Vehicle safety management cooperation system and method based on multi-source data fusion and business closed loop

Abstract

The invention provides a vehicle safety management cooperation system based on multi-source data fusion and business closed loop. The vehicle safety management cooperation system comprises a vehicle-mounted intelligent terminal, a driver file module, a vehicle file module, a business management module, a safety monitoring module and a cooperation processing engine. The vehicle-mounted intelligent terminal is used for collecting Beidou positioning, driving behaviors and vehicle state data. The co-processing engine carries out deep association and fusion analysis on risk events identified by the safety monitoring module and driver archives, vehicle archives, historical rewards and punishment and training records in the driver archive module and the vehicle archive module through a built-in rule model to obtain a risk assessment level, and carries out risk assessment according to the risk assessment level. And automatically triggering and driving a corresponding disposal flow in the business management module. The system solves the problems of information island, lagging disposal and strong human factor dependence in the existing vehicle safety management, and realizes the intelligent safety management mode conversion from passive response to active prevention and from post-event punishment to process intervention.

Description

Technical Field

[0001] This invention relates to the field of vehicle safety management technology, and in particular to a collaborative system and method for vehicle safety management based on multi-source data fusion and business closed loop. Background Technology

[0002] With the popularization of the BeiDou Navigation Satellite System and the development of vehicle-to-everything (V2X) technology, BeiDou-based vehicle safety management applications are becoming increasingly widespread. Existing technologies mostly focus on implementing single safety functions, such as the "Fatigue Driving Monitoring System Based on BeiDou Satellite Positioning and Timing Technology" disclosed in patent CN110148281A, which guides drivers to rest or issues an alarm after exceeding the permitted driving time through precise timing, effectively solving the problem of single-point monitoring of fatigue driving. Another type of patent focuses on route planning optimization.

[0003] However, in actual enterprise and logistics fleet vehicle management, safety management is a systematic project involving multiple dimensions such as "people, vehicles, tasks, environment, and management." Currently, the following pain points are prevalent: 1. Information silos: Real-time vehicle data (location, status), driver static files (training, rewards and punishments), vehicle static files (maintenance, insurance), and safety management data (inspections, education, drills) are scattered across different modules or systems, fragmented and unable to form a comprehensive understanding of safety risks. For example, knowing that a driver is fatigued may not reveal whether they have recently participated in relevant training or whether the vehicle has any braking hazards. 2. Lagging and passive management: Risk handling often relies on manual discovery, judgment, and dispatch processes, resulting in slow response and easy oversights. For example, after a dangerous driving warning occurs, it is usually only recorded, making it difficult to automatically and promptly link it to preventative measures such as driver retraining or additional vehicle inspections. 3. Disjointed handling and lack of closed-loop system: The handling, training, assessment, and reward / punishment of safety incidents are independent of each other and lack linkage, making it impossible to form a closed loop of "event-driven management improvement", resulting in low management efficiency. Summary of the Invention

[0004] The purpose of this invention is to provide a collaborative system and method for vehicle safety management based on multi-source data fusion and business closed loop. By deeply linking real-time risk monitoring with multi-dimensional static files and dynamic business processes, it aims to achieve intelligent assessment of safety risks and improve the overall level of safety management.

[0005] To achieve the above objectives, the present invention provides the following solution: A collaborative vehicle safety management system based on multi-source data fusion and business closed loop includes: Vehicle-mounted intelligent terminal: Deployed on the vehicle to collect the vehicle's BeiDou positioning information, driving status information and driver behavior data; Driver Profile Module: Used to store and manage driver information; driver information includes: personal resume, training records, reward and punishment records, and conduct profile; Vehicle file module: Used to store and manage vehicle information, including: file information, usage and maintenance records, insurance information, and self-inspection records of safety hazards; Business Management Module: Used to implement vehicle management systems, driver education and training, vehicle operation activities, emergency drills, dispatch records, transportation tasks, and driver evaluation based on driver and vehicle information; Safety monitoring module: used to receive data uploaded by the in-vehicle intelligent terminal and identify driving risk events based on rule model; driving risk events include: fatigue driving, dangerous driving behavior and abnormal vehicle status; Collaborative Response Engine: Used to conduct risk assessments based on driver information, vehicle information, and driving risk events, and obtain comprehensive risk assessment results and response recommendations.

[0006] Optionally, the business management module includes: Driver Education and Training Submodule: Used to automatically create and issue a mandatory special training task that is strongly related to driving risk events based on driver and vehicle information; Vehicle safety hazard self-inspection submodule: used to automatically generate additional inspection or maintenance work orders for the target vehicle based on driver and vehicle information; Driver Rewards and Penalties Record Submodule: Used to automatically generate a reward or penalty pending item corresponding to the comprehensive risk assessment result based on driver and vehicle information; Driver Thought Profile Submodule: Used to generate conversation or care prompts based on driver and vehicle information; Vehicle Service Activity Plan Submodule: Used to provide risk warnings or temporary restrictions on the transportation tasks of target drivers.

[0007] Optionally, the specific recognition process of the rule model includes: Data preprocessing is performed on BeiDou positioning information, driving status information, and driver behavior data to obtain preprocessed data. Data preprocessing includes: data cleaning, time alignment, feature extraction, and data fusion. Vehicle safety parameters are calculated based on preprocessed data; vehicle safety parameters include: continuous driving time, rest time, rapid acceleration and rapid deceleration. Based on the inspection conditions, threshold detection is performed on vehicle safety parameters to identify driving risk events and trigger safety warnings.

[0008] Optionally, the formula for calculating continuous driving time is: ;in, This represents the current cumulative valid driving time. This is the cumulative driving time from the previous calculation period. To calculate the driving cycle duration, For effective driving time function; The formula for calculating rest time is: ;in, This represents the current total effective rest time. This is the cumulative rest time from the previous calculation period. To calculate the duration of the rest cycle, For parking and rest functions; The formula for calculating rapid acceleration is: ;in, Accelerate to the current moment, Current vehicle speed The speed of the vehicle at the previous moment. The time interval between two vehicle speed samplings; The calculation process for rapid deceleration and rapid acceleration is completely symmetrical and in opposite directions.

[0009] Optionally, threshold detection is performed on vehicle safety parameters based on inspection conditions to identify driving risk events and trigger safety warnings, including: When the continuous driving time is ≥4 hours, an overtime alarm is triggered; When the parking time is ≥30 minutes, the continuous driving time is recalculated; When the cumulative driving time within 24 hours is ≥8 hours, the driving risk event will be judged as fatigue driving.

[0010] A collaborative method for vehicle safety management based on multi-source data fusion and business closed loop is implemented through a collaborative system for vehicle safety management based on multi-source data fusion and business closed loop, including: Based on driving risk events, multi-dimensional correlation data of the target driver and the target vehicle is obtained; the multi-dimensional correlation data includes: driver facial features, vehicle status, alarm events and driver history records; Risk assessment models are used to analyze driving risk events and multi-dimensional related data to obtain comprehensive risk assessment results and handling recommendations. Based on the comprehensive risk assessment results and disposal recommendations, disposal instructions are generated, and collaborative disposal operations are executed according to the disposal instructions to obtain disposal results; The effectiveness of the treatment results is verified according to the preset completion criteria, and the multi-dimensional related data is updated based on the verification results.

[0011] Optionally, a risk assessment model can be used to analyze driving risk events and multi-dimensional related data to obtain comprehensive risk assessment results and handling recommendations, including: Fatigue risk is calculated based on driving risk events and driver facial features; the formula for calculating fatigue risk is: ;in, The total fatigue risk score, The risk score for eye condition. To assess the risk score for yawning, This is a risk score for driving time; The alarm risk is calculated based on the alarm events; the formula for calculating alarm risk is: ;in, The total score for alarm risk. The attenuation coefficient is... The sum of weighted events; The vehicle speed risk is calculated based on the vehicle's condition. The risk level is calculated based on the driver's historical records; the formula for calculating the risk level is: ;in, The total risk score for the archives. This is a coefficient representing driving experience. This is the violation coefficient. Accident coefficient; The overall comprehensive risk is obtained by weighting and summing the risks of fatigue, alarm, speed, and records.

[0012] According to specific embodiments provided by the present invention, the following technical effects are disclosed: The vehicle safety management collaborative system and method based on multi-source data fusion and business closed loop provided by the present invention includes: an in-vehicle intelligent terminal, deployed on the vehicle, for collecting the vehicle's Beidou positioning information, driving status information, and driver behavior data; a driver file module, for storing and managing driver information; driver information includes: personal resume, training records, reward and punishment records, and ideological archives; a vehicle file module, for storing and managing vehicle information; vehicle information includes: file information, usage and maintenance records, insurance information, and safety hazard self-inspection records; a business management module, for executing vehicle management systems, driver education and training, vehicle operation activities, emergency drills, dispatch records, transportation tasks, and driver evaluation business based on driver information and vehicle information; a safety monitoring module, for receiving data uploaded by the in-vehicle intelligent terminal and identifying driving risk events based on a rule model; driving risk events include: fatigue driving, dangerous driving behavior, and abnormal vehicle status; and a collaborative handling engine, for performing risk assessment based on driver information, vehicle information, and driving risk events to obtain comprehensive risk assessment results and handling suggestions. This system deeply integrates real-time risk monitoring with multi-dimensional static archives and dynamic business processes, enabling intelligent assessment, automatic handling, and closed-loop management of security risks. This transforms passive response into proactive prevention, thereby improving the overall level of security management. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is a schematic diagram of the vehicle safety management collaborative system structure based on multi-source data fusion and business closed loop of the present invention; Figure 2 This is a flowchart of the collaborative method for vehicle safety management based on multi-source data fusion and business closed loop of the present invention; Figure 3 This is a schematic diagram of the overall architecture of Embodiment 1 of the present invention; Figure 4 This is a flowchart of the collaborative processing engine in Embodiment 2 of the present invention. Detailed Implementation

[0015] 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.

[0016] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0017] like Figure 1 As shown, the present invention provides a vehicle safety management collaborative system based on multi-source data fusion and business closed loop, including: an in-vehicle intelligent terminal and a cloud management platform.

[0018] Vehicle-mounted intelligent terminal: Deployed on the vehicle, it integrates a Beidou positioning module, a vehicle CAN bus interface and behavior detection sensors. It is used to collect the vehicle's Beidou positioning information, driving status information (including: location, speed and mileage information) and driver behavior data (including: rapid acceleration, rapid deceleration, sharp turns, etc.), and uploads it to the cloud management platform through the wireless communication network.

[0019] The cloud management platform includes: Driver Profile Module: Used to store and manage driver information; driver information includes: personal resume, training records, reward and punishment records, and conduct profile; Vehicle file module: Used to store and manage vehicle information, including: file information, usage and maintenance records, insurance information, and self-inspection records of safety hazards; The Business Management Module is a workflow engine used to support and execute various vehicle management business processes. It executes vehicle management regulations, driver education and training, vehicle operation activities, emergency drills, dispatch records, transportation tasks, and driver evaluation processes based on driver and vehicle information. Examples include the release and tracking of driver education and training tasks, the execution of vehicle operation activity plans and emergency drill plans, the dispatch and receipt of driver dispatch records and transportation tasks, the organization of the "Red Flag Vehicle Driver Evaluation" activity, and the "Traffic Accident Entry" and processing procedures. Safety monitoring module: used to receive data uploaded by the in-vehicle intelligent terminal and identify driving risk events based on rule model; driving risk events include: fatigue driving, dangerous driving behavior and abnormal vehicle status; Collaborative Response Engine: Used to conduct risk assessments based on driver information, vehicle information, and driving risk events, and obtain comprehensive risk assessment results and response recommendations.

[0020] Specifically, the business management module includes: Driver Education and Training Submodule: Used to automatically create and issue a mandatory special training task that is strongly related to driving risk events based on driver and vehicle information; The Vehicle Safety Hazard Self-Inspection (or Vehicle Use and Maintenance) submodule is used to automatically generate additional inspection or maintenance work orders for the target vehicle based on driver and vehicle information. Driver Rewards and Penalties Record Submodule: Used to automatically generate a reward or penalty pending item corresponding to the comprehensive risk assessment result based on driver and vehicle information; Driver Thought Profile Submodule: Used to generate conversation or care prompts based on driver and vehicle information; The Vehicle Service Activity Plan (or Vehicle Dispatch Record) submodule is used to provide risk warnings or temporary restrictions on the transportation tasks of target drivers.

[0021] Specifically, the specific recognition process of the rule model includes: Data preprocessing is performed on BeiDou positioning information, driving status information, and driver behavior data to obtain preprocessed data. Data preprocessing includes: data cleaning, time alignment, feature extraction, and data fusion. Vehicle safety parameters are calculated based on preprocessed data; vehicle safety parameters include: continuous driving time, rest time, rapid acceleration and rapid deceleration. Based on the inspection conditions, threshold detection is performed on vehicle safety parameters to identify driving risk events and trigger safety warnings.

[0022] Furthermore, the formula for calculating continuous driving time is: ; in, This represents the current cumulative valid driving time. This is the cumulative driving time from the previous calculation period. To calculate the driving cycle duration (2 seconds in some embodiments). The effective driving time function is defined as follows: ... (1) The vehicle's ACC status is ON; (2) The vehicle speed is >5 km / h; (3) The GPS location change is >50 meters; otherwise 0.

[0023] The formula for calculating rest time is: ; in, This represents the current total effective rest time. This is the cumulative rest time from the previous calculation period. To calculate the duration of the rest cycle (2 seconds in some embodiments). Let this be the parking rest function. The parking rest function is defined as follows: The parking rest function is defined if and only if the following conditions are met simultaneously. (1) Vehicle speed = 0; (2) Vehicle ACC status is OFF; otherwise 0.

[0024] The formula for calculating rapid acceleration is: ; in, Accelerate to the current moment, Current vehicle speed The speed of the vehicle at the previous moment. This represents the time interval between two vehicle speed samplings. The calculation processes for rapid deceleration and rapid acceleration are completely symmetrical and in opposite directions.

[0025] More specifically, the conditions for resetting rest time are: > , The minimum effective rest time is 1800 seconds (30 minutes) in some embodiments. The 24-hour fatigue driving warning process is as follows: at any check time, the cumulative value of all effective driving time in the past 24 hours is calculated. If the cumulative value is ≥28800 seconds (8 hours), it is determined to be 24-hour fatigue driving.

[0026] Furthermore, the inspection conditions include: When continuous driving time is ≥4 hours, an overtime alarm is triggered; when parking time is ≥30 minutes, continuous driving time is recalculated; when the cumulative driving time within 24 hours is ≥8 hours, the driving risk event is judged as fatigue driving.

[0027] When the acceleration is ≥3 m / s and the duration of the acceleration is >1 second, the acceleration warning is triggered; when the deceleration is ≥3 m / s and the duration of the deceleration is >1 second, the deceleration warning is triggered; when the lane departure is ≥30 cm and the duration is ≥3 seconds and the turn signal is not used, the lane departure warning is triggered.

[0028] Specifically, the collaborative handling engine communicates with all the aforementioned modules. When the safety monitoring module identifies a risk event (such as a "continuous driving timeout warning"), the collaborative handling engine is triggered. First, based on the vehicle and driver associated with the event, it retrieves multi-dimensional correlation data from the driver and vehicle profile modules (such as the driver's rewards and punishments over the past three months, whether fatigue driving training has been completed recently, and the last mileage of the brake pads on the vehicle). Then, the engine's built-in risk assessment model fuses and analyzes the multi-dimensional correlation data and static profile data to arrive at a comprehensive risk level. The specific steps are: real-time collection of driver facial features (eyelid closure, yawning frequency), monitoring of vehicle status (vehicle speed, acceleration, steering angle), acquisition of alarm events (fatigue, smoking, phone calls, etc.), and reading of the driver's historical profile (driving experience, traffic violation records). Then, fatigue risk, alarm risk, vehicle speed risk, and profile risk are calculated and weighted to obtain the overall comprehensive risk.

[0029] More specifically, fatigue risk is calculated based on driving risk events and driver facial features, using the following formula: ; in, The total fatigue risk score ranges from 0 to 100. The risk score for eye condition ranges from 0 to 100. The risk score for yawning ranges from 0 to 50. This is a risk score for driving time.

[0030] The calculation formula is: ; in, The total time the eyes are closed. The statistical window time is (in some embodiments, it is 180 seconds (3 minutes)). This refers to the actual blinking frequency. This is the normal blinking frequency (15 times / minute in some embodiments).

[0031] The calculation formula is: ; in, This represents the actual frequency of yawning. The baseline yawning frequency (1 time / hour in some embodiments).

[0032] The alarm risk is calculated based on the alarm events, using the following formula: ; ; ; in, The total score for alarm risk. This is the attenuation coefficient (0.1 in some embodiments). For the weighted sum of events, Let be the weight coefficient for the i-th type of event. Let i be the number of times the i-th type of event occurs within the statistics window. The number of times event i occurs within the statistics window. The duration of the statistical window is 1800 seconds (30 minutes) in some embodiments. The specific values ​​of the weighting coefficients are shown in Table 1.

[0033] Table 1. Event Weight Coefficient Values

[0034] The vehicle speed risk is calculated based on the vehicle's status, specifically as follows: when the current speed is less than or equal to the road speed limit, the total speed risk score is 0; when the road speed limit is less than the current speed and less than or equal to 1.2 × the road speed limit, the total speed risk score is 20; when the road speed limit is less than the current speed and less than or equal to 1.5 × the road speed limit, the total speed risk score is 50; and in all other cases, the total speed risk score is 80.

[0035] The risk level is calculated based on the driver's historical records, using the following formula: ; in, The total risk score for the archives. This is a coefficient representing driving experience. This is the violation coefficient. This is the accident coefficient. When the driver's driving experience is ≥10 years, When 5 ≤ driver's driving experience < 10 years, When 2 ≤ driver's driving experience < 5 years, In other cases .

[0036] ,in Let i be the weight score for the i-th type of violation. Let i be the number of violations of type i. This represents the sum of all traffic violations committed within the past year. When the violation is speeding... When the violation is running a red light When the violation is drunk driving When the violation is other Then, the overall comprehensive risk is obtained by weighting and summing the risks of fatigue, alarms, speed, and records. The calculation formula is as follows: At the same time, when When the score is ≥60, the comprehensive risk level will be determined as Level 1; when If the score is ≥40, the overall risk level will be determined as Level 2.

[0037] Finally, the engine automatically generates structured instructions based on the comprehensive risk level and pushes them to the business management module to drive the corresponding business processes to start automatically. In some embodiments, a mandatory learning task on "fatigue driving and safe parking" is automatically created in "Driver Education and Training" and assigned to the driver; at the same time, a work order for "focusing on braking system inspection for vehicles driven for extended periods" is generated in "Vehicle Safety Hazard Self-Inspection"; and the incident and handling suggestions are recorded in "Driver Reward and Punishment Record" for administrator review. The entire process requires no manual intervention, and is automatically initiated, automatically associated, and automatically driven. The engine is also responsible for closed-loop management, monitoring the completion status of triggered business processes (such as whether the driver has completed training, whether the inspection work order has been processed and feedback). When all handling processes are completed as required, the engine marks the risk event as "closed loop" and updates the handling results in the driver and vehicle files, forming a complete management closed loop.

[0038] like Figure 2 As shown, this invention also provides a collaborative method for vehicle safety management based on multi-source data fusion and business closed loop, implemented through a collaborative system for vehicle safety management based on multi-source data fusion and business closed loop, including: Based on driving risk events, multi-dimensional correlation data of the target driver and the target vehicle is obtained; the multi-dimensional correlation data includes: driver facial features, vehicle status, alarm events and driver history records; Risk assessment models are used to analyze driving risk events and multi-dimensional related data to obtain comprehensive risk assessment results and handling recommendations. Based on the comprehensive risk assessment results and disposal recommendations, disposal instructions are generated, and collaborative disposal operations are executed according to the disposal instructions to obtain disposal results; The effectiveness of the treatment results is verified according to the preset completion criteria, and the multi-dimensional related data is updated based on the verification results.

[0039] In Example 1, as Figure 3 As shown, the system consists of an in-vehicle intelligent terminal and a cloud management platform. The in-vehicle terminal communicates with the cloud management platform via a 4G / 5G network. The cloud management platform adopts a microservice architecture, including driver profile services, vehicle profile services, safety monitoring services, a business process engine service (i.e., the core of the business management module), and a collaborative processing engine service. All services exchange data through an enterprise service bus. The workflow of the collaborative processing engine is as follows: Figure 4 As shown, it includes the following steps: S1: Risk Event Awareness. The safety monitoring service analyzes real-time data reported by in-vehicle terminals to identify a risk event, such as "vehicle XXX has engaged in continuous sudden braking (dangerous driving)", and the event is immediately published to the message queue.

[0040] S2: Event Triggering and Data Aggregation. The collaborative processing engine listens to the message queue and captures the risk event. Based on the vehicle ID "XXX" in the event, the engine queries the current driver as "Zhang". Subsequently, the engine concurrently calls the driver profile service and vehicle profile service to obtain Zhang's "multi-dimensional related data", such as: two emergency braking warning records in the past month (historical behavior), a safety assessment grade of B last month (reward and punishment record), 8 hours of routine safety training completed this year (training record), and the self-inspection report of the vehicle "XXX" last week showing that the tire wear is approaching the threshold (vehicle hidden danger).

[0041] S3: Intelligent Analysis and Risk Assessment. The engine's built-in risk assessment model integrates and analyzes the information from S2. The model rules are set as follows: the basic risk event (sudden braking) is a medium risk, but because it is associated with "occurring multiple times in a short period of time" (weight +0.3) and "the vehicle has related hidden dangers" (weight +0.2), the overall risk level is upgraded to "high risk". The proposed action is: "Defensive driving training is required, and an emergency inspection of the vehicle's tires is necessary."

[0042] S4: Business Collaboration and Handling Trigger. Based on the judgment results of S3, the engine generates two structured instructions: (1) Create a training task instruction, including the task type (defensive driving), the assigned person (Zhang), and the required completion time (within 24 hours), and sends it to the "Training Task Creation" interface of the business process engine. (2) Create a vehicle inspection work order instruction, including the vehicle (XXX), the inspection item (tire and braking system), and the inspection level (emergency), and sends it to the "Safety Inspection Work Order" interface of the business process engine.

[0043] S5: Process-Driven and Human-Involved. Upon receiving the instruction, the business process engine automatically creates a pending online course, "Defensive Driving Training," for Zhang in the system and sends app push notifications and SMS messages. Simultaneously, it creates a pending work order, "XXX Emergency Safety Inspection," for the team leader. Both the team leader and the driver see the pending tasks on their respective workbenches.

[0044] S6: Closed-Loop Verification and File Update. Mr. Zhang completes the training via mobile device and passes the online test. The business process engine feeds back the "task completed" status to the collaborative handling engine. The mechanic completes the vehicle inspection and uploads the report. The team leader confirms the work order closure, and the status is also fed back. The collaborative handling engine confirms that both actions are completed, updates the status of this "high-risk - emergency braking" event to "closed-loop," and automatically records the event and completed handling measures in Mr. Zhang's driver file, forming part of his safety profile.

[0045] In Example 2, under the preventative management scenario, the vehicle service activity plan submodule in the business management module can call the "driver adaptability assessment" interface of the collaborative processing engine when assigning a long-distance transportation task. This allows the engine to receive a list of candidate drivers and task information (such as "long-distance, night, highway"), and query the driver's relevant files and recent risk records. For drivers with "fatigue driving" warning records and no recent relevant retraining, the engine returns a "Risk Warning: This driver has a recent fatigue driving record; it is recommended to complete the 'Long-Distance Driving Safety Regulations' study before assigning the task." For vehicle "lifting operation" tasks, the engine checks the driver's "Special Operation Certificate" status and previous "lifting operation" evaluation records to provide intelligent recommendations for task assignment.

[0046] The beneficial effects of this invention are as follows: 1) The collaborative processing engine deeply integrates real-time dynamic risk data with the full-dimensional static file data of drivers and vehicles, changing the previous single and one-sided judgment method of risk events, making risk assessment more comprehensive and accurate, and realizing deep data integration and intelligent analysis. 2) Linking the identification of safety risks with subsequent management actions such as education, training, inspection, and rewards and punishments, and automatically triggering them through preset conditions, greatly improves the speed of management response and execution efficiency, ensures the timeliness and enforceability of risk handling, and forms an automated management closed loop of "perception-analysis-handling-verification-improvement", realizing the automatic collaboration and closed loop of management business processes; 3) This system can not only handle risks that have already occurred, but also provide risk warnings and decision support for scenarios such as "driver evaluation" and "task allocation" through data analysis. For example, when allocating special tasks such as "lifting operations", the system can remind the driver of relevant operation alarm records in the near future, realizing the transformation from "post-event handling" to "pre-event prevention" and "in-event intervention", and improving the proactive prevention capability of safety management. 4) Through the recording and analysis of data throughout the entire process, and the generation of multi-dimensional statistical reports, safety management work becomes quantifiable, assessable, and traceable, providing a scientific basis for optimizing management systems and allocating training resources, while also strengthening data support for management decisions.

[0047] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0048] Specific examples have been used to illustrate the principles and implementation methods of this invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this invention. Furthermore, those skilled in the art will recognize that, based on the ideas of this invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this invention.

Claims

1. A collaborative system for vehicle safety management based on multi-source data fusion and business closed loop, characterized in that, include: Vehicle-mounted intelligent terminal: Deployed on the vehicle, used to collect the vehicle's BeiDou positioning information, driving status information and driver behavior data; Driver profile module: used to store and manage driver information; The driver information includes: personal resume, training records, reward and punishment records, and ideological profile; Vehicle file module: used to store and manage vehicle information, including: file information, usage and maintenance records, insurance information, and self-inspection records of safety hazards; Business Management Module: Used to execute vehicle management systems, driver education and training, vehicle attendance activities, emergency drills, dispatch records, transportation tasks, and driver evaluation based on the driver information and vehicle information; Safety monitoring module: used to receive data uploaded by the vehicle-mounted intelligent terminal and identify driving risk events based on a rule model; the driving risk events include: fatigue driving, dangerous driving behavior and abnormal vehicle status; Collaborative handling engine: Used to perform risk assessment based on the driver information, vehicle information, and driving risk events, and obtain comprehensive risk assessment results and handling suggestions.

2. The vehicle safety management collaborative system based on multi-source data fusion and business closed loop as described in claim 1, characterized in that, The business management module includes: Driver Education and Training Submodule: Used to automatically create and issue a mandatory special training task that is strongly related to the driving risk event based on the driver information and the vehicle information; Vehicle safety hazard self-inspection submodule: used to automatically generate additional inspection or maintenance work orders for the target vehicle based on the driver information and the vehicle information; Driver Rewards and Penalties Record Submodule: Used to automatically generate a reward or penalty pending item corresponding to the comprehensive risk assessment result based on the driver information and the vehicle information; Driver Thought Profile Submodule: Used to generate conversation or care prompts based on the driver information and vehicle information; Vehicle Service Activity Plan Submodule: Used to provide risk warnings or temporary restrictions on the transportation tasks of target drivers.

3. The vehicle safety management collaborative system based on multi-source data fusion and business closed loop as described in claim 1, characterized in that, The specific recognition process of the rule model includes: The BeiDou positioning information, the driving status information, and the driver behavior data are preprocessed to obtain preprocessed data; the data preprocessing includes: data cleaning, time alignment, feature extraction, and data fusion. Vehicle safety parameters are calculated based on the preprocessed data; the vehicle safety parameters include: continuous driving time, rest time, rapid acceleration, and rapid deceleration. Based on the inspection conditions, threshold detection is performed on the vehicle safety parameters to obtain the driving risk event and trigger a safety warning.

4. The vehicle safety management collaborative system based on multi-source data fusion and business closed loop as described in claim 3, characterized in that, The formula for calculating the continuous driving time is: ;in, This is the current cumulative valid driving time. This is the cumulative driving time from the previous calculation period. To calculate the driving cycle duration, For effective driving time function; The formula for calculating the rest time is: ;in, This represents the current total effective rest time. This is the cumulative rest time from the previous calculation period. To calculate the duration of the rest cycle, For parking and rest functions; The formula for calculating rapid acceleration is: ;in, Accelerate to the current moment, Current vehicle speed The speed of the vehicle at the previous moment. The time interval between two vehicle speed samplings; The calculation process for the rapid deceleration is completely symmetrical to that for the rapid acceleration, but in the opposite direction.

5. The vehicle safety management collaborative system based on multi-source data fusion and business closed loop as described in claim 3, characterized in that, Based on the inspection conditions, threshold detection is performed on the vehicle safety parameters to identify the driving risk event and trigger a safety warning, including: When the continuous driving time is ≥4 hours, a timeout alarm is triggered; When the parking time is ≥30 minutes, the continuous driving time is recalculated; When the cumulative driving time within 24 hours is ≥8 hours, the driving risk event is determined to be fatigue driving.

6. A collaborative method for vehicle safety management based on multi-source data fusion and business closed loop, implemented through the aforementioned collaborative system for vehicle safety management based on multi-source data fusion and business closed loop, characterized in that... include: Based on the driving risk events, multi-dimensional correlation data between the target driver and the target vehicle is obtained; The multi-dimensional associated data includes: driver facial features, vehicle status, alarm events, and driver history records; The driving risk events and the multi-dimensional related data are analyzed using a risk assessment model to obtain comprehensive risk assessment results and handling recommendations. Based on the comprehensive risk assessment results and the proposed solutions, a solution instruction is generated, and a collaborative solution operation is performed according to the solution instruction to obtain the solution result. The effectiveness of the treatment results is verified according to the preset completion criteria, and the multi-dimensional correlation data is updated based on the verification results.

7. The collaborative method for vehicle safety management based on multi-source data fusion and business closed loop as described in claim 6, characterized in that, Risk assessment is conducted using a risk assessment model to analyze the driving risk events and the multi-dimensional correlation data, resulting in a comprehensive risk assessment and response recommendations, including: Fatigue risk is calculated based on the driving risk event and the driver's facial features; the formula for calculating fatigue risk is: ;in, The total fatigue risk score, The risk score for eye condition. To assess the risk score for yawning, This is a risk score for driving time; The alarm risk is calculated based on the alarm event; the formula for calculating the alarm risk is: ;in, The total score for alarm risk. The attenuation coefficient is... The sum of weighted events; The vehicle speed risk is calculated based on the vehicle status. The risk level is calculated based on the driver's historical records; the formula for calculating the risk level is: ;in, The total risk score for archives. This is a coefficient representing driving experience. This is the violation coefficient. Accident coefficient; The overall comprehensive risk is obtained by weighting and summing the fatigue risk, the alarm risk, the vehicle speed risk, and the file risk.

Images