Omnibearing intelligent safety monitoring and driving early warning method and system integrating multiple devices

Abstract

The invention provides an omnibearing intelligent safety monitoring and driving early warning method fusing multiple devices, and the method is characterized in that the method comprises the steps: a multi-source data collection step: collecting motorcycle body data, surrounding environment data, driver state data and vehicle-road cooperation data in real time; a risk assessment step of obtaining a stability risk level and a collision risk level based on the ontology data, the surrounding environment data and the vehicle-road cooperation data, the stability risk level reflecting a vehicle falling risk, and the collision risk level reflecting a collision risk between the vehicle and other traffic participants; and an early warning step of deciding an early warning level and an optimal early warning mode according to the stability risk level and the collision risk level and based on the driver state data.

Description

TECHNICAL FIELD

[0001] The present application relates to safety monitoring, in particular to a multi-device integrated all-around intelligent safety monitoring and driving warning method and system. BACKGROUND

[0002] With the continuous growth of the number of motorcycles, the contradiction between its flexible and convenient travel advantage and high accident risk is increasingly prominent. The core risks are concentrated in two scenes of falling off the vehicle out of control and collision. The existing stability control mainly intervenes the vehicle posture through algorithm, such as adjusting the brake force distribution, but does not establish a quantitative falling off the vehicle risk level evaluation system. The collision warning focuses on distance judgment, ignores the dynamic risk brought by vehicle-road cooperation information such as sudden congestion at the front intersection, and is difficult to realize accurate prediction of double risks. SUMMARY

[0003] In view of the defects in the prior art, the purpose of the present application is to provide a multi-device integrated all-around intelligent safety monitoring and driving warning method and system to overcome the above-mentioned defects in the prior art.

[0004] To achieve the above-mentioned purpose, the present application provides the following technical scheme: The multi-device integrated all-around intelligent safety monitoring and driving warning method comprises: A multi-source data acquisition step, real-time acquisition of motorcycle body data, surrounding environment data, driver state data and vehicle-road cooperation data; A risk assessment step, based on the body data, the surrounding environment data and the vehicle-road cooperation data, obtaining a stability risk level and a collision risk level, the stability risk level reflecting the falling off the vehicle risk, and the collision risk level reflecting the collision risk of the vehicle and other traffic participants; A warning step, according to the stability risk level and the collision risk level, and based on the driver state data, deciding the warning level and the best warning mode.

[0005] As a preferred, the risk assessment step includes a stability risk level assessment step, the motorcycle body data includes vehicle body inertial measurement unit data, front sensor data and front and rear wheel speed difference, calculates the stability margin of the current vehicle, and combines the surrounding environment information, the real-time stability coefficient of the vehicle, and generates the stability risk level according to the stability coefficient. When emergency braking is detected and the vehicle inclination angle is close to the upright state, or the inclination angle is too large in the curve and the road surface degree of turbidity is too high, it is determined that the stability risk level is high.

[0006] Preferably, the risk assessment step includes a collision risk level assessment step. The collision detection level assessment step includes collecting the driver's historical driving data, generating a personalized profile of the driver's baseline reaction speed and driving style, acquiring surrounding environment data, vehicle-to-everything (V2X) data, and driver status data, obtaining a basic environmental risk value based on the surrounding environment data and V2X data, comparing the driver's current status data with the personalized profile, obtaining an adjustment coefficient reflecting the driver's real-time response capability, and fusing the basic risk value and the adjustment coefficient to generate a collision risk level.

[0007] Preferably, the warning step includes generating a quantitative assessment based on the driver's state. The quantitative assessment includes distraction level, cognitive compliance, and situational perception accuracy, and obtaining stability risk level and collision risk level. The quantitative assessment is input together with the stability risk level and collision risk level, and a personalized warning strategy is mapped out through preset logical decision-making. The warning strategy includes the warning triggering event, output mode, information content, and warning intensity.

[0008] Preferably, the warning step also includes a personalized avoidance path generation strategy. The personalized avoidance path generation strategy includes obtaining the driver's personalized profile, collecting surrounding environmental information and vehicle-road cooperative information, including the road surface adhesion coefficient and avoidance space. Based on the warning information, multiple candidate avoidance paths are generated. Based on each candidate path, path feasibility simulation is performed to evaluate its risk value. The path with the lowest risk value is selected from the candidate avoidance paths as the avoidance path.

[0009] Preferably, the personalized avoidance path generation strategy also includes a feasibility simulation strategy, constructs a real-time loop map, marks blind spot areas, calculates the blind spot risk probability value for each blind spot based on blind spot information and driver information, and calculates the comprehensive cost for each candidate path. The comprehensive cost includes the known environmental cost and the blind spot risk cost. The blind spot risk cost is determined by the blind spot risk probability value, proximity, and exposure time of the blind spot area adjacent to the path.

[0010] Preferably, it also includes a multi-channel warning execution step, which transmits warning information through one or more combinations of helmet speakers, motorcycle handlebars, and seat vibration motors, depending on the warning mode.

[0011] A comprehensive intelligent safety monitoring and driving warning system integrating multiple devices, including: The multi-source data acquisition module collects real-time data on the motorcycle itself, the surrounding environment, the driver's status, and vehicle-road cooperative data. The risk assessment module, based on ontological data, surrounding environmental data, and vehicle-road cooperative data, obtains stability risk level and collision risk level. The stability risk level reflects the risk of a crash, and the collision risk level reflects the risk of collision between the vehicle and other traffic participants. The warning module determines the warning level and the optimal warning mode based on the stability risk level, collision risk level, and driver status data.

[0012] The beneficial effects of this invention are as follows: By simultaneously collecting motorcycle body data, surrounding environment data, driver status data, and vehicle-to-infrastructure (V2I) data, it overcomes the deficiency of single-dimensional data in existing systems. Utilizing complementary multi-source data, it effectively eliminates errors from single sensors, improves the robustness of risk identification, and can cover hidden risks in complex road conditions. Risks are classified into stability risk levels and collision risk levels, specifically addressing core safety pain points of motorcycles. Specifically, stability risk assessment combines motorcycle body data and V2I data, filling the gap in the quantitative assessment of crash risk in existing technologies; collision risk assessment integrates environmental data and V2I information, achieving full-chain prediction from near-end obstacles to far-end traffic events; and dynamically adjusting warning levels and modes based on driver status data solves the problem of rigid warnings in existing systems. For example, it automatically increases warning intensity when the driver is fatigued, uses multimodal warnings to quickly attract attention when distracted, and reduces warning interference under normal conditions, balancing warning effectiveness and driving comfort. Attached Figure Description

[0013] Figure 1 This is an overall flowchart of the present invention; Figure 2 This is a flowchart of the risk assessment steps of the present invention. Detailed Implementation

[0014] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0015] It should be noted that when a component is described as "fixed to" another component, it can be directly on the other component or may have a component in between. When a component is considered "connected to" another component, it can be directly connected to the other component or may have a component in between. When a component is considered "set on" another component, it can be directly set on the other component or may have a component in between. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this document are for illustrative purposes only.

[0016] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0017] The embodiments of the present invention will be further described in detail below with reference to the accompanying drawings: A comprehensive intelligent safety monitoring and driving warning method integrating multiple devices, including: The multi-source data acquisition process involves real-time collection of motorcycle body data, surrounding environment data, driver status data, and vehicle-road cooperative data. The risk assessment step, based on ontological data, surrounding environmental data, and vehicle-road cooperative data, obtains stability risk level and collision risk level. The stability risk level reflects the risk of a crash, and the collision risk level reflects the risk of collision between the vehicle and other road users. This step takes ontological data, environmental data, vehicle-road cooperative data, and driver status data output from the multi-source data acquisition phase as input, and achieves risk quantification through dual-dimensional parallel assessment and dynamic data fusion. The assessment cycle is synchronized with the data acquisition cycle to ensure real-time performance and decision-making effectiveness.

[0018] The risk assessment process includes a stability risk level assessment step. This step utilizes motorcycle data, including inertial measurement unit (IMU) data, sensor data, and front and rear wheel speed differences, to calculate the vehicle's current stability margin. Combined with surrounding environmental information and the vehicle's real-time stability coefficient, a stability risk level is generated based on this coefficient. An increased stability risk level is determined when emergency braking is detected with the vehicle leaning close to an upright position, or when the lean angle is excessive in a curve and road surface turbidity is low. Focusing on motorcycle dynamic stability and combining the vehicle's physical condition with environmental influences, this approach quantifies stability margin, dynamically adjusts risk thresholds, and ultimately outputs the risk level. It primarily addresses the issues of delayed crash risk prediction and insufficient adaptation to complex road conditions. The system calculates the current vehicle's stability margin and, combined with surrounding environmental information, obtains the vehicle's real-time stability coefficient. The stability margin is a basic crash-resistant indicator based on the vehicle's intrinsic data. It comprehensively considers lateral imbalance and wheel slippage risks, and uses surrounding environmental data to perform scenario-based adjustments to the stability margin, resulting in the real-time stability coefficient. When any of the following conditions are met, regardless of the current risk level, the stability risk level is directly increased by 1 level to cover extreme crash scenarios: emergency braking and upright position, such as sudden braking while driving on a straight road, which is originally medium risk and directly increases to high risk; large lean angle and low adhesion in curves, such as sharp curves in rainy weather, which is originally high risk and directly increases to extremely high risk.

[0019] The risk assessment process includes a collision risk level assessment step. The collision detection level assessment step involves collecting the driver's historical driving data, generating a personalized profile of the driver's baseline reaction speed and driving style, acquiring surrounding environmental data, vehicle-to-everything (V2X) data, and driver status data. Based on the surrounding environmental data and V2X data, a basic environmental risk value is obtained, and the driver's current status data is compared with the personalized profile to obtain an adjustment coefficient reflecting the driver's real-time response capability. The basic risk value and adjustment coefficient are then integrated to generate the collision risk level. This assessment uses environmental risk and driver response capability as its dual cores, achieving personalized risk adaptation for each individual driver by establishing a personalized driver profile. This addresses the problems of existing assessments neglecting individual driver differences and the disconnect between risk prediction and actual response capability. The profile is the core of personalized risk assessment, requiring the collection of the driver's effective driving history data, including baseline reaction speed, driving style, and state-behavior correlation models. A basic environmental risk value is calculated, compared with the driver's current status and personalized profile, and an adjustment coefficient for the driver's response capability is calculated. A collision risk value is then obtained, reflecting the probability of collision with other road users.

[0020] The warning process involves determining the warning level and the optimal warning mode based on the stability risk level, collision risk level, and driver status data.

[0021] The early warning process includes generating a quantitative assessment based on the driver's state. This assessment includes distraction level, cognitive compliance, and situational awareness accuracy, and also obtains stability risk level and collision risk level. The quantitative assessment, along with the stability risk level and collision risk level, are input together and mapped to a personalized early warning strategy through pre-set logical decision-making. The early warning strategy includes the triggering event, output mode, information content, and warning intensity. Based on driver state data from the multi-source data acquisition phase, using the smart helmet camera, physiological sensors, and handlebar sensors, a three-dimensional quantitative assessment model is constructed. The model outputs quantitative values ​​that can be directly used for strategy decision-making, including distraction level, cognitive compliance, and situational awareness deficiency. Distraction level is obtained through the smart helmet camera, handlebar capacitive sensor, and helmet microphone; cognitive compliance is obtained through the smart helmet physiological sensor, vehicle CAN bus, and helmet voice module; and situational awareness deficiency is obtained through the V2X module, event trigger recording, and helmet camera. Combining the driver's three-dimensional quantitative assessment results, stability risk level, and collision risk level, a personalized strategy including trigger timing, output mode, information content, and warning intensity is output through a pre-set multi-input multi-output logical decision table.

[0022] The early warning process also includes a personalized avoidance path generation strategy. This strategy involves acquiring the driver's personalized profile, collecting surrounding environmental information and vehicle-to-infrastructure (V2I) information (including road surface adhesion coefficient and avoidance space), generating multiple candidate avoidance paths based on the early warning information, performing path feasibility simulations for each candidate path, evaluating its risk value, and selecting the path with the lowest risk value as the avoidance path. For medium- to high-risk scenarios, an avoidance path is generated by combining the driver's personalized profile, real-time environment, and V2I data, adapting to the driver's operating habits and road conditions. Among the dynamically feasible candidate paths, the path with the lowest overall cost is selected as the personalized avoidance path. The personalized path avoidance generation strategy also includes a feasibility simulation strategy, constructing a real-time ring map and marking blind spot areas. Based on blind spot information and driver data, a blind spot risk probability value is calculated for each blind spot. For each candidate path, a comprehensive cost is calculated, including known environmental costs and blind spot risk costs. The blind spot risk cost is determined by the blind spot risk probability value, proximity, and exposure time of the areas adjacent to the blind spot. A dual-dimensional simulation evaluation of dynamic feasibility and blind spot risk is performed on each candidate path, outputting the comprehensive cost. A real-time environmental map and blind spot markings are constructed, including sensor blind spots and line-of-sight blind spots. Based on a motorcycle dynamics model, stability indicators during path operation are simulated to determine whether safety constraints are met. For dynamically feasible paths, the sum of known environmental costs and blind spot risk costs is calculated.

[0023] It also includes a multi-channel warning execution step, which transmits warning information through one or more combinations of helmet speakers, motorcycle handlebars, and seat vibration motors, depending on the warning mode.

[0024] A comprehensive intelligent safety monitoring and driving warning system integrating multiple devices, including: The multi-source data acquisition module collects real-time data on the motorcycle itself, the surrounding environment, the driver's status, and vehicle-road cooperative data. The risk assessment module, based on ontological data, surrounding environmental data, and vehicle-road cooperative data, obtains stability risk level and collision risk level. Stability risk level reflects the risk of crash, and collision risk level reflects the risk of collision between the vehicle and other traffic participants. The warning module determines the warning level and optimal warning mode based on stability risk level, collision risk level, and driver status data. Through multi-hardware collaboration involving the helmet speaker, motorcycle handlebar vibration motor, and seat vibration motor, it achieves multi-dimensional, high-perception transmission of warning information, fundamentally addressing the issues of single-channel systems being easily interfered with by environmental factors and experiencing delayed perception. The warning intensity dynamically selects between single-channel and multi-channel combinations based on the driver's status, ensuring a balance between warning effectiveness and driving interference. For example, when generating a personalized avoidance path, the channel combination must match the path direction. For instance, when using a left-side avoidance path, the left handlebar continuously vibrates, the left channel provides voice prompts, indicating a move to the adjacent lane, and the seat vibrates simultaneously to indicate the avoidance is in progress and maintain speed. During execution, if the risk level decreases from high to low due to multi-channel warnings and path prompts, the onboard ECU gradually shuts down the vibration channels, retaining only the low-frequency speaker notification to indicate the risk has been mitigated, and normal driving resumes.

[0025] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principle of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A method for all-round intelligent safety monitoring and driving warning of fusion multi-device, characterized in that, The application relates to a motorcycle risk assessment and early warning system. The application comprises: a multi-source data collection step, which collects motorcycle body data, surrounding environment data, driver state data and vehicle-road coordination data in real time; a risk assessment step, which obtains a stability risk level and a collision risk level based on the body data, the surrounding environment data and the vehicle-road coordination data, wherein the stability risk level reflects a motorcycle falling risk and the collision risk level reflects a collision risk between the motorcycle and other traffic participants; 2. The method for all-around intelligent safety monitoring and driving warning of fusion multi-device according to claim 1, characterized in that, a warning step, which decides a warning level and an optimal warning mode based on the stability risk level and the collision risk level and the driver state data.

3. The method of claim 1, wherein the method further comprises: The risk assessment step comprises a stability risk level assessment step, the motorcycle body data comprises vehicle body inertia measurement unit data, sensor data and front and rear wheel speed differences, the stability margin of the current vehicle is calculated, the real-time stability coefficient of the vehicle is combined with the surrounding environment information, and the stability risk level is generated according to the stability coefficient; when emergency braking is detected and the vehicle inclination angle is close to the upright state, or the inclination angle is too large in a curve and the road surface degree of turbidity is high, it is determined that the stability risk level is high.

4. The method of claim 1, wherein the method further comprises: The risk assessment step comprises a collision risk level assessment step, the collision risk level assessment step comprises collecting historical driving data of the driver, generating a reference reaction speed and a driving style of the driver to generate a personalized profile, obtaining surrounding environment data, vehicle coordination data and driver state data, obtaining a basic environment risk value based on the surrounding environment data and the vehicle coordination data, comparing the current state data of the driver with the personalized profile to obtain an adjustment coefficient reflecting the real-time response ability of the driver, and fusing the basic risk value and the adjustment coefficient to generate the collision risk level.

5. The method of claim 1, wherein the method further comprises: The warning step comprises generating a quantitative assessment according to the driver state, the quantitative assessment comprises a distraction level, cognitive compliance and situational awareness certainty, the stability risk level and the collision risk level are obtained, the quantitative assessment, the stability risk level and the collision risk level are jointly input, a personalized warning strategy is mapped through a preset logic decision, and the warning strategy comprises a triggering event of the warning, an output mode, information content and a warning intensity. The warning step further comprises a personalized avoidance path generation strategy, the personalized avoidance path generation strategy comprises obtaining a personalized profile of the driver, collecting surrounding environment information and vehicle coordination information, the vehicle coordination information comprises a road surface adhesion coefficient and an avoidance space, a plurality of candidate avoidance paths are generated based on the warning information, the risk value of each candidate path is simulated based on the candidate path, the risk value is evaluated, and the path with the lowest risk value is selected as the avoidance path from the candidate avoidance paths.

6. The omnibearing intelligent safety monitoring and driving warning method for fusing multiple devices according to claim 5, characterized in that, The personalized avoidance path generation strategy further includes a feasibility simulation strategy, which constructs a real-time ring map, marks blind area regions, calculates blind area risk probability values for each blind area based on blind area information and driver information, and calculates a comprehensive cost for each candidate path, the comprehensive cost including a known environment cost and a blind area risk cost determined by the blind area risk probability values, proximity, and exposure time of the regions adjacent to the blind area of the path.

7. The method of claim 1, wherein the method further comprises: The strategy further includes a multi-channel early warning execution step, which transmits early warning information through one or more combinations of a helmet speaker, a motorcycle handle, and a vibrating motor in a seat cushion according to an early warning mode.

8. A comprehensive intelligent safety monitoring and driving warning system fusing multi-devices, characterized in that, The strategy includes: a multi-source data acquisition module that acquires motorcycle body data, surrounding environment data, driver state data, and vehicle-road cooperation data in real time; a risk assessment module that obtains a stability risk level and a collision risk level based on the body data, the surrounding environment data, and the vehicle-road cooperation data, the stability risk level reflecting the risk of falling off the motorcycle, and the collision risk level reflecting the risk of collision with other traffic participants; an early warning module that determines an early warning level and an optimal early warning mode based on the stability risk level and the collision risk level and based on the driver state data.

Images