Curve rollover prevention monitoring method and system

Abstract

The invention relates to a curve rollover prevention monitoring method and system, and the method comprises the steps: enabling a vehicle-mounted terminal to collect the load distribution information and driving parameters of a vehicle, and uploading the information and parameters to a cloud server; the cloud server receives front road information provided by the road side unit; the cloud server calculates a rollover risk value of the vehicle on the front curve based on the load distribution information, the driving parameters and the road information; the cloud server determines a response strategy according to the rollover risk value and transmits the response strategy to the vehicle-mounted terminal; the vehicle-mounted terminal executes a response strategy before the vehicle arrives at the curve, and the response strategy at least comprises risk warning and / or vehicle speed control. Through cooperation of the vehicle-mounted terminal and the cloud server, the rollover risk can be accurately evaluated before the vehicle enters the curve, a risk alarm is given in advance, and / or the vehicle speed is directly and automatically controlled, so that enough response time is provided for a driver, and the possibility of rollover accidents is greatly reduced.

Description

Technical Field

[0001] The present invention relates to the field of safety assisted driving technology, and in particular to a curve anti-rollover monitoring method and system. Background Art

[0002] Due to the heavy load capacity of commercial vehicles, they are prone to body tilt when the load is unbalanced. When the load mass on the left and right sides of the vehicle is unbalanced, especially when turning at high speed, it is very easy to cause the vehicle to roll over, resulting in serious casualties and property losses.

[0003] The fundamental cause of vehicle rollovers is uneven loading on the left and right sides of the cargo box. This unevenness can result from an unbalanced distribution during initial loading or from cargo shifting during turns. Existing technologies primarily rely on real-time calculations to determine the safe speed and set a speed limit while the vehicle is cornering. However, this approach has significant drawbacks.

[0004] First, there's an inevitable delay in the vehicle system's response. In extreme roll scenarios, this delay may not be enough to offset the instantaneous rollover torque. Second, existing technologies largely rely on onboard equipment to independently perform calculations and judgments, requiring high computing power from the onboard controller and increasing system costs. Furthermore, due to a lack of predictive ability for road conditions ahead, vehicles often fail to provide early warnings and sufficient driver reaction time, reducing the effectiveness of rollover mitigation systems. Furthermore, existing technologies lack effective fault handling mechanisms in areas with poor road conditions or unstable network signals, making them susceptible to system failure due to sensor anomalies or communication interruptions. Summary of the Invention

[0005] The present invention discloses a curve anti-rollover monitoring method and system, aiming to solve the technical problems existing in the prior art.

[0006] The present invention adopts the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides a method for monitoring a curve to prevent rollover, the method comprising:

[0008] The on-board terminal collects vehicle load distribution information and driving parameters and uploads them to the cloud server;

[0009] The cloud server receives the road ahead information provided by the roadside unit;

[0010] The cloud server calculates the vehicle's rollover risk on the upcoming curve based on load distribution information, driving parameters, and road information.

[0011] The cloud server determines the response strategy based on the rollover risk value and transmits it to the vehicle terminal;

[0012] Before the vehicle reaches the curve, the on-board terminal executes a response strategy, which at least includes risk warning and / or vehicle speed control.

[0013] As a preferred technical solution, the method further includes a step of determining whether to activate rollover risk value calculation, specifically including:

[0014] Rollover risk calculation is not activated when the following conditions are detected: the vehicle is unloaded, is being loaded or unloaded, or is loaded but not in driving condition;

[0015] When it is detected that the vehicle is loaded with cargo and the driving speed exceeds a preset threshold, the rollover risk calculation is activated.

[0016] As a preferred technical solution, load distribution information is collected by multiple pressure sensors arranged on the left and right sides of the vehicle; driving parameters include at least vehicle speed, vehicle position, lateral acceleration and yaw angular velocity.

[0017] As a preferred technical solution, the method further includes a step of detecting an abnormality of the pressure sensor, specifically including:

[0018] Monitor the vehicle's acceleration and the output values ​​of multiple pressure sensors;

[0019] When the vehicle acceleration is less than or equal to the preset acceleration threshold, and the output value change rate of any pressure sensor is detected to be greater than the preset pressure change threshold, the pressure sensor is determined to be abnormal;

[0020] The output value of the abnormal pressure sensor is replaced by the output average of the remaining pressure sensors on the side where the abnormal pressure sensor is located, and is used to calculate the load distribution information.

[0021] As a preferred technical solution, the road ahead information includes curve radius, road surface friction coefficient and curve position information.

[0022] As a preferred technical solution, the rollover risk value is determined by a weighted combination of a turning risk factor, a sharp turn risk factor, and a lateral load risk factor;

[0023] Among them, the turning risk factor is determined based on the load imbalance rate, vehicle speed, curve radius and turning coefficient; the sharp turn risk factor is determined based on vehicle speed, yaw angular velocity and emergency steering coefficient; and the lateral load risk factor is determined based on lateral acceleration and lateral overload coefficient.

[0024] As a preferred technical solution, the load imbalance rate is the ratio of the pressure difference between the left and right sides of the vehicle to the total pressure value of the vehicle;

[0025] The pressure difference between the left and right sides of the vehicle is obtained by the difference between the total pressure value of multiple pressure sensors on the left side and the total pressure value of multiple pressure sensors on the right side; the total pressure value of the vehicle is the sum of the total pressure value on the left side and the total pressure value on the right side.

[0026] As a preferred technical solution, the turning coefficient, emergency steering coefficient and lateral overload coefficient adopt basic weight values ​​and are dynamically adjusted according to the road friction coefficient. The dynamic adjustment includes at least:

[0027] When the road friction coefficient changes, the weight values ​​of each coefficient are adjusted according to the preset rules;

[0028] And, adaptive optimization through historical rollover risk data and machine learning algorithms.

[0029] As a preferred technical solution, the response strategy is determined according to the size of the rollover risk value;

[0030] When the rollover risk value is less than the first risk threshold, executing a weak risk prompt response;

[0031] When the rollover risk value is greater than or equal to the first risk threshold and less than the second risk threshold, executing a medium risk warning response;

[0032] When the rollover risk value is greater than or equal to a second risk threshold, executing a strong risk intervention response;

[0033] Response strategies include but are not limited to driving risk warnings, maximum safe speed push, and automatic deceleration intervention.

[0034] As a preferred technical solution, the maximum safe vehicle speed is determined based on the load imbalance rate, curve radius and road friction coefficient.

[0035] As a preferred technical solution, the method further includes the steps of determining and handling network failures, specifically including:

[0036] When it is detected that the network communication delay between the cloud server and the vehicle terminal exceeds the preset delay threshold, the rollover risk value is continued to be calculated using the load distribution information and driving parameters valid at the last moment;

[0037] When the network communication between the cloud server and the vehicle terminal is detected to be interrupted and exceeds the preset interruption time limit, the vehicle terminal calls the locally stored historical data to perform rollover risk value calculation and response strategy determination.

[0038] On the other hand, an embodiment of the present invention further provides a curve anti-rollover monitoring system, comprising a vehicle-mounted terminal and a cloud server;

[0039] The vehicle-mounted terminal is used to collect vehicle load distribution information and driving parameters and upload them to the cloud server;

[0040] The cloud server is used to receive the road ahead information provided by the roadside unit;

[0041] The cloud server is further configured to calculate a rollover risk value of the vehicle on a forward curve based on the load distribution information, the driving parameters, and the road information;

[0042] The cloud server is further configured to determine a response strategy based on the rollover risk value and transmit the result to the vehicle-mounted terminal;

[0043] The vehicle-mounted terminal is further configured to execute the response strategy before the vehicle reaches a curve, wherein the response strategy at least includes risk warning and / or vehicle speed control.

[0044] One embodiment of the above invention has the following advantages or beneficial effects:

[0045] The embodiment of the present invention mainly provides a curve anti-rollover monitoring method, which combines vehicle load distribution information with driving parameters and front road information to achieve early prediction and active prevention and control of vehicle rollover risks.

[0046] Specifically, traditional technologies typically calculate and limit safe speeds only while the vehicle is cornering. This results in significant system response delays and makes it difficult to cope with the instantaneous rollover torque in extreme rollover scenarios. However, the embodiments of the present invention, through the coordinated collaboration of an onboard terminal and a cloud server, can accurately assess rollover risk before the vehicle enters a curve, issuing a risk warning in advance and / or automatically controlling the vehicle's speed, thereby providing the driver with ample reaction time and significantly reducing the likelihood of a rollover accident.

[0047] In addition, the embodiment of the present invention places computing-intensive tasks on the cloud server for execution, and the on-board terminal only stores necessary result data, effectively reducing the computing burden and cost of the vehicle-side controller. At the same time, through a multi-level response strategy, it realizes gradient prevention and control from weak risk warnings to strong risk interventions, ensuring safety while taking into account the driving experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments, which constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions are provided to explain the present invention and do not constitute an undue limitation of the present invention. In the drawings:

[0049] Figure 1 A flowchart of the steps of a curve anti-rollover monitoring method provided by one embodiment of the present invention;

[0050] Figure 2 A schematic diagram of information interaction of a curve anti-rollover monitoring system provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0051] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with specific embodiments of the present invention and corresponding drawings. In the description of the present invention, it should be noted that the term "or" is generally used in the sense of including "and / or" unless the content clearly indicates otherwise.

[0052] Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0053] refer to Figure 1 、 Figure 2 To address the deficiencies of the prior art, the present invention provides a method for monitoring a rollover prevention on a curved road. In a preferred embodiment, the method includes at least steps S110 to S150, as follows:

[0054] In step S110 , the vehicle terminal collects vehicle load distribution information and driving parameters and uploads them to the cloud server.

[0055] In a preferred embodiment, the load distribution information is collected by a plurality of pressure sensors disposed on the left and right sides of the vehicle.

[0056] Preferably, at least three pressure film sensors are arranged on the left and right sides of the vehicle's cargo platform, and the pressure value obtained by the pressure sensor is P i , where i = 1, 2, 3…6. Specifically, the pressure values ​​obtained by the left pressure sensor are recorded as P1, P2, and P3, and the pressure values ​​obtained by the right pressure sensor are recorded as P4, P5, and P6. The pressure sensors are evenly spaced along the longitudinal direction to form a bilaterally symmetrical pressure monitoring network. Preferably, the pressure sensor adopts a high-frequency sampling period of 10ms to ensure the real-time and accuracy of the load data. By arranging multiple pressure sensors on the left and right sides of the vehicle, the impact of aging phenomena such as performance degradation, sensitivity reduction, and zero drift caused by long-term use of the sensors on the load calculation can be effectively offset, significantly improving the measurement accuracy of the load distribution information and the stability of the system.

[0057] Specifically, the vehicle terminal receives the digital signals of the six pressure sensors through the CAN bus, and calculates the total pressure value P on the left side after filtering. 左 =P1+P2+P3 and the total pressure value on the right side P 右 =P4+P5+P6, and the total load P is obtained accordingly 总 =P 左 +P右 , and the load distribution state. Preferably, in order to eliminate the drift error and temperature effect of the pressure sensor, a Kalman filter algorithm is used to correct the raw data to ensure the reliability of the load distribution information.

[0058] In a preferred embodiment, driving parameters include at least vehicle speed, vehicle position, lateral acceleration, and yaw rate. Specifically, these parameters are acquired through an integrated vehicle status monitoring system. Vehicle speed information is calculated by integrating high-resolution wheel speed sensors with the vehicle body motion control unit; vehicle position information is acquired through a positioning module; lateral acceleration is measured by a triaxial accelerometer; and yaw rate is monitored in real time by a gyroscope. All sensors are connected to the vehicle terminal via standardized interfaces for synchronized data acquisition.

[0059] The vehicle-mounted terminal preferably pre-processes the collected load distribution information and driving parameters, including outlier removal, data normalization, and feature extraction, before uploading the data to the cloud server via the vehicle-mounted communication module. Furthermore, the vehicle-mounted terminal preferably also performs a local backup of the uploaded data to mitigate network interruptions.

[0060] In a preferred embodiment, step S110 further includes a step S111 of detecting an abnormality of the pressure sensor, which is specifically implemented as follows:

[0061] Monitor the vehicle's overall acceleration and the output values ​​of multiple pressure sensors; when the vehicle's overall acceleration is less than or equal to a preset acceleration threshold, and when the output value change rate of any pressure sensor is detected to be greater than a preset pressure change threshold, the pressure sensor is determined to be abnormal; the output value of the abnormal pressure sensor is replaced by the output mean of the remaining pressure sensors on the side where the abnormal pressure sensor is located, and is used to calculate the load distribution information.

[0062] Specifically, the on-board terminal monitors the vehicle's acceleration vector in real time through a three-axis accelerometer installed on the vehicle body, and calculates the modulus value |a| of the vehicle's acceleration. At the same time, the on-board terminal records the pressure output value sequence of each pressure sensor within n consecutive sampling cycles.

[0063] The onboard terminal compares the modulus of the vehicle's acceleration, |a|, with a preset acceleration threshold, a0. When |a| ≤ a0, the output of each pressure sensor should theoretically remain relatively stable. Simultaneously, the onboard terminal calculates the rate of change of each pressure sensor's output within a time window. If the rate of change of any pressure sensor exceeds a preset pressure change threshold, the sensor is deemed abnormal. This threshold is set based on the vehicle's load characteristics.

[0064] When the vehicle terminal determines that a pressure sensor is abnormal, it immediately activates the pressure sensor data compensation mechanism. For an abnormal pressure sensor on the left side, the vehicle terminal uses the arithmetic mean of the output values ​​of the remaining normal left pressure sensors as the replacement value. Similarly, for an abnormal pressure sensor on the right side, the arithmetic mean of the output values ​​of the remaining normal right pressure sensors is used as the replacement value. If multiple pressure sensors on the same side are abnormal at the same time, the vehicle terminal will estimate a reasonable value based on historical data models or trigger a higher-level system warning.

[0065] Specifically, the abnormality detection and compensation process of the pressure sensor is automatically executed in each data collection cycle, and the compensated pressure sensor data is used for subsequent load distribution information calculations to ensure that even in the event of single or multiple pressure sensor failures, the on-board terminal can still maintain the necessary functional stability and data reliability, thereby providing accurate basic parameters for rollover risk assessment.

[0066] In step S120 , the cloud server receives the road ahead information provided by the roadside unit.

[0067] In a preferred embodiment, the forward road information includes curve radius, road surface friction coefficient, and curve position information.

[0068] Preferably, the roadside units (RSUs) in this step include, but are not limited to, intelligent traffic signal systems, roadside sensing equipment, intelligent road infrastructure, and regional traffic control centers. These facilities are connected to the traffic infrastructure cloud network via wired or wireless means, forming a roadside information collection network covering major roads. Specifically, the RSUs collect and process data on road geometry, road conditions, and traffic environment within their monitoring areas, regularly updating the road information database on the cloud server.

[0069] In a preferred embodiment, the cloud server spatially matches the received vehicle location information with the curve location information in the road network database to identify all possible curves within 1-5 kilometers ahead of the vehicle. For each identified curve, the cloud server will extract its complete geometric parameters and road condition information, and sort them according to the estimated time the vehicle will arrive at the curve.

[0070] The cloud server preferably updates road information in real time. For key parameters, such as the road friction coefficient, the update frequency can be increased during unusual weather conditions (e.g., rain, ice, etc.). Furthermore, the cloud server maintains a database linking historical vehicle driving records with road information to optimize the accuracy of curve recognition and the relevance of response strategies. By accurately matching and predicting this position information, the cloud server can provide sufficient warning time before the vehicle actually reaches a curve, facilitating subsequent rollover risk assessment and implementation of preventive measures.

[0071] In step S130 , the cloud server calculates the rollover risk value of the vehicle on the upcoming curve based on the load distribution information, driving parameters, and road information.

[0072] In a preferred embodiment, while the cloud server can identify all curves within a predetermined distance ahead of the vehicle, it preferably employs a curve-by-curve approach. That is, at any given moment, rollover risk calculation is performed only for the first curve the vehicle will traverse. For example, for curves with large radii or low vehicle speeds, the safety risk is extremely low, and rollover risk calculation is unnecessary. Furthermore, as the vehicle negotiates a curve, centrifugal force and vehicle body tilt temporarily alter the load distribution on both sides of the vehicle. This change can significantly alter the initial load distribution. Therefore, rollover risk assessments for subsequent curves must be based on the most recent load distribution information, rather than historical data from before the curve was traversed.

[0073] In a preferred embodiment, when there are several consecutive curves in front of the vehicle, the relative relationship between the vehicle position and the current target curve is analyzed to determine whether the vehicle has completed the curve. Once it is confirmed that the vehicle has passed the current curve, the cloud server will receive the latest load distribution information collected by the on-board terminal and immediately start the rollover risk assessment process for the next curve that meets the calculation conditions.

[0074] In a preferred embodiment, before executing the rollover risk value calculation, a step of determining whether to activate the rollover risk value calculation is included to identify key scenarios where rollover risk calculation needs to be performed and avoid unnecessary calculations in low-risk or non-relevant operating conditions. The determination step specifically includes:

[0075] The rollover risk calculation will not be activated when the following situations are detected: the vehicle is not loaded with cargo, is being loaded or unloaded, or is loaded but not driving; the rollover risk calculation will be activated when it is detected that the vehicle is loaded with cargo and the driving speed exceeds the preset threshold.

[0076] In a preferred embodiment, according to the vehicle's own weight characteristics, the preset load judgment threshold is set to 500 kg; the load change rate is defined as P' 总 , which represents the total load P 总 The derivative with respect to time t, that is, P' 总 =f'(t), when P' 总 = 0, indicating that the vehicle load remains stable; when P' 总 When ≠0, it means that the vehicle load is changing, which usually corresponds to the loading and unloading process or unstable loading state.

[0077] Specifically, when P' 总 =0, and P 总When the load is ≤500kg, the vehicle is considered to be unladen. In this case, the rollover risk can be ignored and the rollover risk calculation process is not activated.

[0078] When P' 总 ≠0, and P 总 When the load is greater than 500kg, the vehicle is judged to be loading or unloading, and the load distribution is in a dynamic change process, so the rollover risk calculation process is not activated.

[0079] When P' 总 =0,P 总 If the vehicle's weight is greater than 500 kg and its speed is ≤ 5 km / h, the vehicle is deemed loaded but parked or moving at a low speed, and the rollover risk calculation process is not activated. Specifically, the 5 km / h speed threshold serves as the minimum standard for determining driving status; speeds below this threshold are not included in the rollover risk assessment. Persons skilled in the art may adjust this speed threshold based on actual driving needs.

[0080] When P' 总 =0,P 总 When the vehicle's load is greater than 500kg and its speed is greater than 5km / h, the vehicle is considered to be carrying cargo. The rollover risk calculation function is activated, performing a complete risk assessment of the curve ahead. In this state, the cloud server calculates the vehicle's rollover risk as it negotiates the curve ahead based on load distribution information, driving parameters, and road conditions.

[0081] In a preferred embodiment, the cloud server is preset with a rollover risk value prediction model, which is calculated by a weighted combination of a turning risk factor, a sharp turn risk factor and a lateral load risk factor to obtain a rollover risk value. The turning risk factor is determined based on the load imbalance rate, vehicle speed, curve radius and turning coefficient; the sharp turn risk factor is determined based on the vehicle speed, yaw angular velocity and emergency steering coefficient; and the lateral load risk factor is determined based on the lateral acceleration and the lateral overload coefficient.

[0082] Preferably, the turning risk factor is K1*η*V 2 / R, the risk factor for sharp turn is K2*V*ω, and the risk factor for lateral load is K3*a y / a ymax , so the rollover risk value prediction model is:

[0083] P=(K1*η*V 2 / R)+K2*V*ω+K3*a y / a ymax

[0084] Among them, P represents the comprehensive rollover risk value, which is dimensionless and ranges from 0 to 1. The larger the value, the higher the rollover risk. K1 is the turning coefficient, which is used to adjust the weight of the turning risk factor. K2 is the emergency steering coefficient, which is used to adjust the weight of the emergency steering risk factor. K3 is the lateral overload coefficient, which is used to adjust the weight of the lateral load risk factor. η is the load imbalance rate, which reflects the degree of imbalance in the load distribution on the left and right sides of the vehicle. V is the vehicle speed. R is the curve radius. ω is the yaw rate, which represents the angular velocity of the vehicle around the vertical axis. y is the lateral acceleration, which indicates the acceleration of the vehicle in the lateral direction; α ymax is the maximum lateral acceleration, which is the theoretical maximum safe lateral acceleration determined based on the vehicle type and road friction coefficient.

[0085] Preferably, the load imbalance ratio η is a key parameter for measuring the unevenness of the load distribution on the left and right sides of the vehicle. It is calculated based on the ratio of the pressure difference measured by the left and right pressure sensors to the total pressure. Specifically, the load imbalance ratio η is:

[0086]

[0087] Among them, P i (i=1,2,3) represents the output values ​​of the three pressure sensors on the left, p i+3 (i=1,2,3) represents the output values ​​of the three pressure sensors on the right side, corresponding to the left sensor. The numerator of the above formula calculates the sum of the absolute differences between the output values ​​of the corresponding left and right sensors, indicating the degree of load asymmetry; the denominator is the sum of the output values ​​of all six sensors, representing the total vehicle load. Through normalization of the above formula, the value of η typically ranges from 0 to 1, with η=0 indicating a perfectly balanced load distribution. Larger values ​​of η indicate a more unbalanced load distribution and a higher risk of rollover.

[0088] In a preferred embodiment, the turning coefficient K1, the emergency steering coefficient K2 and the lateral overload coefficient K3 adopt basic weight values, which are all determined through actual vehicle testing and data analysis, and are dynamically adjusted according to the road friction coefficient to improve the accuracy of rollover risk assessment.

[0089] Preferably, the dynamic adjustment of K1, K2, and K3 includes at least:

[0090] When the road friction coefficient changes, the weight values ​​of each coefficient are adjusted according to preset rules, and adaptive optimization is performed through historical rollover risk data and machine learning algorithms.

[0091] In a preferred embodiment, under low-friction road conditions, such as rain, snow, or icy roads, the weight of the turning coefficient K1 is increased, while the weight of the emergency steering coefficient K2 is appropriately reduced to reflect the vehicle's greater sensitivity to lateral forces under low-friction conditions. The specific adjustment range is determined by the change in friction coefficient and the vehicle speed range, ensuring the risk assessment model's responsiveness to changing road conditions.

[0092] In a preferred embodiment, the cloud server continuously collects and stores vehicle driving data, load distribution information, and corresponding rollover risk assessment results under different operating conditions to build a rollover risk knowledge base. By deploying machine learning algorithms such as gradient boosting decision trees, the cloud server regularly analyzes risk assessment deviations in historical data, identifies differences between model parameters and actual risk conditions, and uses this to select the optimal weight coefficient combination.

[0093] In step S140 , the cloud server determines a response strategy based on the rollover risk value and transmits it to the vehicle terminal.

[0094] In a preferred embodiment, the response strategy is determined based on the size of the rollover risk value, ensuring that the vehicle can implement corresponding safety measures for different risk levels.

[0095] When the rollover risk value is less than the first risk threshold, the vehicle is determined to be in a low-risk state, and a low-risk warning response or no response is executed. Preferably, the first risk threshold can be set to 0.4, at which point the vehicle terminal can alert the driver through non-intrusive information prompts. For example, the vehicle's display screen displays information about the upcoming curve as a green icon, accompanied by text indicating the current load status and curve characteristics; the vehicle's audio system emits a low-intensity warning tone to alert the driver to the road conditions ahead; and the instrument panel activates a low-risk indicator light to visually draw the driver's attention. Optionally, since the rollover risk is low at this point, the risk prompt may be ignored.

[0096] When the rollover risk value is greater than or equal to the first risk threshold and less than the second risk threshold, the vehicle is determined to be in a medium-risk state and a medium-risk warning response is implemented. Preferably, the second risk threshold can be set to 0.6, at which point more specific warning measures are implemented. The medium-risk warning response includes: highlighting the upcoming curve risk information in yellow on the on-board display screen and pushing a comparison of the current vehicle speed with the maximum safe speed calculated by the cloud server; emitting a moderate-intensity warning sound through the on-board audio system; activating the medium-risk warning light on the instrument panel, flashing a yellow cursor to alert the driver; and triggering slight vibration feedback on the steering wheel to tactilely remind the driver to slow down. Simultaneously, the system pushes the maximum safe speed to the on-board terminal for the driver's reference.

[0097] When the rollover risk value is greater than or equal to the second risk threshold, the vehicle is determined to be in a high-risk state, and a high-risk intervention response is executed. Preferably, the high-risk intervention response includes: highlighting the high-risk warning message in red on the vehicle's display screen and forcibly displaying the specific value by which the current vehicle speed exceeds the safe speed; emitting a high-intensity alarm sound through the vehicle's audio system; activating the high-risk warning light on the instrument panel to rapidly flash a red cursor to alert the driver; triggering strong vibration feedback in the steering wheel; and, at the same time, the vehicle terminal sends a deceleration command to the vehicle's power control unit, automatically reducing the vehicle speed to below the safe speed by limiting throttle response, actively downshifting, and moderate braking. In extreme cases, the vehicle terminal can also activate emergency brake assist to ensure the vehicle safely navigates high-risk curves.

[0098] In a preferred embodiment, the maximum safe vehicle speed is determined based on the load imbalance rate, the curve radius and the road friction coefficient. Specifically, the maximum safe vehicle speed V max The calculation formula is:

[0099]

[0100] Where μ is the road friction coefficient, which reflects the adhesion of the road surface. Its value range is usually 0.1 to 0.9, where 0.1 represents a slippery road surface caused by rain or snow, and 0.9 represents a dry asphalt road surface. g is the gravitational acceleration constant, which is 9.8 m / s. 2 ; R is the curve radius.

[0101] Specifically, the term (1-η) in the above formula reflects the weakening effect of load imbalance on safe vehicle speed. When the load distribution is completely balanced (η=0), the vehicle can reach the theoretical maximum safe passing speed. As the degree of load distribution imbalance increases, the η value increases, resulting in a corresponding decrease in the maximum safe vehicle speed.

[0102] In step S150 , the vehicle-mounted terminal executes a response strategy before the vehicle reaches a curve, and the response strategy at least includes risk warning and / or vehicle speed control.

[0103] In a preferred embodiment, the baseline response distance is one kilometer. This means that under normal driving conditions, the corresponding level of response strategy begins one kilometer from the vehicle's distance to a risky curve. This distance is based on the average driver perception-decision-action time (typically 3-5 seconds) and the deceleration performance characteristics of commercial vehicles under different load conditions, ensuring that the driver has sufficient time to take necessary deceleration measures.

[0104] In another more preferred embodiment, the cloud server or vehicle terminal can further optimize the response distance by taking into account factors such as road slope, sight distance before a curve, and vehicle load. The response distance can be appropriately increased on downhill roads, at high speeds, or in situations with limited sight distance. For fully loaded vehicles, the response distance can also be extended to ensure a safety margin, given their reduced deceleration performance.

[0105] In a preferred embodiment, for weak risk prompt responses, non-invasive visual and sound prompts are used to alert the driver; for medium risk warning responses, the on-board terminal activates a clearer warning mechanism when it reaches the response distance from the curve, including multi-modal feedback such as a yellow warning icon on the display, intermittent warning sounds, and slight vibration of the steering wheel.

[0106] In a preferred embodiment, for high-risk intervention responses, the vehicle terminal immediately initiates a full range of warning signals upon reaching the curve's response distance, including a red warning icon on the display, a continuous high-frequency alarm tone, and strong steering wheel vibration. Simultaneously, the cloud server automatically sends intervention instructions to the vehicle control unit, determining the required deceleration rate.

[0107] Specifically, the cloud server can calculate the theoretical deceleration required to achieve safe deceleration based on the current vehicle speed, maximum safe speed and available deceleration distance. Then, the cloud server combines this theoretical value with factors such as vehicle type, load status and road adhesion conditions to determine the actual executable deceleration instruction.

[0108] Preferably, in special circumstances, such as when the on-board terminal detects that the driver has not responded and the distance to the curve is not enough to complete normal deceleration, the on-board terminal will activate the emergency deceleration mode, directly intervene in the deceleration process, and apply a greater deceleration to ensure vehicle safety.

[0109] In a preferred embodiment, the aforementioned curve rollover prevention monitoring method further includes steps for determining and handling network failures, ensuring that the rollover risk monitoring function can maintain basic operation and ensure vehicle driving safety even when communication conditions are unstable or interrupted. Specifically, it includes:

[0110] When it is detected that the network communication delay between the cloud server and the vehicle terminal exceeds the preset delay threshold, the vehicle terminal continues to calculate the rollover risk value using the load distribution information and driving parameters valid at the last moment;

[0111] When the network communication between the cloud server and the vehicle terminal is detected to be interrupted and exceeds the preset interruption time limit, the vehicle terminal calls the locally stored historical data to perform rollover risk value calculation and response strategy determination.

[0112] Specifically, the on-board terminal evaluates the network connection status in real time by monitoring the communication quality parameters between the terminal and the cloud server, including but not limited to the round-trip time of data packets, packet loss rate and signal strength. When it is detected that the network communication delay exceeds the preset threshold, the mild network anomaly handling process is triggered. In this state, the on-board terminal no longer waits for the real-time calculation results from the cloud, but starts the local rollover risk calculation based on the effective load distribution information successfully received at the last moment, the current vehicle driving parameters and the locally stored road data, and continues to perform risk assessment to ensure the continuous operation capability of the on-board terminal in the case of slight network fluctuations. Preferably, the on-board terminal can adopt a data prediction algorithm, such as a Kalman filter or a support vector regression algorithm, to make a reasonable estimate of the current load status by analyzing the load change trend in the recent period, thereby reducing the risk assessment error that may be caused by data lag.

[0113] In a preferred embodiment, when the vehicle terminal detects a complete network communication interruption that lasts longer than a preset interruption time limit, a severe network anomaly handling process is triggered. At this point, the vehicle terminal activates local emergency mode, calls upon historical driving data and load patterns stored in a local database, and performs an approximate calculation of rollover risk based on the vehicle's current state. Specifically, the local database stores the vehicle's typical load distribution patterns under different load conditions over the past 30 days, as well as historical risk assessment results for the corresponding road sections. Using a pattern matching algorithm, the historical pattern most similar to the current state is selected as the basis for calculation.

[0114] Preferably, the local emergency mode adopts a conservative strategy for risk assessment. That is, as uncertainty increases, the vehicle terminal tends to appropriately increase the risk level to increase the safety margin. At the same time, the vehicle terminal dynamically adjusts the risk threshold based on the duration of the network outage. The longer the outage, the more conservative the risk assessment.

[0115] Preferably, when network communication is re-established, the on-board terminal immediately sends the vehicle status data collected during the offline period and the risk assessment results of the local calculation to the cloud server. The cloud server analyzes and compares these data to evaluate the accuracy of the local calculation, and sends the latest model parameters and algorithm updates to the on-board terminal to optimize the performance of the local computing module.

[0116] Compared with the existing technology, the embodiments of the present invention, through the coordinated cooperation between the vehicle-mounted terminal and the cloud server, can accurately assess the rollover risk before the vehicle enters a curve, and issue a risk warning in advance and / or automatically control the vehicle speed directly, thereby providing the driver with sufficient reaction time and significantly reducing the possibility of rollover accidents.

[0117] In addition, the embodiment of the present invention places computing-intensive tasks on the cloud server for execution, and the on-board terminal only stores necessary result data, effectively reducing the computing burden and cost of the vehicle-side controller. At the same time, through a multi-level response strategy, it realizes gradient prevention and control from weak risk warnings to strong risk interventions, ensuring safety while taking into account the driving experience.

[0118] refer to Figure 2 In one embodiment of the present invention, the embodiment of the present invention also provides a curve anti-rollover monitoring system, including a vehicle-mounted terminal and a cloud server; wherein the vehicle-mounted terminal is used to collect vehicle load distribution information and driving parameters, and upload them to the cloud server; the cloud server is used to receive the front road information provided by the roadside unit, and based on the load distribution information, driving parameters and road information, calculate the rollover risk value of the vehicle in the front curve; the cloud server is further used to determine a response strategy according to the rollover risk value, and transmit it to the vehicle-mounted terminal; the vehicle-mounted terminal is also used to execute the response strategy before the vehicle reaches the curve, and the response strategy at least includes risk warning and / or vehicle speed control.

[0119] In the system provided in this embodiment, through the collaborative work of the vehicle terminal and the cloud server, forward-looking assessment and active prevention of the risk of vehicle rollover on curves can be achieved.

[0120] The vehicle-mounted terminal is preferably installed on large vehicles such as commercial trucks and buses. It is particularly suitable for transport vehicles with frequent load fluctuations, high centers of gravity, and complex road conditions. The vehicle-mounted terminal is equipped with a multi-channel pressure sensor array, located at key support points on the vehicle chassis, to collect real-time information on the vehicle's load distribution. It also integrates driving status monitoring equipment such as vehicle speed sensors, yaw rate sensors, and lateral acceleration sensors to comprehensively obtain vehicle dynamic parameters.

[0121] Preferably, the cloud server is deployed on a highly reliable cloud computing platform and is equipped with corresponding computing power and data processing capabilities.

[0122] It should be noted that the curve anti-rollover monitoring system provided by the embodiment of the present invention and the aforementioned curve anti-rollover monitoring method are derived from the same inventive concept. The vehicle-mounted terminal and the cloud server in the embodiment of this system respectively assume the responsibilities of the execution of each step in the method embodiment. For example, the vehicle-mounted terminal is responsible for implementing the data collection, uploading and response execution links in the method; the cloud server is responsible for implementing the data reception, risk calculation and strategy generation links in the method. The two work closely together through a wireless communication network to form a closed-loop rollover risk monitoring and prevention mechanism.

[0123] Specifically, the system and method embodiments share a common set of technical parameters and algorithmic models. Core technical content, including the load imbalance ratio calculation method, rollover risk assessment model, maximum safe speed determination formula, and dynamic response distance adjustment mechanism, detailed previously, is fully implemented in the system embodiment. Similarly, key technical points discussed previously, such as the network anomaly handling strategy, data prediction algorithm selection, and hierarchical response mechanism, are also supported by corresponding functional modules in the system architecture and will not be detailed here.

[0124] Although example embodiments have been described herein with reference to the accompanying drawings, it should be understood that the above example embodiments are merely illustrative and are not intended to limit the scope of the present application. Various changes and modifications may be made therein by those skilled in the art without departing from the scope and spirit of the present application. All such changes and modifications are intended to be included within the scope of the present application as required by the appended claims.

[0125] In the description provided herein, a large number of specific details are described. However, it is understood that the embodiments of the present application can be practiced without these specific details. In some instances, well-known methods, structures, and techniques are not shown in detail so as not to obscure the understanding of this description.

[0126] Similarly, it should be understood that in order to streamline the present application and aid in understanding one or more of the various inventive aspects, in the description of the exemplary embodiments of the present application, the various features of the present application are sometimes grouped together into a single embodiment, figure, or description thereof. However, this approach of the present application should not be interpreted as reflecting the intention that the application claimed for protection requires more features than those explicitly recited in each claim. More precisely, as reflected in the corresponding claims, the inventive point is that the corresponding technical problem can be solved with fewer features than all the features of a single disclosed embodiment. Therefore, the claims following the detailed description are hereby expressly incorporated into the detailed description, with each claim itself serving as a separate embodiment of the present application.

[0127] Those skilled in the art will understand that, except where mutually exclusive, all features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all processes or units of any method or apparatus disclosed herein may be combined in any combination. Unless expressly stated otherwise, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) may be replaced by an alternative feature that provides the same, equivalent, or similar purpose.

[0128] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which are protected by the present invention.

Claims

1. A curve anti-rollover monitoring method, characterized in that: The method comprises: The on-board terminal collects vehicle load distribution information and driving parameters and uploads them to the cloud server; The cloud server receives the road ahead information provided by the roadside unit; The cloud server calculates a rollover risk value of the vehicle on a forward curve based on the load distribution information, the driving parameters, and the road information; The cloud server determines a response strategy according to the rollover risk value and transmits the response strategy to the vehicle terminal; The vehicle-mounted terminal executes the response strategy before the vehicle reaches a curve, and the response strategy at least includes risk warning and / or vehicle speed control.

2. The curve anti-rollover monitoring method according to claim 1, characterized in that: The method further includes a step of determining whether to activate rollover risk value calculation, specifically comprising: Rollover risk calculation is not activated when the following conditions are detected: the vehicle is unloaded, is being loaded or unloaded, or is loaded but not in driving condition; When it is detected that the vehicle is loaded with cargo and the driving speed exceeds a preset threshold, the rollover risk calculation is activated.

3. The curve anti-rollover monitoring method according to claim 1, characterized in that: The load distribution information is collected by multiple pressure sensors arranged on the left and right sides of the vehicle; the driving parameters include at least vehicle speed, vehicle position, lateral acceleration and yaw angular velocity.

4. The curve anti-rollover monitoring method according to claim 3, characterized in that: The method further includes a step of detecting an abnormality of the pressure sensor, specifically comprising: Monitor the vehicle's acceleration and the output values ​​of multiple pressure sensors; When the vehicle acceleration is less than or equal to a preset acceleration threshold, and when it is detected that the output value change rate of any pressure sensor is greater than a preset pressure change threshold, the pressure sensor is determined to be abnormal; The output value of the abnormal pressure sensor is replaced by the output average of the remaining pressure sensors on the side where the abnormal pressure sensor is located, and is used to calculate the load distribution information.

5. The curve anti-rollover monitoring method according to claim 3, characterized in that: The forward road information includes curve radius, road surface friction coefficient and curve position information.

6. The curve anti-rollover monitoring method according to claim 5, characterized in that: The rollover risk value is determined by a weighted combination of a turning risk factor, a sharp turn risk factor, and a lateral load risk factor; Among them, the turning risk factor is determined based on the load imbalance rate, vehicle speed, curve radius and turning coefficient; the sharp turn risk factor is determined based on vehicle speed, yaw angular velocity and emergency steering coefficient; and the lateral load risk factor is determined based on lateral acceleration and lateral overload coefficient.

7. The curve anti-rollover monitoring method according to claim 6, characterized in that: The load imbalance rate is the ratio of the pressure difference between the left and right sides of the vehicle to the total pressure value of the vehicle; The pressure difference between the left and right sides of the vehicle is obtained by the difference between the total pressure value of multiple pressure sensors on the left side and the total pressure value of multiple pressure sensors on the right side; the total pressure value of the vehicle is the sum of the total pressure value on the left side and the total pressure value on the right side.

8. The curve anti-rollover monitoring method according to claim 6, characterized in that: The turning coefficient, the emergency steering coefficient, and the lateral overload coefficient use basic weight values ​​and are dynamically adjusted according to the road friction coefficient. The dynamic adjustment includes at least: When the road friction coefficient changes, the weight values ​​of each coefficient are adjusted according to the preset rules; And, adaptive optimization through historical rollover risk data and machine learning algorithms.

9. The curve anti-rollover monitoring method according to claim 7, characterized in that: The response strategy is determined according to the size of the rollover risk value; When the rollover risk value is less than a first risk threshold, executing a weak risk prompt response; When the rollover risk value is greater than or equal to the first risk threshold and less than a second risk threshold, executing a medium risk warning response; When the rollover risk value is greater than or equal to the second risk threshold, executing a strong risk intervention response; The response strategy includes but is not limited to driving risk prompts, maximum safe speed push, and automatic deceleration intervention.

10. The curve anti-rollover monitoring method according to claim 9, characterized in that: The maximum safe vehicle speed is determined based on the load imbalance rate, the curve radius and the road friction coefficient.

11. The curve anti-rollover monitoring method according to claim 1, characterized in that: The method also includes the steps of determining and handling network failures, specifically including: When it is detected that the network communication delay between the cloud server and the vehicle terminal exceeds a preset delay threshold, the rollover risk value is continued to be calculated using the load distribution information and the driving parameters valid at the last moment; When it is detected that the network communication between the cloud server and the vehicle-mounted terminal is interrupted and exceeds a preset interruption time limit, the vehicle-mounted terminal calls locally stored historical data to perform rollover risk value calculation and response strategy determination.

12. A curve anti-rollover monitoring system, characterized in that: Including vehicle terminal and cloud server; The vehicle-mounted terminal is used to collect vehicle load distribution information and driving parameters and upload them to the cloud server; The cloud server is used to receive the road ahead information provided by the roadside unit; The cloud server is further configured to calculate a rollover risk value of the vehicle on a forward curve based on the load distribution information, the driving parameters, and the road information; The cloud server is further configured to determine a response strategy based on the rollover risk value and transmit the result to the vehicle-mounted terminal; The vehicle-mounted terminal is further configured to execute the response strategy before the vehicle reaches a curve, wherein the response strategy at least includes risk warning and / or vehicle speed control.

Images