A method for evaluating and analyzing big data of smart car user behavior

Abstract

The present invention belongs to the technical field of automobile user behavior analysis, and relates to an intelligent automobile user behavior big data evaluation and analysis method, which defines the overall driving style of the target user and determines the corresponding basic driving risk coefficient through the historical driving behavior data of the target user, further predicts the driving behavior risk coefficient of the target user for each road section to be traveled, combines the two and uses weather conditions, traffic conditions and user conditions as correction factors, analyzes the comprehensive driving risk coefficient of the target user for each road section to be traveled, screens each prompt road section and retrieves the potential driving risk items corresponding to the target user, and performs corresponding voice broadcast prompts to the target user at the arrival time of the starting point of each prompt road section. After the target user finishes the current driving navigation route, the driving behavior correction execution ability of the target user is evaluated and fed back, effectively improving the existing driving behavior standard constraint mechanism, thereby ensuring the driving safety of automobile users.

Description

Technical Field

[0001] The present invention belongs to the technical field of automobile user behavior analysis and relates to a method for evaluating and analyzing big data of intelligent automobile user behavior. Background Art

[0002] In the rapidly developing transportation sector, smart cars are a new travel option, and the driving behavior of their users plays a decisive role in road safety, traffic efficiency, and vehicle performance and life. Irrational driving behavior not only increases the risk of accidents, but also leads to increased energy consumption, increased emissions, and increased vehicle wear. Given the current lack of driving behavior constraint mechanisms, it is urgent to build a smart car user behavior big data evaluation and analysis method to accurately evaluate driving behavior and promote safe, green, and efficient travel.

[0003] Although the existing technology for user driving restraint through automobile user behavior analysis can identify and correct bad driving behavior to a certain extent and improve road safety and traffic efficiency, it still has several limitations, which are as follows: 1. The data coverage and accuracy of the existing technology for user driving behavior analysis are incomplete and insufficient. It is limited by the lane change data and speed data during driving, and ignores the analysis of data related to following distance, line crossing and turning, which affects the accuracy and comprehensiveness of the user driving behavior analysis results.

[0004] 2. Existing technologies mostly focus on the management of behavioral constraints throughout the driving route, regulating driving behavior by establishing clear behavioral constraints. However, this also exposes the problem of insufficient analysis of the constraint requirements of each section within the route. On the one hand, for drivers in familiar sections or safe driving environments, overly strict constraints may cause unnecessary disgust or resistance. On the other hand, in a complex and changeable traffic environment, clear behavioral constraints may not fully meet the actual driving conditions, which may lead to excessive intervention in the user's driving behavior. Summary of the invention

[0005] In view of this, in order to solve the problems raised in the above background technology, a method for evaluating and analyzing big data of smart car user behavior is proposed.

[0006] The purpose of the present invention can be achieved through the following technical solutions: The present invention provides a method for evaluating and analyzing big data of user behavior of smart cars, including: S1. Retrieving historical driving behavior data for a target user in a vehicle terminal, including speed-time history curves, average following distance, lane change frequency, number of line crossings, turning amplitude and angular velocity for each driving trip process for each driving section within a historical preset period, defining the overall driving style of the target user and determining its corresponding basic driving risk coefficient.

[0007] S2. Divide the target user's current navigation route into sections to be driven according to a preset path distance, and predict the target user's driving behavior risk coefficient for each section to be driven based on the target user's historical driving behavior data.

[0008] S3. Using weather conditions, traffic conditions and user conditions as correction factors, analyze the comprehensive driving risk coefficient of the target user for each to-be-traveled section, and mark each to-be-traveled section whose comprehensive driving risk coefficient is greater than a preset value as a prompt section.

[0009] S4. Retrieve the potential driving risk items of the target user for each prompted road section, and give the target user a corresponding voice broadcast prompt at the time of arrival at the starting point of each prompted road section.

[0010] S5. After the target user finishes the current navigation route, evaluate and provide feedback on the target user's driving behavior correction execution efforts.

[0011] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) The present invention comprehensively and deeply defines the overall driving style of the target user from three levels: driving standard index, driving stability index, and driving robustness index by combining multiple driving behavior parameters such as lane change, speed, acceleration, following distance, steering amplitude, and angular velocity. It can more comprehensively and accurately reflect the driver's driving habits and style, which is more scientific than a single indicator and avoids errors caused by one-sided judgment.

[0012] (2) The present invention is based on the historical driving behavior data of the target user, focuses on the reference performance of historical driving trips or similar driving trips within a preset historical period of each section to be driven, and predicts the driving behavior risk coefficient of the target user for each section to be driven, effectively improving the accuracy and practicality of the prediction results, and providing data support for the subsequent screening of prompted sections.

[0013] (3) The present invention introduces weather conditions, traffic conditions and user conditions as correction factors to effectively and comprehensively analyze the comprehensive driving risk coefficient of the target user for each road section to be driven, ensuring that the analysis results are closer to the actual situation, thereby enhancing adaptability and practicality.

[0014] (4) The present invention provides corresponding voice broadcast prompts to the target user through the arrival time of the starting point of each prompt road section. The non-intrusive prompts not only avoid excessive interference of over-specific constraints in the driving process, but also improve the overall user experience.

[0015] (5) After the target user finishes the current driving navigation route, the present invention evaluates and provides feedback on the target user's driving behavior correction execution efforts, helping to develop a mechanism for continuous tracking and improvement, thereby helping the user to gradually form good driving habits and maintain a high level of safety in future driving processes. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0017] Figure 1 The present invention is a flowchart of the steps for implementing the method.

[0018] Figure 2 It is a reference schematic diagram of a speed-time history curve of a certain driving trip process for a certain driving section within a preset historical period of a target user of the present invention. DETAILED DESCRIPTION

[0019] 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.

[0020] See also Figure 1 As shown, the present invention provides a method for evaluating and analyzing big data of user behavior of smart cars, including: S1. Retrieving historical driving behavior data for a target user in a vehicle terminal, including speed-time history curves, average following distance, lane change frequency, number of line crossings, turning amplitude and angular velocity for each driving trip process for each driving section within a historical preset period, defining the overall driving style of the target user and determining its corresponding basic driving risk coefficient.

[0021] It should be noted that the above speed time history curve represents the history value recorded as the speed changes over time, and each driving trip process within the above historical preset period is also obtained by dividing each driving section according to the preset path distance.

[0022] Reference Figure 2 As shown, specifically, the definition of the overall driving style of the target user includes: extracting the speed time history curve of each driving trip process of the target user in the preset historical period for each driving section and substituting it into a rectangular coordinate system, and obtaining the speed limit threshold of each driving section of each driving trip process of the target user in the preset historical period. , draw a straight line corresponding to the speed limit threshold in the same rectangular coordinate system, identify the abnormal speeding behavior of the target user's driving trip process for each driving section within the historical preset period, and analyze the driving standard index of the target user's driving trip process for each driving section within the historical preset period ,in is the number of each driving trip within the historical preset period, , is the number of each driving section, .

[0023] It should be noted that the above The specific analysis process is as follows: identifying abnormal speeding behavior at each driving time point where the speed value on the speed time history curve of each driving section during each driving trip of the target user within the preset historical period is greater than the speed limit threshold of the corresponding driving section, performing difference calculation between the speed value corresponding to each speeding time point on the speed time history curve of each driving section during each driving trip of the target user within the preset historical period and the speed limit threshold of the corresponding driving section, obtaining the speeding amplitude corresponding to each speeding time point of each driving section during each driving trip of the target user within the preset historical period, and counting the cumulative speeding time of each driving section during each driving trip of the target user within the preset historical period and the accumulated overspeed amplitude , according to the formula Obtain the driving norm index of each driving trip process for each driving section within the historical preset period of the target user, where The target user's history within a preset period recorded by the vehicle terminal The driving trip process is The driving time for each driving segment.

[0024] Obtain the best fitting function corresponding to the speed time history curve of each driving section of the target user's historical preset period and perform first-order derivative, identify the sudden acceleration and braking behaviors of each driving section of the target user's historical preset period, and combine the steering amplitude of each driving section of the target user's historical preset period. and angular velocity , analyze the driving stability index of each driving section during each driving trip process of the target user within the historical preset period .

[0025] It should be noted that the best fitting function corresponding to the speed time history curve of each driving section of each driving trip process within the above-mentioned target user's historical preset period is obtained by substituting the speed time history curve into Matlab software and using the best fitting tool of the software.

[0026] Above The specific analysis process is as follows: each driving time point on the speed time history curve of each driving section in each driving trip process of the target user in the preset historical period is substituted into the best fitting function after its first-order derivative, and the corresponding acceleration of each driving time point in each driving section in each driving trip process of the target user in the preset historical period is obtained. If the corresponding acceleration of a driving time point is greater than or equal to the preset acceleration threshold of the sudden acceleration behavior, then it is recognized that there is a sudden acceleration behavior at the driving time point; if the corresponding acceleration of the driving time point is less than or equal to the preset acceleration threshold of the sudden braking behavior, then it is recognized that there is a sudden braking behavior at the driving time point; the acceleration deviation ratios of each sudden acceleration time point and each sudden braking time point in each driving trip process of the target user in the preset historical period are summarized, and the sudden acceleration degree of each driving section in each driving trip process of the target user in the preset historical period is counted. and emergency braking , combined with the WEB cloud preset safe driving conditions corresponding to the reasonable steering reference range and reasonable reference angular velocity , according to the formula Analyze the driving stability index of each driving trip process for each driving section within the historical preset period of the target user.

[0027] For example, the preset acceleration threshold for the sudden acceleration behavior may be 4 The preset acceleration threshold for emergency braking behavior can be -5 .

[0028] Extract the average following distance of each driving trip process for each driving section during the target user's historical preset period and lane change frequency , analyze the driving robustness index of each driving trip process for each driving section within the target user's historical preset period .

[0029] It should be noted that the above The specific analysis process is as follows: extract the reasonable following reference distance corresponding to the safe driving conditions preset in the WEB cloud and reasonable lane change reference frequency under the preset path distance , according to the formula The driving robustness index of each driving trip process for each driving section within the historical preset period of the target user is analyzed.

[0030] By formula Obtain the target user's driving behavior evaluation index for the preset historical period, where is the number of driving trips within a preset historical period, is the number of driving sections, The preset driving standard index, driving stability index, and driving steadiness index correspond to the weight ratios respectively. According to the preset driving behavior evaluation index intervals of each driving style stored in the WEB cloud, the driving style corresponding to the preset driving behavior evaluation index interval of the target user's historical preset period is defined as the overall driving style of the target user.

[0031] For example, the above Can be specifically .

[0032] Specifically, determining the basic driving risk coefficient corresponding to the overall driving style of the target user includes: presetting basic driving risk coefficients for various driving styles stored in the WEB cloud, and extracting and determining the preset basic driving risk coefficient corresponding to the overall driving style of the target user.

[0033] The embodiment of the present invention comprehensively and deeply defines the overall driving style of the target user from three levels: driving standard index, driving stability index, and driving robustness index by combining multiple driving behavior parameters such as lane change, speed, acceleration, following distance, steering amplitude, and angular velocity. It can more comprehensively and accurately reflect the driver's driving habits and style, is more scientific than a single indicator, and avoids errors caused by one-sided judgment.

[0034] S2. Divide the target user's current navigation route into sections to be driven according to a preset path distance, and predict the target user's driving behavior risk coefficient for each section to be driven based on the target user's historical driving behavior data.

[0035] Specifically, the predicted driving behavior risk coefficient of the target user for each road section to be traveled includes: comparing each road section to be traveled with each driving section of each driving trip process of the target user within a historical preset period, and determining whether each road section to be traveled is a road section that the target user has traveled in the past, thereby screening each road section to be traveled into known sections and unknown sections.

[0036] For a known road section, extract and summarize the target user's driving norm index for each driving trip process of the known road section in the historical preset period , Driving Stability Index , Driving Stability Index And the number of pressing lines ,in is the number of each driving trip on the known road section within the historical preset period. , according to the formula Get the target user's driving behavior risk coefficient for the known road section, where It is the number of driving trips on the known road section within a preset historical period.

[0037] For a certain unknown road section, a reference driving section with a high similarity to the unknown road section is selected from each driving section of each driving trip process within a preset historical period of the target user, so as to extract and summarize the driving standard index, driving stability index, driving robustness index and number of line crossings of each driving trip process of the reference driving section for the unknown road section within the preset historical period of the target user, and similarly calculate the driving behavior risk coefficient of the target user for the unknown road section.

[0038] Based on the above predictions, the target user's driving behavior risk coefficient for each road section to be driven is obtained.

[0039] Specifically, the screening of reference driving sections that are highly similar to the unknown section includes: summarizing the same driving sections among the driving sections of each driving trip process within a preset historical period of the target user to obtain each driving trip section within the preset historical period of the target user.

[0040] The road characteristic information of the unknown road section and each driving travel section within the preset historical period of the target user is collected respectively, including road type, speed limit range, number of lanes, lane width, road slope and average road traffic flow.

[0041] The deviations of various parameters in the road feature information between the unknown road section and each driving travel section within the target user's historical preset period are calculated and accumulated, and the inverse of the accumulated value is used as the road similarity between the unknown road section and each driving travel section within the target user's historical preset period. The driving travel section corresponding to the maximum road similarity is extracted as a reference driving section that is highly similar to the unknown road section.

[0042] It should be noted that a preset constant is introduced in the process of accumulating the deviations of the parameters in the road characteristic information of the unknown road section and each driving travel road section in the preset historical period of the target user, and the reference value range can be , to ensure that the denominator is not 0 in the subsequent cumulative value inversion process.

[0043] The embodiment of the present invention is based on the historical driving behavior data of the target user, focuses on the reference performance of historical driving trips or similar driving trips within a preset historical period of each section to be driven, predicts the driving behavior risk coefficient of the target user for each section to be driven, effectively improves the accuracy and practicality of the prediction results, and provides data support for the screening of subsequent prompt sections.

[0044] S3. Using weather conditions, traffic conditions and user conditions as correction factors, analyze the comprehensive driving risk coefficient of the target user for each to-be-traveled section, and mark each to-be-traveled section whose comprehensive driving risk coefficient is greater than a preset value as a prompt section.

[0045] Specifically, the analysis of the comprehensive driving risk coefficient of the target user for each road section to be traveled includes: extracting the driving behavior risk coefficient of the target user for each road section to be traveled And the basic driving risk coefficient corresponding to the overall driving style of the target user ,in is the number of each road section to be traveled. , introduce the correction factors of the weather conditions, traffic conditions and user conditions of each road section to be driven, which are recorded as , according to the formula Get the comprehensive driving risk coefficient of the target user for each road section to be driven. for .

[0046] Specifically, the process of obtaining the correction factor of the weather condition of each current section to be traveled includes: retrieving the meteorological information of the target user's current area through the meteorological station, including rainfall, snowfall and air visibility, and comparing it with the permitted rainfall threshold, permitted snowfall threshold and permitted air visibility threshold corresponding to the driving impact of the reasonable meteorological environment stored in the WEB cloud. If the rainfall, snowfall and air visibility in the target user's current area are all less than or equal to their corresponding permitted thresholds, the correction factors of the weather conditions of each current section to be traveled are set to 0. Conversely, if any meteorological parameter among the rainfall, snowfall and air visibility in the target user's current area is greater than its corresponding permitted threshold, the deviation ratio of each meteorological parameter greater than its corresponding permitted threshold is calculated, and the accumulated value is used as the correction factor of the weather condition of each current section to be traveled.

[0047] It should be noted that, if the rainfall in the target user's current area is greater than its corresponding permitted threshold, the rainfall in the target user's current area is subtracted from the corresponding permitted threshold, and the ratio of the difference to the corresponding permitted threshold is taken as the deviation ratio of the rainfall in the target user's current area. Similarly, when the snowfall or air visibility in the target user's current area is greater than its corresponding permitted threshold, the deviation ratio of the snowfall or air visibility in the target user's current area is analyzed.

[0048] Specifically, the process of obtaining the correction factor of the traffic conditions of each current section to be traveled includes: retrieving the traffic density of each current section to be traveled by the target user through the navigation software, obtaining the traffic congestion level of each current section to be traveled by the target user based on the road traffic density interval of each traffic congestion level specification stored in the WEB cloud, setting the driving impact weight corresponding to each traffic congestion level, so as to obtain the driving impact weight corresponding to the traffic congestion level of each current section to be traveled by the target user, and further using it as the correction factor of the traffic conditions of each current section to be traveled.

[0049] It should be noted that the basis for setting the driving impact weights corresponding to the above-mentioned traffic congestion levels is that the traffic congestion levels can refer to the classification specifications of the urban traffic management evaluation index system, specifically into four levels: smooth, lightly congested, congested and severely congested. Under different traffic congestion levels, vehicle speed, driving operation difficulty, driving psychological pressure and accident risk are very different, and the corresponding driving impact weights gradually increase.

[0050] For example, the driving impact weights corresponding to the above-mentioned traffic congestion levels can be specifically set to 0.1 for smooth traffic, 0.2 for light congestion, 0.3 for congestion, and 0.4 for severe congestion.

[0051] Specifically, the process of obtaining the correction factor of the user status of each current section to be traveled includes: sensing the driving time of the target user's current navigation route through the vehicle-mounted equipment, and accumulating the cumulative driving time of the target user to arrive at each section to be traveled based on the arrival time of the target user at each section to be traveled predicted by the navigation software, and using the ratio of the cumulative driving time to the preset user fatigue driving time as the correction factor of the user status of each current section to be traveled.

[0052] The embodiment of the present invention introduces weather conditions, traffic conditions and user conditions as correction factors to effectively and comprehensively analyze the comprehensive driving risk coefficient of the target user for each road section to be driven, ensuring that the analysis result is closer to the actual situation, thereby enhancing adaptability and practicality.

[0053] S4. Retrieve the potential driving risk items of the target user for each prompted road section, and give the target user a corresponding voice broadcast prompt at the time of arrival at the starting point of each prompted road section.

[0054] It should be noted that the above-mentioned retrieval process of the potential driving risk items of the target user for each prompted road section includes: calculating the average of the following distance, lane change frequency, line crossing number of times, turning amplitude and angular velocity of each driving trip process for each driving section within the historical preset period of the target user, and obtaining the following distance, lane change frequency, line crossing number of times, turning amplitude and angular velocity of the target user's driving trip in the historical preset period. If a certain prompted road section is a known road section, the average following distance, lane change frequency, line crossing number of times, turning amplitude and angular velocity of each driving trip process in the historical preset period of the prompted road section are extracted, and compared with the average reference value of the corresponding parameter of the driving trip in the historical preset period of the target user. If any parameter is greater than the average reference value of the driving trip in the historical preset period, it indicates that there is a potential driving risk, and the corresponding parameter item is used as a potential driving risk item, and the speed time history curve of each driving trip process in the historical preset period of the prompted road section is extracted to identify whether there is abnormal speeding behavior, sudden braking behavior, and sudden acceleration behavior. If any behavior exists, it is also used as a potential driving risk item.

[0055] If a certain prompted road section is an unknown road section, the speed-time history curve, average following distance, lane change frequency, number of line crossings, turning amplitude and angular velocity of each driving trip process within a preset historical period of the reference driving section of the prompted road section are extracted, and potential driving risk items are retrieved similarly.

[0056] The embodiment of the present invention provides corresponding voice broadcast prompts to the target user through the arrival time of the starting point of each prompt road section. The non-intrusive prompts not only avoid excessive intervention of over-specific constraints in the driving process, but also improve the overall user experience.

[0057] S5. After the target user finishes the current navigation route, evaluate and provide feedback on the target user's driving behavior correction execution efforts.

[0058] Specifically, the evaluation of the target user's driving behavior correction execution strength includes: collecting the speed time history curve, average following distance, lane change frequency, line crossing number, turning amplitude and angular velocity of each prompted section of the target user, and calculating the actual driving behavior risk coefficient of each prompted section of the target user.

[0059] The predicted target user's driving behavior risk coefficient for each prompted road section is extracted, and the difference is made from the target user's actual driving behavior risk coefficient for each prompted road section. The difference is accumulated and the accumulated value is used as an evaluation indicator of the target user's driving behavior correction execution strength.

[0060] After the target user finishes the current driving navigation route, the embodiment of the present invention evaluates and provides feedback on the target user's driving behavior correction execution strength, helping to develop a mechanism for continuous tracking and improvement, thereby helping the user to gradually form good driving habits and maintain a high level of safety in future driving processes.

[0061] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they shall all fall within the protection scope of the present invention.

Claims

1. A method for evaluating and analyzing big data of smart car user behavior, characterized in that: The method includes: S1. Retrieve the historical driving behavior data of the target user in the vehicle terminal, including the speed-time history curve, average following distance, lane change frequency, line crossing times, steering amplitude and angular velocity of each driving trip process for each driving section within the historical preset period, define the overall driving style of the target user and determine the corresponding basic driving risk coefficient; S2. Divide the target user's current navigation route into sections to be driven according to the preset path distance, and predict the target user's driving behavior risk factor for each section to be driven based on the target user's historical driving behavior data; S3. Using weather conditions, traffic conditions and user conditions as correction factors, analyzing the comprehensive driving risk coefficient of the target user for each section to be driven, and marking each section to be driven where the comprehensive driving risk coefficient is greater than a preset value as each prompt section; S4. Retrieve the target user's potential driving risk items for each prompted section, and give the target user a corresponding voice broadcast prompt at the arrival time of each prompted section starting point; S5. After the target user finishes the current navigation route, evaluate and provide feedback on the target user's driving behavior correction execution; The predicting of the target user's driving behavior risk coefficient for each road section to be traveled includes: comparing each road section to be traveled with each driving section of each driving trip process of the target user in a historical preset period, determining whether each road section to be traveled is a road section that the target user has traveled in the past, thereby screening each road section to be traveled into each known road section and each unknown road section; For a known road section, extract and summarize the target user's driving norm index for each driving trip process of the known road section in the historical preset period , Driving Stability Index , Driving Stability Index And the number of pressing lines ,in is the number of each driving trip on the known road section within the historical preset period. , according to the formula Get the target user's driving behavior risk coefficient for the known road section, where The number of driving trips on the known road section within a preset historical period; For a certain unknown road section, a reference driving section with a high similarity to the unknown road section is selected from each driving section of each driving trip process in the target user's historical preset period, so as to extract and summarize the driving standard index, driving stability index, driving robustness index and number of crossing the line of each driving trip process of the reference driving section of the unknown road section in the target user's historical preset period, and similarly calculate the target user's driving behavior risk coefficient for the unknown road section; Based on the above predictions, the target user's driving behavior risk coefficient for each road section to be driven is obtained; The analysis of the comprehensive driving risk coefficient of the target user for each road section to be traveled includes: extracting the driving behavior risk coefficient of the target user for each road section to be traveled And the basic driving risk coefficient corresponding to the overall driving style of the target user ,in is the number of each road section to be traveled. , introduce the correction factors of the weather conditions, traffic conditions and user conditions of each road section to be driven, which are recorded as , according to the formula Get the comprehensive driving risk coefficient of the target user for each road section to be driven. for .

2. The method for evaluating and analyzing big data of smart car user behavior according to claim 1, characterized in that: The method of defining the overall driving style of the target user includes: extracting the speed time history curve of each driving trip process of the target user in the preset historical period for each driving section and substituting it into a rectangular coordinate system, obtaining the speed limit threshold of each driving section of each driving trip process of the target user in the preset historical period, drawing a straight line corresponding to the speed limit threshold in the same rectangular coordinate system, identifying abnormal speeding behavior of each driving trip process of the target user in the preset historical period for each driving section, and analyzing the driving standard index of each driving trip process of the target user in the preset historical period for each driving section. ,in is the number of each driving trip within the historical preset period, , is the number of each driving section, ; Obtain the best fitting function corresponding to the speed time history curve of each driving trip process for each driving section in the target user's historical preset period and perform first-order derivative, identify the sudden acceleration behavior and sudden braking behavior of each driving trip process for each driving section in the target user's historical preset period, and analyze the driving stability index of each driving trip process for each driving section in the target user's historical preset period by combining the turning amplitude and angular velocity of each driving trip process for each driving section in the target user's historical preset period ; Extract the average following distance and lane change frequency of each driving trip process for each driving section during the target user's historical preset period, and analyze the driving robustness index of each driving trip process for each driving section during the target user's historical preset period ; Will , , The products of the preset weights are accumulated respectively, and the accumulated values ​​are further averaged to obtain the driving behavior evaluation index of the target user in the preset historical period. According to the preset driving behavior evaluation index intervals of each driving style stored in the WEB cloud, the driving style corresponding to the preset driving behavior evaluation index interval of the driving behavior evaluation index of the target user in the preset historical period is defined as the overall driving style of the target user.

3. The method for evaluating and analyzing big data of smart car user behavior according to claim 1, characterized in that: The determining of the basic driving risk coefficient corresponding to the overall driving style of the target user includes: extracting and determining the preset basic driving risk coefficient corresponding to the overall driving style of the target user according to the preset basic driving risk coefficients of each driving style stored in the WEB cloud.

4. The method for evaluating and analyzing big data of smart car user behavior according to claim 1, characterized in that: The screening of the reference driving sections that are highly similar to the unknown section includes: summarizing the same driving sections among the driving sections of each driving trip process of the target user within the historical preset period to obtain the driving trip sections of the target user within the historical preset period; Collect the road characteristic information of the unknown road section and each driving travel section of the target user in the preset historical period, including road type, speed limit range, number of lanes, lane width, road slope and average road traffic volume; The deviations of various parameters in the road feature information between the unknown road section and each driving travel section within the target user's historical preset period are calculated and accumulated, and the inverse of the accumulated value is used as the road similarity between the unknown road section and each driving travel section within the target user's historical preset period. The driving travel section corresponding to the maximum road similarity is extracted as a reference driving section that is highly similar to the unknown road section.

5. The method for evaluating and analyzing big data of smart car user behavior according to claim 1, characterized in that: The process of obtaining the correction factor of the weather condition of each current section to be traveled includes: retrieving the meteorological information of the target user's current area through the meteorological station, including rainfall, snowfall and air visibility, and comparing it with the permitted rainfall threshold, permitted snowfall threshold and permitted air visibility threshold corresponding to the driving impact of the reasonable meteorological environment stored in the WEB cloud. If the rainfall, snowfall and air visibility in the target user's current area are all less than or equal to their corresponding permitted thresholds, the correction factors of the weather conditions of each current section to be traveled are set to 0. Conversely, if any meteorological parameter among the rainfall, snowfall and air visibility in the target user's current area is greater than its corresponding permitted threshold, the deviation ratio of each meteorological parameter greater than its corresponding permitted threshold is calculated, and the accumulated value is used as the correction factor of the weather condition of each current section to be traveled after accumulation.

6. The method for evaluating and analyzing big data of smart car user behavior according to claim 1, characterized in that: The process of obtaining the correction factor of the traffic conditions of each current section to be traveled includes: retrieving the traffic density of each current section to be traveled by the target user through the navigation software, obtaining the traffic congestion level of each current section to be traveled by the target user based on the road traffic density interval of each traffic congestion level specification stored in the WEB cloud, setting the driving impact weight corresponding to each traffic congestion level, so as to obtain the driving impact weight corresponding to the traffic congestion level of each current section to be traveled by the target user, and further using it as the correction factor of the traffic conditions of each current section to be traveled.

7. The method for evaluating and analyzing big data of smart car user behavior according to claim 1, characterized in that: The process of obtaining the correction factor of the user status of each current section to be traveled includes: sensing the driving time of the target user's current navigation route through the vehicle-mounted equipment, accumulating the cumulative driving time of the target user to arrive at each section to be traveled based on the arrival time of the target user at each section to be traveled predicted by the navigation software, and using the ratio of the cumulative driving time to the preset user fatigue driving time as the correction factor of the user status of each current section to be traveled.

8. The method for evaluating and analyzing big data of smart car user behavior according to claim 1, characterized in that: The evaluation of the target user's driving behavior correction execution strength includes: collecting the target user's speed time history curve, average following distance, lane change frequency, line crossing times, steering amplitude and angular velocity for each prompted road section, and calculating the target user's actual driving behavior risk coefficient for each prompted road section; The predicted target user's driving behavior risk coefficient for each prompted road section is extracted, and the difference is made from the target user's actual driving behavior risk coefficient for each prompted road section. The difference is accumulated and the accumulated value is used as an evaluation indicator of the target user's driving behavior correction execution strength.

Images