Risk assessment system based on block chain

Abstract

The invention discloses a risk assessment system based on a block chain, and the system comprises a driving data collection and processing module, a driving behavior state recognition module, and a driving behavior risk assessment early warning module. The driving behavior state recognition module is used for establishing a driving behavior state recognition model according to a driving track to perform driving behavior state recognition, and the driving behavior risk assessment and early warning module is used for performing risk analysis and early warning triggering based on feature parameter extraction corresponding to different recognized driving behavior states. According to the invention, risk analysis is carried out based on feature parameter extraction corresponding to the identified different driving behavior states, and the collision risk of the vehicle in the different driving behavior states is assessed, judged and pre-warned through the minimum safety distance model, so that the problem that the interactivity between the vehicles and the real-time applicability of risk assessment are not considered is solved; and the accuracy of driving behavior risk assessment is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of risk assessment, and in particular to a risk assessment system based on blockchain. Background Art

[0002] The intelligent vehicle system is a comprehensive system that integrates multidisciplinary advanced technologies. Through advanced driving assistance and automatic driving systems, traffic accidents can be effectively reduced and avoided. It is the core means to solve road traffic safety problems. However, some driving users who operate improperly fail to turn on the turn signal in time before actually performing the turning action. Also, due to the limitations of the sensor detection range and the technical level of the warning algorithm, the lane change warning misses and false alarms occur frequently, and assisted driving is not enough to truly assist the driver in driving. At present, the targets of most intelligent vehicles with longitudinal control systems are limited to the current lane and cannot respond accurately to merging vehicles. The trajectory prediction method based on the vehicle physical model is suitable for short-term motion trajectory prediction. Although the trajectory prediction combined with driving intention can track the changes in the trajectory in the long-term domain, most of them do not consider the interactivity of multiple vehicles, and the real-time applicability of risk assessment is not strong. Therefore, it is necessary to design a blockchain-based risk assessment system to improve the accuracy of assisted driving risk assessment and expand the comprehensiveness of risk prediction analysis. Summary of the invention

[0003] The purpose of the present invention is to provide a risk assessment system based on blockchain to solve the problems raised in the above background technology.

[0004] In order to solve the above technical problems, the present invention provides the following technical solutions: a risk assessment system based on blockchain, comprising a driving data collection and processing module, a driving behavior state recognition module, and a driving behavior risk assessment and early warning module, characterized in that: the driving data collection and processing module is used to collect the driving-related data of the vehicle at the current moment based on the on-board sensor; the driving behavior state recognition module is used to establish a state recognition model of driving behavior according to the driving trajectory to perform driving behavior state recognition; the driving behavior risk assessment and early warning module is used to perform risk analysis and early warning triggering based on the extraction of corresponding feature parameters of different identified driving behavior states.

[0005] According to the above technical solution, the driving data collection and processing module includes: a sensor monitoring and collection module, an information data preprocessing module, and a feature information extraction and classification module. The sensor monitoring and collection module is used to use each sensor to monitor and collect relevant data corresponding to the vehicle driving behavior; the information data preprocessing module is used to perform local preprocessing on the vehicle information collected by each sensor; the feature information extraction and classification module is used to extract feature data that can represent the overall information from the original data collected by the sensor.

[0006] According to the above technical solution, the driving behavior state recognition module includes: a driving behavior recognition module, a vehicle driving trajectory analysis and prediction module, and a driving behavior decision module. The driving behavior recognition module is used to identify the driving behavior using a machine learning algorithm; the vehicle driving trajectory analysis and prediction module is used to use the trajectory of the vehicle at the current moment to predict the vehicle driving trajectory at the next moment; the driving behavior decision module is used to form an optional driving behavior decision by extracting historical trajectory data from a data set.

[0007] According to the above technical solution, the driving behavior risk assessment and warning module includes: a driving behavior state recognition module, a driving state judgment module, a risk quantification assessment establishment module, and a collision risk analysis module. The driving behavior state recognition module is used to establish a driving behavior state recognition model to identify the driving behavior state; the driving state judgment module is used to use the vehicle driving state data to judge the current vehicle driving state; the risk quantification assessment establishment module is used to establish a risk quantification assessment model under different driving states; the collision risk analysis module is used to evaluate the collision probability through the vehicle geometric model to obtain the overall collision risk; the warning mechanism triggering module is used to calculate the vehicle's driving evaluation value, and compare the value with the standard value to determine whether to trigger the warning mechanism.

[0008] According to the above technical solution, the operation method of the risk assessment system includes:

[0009] The current driving data of the vehicle collected by the on-board sensors;

[0010] Conduct qualitative analysis on the vehicle's current driving behavior and predict the trajectory distribution of the vehicle's driving behavior;

[0011] Establish a state recognition model of driving behavior based on driving trajectory;

[0012] Risk analysis is performed based on the extraction of corresponding characteristic parameters of different driving behavior states;

[0013] The minimum safety distance model is used to evaluate, judge and warn of collision risks under different driving behavior states of the vehicle.

[0014] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention, through the driving-related data of the vehicle at the current moment collected by the on-board sensor; performs a qualitative analysis on the driving behavior of the vehicle at the current moment, predicts the trajectory distribution of the vehicle's driving behavior, establishes a state recognition model of the driving behavior according to the driving trajectory, extracts corresponding characteristic parameters based on the different driving behavior states identified to perform risk analysis, evaluates and judges the collision risk of the vehicle under different driving behavior states through the minimum safety distance model, uses the risk assessment and warning model to obtain the perception data information and operation scene data information around the vehicle through sensors, analyzes and evaluates the influence of other traffic participants on the vehicle's driving behavior, and warns of the operation risk, thereby improving the accuracy of risk assessment of the vehicle's driving behavior and driving state. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0016] Figure 1 A schematic diagram of the module composition of a blockchain-based risk assessment system provided in Example 2 of the present invention. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] Embodiment 1: This embodiment can be applied to the scenario of risk assessment of intelligent vehicle assisted driving. The method can be executed by a blockchain-based risk assessment system provided by this embodiment. The method specifically includes the following steps:

[0019] Step 1: driving-related data of the vehicle at the current moment collected by the vehicle-mounted sensors;

[0020] Step 2: Conduct a qualitative analysis of the vehicle's current driving behavior and predict the trajectory distribution of the vehicle's driving behavior;

[0021] Step 3: Establish a state recognition model of driving behavior based on the driving trajectory;

[0022] Step 4: Extract corresponding characteristic parameters based on the identified different driving behavior states for risk analysis;

[0023] Step 5: Use the minimum safety distance model to assess, judge and warn of the collision risk under different driving behavior states of the vehicle.

[0024] In an embodiment of the present invention, step one further includes the following steps:

[0025] Step 11: Use visual sensors to collect environmental information on the road the vehicle is traveling on, obtain obstacle data for vehicle obstacle avoidance, and use ultrasonic sensors installed on the right, middle, and left sides of the front and rear ends of the vehicle to measure the distance data of obstacles on the right, middle, and left sides in front of the vehicle and the distance data between the vehicle and the vehicle behind.

[0026] Step 12: Collect the driving state data of the vehicle through the vehicle-mounted sensor, that is, establish a driving state recognition model of the vehicle according to the original road driving state and lane-changing driving state of the vehicle, and collect the driving data of the vehicle under different driving states in real time;

[0027] Step 13: Perform local preprocessing on the vehicle information collected by each sensor, compress the local preprocessed information to obtain compressed data, and then transmit the compressed data to the information fusion center. At the same time, extract the feature data that can represent the overall information from the original data collected by the sensor to form the feature information of the collected data, and classify and merge them;

[0028] Exemplarily, the vehicle's driving condition is subject to the joint constraints of the current environment and the road. The vehicle's assisted driving end selects an optimal driving behavior based on the current driving environment and executes this driving operation. It simultaneously and randomly generates an updated driving state, and associates the updated driving state with the driving behavior selection at the previous moment. The current driving state affects the driving behavior at the next moment, and the driving selection operation of repeatedly updating the driving behavior is performed.

[0029] In this embodiment of the present invention, step 2 further includes the following steps:

[0030] Step 21: By analyzing the vehicle driving state data, determine the features most relevant to the vehicle driving behavior, use a machine learning algorithm to identify the driving behavior, and use the vehicle's steering wheel angle information to verify the driving behavior identification model;

[0031] Step 22: Based on the vehicle driving behavior targets detected by the sensor, define the vehicle driving behavior targets to be tracked, set the starting point of the vehicle driving behavior tracking trajectory, perform data association on the vehicle driving behavior targets collected and detected after the vehicle driving behavior starting point, and divide the successfully associated vehicle driving behavior targets into the driving behavior tracking trajectory;

[0032] Step 23: Use the coordinates (x i ,y i ) is used to represent the trajectory of the vehicle in a specified time period, and then the trajectory of the vehicle at the next moment is predicted by using the trajectory of the vehicle at the current moment;

[0033] Step 24: Based on the predicted trajectory information of the vehicle at the next moment, the vehicle trajectory information and the lane marking position of the vehicle are used to define the predicted vehicle trajectory range through the lane marking position to form an optional driving behavior decision.

[0034] In an embodiment of the present invention, step three further includes the following steps:

[0035] Step 31: Divide the vehicle driving state into a normal driving state and an abnormal driving state. The driving behaviors involved include normal driving, following driving and lane changing. By identifying the driving state at each moment, the lane changing driving trajectory data and non-lane changing driving trajectory data of the vehicle are determined based on the collected data set. Multiple driving trajectory feature data of the vehicle are selected to construct a vehicle driving sample data set, and a driving behavior state recognition model is established.

[0036] Step 32: By inputting the vehicle driving state data including the speed, acceleration, angular velocity and tilt degree of continuous variables, the values ​​of the vehicle and the front or side vehicle and the following vehicle in the target lane are calculated by using the time difference between the invaded vehicle leaving the potential collision zone and the rear invading vehicle arriving at the zone, and the feasibility of lane change is determined. When the value of the time difference is larger, the possibility of collision between the two vehicles is smaller, the driving safety of the two vehicles is higher, and the current vehicle driving state is a normal driving state;

[0037] Step 33: When the distance between the vehicle and the vehicle in front is greater than the minimum lane-changing safety distance, a judgment is made based on the vehicle speed. When the speed of the vehicle in front of the target lane is greater than the vehicle speed and greater than the speed of the vehicle behind the target lane, it is judged that the conditions for safe lane changing are met. At this time, the driving state of the vehicle is normal. When the speed condition cannot be directly judged, the time difference between the two vehicles in front and behind the target lane and the vehicle is calculated. When both are greater than the set safety threshold, it is a lane-changing driving behavior under normal driving conditions.

[0038] In this embodiment of the present invention, step 4 further includes the following steps:

[0039] Step 41: judging whether there is a vehicle ahead according to the vehicle-mounted sensor data; the lane-changing state when the vehicle ahead exists is a following-type lane-changing state, and the lane-changing state when the vehicle ahead does not exist is a free-type lane-changing state; analyzing each driving state respectively, defining the corresponding dangerous driving behavior, extracting the characteristic parameters of each dangerous driving behavior, using the driving state recognition model to distinguish the lane-changing state, and establishing a risk quantification assessment model under different driving states;

[0040] Step 42: Based on the single-input and single-output free driving state driving risk assessment model, the absolute value of the longitudinal acceleration, a characteristic parameter of the free driving state, is used as an input variable, and the driving risk obtained by fuzzy reasoning is output; based on the two-input and single-output following state driving risk assessment model, the two variables of the following state characteristic parameters, headway and longitudinal acceleration, are respectively input, and the driving risk obtained by fuzzy reasoning is output;

[0041] Step 43: Based on the current state of surrounding vehicles, predict the behavior trajectory of each vehicle within the detectable range around the vehicle, and calculate all possible driving ranges that the vehicle can reach from time t0 to time t1 based on the current maximum acceptable deceleration and acceleration of the vehicle;

[0042] Step 44: The drivable range of the vehicle is represented by a dot matrix, and each dot in the dot matrix represents a state point of the vehicle. The predicted trajectory of the vehicle is decomposed into position points corresponding to the vehicle's drivable range at different times. The collision time and headway between the predicted position point of the vehicle at time t1 and each state point in the vehicle's drivable range dot matrix are calculated to obtain the risk of the traffic environment around the vehicle at time t1. The risk values ​​corresponding to the dot matrix at each time contained in the time domain T are accumulated and averaged to obtain a vehicle risk assessment model in the future time domain T.

[0043] Step 45: Combined with the influencing factor data of the lane changing decision of the driving vehicle, a lane changing necessity discrimination model is established to generate a candidate intention set. Different driving intentions in the candidate intention set are used to analyze the positions of vehicles with potential intention conflicts, generate an intention conflict set, analyze and obtain the driving trajectory corresponding to the conflicting intention of the target vehicle that conflicts with the vehicle intention, and use the acceleration method to generate a candidate trajectory cluster of the vehicle, analyze the risks of the target vehicle and the vehicle driving behavior, and make driving behavior decisions based on the quantified results of the risks.

[0044] In the embodiment of the present invention, step five includes the following steps:

[0045] Based on the predicted vehicle trajectory at the next moment, collision detection is performed on each predicted trajectory, the vehicle collision risk coefficient is solved using the Monte Carlo algorithm, and the collision probability is evaluated through the vehicle geometric model to obtain the overall collision risk;

[0046] The solution method for the vehicle collision risk coefficient based on the Monte Carlo algorithm is as follows: for the given vehicle driving data, multiple driving trajectories are generated according to the Gaussian distribution and driving model. For each driving trajectory, it is judged whether a collision occurs between the target vehicle and the surrounding traffic vehicles at each sampling moment, and the number of collisions is counted. The number of trajectory conflicts or collisions is divided by the total number of vehicle trajectories to calculate the collision risk coefficient.

[0047] Using the data set of driving behavior and driving status, the real-time driving behavior data in the data is subtracted from the output of each risk model and the absolute value is taken to obtain the deviation value of each behavior from the risk model at different time points. Clustering is used to obtain the risk range data of each driving behavior with similar attributes. According to the risk type and weight of the behavior, the driving assessment value of the vehicle at this time is calculated, and the value is compared with the standard value to determine whether the early warning mechanism is triggered.

[0048] Embodiment 2: Embodiment 2 of the present invention provides a risk assessment system based on blockchain. Figure 1 A schematic diagram of the module composition of a risk assessment system based on blockchain provided in Embodiment 2 of the present invention is shown in FIG. Figure 1 As shown, the system includes:

[0049] A driving data collection and processing module is used to collect the current driving-related data of the vehicle based on the on-board sensors;

[0050] A driving behavior state recognition module is used to establish a driving behavior state recognition model according to the driving trajectory to perform driving behavior state recognition;

[0051] The driving behavior risk assessment and warning module is used to perform risk analysis and warning triggering based on the extraction of corresponding feature parameters of different identified driving behavior states.

[0052] In some embodiments of the present invention, the driving data collection and processing module includes:

[0053] The sensor monitoring and acquisition module is used to monitor and collect relevant data corresponding to the vehicle driving behavior using various sensors;

[0054] An information data preprocessing module is used to perform local preprocessing on the vehicle information collected by each sensor;

[0055] The feature information extraction and classification module is used to extract feature data that can represent the overall information from the original data collected by the sensor.

[0056] In some embodiments of the present invention, the driving behavior state recognition module includes:

[0057] A driving behavior recognition module, which is used to identify driving behavior using a machine learning algorithm;

[0058] The vehicle driving trajectory analysis and prediction module is used to predict the vehicle driving trajectory at the next moment by using the vehicle trajectory at the current moment;

[0059] The driving behavior decision module is used to form selectable driving behavior decisions by extracting historical trajectory data from the data set.

[0060] In some embodiments of the present invention, the driving behavior risk assessment and warning module includes:

[0061] A driving behavior state recognition module is used to establish a driving behavior state recognition model to recognize the driving behavior state;

[0062] A driving state determination module is used to determine the current vehicle driving state using the vehicle driving state data;

[0063] A risk quantification assessment establishment module is used to establish a risk quantification assessment model under different driving conditions;

[0064] The collision risk analysis module is used to evaluate the collision probability through the vehicle geometry model to obtain the overall collision risk;

[0065] The early warning mechanism triggering module is used to calculate the driving evaluation value of the vehicle and compare the value with the standard value to determine whether to trigger the early warning mechanism.

[0066] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0067] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A risk assessment system based on blockchain, comprising a driving data collection and processing module, a driving behavior state recognition module, and a driving behavior risk assessment and early warning module, characterized in that: The driving data collection and processing module is used to collect the driving-related data of the vehicle at the current moment based on the vehicle-mounted sensors; The driving behavior state recognition module is used to establish a driving behavior state recognition model according to the driving trajectory to perform driving behavior state recognition; the driving behavior risk assessment and early warning module is used to perform risk analysis and early warning triggering based on the extraction of corresponding feature parameters of different identified driving behavior states.

2. A blockchain-based risk assessment system according to claim 1, characterized in that: The driving data collection and processing module includes: a sensor monitoring and collection module, an information data preprocessing module, and a feature information extraction and classification module. The sensor monitoring and collection module is used to use each sensor to monitor and collect relevant data corresponding to the vehicle driving behavior; the information data preprocessing module is used to perform local preprocessing on the vehicle information collected by each sensor; the feature information extraction and classification module is used to extract feature data that can represent the overall information from the original data collected by the sensor.

3. A blockchain-based risk assessment system according to claim 2, characterized in that: The driving behavior state recognition module includes: a driving behavior recognition module, a vehicle driving trajectory analysis and prediction module, and a driving behavior decision module. The driving behavior recognition module is used to identify driving behavior using a machine learning algorithm; the vehicle driving trajectory analysis and prediction module is used to use the vehicle's trajectory at the current moment to predict the vehicle's driving trajectory at the next moment; the driving behavior decision module is used to form an optional driving behavior decision by extracting historical trajectory data from a data set.

4. A blockchain-based risk assessment system according to claim 3, characterized in that: The driving behavior risk assessment and early warning module includes: a driving behavior state recognition module, a driving state judgment module, a risk quantification assessment establishment module, and a collision risk analysis module. The driving behavior state recognition module is used to establish a driving behavior state recognition model to identify the driving behavior state; the driving state judgment module is used to use the vehicle driving state data to judge the current vehicle driving state; the risk quantification assessment establishment module is used to establish a risk quantification assessment model under different driving states; the collision risk analysis module is used to evaluate the collision probability through the vehicle geometric model to obtain the overall collision risk; the early warning mechanism triggering module is used to calculate the vehicle's driving evaluation value, and compare the value with the standard value to determine whether to trigger the early warning mechanism.

5. A blockchain-based risk assessment system according to claim 4, characterized in that: The operation method of the risk assessment system includes: The current driving data of the vehicle collected by the on-board sensors; Conduct qualitative analysis on the vehicle's current driving behavior and predict the trajectory distribution of the vehicle's driving behavior; Establish a state recognition model of driving behavior based on driving trajectory; Risk analysis is performed based on the extraction of corresponding characteristic parameters of different driving behavior states; The minimum safety distance model is used to evaluate, judge and warn of collision risks under different driving behavior states of the vehicle.

6. A blockchain-based risk assessment system according to claim 5, characterized in that: The data collection based on the vehicle-mounted sensor includes: Use visual sensors to collect environmental information on the road the vehicle is traveling on, obtain obstacle data for vehicle obstacle avoidance, and collect vehicle driving status data through on-board sensors. That is, establish a vehicle driving status recognition model based on the vehicle's original road driving status and lane change driving status, and collect vehicle driving data under different driving states in real time; The vehicle information collected by each sensor is locally preprocessed, and at the same time, the feature data that can represent the overall information in the original data collected by the sensor is extracted to form the feature information of the collected data, which is then classified and merged.

7. A blockchain-based risk assessment system according to claim 6, characterized in that: The trajectory distribution of the predicted vehicle driving behavior includes: By analyzing the vehicle driving status data, the most relevant features of the vehicle driving behavior are determined, the driving behavior is identified using a machine learning algorithm, and the driving behavior identification model is verified using the vehicle's steering wheel angle information; Based on the vehicle driving behavior targets detected by the sensor, define the vehicle driving behavior targets that need to be tracked, set the starting point of the vehicle driving behavior tracking trajectory, perform data association on the vehicle driving behavior targets collected and detected after the vehicle driving behavior starting point, and divide the successfully associated vehicle driving behavior targets into the driving behavior tracking trajectory; The position of the vehicle on the road at time t is expressed by coordinates (x i ,y i ) is used to represent the trajectory of the vehicle in a specified time period, and then the trajectory of the vehicle at the next moment is predicted by using the trajectory of the vehicle at the current moment; Based on the predicted trajectory information of the vehicle at the next moment, the vehicle trajectory information and the lane marking position of the vehicle are used to define the predicted vehicle trajectory range through the lane marking position. By extracting historical trajectory data from the data set, an optional driving behavior decision is formed.

8. A blockchain-based risk assessment system according to claim 7, characterized in that: The state recognition model of driving behavior established according to the driving trajectory includes: The vehicle driving state is divided into normal driving state and abnormal driving state. The driving behaviors involved include normal driving, following driving and lane changing. By identifying the driving state at each moment, the vehicle lane changing driving trajectory data and non-lane changing driving trajectory data are determined based on the collected data set. Multiple driving trajectory feature data of the vehicle are selected to construct a vehicle driving sample data set and establish a driving behavior state recognition model. By inputting the vehicle driving state data including speed, acceleration, angular velocity, and tilt degree as continuous variables, the time difference between the invaded vehicle leaving the potential collision zone and the subsequent invading vehicle arriving at the zone is used to calculate the value of the vehicle and the front or side vehicle and the following vehicle in the target lane, and judge the feasibility of lane change. The larger the value of the time difference, the smaller the possibility of collision between the two vehicles, the higher the driving safety of the two vehicles, and the current vehicle driving state is a normal driving state; When the distance between the vehicle and the vehicle in front is greater than the minimum safe distance for lane changing, a judgment is made based on the vehicle speed. When the speed of the vehicle in front of the target lane is greater than the vehicle speed and greater than the speed of the vehicle behind the target lane, it is judged that there are safe lane changing conditions. At this time, the vehicle's driving state is normal. When the speed condition cannot be directly judged, the time difference between the two vehicles in front and behind the target lane and the vehicle is calculated. When both are greater than the set safety threshold, it is a lane changing behavior under normal driving conditions.

9. A blockchain-based risk assessment system according to claim 8, characterized in that: The risk analysis includes: According to the on-board sensor data, it is judged whether there is a vehicle ahead. The lane-changing state of the vehicle ahead is the following-type lane-changing state, and the lane-changing state of the vehicle ahead is the free-type lane-changing state. Each driving state is analyzed separately, the corresponding dangerous driving behavior is defined, and the characteristic parameters of each dangerous driving behavior are extracted. The driving state recognition model is used to distinguish the lane-changing state, and a risk quantification assessment model under different driving states is established; Based on the single-input and single-output free driving state driving risk assessment model, the absolute value of the longitudinal acceleration, a characteristic parameter of the free driving state, is used as the input variable, and the driving risk obtained by fuzzy reasoning is output. Based on the two-input and single-output following state driving risk assessment model, the two variables of the following state characteristic parameters, headway and longitudinal acceleration, are input respectively, and the driving risk obtained by fuzzy reasoning is output. Based on the current state of surrounding vehicles, the behavior trajectory of each vehicle within the detectable range around the vehicle is predicted, and the possible driving range of the vehicle from time to time is calculated based on the vehicle's current maximum acceptable deceleration and acceleration; The dot matrix represents the drivable range of the vehicle, and each dot in the dot matrix represents a state point of the vehicle. The predicted vehicle trajectory is decomposed into position points corresponding to the vehicle's driving range at different times. The collision time and headway between the predicted vehicle position point at the moment and each state point in the vehicle's driving range dot matrix are calculated to obtain the risk of the traffic environment around the vehicle at the moment. The risk values ​​corresponding to each dot matrix at each moment in the time domain T are accumulated and averaged to obtain the vehicle risk assessment model in the future time domain T. Combined with the data of influencing factors of driving vehicle lane changing decisions, a lane changing necessity discrimination model is established to generate a candidate intention set. The different driving intentions in the candidate intention set are used to analyze the positions of vehicles with potential intention conflicts, generate an intention conflict set, and analyze the driving trajectory corresponding to the conflicting intention of the target vehicle that conflicts with the vehicle intention. The acceleration method is used to generate a candidate trajectory cluster of the vehicle, analyze the risks of the target vehicle and the vehicle driving behavior, and make driving behavior decisions based on the quantified results of the risk.

10. A blockchain-based risk assessment system according to claim 9, characterized in that: The assessment and early warning judgment include: Based on the predicted vehicle trajectory at the next moment, collision detection is performed on each predicted trajectory, the vehicle collision risk coefficient is solved using the Monte Carlo algorithm, and the collision probability is evaluated through the vehicle geometric model to obtain the overall collision risk; Using the data set of driving behavior and driving status, the real-time driving behavior data in the data is subtracted from the output of each risk model and the absolute value is taken to obtain the deviation value of each behavior from the risk model at different time points. Clustering is used to obtain the risk range data of each driving behavior with similar attributes. According to the risk type and weight of the behavior, the driving assessment value of the vehicle at this time is calculated, and the value is compared with the standard value to determine whether the early warning mechanism is triggered.

Images