assistance execution plan from the preset anti-rollover assistance strategy library. The anti-rollover assistance strategy library contains multi-level assistance execution plans from the active control layer to the passive prompting layer. The level of each assistance execution plan is positively correlated with its corresponding sensitivity level to receiving the plan, and the intensity of assistance intervention decreases as the level increases.

[0096] In this step, the rollover prevention assist strategy library contains multi-level preset schemes, ranging from active control (such as automatic deceleration) to passive prompts (such as audible warnings). Each level is positively correlated with the sensitivity level, and the intervention intensity decreases as the level increases. Dynamic selection ensures that the assist measures match the driver's state, effectively controlling risks while avoiding excessive interference.

[0097] Here is a simple example of such a rollover prevention assist strategy library:

[0098] Level 1 (low scheme reception sensitivity, high auxiliary intervention intensity): active control, such as automatic adjustment of vehicle speed or steering angle, with high intervention intensity;

[0099] Level 2 (Medium reception sensitivity, medium intervention intensity): Semi-active assistance, such as slight braking accompanied by dashboard warnings, with moderate intervention intensity;

[0100] Level 3 (High scheme reception sensitivity, low auxiliary intervention intensity): Passive prompts, such as buzzer alarms or voice reminders, with low intervention intensity.

[0101] Specific implementation example: In the case of mountain curves, the driver has a high cognitive load and high sensitivity. The system selects the level 3 solution from the strategy library and plays "Please slow down, be aware of the risk of rollover" through the car audio in the first node to avoid direct intervention on the steering wheel.

[0102] Step e: Implement a dynamic selection-assisted execution scheme for the driver.

[0103] In this step, the selected assistance program is implemented to reduce rollover risk through appropriate intervention, while ensuring that the driver understands and cooperates with the system's behavior. This step translates the strategy into concrete actions that are applied to the vehicle or driver in real time.

[0104] Specific implementation example: For Level 3 scheme, the system activates the in-vehicle audio system, plays a pre-recorded voice prompt, and displays a red warning icon on the instrument panel. For Level 1 scheme, the electronic control unit (ECU) may reduce the vehicle speed to 50 km / h and fine-tune the steering angle. Step f: When the similarity between the driver's active operation behavior vector and the auxiliary target vector of the auxiliary execution scheme executed on the driver exceeds a threshold, the execution of the corresponding auxiliary execution scheme on the driver is stopped.

[0105] In this step, the system compares the driver's actual actions with the expected behavior of the assisted target to determine whether the driver has effectively responded to the risk. When the similarity between the two exceeds a preset threshold, it indicates that the driver has taken control of the situation, and the system can stop intervening to avoid redundancy.

[0106] Specific implementation example: Active operation behavior vectors are constructed using parameters such as vehicle speed changes and steering angle adjustments. For example, if the driver decelerates by 5 km / h and adjusts the steering angle by 2 degrees, the voice prompt's goal is to decelerate and maintain stability. An auxiliary target vector is constructed using parameters such as a speed reduction of 5-10 km / h and a steering angle adjustment of 1-3 degrees. Cosine similarity is used to calculate the similarity between the active operation behavior vector and the auxiliary target vector. If the result exceeds a preset threshold of 0.9, assistance is stopped.

[0107] In summary, steps a to f, by defining a safety critical time window, assessing cognitive load in real time, dynamically selecting and executing auxiliary strategies, and combining driver behavior feedback to form a closed-loop optimization mechanism, ensure the accuracy and efficiency of vehicle rollover risk warning. This method significantly improves safety and driver experience in complex driving scenarios, further providing an innovative solution for the safety of intelligent electric vehicles.

[0108] In some embodiments, the step of obtaining the safety critical time window includes:

[0109] The time window from when the risk of vehicle rollover is determined to the future target time is taken as the safety critical time window.

[0110] Among them, the future target time is the smaller value between the physical rollover prevention limit time and the driver's cognitive buffer time;

[0111] Among them, the physical rollover prevention limit time includes: the conservative time for an electric vehicle to roll over irreversibly in the future, determined based on the real-time updated vehicle roll dynamics model;

[0112] Cognitive buffer time includes: the maximum permissible time for a driver to complete the cognitive perception and operational response to vehicle rollover risk, determined based on the driver's cognitive load during the time spent analyzing and determining the vehicle rollover risk.

[0113] The critical safety time window is the foundation of the early warning system. It needs to be determined by combining the vehicle's physical limits with the driver's cognitive ability, taking the smaller value between the physical rollover prevention limit time and the cognitive buffer time, to ensure that the system takes action before the risk becomes irreversible.

[0114] By using a vehicle roll dynamics model (considering vehicle speed, roll angle, center of gravity height, etc.), a conservative time for irreversible rollover can be predicted. Technicians can then experiment or set physical rollover prevention limits for different vehicle rollover risks as needed. For example, at a vehicle speed of 80 km / h and a roll angle of 5 degrees, the model calculates a limit time of 4 seconds.

[0115] Based on the driver's current cognitive load and historical response data, the maximum permissible time for the driver to identify and act upon the risk can be estimated. Technicians can also determine the cognitive buffer time for different driver cognitive loads in relation to different vehicle rollover risks through experiments. For example, a load of 0.75 corresponds to a buffer time of 3 seconds.

[0116] The smaller of the physical rollover prevention limit time and the driver's cognitive buffer time is taken as the future target time, and the time window from the time when the vehicle rollover risk is analyzed and determined to the future target time is taken as the safety critical time window.

[0117] In summary, the above technical solutions provide a scientific basis for system decision-making by using both physical and cognitive constraints to obtain the safety critical time window.

[0118] This application provides a vehicle rollover early warning method based on state prediction and working condition clustering, including:

[0119] Predicting the motion state of electric vehicles using neural networks;

[0120] The current operating condition of electric vehicles is determined by clustering;

[0121] The vehicle roll dynamics model of the electric vehicle is updated in real time based on the predicted motion state and the determined current operating conditions.

[0122] Based on the real-time updated vehicle rollover dynamics model, the risk of vehicle rollover is analyzed and a vehicle rollover risk warning is issued.

[0123] In some embodiments, the method of providing vehicle rollover risk warning includes:

[0124] Step a: Divide the critical time window into at least two consecutive decision nodes, and repeat steps b to f within each consecutive decision node;

[0125] Step b: Real-time collection of cognitive load assessment data for electric vehicle drivers;

[0126] Step c: Based on the cognitive load assessment, determine the driver's current cognitive load and match the corresponding solution reception sensitivity level; wherein, the solution reception sensitivity level is negatively correlated with the maximum intensity of auxiliary intervention that the driver can accept;

[0127] Step d: Based on the driver's current sensitivity level to the scheme reception, dynamically select the corresponding level of the auxiliary execution scheme from the preset anti-rollover assist strategy library; wherein, the anti-rollover assist strategy library contains multi-level auxiliary execution schemes from the active control layer to the passive prompting layer, the level of each level of the auxiliary execution scheme is positively correlated with its corresponding scheme reception sensitivity level, and the auxiliary intervention intensity of each level of the auxiliary execution scheme decreases as the level increases;

[0128] Step e: Implement a dynamic selection-assisted execution plan for the driver;

[0129] Step f: When the similarity between the active operation behavior vector of the driver operating the electric vehicle and the auxiliary target vector of the auxiliary execution scheme executed on the driver exceeds a threshold, stop executing the corresponding auxiliary execution scheme on the driver.

[0130] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A vehicle rollover early warning system based on state prediction and working condition clustering, characterized in that, include: The state prediction module is used to predict the motion state of an electric vehicle through a neural network. The operating condition clustering module is used to determine the current operating condition of electric vehicles through clustering. The model update module is used to update the vehicle roll dynamics model of electric vehicles in real time based on the predicted motion state and the determined current operating conditions. The rollover warning module is used to analyze vehicle rollover risk and provide early warning of vehicle rollover risk based on the real-time updated vehicle rollover dynamics model. The aforementioned vehicle rollover risk warning includes: Step a: Divide the critical time window into at least two consecutive decision nodes, and repeat steps b to f within each consecutive decision node; Step b: Real-time collection of cognitive load assessment data for electric vehicle drivers; Step c: Based on the cognitive load assessment, determine the driver's current cognitive load and match the corresponding solution reception sensitivity level; wherein, the solution reception sensitivity level is negatively correlated with the maximum intensity of auxiliary intervention that the driver can accept; Step d: Based on the driver's current sensitivity level to the scheme reception, dynamically select the corresponding level of the auxiliary execution scheme from the preset anti-rollover assist strategy library; wherein, the anti-rollover assist strategy library contains multi-level auxiliary execution schemes from the active control layer to the passive prompting layer, the level of each level of the auxiliary execution scheme is positively correlated with its corresponding scheme reception sensitivity level, and the auxiliary intervention intensity of each level of the auxiliary execution scheme decreases as the level increases; Step e: Implement a dynamic selection-assisted execution plan for the driver; Step f: When the similarity between the active operation behavior vector of the driver operating the electric vehicle and the auxiliary target vector of the auxiliary execution scheme executed on the driver exceeds a threshold, stop executing the corresponding auxiliary execution scheme on the driver.

2. The vehicle rollover early warning system based on state prediction and working condition clustering as described in claim 1, characterized in that, The state prediction module predicts the motion state of the electric vehicle through a neural network, including: A pre-trained recurrent neural network is used as the neural network; The neural network is fed with vehicle parameters of an electric vehicle, including at least vehicle wheel speed, steering wheel angle, yaw rate, longitudinal acceleration, lateral acceleration, body roll angle, suspension travel sensor data, and motor torque / speed. The predicted motion state of an electric vehicle is based on at least the predicted vehicle yaw rate, predicted vehicle lateral acceleration, predicted vehicle roll angle, and predicted tire vertical load, as output by a neural network.

3. The vehicle rollover early warning system based on state prediction and working condition clustering as described in claim 1, characterized in that, The operating condition clustering module determines the current operating condition of the electric vehicle through clustering, including: Based on electric vehicles, a feature vector set for clustering is constructed, including at least the current vehicle speed, steering wheel angle, yaw rate, longitudinal acceleration, lateral acceleration, body roll angle, tire slip ratio, and motor output torque. Based on the K-means algorithm, the working conditions are clustered according to the feature vector set used for clustering to obtain the current working conditions. The current operating conditions include at least the following categories: straight driving, steady-state steering, transient steering, acceleration, braking, driving conditions under different road surface adhesion coefficients, and different road types.

4. The vehicle rollover early warning system based on state prediction and working condition clustering as described in claim 1, characterized in that, The model update module updates the vehicle roll dynamics model of the electric vehicle in real time based on the predicted motion state and the determined current operating conditions, including: Based on the predicted motion state and the determined current operating conditions, adjust the corresponding parameters or coefficients in the vehicle roll dynamics model; Among them, the parameters or coefficients include at least: roll stiffness, roll damping, roll moment of inertia, and tire lateral stiffness.

5. The vehicle rollover early warning system based on state prediction and working condition clustering as described in claim 1, characterized in that, The rollover warning module analyzes vehicle rollover risk based on a real-time updated vehicle rollover dynamics model, including: For each preset vehicle rollover risk index, the index value of the real-time updated vehicle rollover dynamics model under that vehicle rollover risk index is calculated and compared with the corresponding preset index threshold. The vehicle rollover risk is determined based on the comparison result. Among them, vehicle rollover risk indicators include at least: roll angle, load transfer rate, and rollover coefficient.

6. The vehicle rollover early warning system based on state prediction and working condition clustering as described in claim 1, characterized in that, The steps for obtaining the safety critical time window include: The time window from when the risk of vehicle rollover is determined to the future target time is taken as the safety critical time window. Among them, the future target time is the smaller value between the physical rollover prevention limit time and the driver's cognitive buffer time; Among them, the physical rollover prevention limit time includes: the conservative time for an electric vehicle to roll over irreversibly in the future, determined based on the real-time updated vehicle roll dynamics model; Cognitive buffer time includes: the maximum permissible time for a driver to complete the cognitive perception and operational response to vehicle rollover risk, determined based on the driver's cognitive load during the time spent analyzing and determining the vehicle rollover risk.

7. The vehicle rollover early warning system based on state prediction and working condition clustering as described in claim 1, characterized in that, The cognitive load assessment is based on: the driver's eye gaze trajectory, changes in steering wheel grip pressure, and the history of operation response delays within a previously preset time period.

8. A vehicle rollover early warning method based on state prediction and working condition clustering, characterized in that, include: Predicting the motion state of electric vehicles using neural networks; The current operating condition of electric vehicles is determined by clustering; The vehicle roll dynamics model of the electric vehicle is updated in real time based on the predicted motion state and the determined current operating conditions. Based on the real-time updated vehicle rollover dynamics model, the risk of vehicle rollover is analyzed and a vehicle rollover risk warning is issued. The aforementioned vehicle rollover risk warning includes: Step a: Divide the critical time window into at least two consecutive decision nodes, and repeat steps b to f within each consecutive decision node; Step b: Real-time collection of cognitive load assessment data for electric vehicle drivers; Step c: Based on the cognitive load assessment, determine the driver's current cognitive load and match the corresponding solution reception sensitivity level; wherein, the solution reception sensitivity level is negatively correlated with the maximum intensity of auxiliary intervention that the driver can accept; Step d: Based on the driver's current sensitivity level to the scheme reception, dynamically select the corresponding level of the auxiliary execution scheme from the preset anti-rollover assist strategy library; wherein, the anti-rollover assist strategy library contains multi-level auxiliary execution schemes from the active control layer to the passive prompting layer, the level of each level of the auxiliary execution scheme is positively correlated with its corresponding scheme reception sensitivity level, and the auxiliary intervention intensity of each level of the auxiliary execution scheme decreases as the level increases; Step e: Implement a dynamic selection-assisted execution plan for the driver; Step f: When the similarity between the active operation behavior vector of the driver operating the electric vehicle and the auxiliary target vector of the auxiliary execution scheme executed on the driver exceeds a threshold, stop executing the corresponding auxiliary execution scheme on the driver.

Images