Driving behavior evaluation method and device based on driving habit and electronic equipment

Abstract

Embodiments of the present application provide a driving behavior evaluation method and device based on driving habits and electronic equipment. The method comprises: determining a driving scene corresponding to a target trip and a target trip feature according to vehicle dynamic data of the target trip; the target trip is a trip completed by a driver most recently; determining a driving habit fingerprint feature of the driver according to the target trip feature and historical trip features corresponding to a plurality of historical trips of the driver; determining a driving habit entropy value of the driver based on the target trip feature and the driving habit fingerprint feature; determining a target trip score and a long-term score based on the target trip feature, the driving habit fingerprint feature and the driving scene; and determining a driving behavior score according to the driving habit entropy value, the target trip score and the long-term score. The method is used to improve the accuracy of the driving behavior evaluation result.

Description

Technical Field

[0001] This application relates to the field of big data processing technology, and in particular to a method, device and electronic device for evaluating driving behavior based on driving habits. Background Technology

[0002] A driver's driving behavior has a significant impact on road safety. Currently, by scoring drivers' driving behavior, the evaluation of driving behavior can be quantified, providing a direct understanding of the quality of driving behavior.

[0003] In related technologies, driving behavior is evaluated by recording specific driving behaviors, such as not wearing a seatbelt or the duration of fatigued driving. This static driving behavior evaluation method results in driving behavior scores that cannot reflect the driver's true driving level, leading to inaccurate driving behavior evaluations. Summary of the Invention

[0004] This application provides a driving behavior evaluation method, device, and electronic device based on driving habits, which can improve the accuracy of driving behavior evaluation results.

[0005] In a first aspect, embodiments of this application provide a driving behavior evaluation method based on driving habits, including:

[0006] Based on the vehicle dynamic data of the target trip, determine the driving scenario and characteristics of the target trip; the target trip is the trip most recently completed by the driver.

[0007] Based on the characteristics of the target trip and the characteristics of the historical trips corresponding to multiple historical trips of the driver, the driver's driving habit fingerprint characteristics are determined;

[0008] Based on the target trip characteristics and driving habit fingerprint characteristics, the driver's driving habit entropy value is determined;

[0009] Based on target trip characteristics, driving habit fingerprint characteristics, and driving scenarios, target trip scores and long-term scores are determined.

[0010] A driving behavior score is determined based on driving habit entropy, target trip score, and long-term score.

[0011] In some embodiments, the driving scenario and characteristics corresponding to the target trip are determined based on vehicle dynamic data of the target trip, including:

[0012] Based on the vehicle dynamic data of the target trip, the scenario is classified to obtain at least one driving scenario corresponding to the target trip;

[0013] Identify at least one driving scenario-related scenario dynamic data from the vehicle dynamic data;

[0014] Based on the scene dynamic data corresponding to at least one driving scenario, determine the target scene features corresponding to at least one driving scenario;

[0015] Based on the target scenario features corresponding to at least one driving scenario, determine the driver's target trip features.

[0016] In some embodiments, determining the driver's driving habit entropy value based on target trip features and driving habit fingerprint features includes:

[0017] Based on the characteristics of the target trip and the fingerprint characteristics of driving habits, the target fluctuation score of the target trip is determined;

[0018] The standard deviation of the score is determined based on the target volatility score and the historical volatility scores of multiple historical trips;

[0019] The driver's driving habit entropy value is determined based on the standard deviation of the score.

[0020] In some embodiments, determining the target fluctuation score of the target trip based on target trip characteristics and driving habit fingerprint characteristics includes:

[0021] Based on the driving time of the driving scenarios corresponding to the target trip, identify key scenarios within the driving scenarios;

[0022] Identify the key travel features corresponding to key scenarios from the target travel features;

[0023] Identify key fingerprint features corresponding to key scenarios from driving habit fingerprint features;

[0024] The target fluctuation score of the target journey is determined based on the similarity between key journey features and key fingerprint features.

[0025] In some embodiments, determining the target fluctuation score of a target journey based on the similarity between key journey features and key fingerprint features includes:

[0026] Based on key journey features and key fingerprint features, similarity is determined under multiple evaluation indicators;

[0027] Determine the key weights corresponding to multiple evaluation indicators based on key scenarios;

[0028] Based on the key weights, the similarity scores corresponding to multiple evaluation indicators are weighted and summed to obtain the target fluctuation score of the target journey.

[0029] In some embodiments, a target trip score and a long-term score are determined based on target trip characteristics, driving habit fingerprint characteristics, and driving scenarios, including:

[0030] Based on the driving time of the driving scenario corresponding to the target trip, determine the dynamic weight of the driving scenario under multiple evaluation indicators;

[0031] The target trip score is determined based on the target trip characteristics, the baseline characteristics corresponding to the driving scenario, and the dynamic weights under multiple evaluation indicators.

[0032] Long-term scores are determined based on driving habit fingerprint characteristics, baseline characteristics, and dynamic weights under multiple evaluation indicators.

[0033] In some embodiments, the target trip score is determined based on the target trip characteristics, the baseline characteristics corresponding to the driving scenario, and the dynamic weights under multiple evaluation indicators, including:

[0034] Determine the target scenario features corresponding to the driving scenario from the target trip features;

[0035] Based on the baseline and target scenario characteristics corresponding to the driving scenario, determine the target deviation score of the driving scenario under multiple evaluation indicators;

[0036] Based on the dynamic weights of multiple evaluation indicators, the target deviation scores under multiple evaluation indicators are weighted and summed to obtain the target travel score.

[0037] In some embodiments, a long-term score is determined based on driving habit fingerprint features, baseline features, and dynamic weights under multiple evaluation indicators, including:

[0038] Determine the driving habit scenario features corresponding to the driving scenario from the driving habit fingerprint features;

[0039] Based on the baseline characteristics and driving habit scenario characteristics corresponding to the driving scenario, determine the driving habit deviation score under multiple evaluation indicators for the driving scenario;

[0040] Based on the dynamic weights of multiple evaluation indicators, the driving habit deviation scores under multiple evaluation indicators are weighted and summed to obtain a long-term score.

[0041] In some embodiments, a driving behavior score is determined based on driving habit entropy, target trip score, and long-term score, including:

[0042] The target weight and long-term weight are determined based on the driving habit entropy value; the driving habit entropy value is positively correlated with the target weight.

[0043] Based on the target weight and long-term weight, the target trip score and the long-term score are weighted and summed to obtain the driving behavior score.

[0044] Secondly, embodiments of this application provide a driving behavior evaluation device based on driving habits, the device comprising:

[0045] The target trip processing module is used to determine the driving scenario and target trip characteristics corresponding to the target trip based on the vehicle dynamic data of the target trip; the target trip is the trip most recently completed by the driver.

[0046] The driving habit fingerprint feature determination module is used to determine the driver's driving habit fingerprint features based on the target trip features and the historical trip features corresponding to multiple historical trips of the driver.

[0047] The driving habit entropy determination module is used to determine the driver's driving habit entropy value based on the target trip features and driving habit fingerprint features.

[0048] The first evaluation module is used to determine the target trip score and long-term score based on the target trip characteristics, driving habit fingerprint characteristics and driving scenario;

[0049] The second evaluation module is used to determine the driving behavior score based on the driving habit entropy value, target trip score, and long-term score.

[0050] Thirdly, embodiments of this application provide an electronic device, including: a memory and a processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0051] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.

[0052] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0053] The driving behavior evaluation method, electronic device, storage medium, and program product based on driving habits provided in this application determine the driving scenario and target trip characteristics based on vehicle dynamic data of the target trip. Based on the driver's target trip characteristics and multiple historical trip characteristics, driving habit fingerprint characteristics are determined. These driving habit fingerprint characteristics comprehensively reflect the driver's long-term, deep-seated driving habit characteristics. Based on the target trip characteristics and driving habit fingerprint characteristics, the driver's driving habit entropy value is determined, which reflects the degree of fluctuation in driving behavior. The driver's driving behavior in the target trip is evaluated according to the driving scenario to obtain a target trip score. The driver's long-term driving behavior in multiple historical trips is also evaluated according to the driving scenario to obtain... The target trip score evaluates driving behavior based on driving scenarios, making the target trip score and long-term score closer to actual driving scenarios. Compared to evaluating driving behavior in the same way in different driving scenarios, this improves the accuracy of the target trip score and long-term score. Furthermore, the driving behavior score is determined based on the driving habit entropy value, target trip score, and long-term score. When driving behavior fluctuations are small, the driving behavior score focuses more on determining the long-term score, reducing interference from short-term driving behavior and making the driving behavior score more stable and reliable. When driving behavior fluctuations are large, the driving behavior score focuses more on determining the current target score, responding to the latest changes in driving behavior and making the driving behavior score more sensitive and accurate, thus improving the accuracy and stability of the driving behavior score. Attached Figure Description

[0054] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0055] Figure 1 A flowchart illustrating the driving behavior evaluation method based on driving habits provided in this application;

[0056] Figure 2 A schematic diagram illustrating the characteristics of the driving scenario and target journey provided for this application;

[0057] Figure 3 A schematic diagram illustrating the fingerprint features used to determine driving habits in this application;

[0058] Figure 4 A schematic diagram for determining the entropy value of driving habits provided in this application;

[0059] Figure 5 A schematic diagram for determining driving behavior scores provided in this application;

[0060] Figure 6 Another flowchart illustrating the driving behavior evaluation method based on driving habits provided in this application;

[0061] Figure 7 This application provides a map for determining driving ability, a plasticity index, and a diagram illustrating the delivery of driving challenge tasks;

[0062] Figure 8 A schematic diagram of the driving capability map provided in this application;

[0063] Figure 9 A schematic diagram of the driving behavior evaluation device based on driving habits provided in this application;

[0064] Figure 10 A schematic diagram of the structure of the electronic device provided in this application.

[0065] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0066] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0067] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0068] Figure 1 This is a flowchart illustrating the driving behavior evaluation method based on driving habits provided in this application. This method can be applied to electronic devices, such as cloud-based devices. Figure 1 As shown, driving behavior evaluation methods based on driving habits include:

[0069] S101. Based on the vehicle dynamic data of the target trip, determine the driving scenario and characteristics of the target trip; the target trip is the trip most recently completed by the driver.

[0070] Vehicle dynamic data includes basic vehicle data, environmental data, and perception data.

[0071] Basic vehicle data comes from the vehicle control unit (VCU), motor control unit (MCU), and battery management system (BMS). For example, basic vehicle data includes, but is not limited to: accelerator pedal opening, braking pressure, vehicle acceleration, and battery output power.

[0072] Environmental data includes road condition data and weather data; road condition data can be obtained by querying vehicle location, which can be obtained from the vehicle-to-everything (T-Box) terminal or cloud server; for example, environmental data includes: traffic congestion index, ambient temperature, weather, etc.

[0073] Perception data includes, but is not limited to: following distance, lane curvature, and obstacle recognition results. Perception data can be acquired by the Advanced Driver Assistance System (ADAS) controller, which determines the perception data based on the vehicle's surrounding data collected by onboard sensors.

[0074] Among them, driving scenario is the usage scenario determined based on vehicle dynamic data. For example, driving scenario can be urban congestion scenario, highway cruising scenario, mountain winding road scenario, rainy driving scenario, etc.; driving scenario is the usage scenario determined based on vehicle dynamic data of the target trip.

[0075] Among them, the travel characteristics include, but are not limited to: average following distance (seconds), average deceleration before stopping (m / s²), standard deviation of speed fluctuation (km / h), average acceleration throughout the journey (m / s²), vehicle energy consumption (kWh / 100km or L / 100km), vehicle steering angular velocity (rad / s), average vehicle deceleration (m / s²), vehicle slip ratio (%), and average vehicle deceleration (m / s²).

[0076] Average following distance is the average time interval between the rear of the preceding vehicle passing a fixed point and the front of the following vehicle reaching that point within the same lane; average deceleration before stopping is the deceleration calculated based on the time taken from the vehicle's initial speed after the brake pedal is applied until the vehicle comes to a complete stop at 0; standard deviation of speed fluctuation indicates the degree of speed fluctuation; average acceleration over the entire journey is the average acceleration calculated based on the vehicle's speed and travel time; vehicle energy consumption is the average electricity consumption (or average fuel consumption) calculated based on the vehicle's electricity (or fuel) consumption and travel distance; vehicle steering angular velocity is calculated based on the vehicle's steering angle and the time taken to turn; vehicle slip ratio is calculated based on vehicle speed, wheel angular velocity, and effective wheel rolling radius, for example, vehicle slip ratio = (vehicle speed - wheel angular velocity × effective wheel rolling radius) ÷ vehicle speed × 100%.

[0077] Target trip characteristics, which are the characteristics of the driver's most recently completed trip.

[0078] Specifically, during driving, the VCU records vehicle dynamic data in real time (including basic vehicle data, environmental data, and perception data at each recorded moment). At the end of the trip, it obtains the vehicle dynamic data at each moment of the trip. Based on the vehicle dynamic data of the trip, it determines the driving scenario at each moment of the trip and the trip characteristics of the trip.

[0079] It is understandable that a trip may include multiple different driving scenarios. For example, in a trip, the first driving scenario is urban congestion, the second driving scenario is highway cruising, and the third driving scenario is urban congestion.

[0080] It should be noted that after each trip, the driving scenario and trip characteristics corresponding to that trip are determined according to the above process. The driving scenario and trip characteristics determined based on the vehicle dynamic data of the driver's most recent completed trip are the driving scenario and target trip characteristics corresponding to the target trip.

[0081] For example, when it is necessary to evaluate a driver's driving behavior, the driver's most recent completed trip is taken as the target trip, and the driving scenario and target trip characteristics are determined based on the vehicle dynamic data at each moment in the target trip.

[0082] Specifically, the driving scenario is classified based on the vehicle dynamic data at each moment in the target trip to obtain the driving scenario at each moment. The vehicle dynamic data at each moment with the same driving scenario are spliced ​​together to obtain the scenario dynamic data corresponding to the driving scenario. Based on the scenario dynamic data corresponding to each driving scenario, the target scenario features corresponding to each driving scenario are determined. The target trip features include the target scenario features corresponding to each driving scenario.

[0083] In some embodiments, such as Figure 2 As shown, S101, based on the vehicle dynamic data of the target trip, determines the driving scenario and target trip features corresponding to the target trip, including: S1011, classifying scenarios based on the vehicle dynamic data of the target trip to obtain at least one driving scenario corresponding to the target trip; S1012, determining the scenario dynamic data corresponding to at least one driving scenario in the vehicle dynamic data; S1013, determining the target scenario features corresponding to at least one driving scenario based on the scenario dynamic data corresponding to at least one driving scenario; S1014, determining the driver's target trip features based on the target scenario features corresponding to at least one driving scenario.

[0084] Optionally, for the vehicle dynamic data at each moment in the target journey, the vehicle dynamic data at that moment is input into the scene classification model, and the driving scene at that moment is determined by the scene classification model.

[0085] The scene classification model can include a convolutional neural network and a classification network. The convolutional neural network is used to extract dynamic features from vehicle dynamic data, and the classification network is used to classify the dynamic features to obtain the driving scene. The scene classification model can be obtained by iteratively training an initial classification model based on dynamic data samples and corresponding driving scene labels. The initial classification model and the scene classification model have the same model structure but different model parameters.

[0086] Optionally, the vehicle dynamic data at each moment during the target journey includes traffic congestion index, road type, distance to the vehicle in front, vehicle speed, lane curvature, and weather information; the congestion level is determined based on the traffic congestion index, the following distance type is determined based on the distance to the vehicle in front, the vehicle speed type is determined based on the vehicle speed, the road curvature level is determined based on the lane curvature, and the driving scenario is determined based on the congestion level, road type, following distance type, vehicle speed type, road curvature level, and weather information.

[0087] For example, the mapping relationship between congestion level, road type, following distance type, vehicle speed type, road curvature level, weather information, and driving scenario is shown in Table 1.

[0088] Table 1

[0089]

[0090] The driving scenarios listed in Table 1 and the embodiments of this application are only illustrative examples. In practical applications, more driving scenarios can be defined.

[0091] Specifically, after determining the driving scenario at each moment in the target journey, the vehicle dynamic data at each moment with the same driving scenario are spliced ​​together to obtain the scene dynamic data corresponding to the driving scenario; for each driving scenario, the target scene features of the driving scenario are determined based on the scene dynamic data corresponding to the driving scenario.

[0092] For example, the target scene features in urban congestion scenarios include: average following distance and average deceleration before stopping; that is, based on the scene dynamic data in urban congestion scenarios, the average following distance and average deceleration before stopping are determined to obtain the target scene features in urban congestion scenarios.

[0093] The target scenario characteristics under high-speed cruising include: speed fluctuation standard deviation, average acceleration over the entire journey, and vehicle energy consumption; that is, based on the scenario dynamic data under high-speed cruising, the speed fluctuation standard deviation, average acceleration over the entire journey, and vehicle energy consumption are determined to obtain the target scenario characteristics under high-speed cruising.

[0094] The target scene features in mountain curve scenarios include the vehicle's steering angular velocity and average deceleration; that is, based on the scene dynamic data in mountain curve scenarios, the vehicle's steering angular velocity and average deceleration are determined to obtain the target scene features in mountain curve scenarios.

[0095] The target scene features for driving in rainy weather include the vehicle slip ratio and the vehicle average deceleration; that is, based on the scene dynamic data of driving in rainy weather, the vehicle slip ratio and the vehicle average deceleration are determined to obtain the target scene features for driving in rainy weather.

[0096] Based on the target scenario features of at least one driving scenario corresponding to the target trip, the driver's target trip features are defined; for example, if the target trip involves two driving scenarios, the target scenario features of the two driving scenarios are respectively represented as follows: and Then the driver's target trip characteristics are represented as .

[0097] In some embodiments, the driving scenario and characteristics corresponding to the target trip are encrypted and stored in the driver's driving data area, and the storage is updated synchronously at the end of each trip.

[0098] In the above embodiments, the target trip is divided into driving scenarios, the characteristics of the target scenario are determined according to the vehicle dynamic data of the driving scenario, and the characteristics of the target trip are determined according to the characteristics of the driving scenario. This allows for the evaluation of the driver's driving behavior based on different characteristics for different driving scenarios. For example, in rainy driving scenarios, driving behavior is evaluated by controlling the vehicle slip rate, and in urban congestion scenarios, driving behavior is evaluated by following distance. This makes the evaluation of driving behavior more targeted and improves the accuracy of driving behavior evaluation.

[0099] S102. Based on the target trip characteristics and the historical trip characteristics corresponding to multiple historical trips of the driver, determine the driver's driving habit fingerprint characteristics.

[0100] Among them, historical trips are trips completed by the driver before the target trip. For example, multiple trips of the driver are sorted by time from earliest to latest as follows: Among them, several historical events include: The target itinerary is .

[0101] Among them, multiple historical trips can be determined based on the number of times or the number of days. For example, multiple historical trips are a preset number of historical trips (e.g., 50), or multiple historical trips are historical trips completed within a preset number of days (e.g., 30 days).

[0102] Among them, historical travel characteristics are travel characteristics determined based on the vehicle dynamic data corresponding to historical travel.

[0103] It should be noted that after each historical trip ends, the driving scenario and historical trip features corresponding to the historical trip are determined in the same way as in step S101; therefore, after the target trip ends, the historical trip features corresponding to multiple historical trips can be directly obtained.

[0104] Among them, the driving habit fingerprint feature is determined based on multiple travel characteristics of the driver and is used to reflect the driver's long-term driving habit characteristics.

[0105] Specifically, driving habit fingerprint features are obtained by averaging the target trip features and multiple historical trip features.

[0106] Optionally, each of the target trip features and multiple historical trip features includes feature values ​​under multiple features, including but not limited to: average following distance, average deceleration before stopping, speed fluctuation standard deviation, average acceleration throughout the trip, vehicle energy consumption, vehicle steering angular velocity, vehicle average deceleration, vehicle slip ratio, and vehicle average deceleration. For each feature, the average feature value of the feature is obtained by averaging the feature values ​​of the feature in the target trip features and multiple historical trip features. The driving habit fingerprint feature is determined based on the average feature value under multiple features.

[0107] For example, multiple historical trips include trip B1, trip B2, ..., trip Bn-1, and the target trip is trip Bn. The driving habit fingerprint feature is obtained by averaging the characteristics of the target trip and the characteristics of multiple historical trips. As shown in Table 2.

[0108] Table 2

[0109]

[0110] Optionally, the target trip features include at least one target scene feature under a driving scenario, and the historical trip features include at least one historical scene feature under a driving scenario; such as Figure 3 As shown, S102, based on the target trip features and the historical trip features corresponding to multiple historical trips of the driver, determine the driver's driving habit fingerprint features, including: S1021, for each driving scenario, based on the target trip features and multiple historical trip features, determine multiple scenario features under that driving scenario; S1022, average the multiple scenario features under that driving scenario to obtain the driving habit scenario features under that driving scenario; S1023, based on the driving habit scenario features under at least one driving scenario, determine the driving habit fingerprint features.

[0111] For example, multiple historical trips include trip C1, trip C2, and trip C3, with the target trip being trip C4; the historical trip features of trip C1 include historical scene features Da1 under urban congestion scenarios and historical scene features Db1 under highway cruising scenarios; the historical trip features of trip C2 include historical scene features Da2 under urban congestion scenarios and historical scene features Dc2 under mountain curve scenarios; the historical trip features of trip C3 include historical scene features Da3 under urban congestion scenarios and historical scene features Db3 under highway cruising scenarios; the target trip features of trip C3 include target scene features Db4 under highway cruising scenarios and target scene features Dc4 under mountain curve scenarios.

[0112] The driving habit scene feature Da under urban congestion scenario is obtained by averaging the historical scene features Da1, Da2, and Da3 under urban congestion scenario; the driving habit scene feature Db under highway cruising scenario is obtained by averaging the historical scene features Db1 and Db3 under highway cruising scenario; and the driving habit scene feature Dc under highway cruising scenario is obtained by averaging the historical scene features Dc2 and Dc4 under mountain curve scenario.

[0113] Driving habit fingerprint features include: driving habit scenario features Da in urban congestion scenarios, driving habit scenario features Db in highway cruising scenarios, and driving habit scenario features Dc in highway cruising scenarios.

[0114] It should be noted that the driving habit fingerprint feature is a multi-dimensional feature vector. The driving habit fingerprint feature includes average following data, vehicle steering angular velocity and vehicle slip ratio, which can reflect the driver's risk preference. The average deceleration before stopping, the average acceleration throughout the journey and the standard deviation of speed fluctuation can reflect the driver's operational finesse. The vehicle energy consumption can reflect the driver's energy consumption preference. In other words, the driving habit fingerprint feature is a feature extracted from the driver's long-term travel data to comprehensively reflect the driver's deep-seated driving habits.

[0115] Optionally, after each trip, the driving habit fingerprint features are recalculated using the trip features of that trip and the features of multiple previous historical trips, in the same manner as in step S102.

[0116] Optionally, the driving habit fingerprint features can be encrypted and then updated in the driver's driving data area.

[0117] S103. Based on the target trip features and driving habit fingerprint features, determine the driver's driving habit entropy value.

[0118] Among them, the driving habit entropy value is used to reflect the degree of fluctuation of the driver's driving habits. The larger the driving habit entropy value, the greater the fluctuation of the driver's driving habits and the unstable driving habits. The smaller the driving habit entropy value, the smaller the fluctuation of the driver's driving habits and the more stable driving habits.

[0119] In some embodiments, a first short-term score is determined based on the characteristics of the target trip, and a first long-term score is determined based on the fingerprint characteristics of driving habits. The first short-term score and the first long-term score are weighted and summed to obtain the fluctuation score of the target trip. After the end of each historical trip, the fluctuation score of that historical trip is determined according to the above process. Therefore, the fluctuation scores of multiple historical trips can be obtained directly. The standard deviation is calculated based on the fluctuation score of the target trip and the fluctuation scores of multiple historical trips, and the driving habit entropy value is determined by the standard deviation.

[0120] In some embodiments, as Figure 4 shown, S103. Determine the driving habit entropy value of the driver based on the target trip feature and the driving habit fingerprint feature, including: S1031. Determine the target fluctuation score of the target trip based on the target trip feature and the driving habit fingerprint feature; S1032. Determine the score standard deviation according to the target fluctuation score and the historical fluctuation scores of multiple historical trips; S1033. Determine the driving habit entropy value of the driver according to the score standard deviation.

[0121] Among them, the target fluctuation score is used to reflect the driving behavior quality of the target trip; the historical fluctuation score is used to reflect the driving behavior quality of the historical trip; the score standard deviation is used to reflect the fluctuation degree of the driver's driving behavior in multiple historical trips and the target trip.

[0122] Optionally, calculate the similarity according to the target trip feature and the driving habit fingerprint feature to obtain the target fluctuation score.

[0123] It should be noted that after each trip (including historical trips and target trips), the fluctuation score of this trip is determined in the same way as determining the target fluctuation score, and the fluctuation scores of historical trips are used as historical fluctuation scores. Therefore, the historical fluctuation scores of multiple historical trips can be directly obtained.

[0124] Determine the score standard deviation according to the target fluctuation score and the historical fluctuation scores of multiple historical trips. Exemplarily, calculate the score standard deviation through formula (1).

[0125] Formula (1): ;

[0126] Among them, is the score standard deviation, is the total number of multiple historical trips and the target trip, is the fluctuation score of the i-th trip. When i = N, is the target fluctuation score. When i < N, is the historical fluctuation score; is the average score of the target fluctuation score and the historical fluctuation scores of multiple historical trips.

[0127] Normalize the score standard deviation to obtain the driving habit entropy value; that is, the driving habit entropy value is a value between 0 and 1; the score standard deviation can be normalized by the maximum standard deviation. Among them, the maximum standard deviation can be determined from the driver's cloud data. For example, use the upper quartile of multiple score standard deviations of the driver within 30 days as the maximum standard deviation.

[0128] Exemplarily, calculate the driving habit entropy value through formula (2).

[0129] Formula (2): ;

[0130] in, It is the entropy value of driving habits. It is the standard deviation of the score. It is the maximum standard deviation; .

[0131] It should be noted that if the entropy value of driving habits... A value approaching 0 indicates that the driver's fluctuation score is stable and small over a period of time across multiple historical trips and the target trip, suggesting relatively fixed driving habits; if the driving habit entropy value... A score approaching 1 indicates that the driver's performance over multiple historical and target trips is unstable and highly volatile, suggesting the driver may be in a learning or fatigued phase, or strongly influenced by external factors. If the driving habit entropy value... A value in the middle (e.g., close to 0.5) indicates that the driver's driving behavior has certain habits, while at the same time some driving behaviors are being changed.

[0132] Optionally, the driving habit entropy value can be encrypted and then updated in the driver's driving data area.

[0133] In the above embodiments, a driving habit entropy value is determined based on the target fluctuation score of the target trip and the historical fluctuation score of the historical trip. The driving habit entropy value reflects the volatility of the driver's driving behavior, so as to facilitate subsequent evaluation of the driver's driving behavior from the perspective of the driver's long-term driving habits and improve the comprehensiveness of the driving behavior evaluation.

[0134] S104. Based on the target trip characteristics, driving habit fingerprint characteristics, and driving scenarios, determine the target trip score and long-term score.

[0135] Among them, the target trip score is a quality score obtained by evaluating the driver's driving behavior during the target trip; the long-term score is a quality score obtained by evaluating the driver's long-term driving behavior over multiple historical trips.

[0136] Specifically, the baseline features corresponding to the driving scenario are obtained. These baseline features are pre-defined and include ideal reference values ​​for each feature under the driving scenario.

[0137] For each driving scenario involved in the target trip features, the short-term quality score corresponding to the driving scenario is determined based on the deviation between the target scenario features included in the target trip features and the baseline features corresponding to the driving scenario; the target trip score is determined based on the deviation between the short-term quality scores corresponding to the multiple driving scenarios involved in the target trip.

[0138] For each driving scenario involved in the target trip features, the long-term quality score corresponding to the driving scenario is determined based on the deviation between the driving habit scenario features included in the driving habit fingerprint features and the baseline features corresponding to the driving scenario; the long-term score is determined based on the long-term quality scores corresponding to each of the multiple driving scenarios involved in the target trip.

[0139] It is evident that the closer the target trip features are to the baseline features, the higher the target trip score. Similarly, the closer the driving habit fingerprint features are to the baseline features, the higher the long-term score.

[0140] S105. Determine the driving behavior score based on the driving habit entropy value, target trip score, and long-term score.

[0141] Among them, the driving behavior score can reflect the comprehensive score of the driver's driving behavior in multiple historical trips and the most recently completed trip. It should be noted that the driving behavior in multiple historical trips can reflect the driver's driving habits. Based on the driving habits, the driving behavior score changes dynamically with the most recently completed trip.

[0142] Specifically, the weights of the target trip score and the long-term score are determined based on the driving habit entropy value, and the driving behavior score is determined by applying the target trip score and the long-term score according to the weights.

[0143] In some embodiments, such as Figure 5 As shown, S105, determine the driving behavior score based on the driving habit entropy value, target trip score, and long-term score, including: S1051, determine the target weight and long-term weight based on the driving habit entropy value; the driving habit entropy value is positively correlated with the target weight; S1052, calculate the weighted sum of the target trip score and long-term score based on the target weight and long-term weight to obtain the driving behavior score.

[0144] Among them, the larger the driving habit entropy value, the greater the fluctuation of driving behavior and the greater the target weight. That is, when the driving behavior fluctuates greatly, more attention is paid to recent driving behavior. The smaller the driving habit entropy value, the smaller the fluctuation of driving behavior and the smaller the target weight. That is, when the driving behavior fluctuates less, more attention is paid to long-term stable driving habits.

[0145] The mapping relationship between driving habit entropy, target weight, and long-term weight is preset, and the corresponding target weight and long-term weight can be directly found based on the driving habit entropy.

[0146] Specifically, the driving behavior score is determined according to formula (3).

[0147] Formula (3): ;

[0148] in, It is a driving behavior score. It is the target weight. It is the target itinerary score. It is a long-term weight. It is a long-term rating.

[0149] For example, if the driving habit entropy value H > 0.7 (indicating large fluctuations in driving behavior), then , If the driving habit entropy value H < 0.3 (indicating small fluctuations in driving behavior), then , If the driving habit entropy value H is between 0.3 and 0.7, then , .

[0150] In the above embodiments, the target weight of the target trip score and the long-term weight of the long-term score are determined by the driving habit entropy value. When the driving behavior fluctuates less, the driving behavior score focuses more on determining the long-term score, reducing the interference of short-term driving behavior and making the driving behavior score more stable and reliable. When the driving behavior fluctuates more, the driving behavior score focuses more on determining the current target score, which can respond to the latest changes in driving behavior and make the driving behavior score more sensitive and accurate, thus improving the accuracy and stability of the driving behavior score.

[0151] The driving behavior evaluation method based on driving habits provided in this application determines the driving scenario and target trip characteristics based on vehicle dynamic data of the target trip. Based on the driver's target trip characteristics and multiple historical trip characteristics, driving habit fingerprint characteristics are determined. These driving habit fingerprint characteristics comprehensively reflect the driver's long-term, deep-seated driving habit characteristics. Based on the target trip characteristics and driving habit fingerprint characteristics, the driver's driving habit entropy value is determined, which reflects the degree of fluctuation in driving behavior. The driver's driving behavior in the target trip is evaluated according to the driving scenario to obtain a target trip score. The driver's long-term driving behavior in multiple historical trips is also evaluated according to the driving scenario to obtain a target trip score. Evaluating driving behavior within specific driving scenarios makes the target trip score and long-term score more closely reflect actual driving conditions. Compared to evaluating driving behavior in the same way across different driving scenarios, this improves the accuracy of the target trip score and long-term score. Furthermore, the driving behavior score is determined based on the driving habit entropy value, target trip score, and long-term score. When driving behavior fluctuations are small, the driving behavior score focuses more on determining the long-term score, reducing interference from short-term driving behavior and making the driving behavior score more stable and reliable. When driving behavior fluctuations are large, the driving behavior score focuses more on determining the current target score, responding to the latest changes in driving behavior and making the driving behavior score more sensitive and accurate, thus improving the accuracy and stability of the driving behavior score.

[0152] In some embodiments, S1031, determining the target fluctuation score of the target trip based on the target trip features and driving habit fingerprint features, includes: S1031a, identifying key scenarios in the driving scenario according to the driving duration of the driving scenario corresponding to the target trip; S1031b, identifying key trip features corresponding to the key scenarios in the target trip features; S1031c, identifying key fingerprint features corresponding to the key scenarios in the driving habit fingerprint features; S1031d, determining the target fluctuation score of the target trip based on the similarity between the key trip features and the key fingerprint features.

[0153] Specifically, the target trip corresponds to at least one driving scenario. The driving time of the target trip in each driving scenario is determined, and the driving scenario with the longest driving time is taken as the key scenario. The key scenario is the main scenario with the longest driving time in the target trip.

[0154] The target trip features include the scene features of each driving scenario involved in the target trip, and the scene features of the key scenarios included in the target trip features are taken as key trip features.

[0155] Driving habit fingerprint features include driving habit scenario features for each driving scenario involved in the historical trip. The driving habit scenario features in key scenarios included in the driving habit fingerprint features are used as key fingerprint features.

[0156] The similarity between key travel features and key fingerprint features is calculated to obtain the target fluctuation score of the target travel.

[0157] In the above embodiments, key scenarios are determined by the driving duration of driving scenarios. Based on the characteristics of key scenarios in the target journey and the characteristics of key scenarios under long-term driving habits, the target fluctuation score is determined. Since key scenarios are the scenarios with the longest driving time, the key journey characteristics are more comprehensive. The focus is on the fluctuation of driving behavior in key driving scenarios, which improves the accuracy of the target fluctuation score and reduces the amount of data used for calculation, thereby improving the evaluation efficiency.

[0158] In some embodiments, S1031d, determining the target fluctuation score of the target journey based on the similarity between key journey features and key fingerprint features, includes: S1031d1, determining the similarity under multiple evaluation indicators based on key journey features and key fingerprint features; S1031d2, determining the key weights corresponding to multiple evaluation indicators based on key scenarios; S1031d3, performing a weighted summation of the similarities corresponding to multiple evaluation indicators based on the key weights to obtain the target fluctuation score of the target journey.

[0159] The evaluation indicators include safety indicators, energy-saving indicators, and vehicle care indicators; the key weights corresponding to these indicators are used to reflect the importance of each indicator in key scenarios.

[0160] In practical applications, a mapping relationship between driving scenarios and the weights of multiple evaluation indicators is pre-defined. Therefore, the key weights of multiple evaluation indicators in key scenarios can be determined based on the mapping relationship.

[0161] For example, the mapping relationship between driving scenarios and the weights of multiple evaluation indicators is shown in Table 3.

[0162] Table 3

[0163]

[0164] Specifically, key travel characteristics include travel characteristic values ​​under multiple characteristics, which are divided according to multiple evaluation indicators to obtain travel characteristic values ​​under the characteristics corresponding to energy saving indicators, travel characteristic values ​​under the characteristics corresponding to safety indicators, and travel characteristic values ​​under the characteristics corresponding to vehicle care indicators.

[0165] Similarly, key fingerprint features include fingerprint feature values ​​under multiple features. They are divided according to multiple evaluation indicators to obtain fingerprint feature values ​​under the features corresponding to energy saving indicators, fingerprint feature values ​​under the features corresponding to safety indicators, and fingerprint feature values ​​under the features corresponding to vehicle care indicators.

[0166] For example, the characteristics corresponding to energy-saving indicators may include, but are not limited to, vehicle energy consumption, average acceleration throughout the journey, and speed fluctuation standard deviation; the characteristics corresponding to safety indicators may include, but are not limited to, average following distance, average deceleration before stopping, vehicle steering angular velocity, and vehicle slip ratio; the characteristics corresponding to vehicle care indicators may include, but are not limited to, vehicle average deceleration, average deceleration before stopping, vehicle slip ratio, and vehicle steering angular velocity.

[0167] The similarity under safety indicators is determined based on the travel feature values ​​and fingerprint feature values ​​corresponding to the safety indicators; the similarity under energy-saving indicators is determined based on the travel feature values ​​and fingerprint feature values ​​corresponding to the energy-saving indicators; and the similarity under vehicle care indicators is determined based on the travel feature values ​​and fingerprint feature values ​​corresponding to the vehicle care indicators.

[0168] Based on the key weights under the safety indicators, energy-saving indicators, and vehicle care indicators, the similarity scores under the safety indicators, energy-saving indicators, and vehicle care indicators are weighted and summed to obtain the target fluctuation score for the target trip.

[0169] For example, the target fluctuation score is determined by formula (4).

[0170] Formula (4): ;

[0171] in, This is the fluctuation score for the Nth trip. Since the Nth trip is the target trip, therefore... It is the target fluctuation score; This indicates the key scenario; J=3 represents three evaluation metrics. These are key weights for safety, energy efficiency, and vehicle care indicators in critical scenarios. It is a similarity function, which is Target travel characteristics It is a fingerprint feature of driving habits (calculated based on the characteristics of the target trip). This represents the similarity between key journey features and key fingerprint features under evaluation metric j.

[0172] For example, driving habit entropy values ​​are calculated based on the target fluctuation score and the historical fluctuation scores of multiple historical trips, as shown in Table 4.

[0173] Table 4

[0174]

[0175] In Table 4, This refers to the target travel feature; the remaining travel features are as follows: , … are characteristics of the historical process; correspondingly, The corresponding volatility score is the target volatility score. , The fluctuation scores corresponding to 、, 、 are historical fluctuation scores.

[0176] Based on historical travel characteristics The driving habit fingerprint characteristics are determined by other historical travel features in the preceding sequence; similarly, , It is a driving habit fingerprint feature determined based on the corresponding trip characteristics.

[0177] For example, based on historical travel characteristics Fingerprint characteristics of driving habits And the key weights in the corresponding key scenarios, determine The corresponding historical fluctuation score is 82; in Table 4, the mean of the fluctuation scores for the driver's N trips is 80, and the calculated standard deviation of the scores is 5.

[0178] Maximum standard deviation Therefore, the entropy value of driving habits H = 5.0 / 15.0 ≈ 0.33.

[0179] Therefore, the driver's driving habit entropy value H is 0.33, indicating that his driving habits are relatively stable (lower than the median value of 0.5).

[0180] In the above embodiments, the evaluation indicators focused on may differ under different key scenarios. For example, in urban congestion scenarios, safety indicators may be more important, while in highway cruising scenarios, energy-saving indicators may be more important. After determining the similarity of key trip features and key fingerprint features under multiple evaluation indicators, a weighted sum is performed according to the weight of the key scenario under multiple evaluation indicators. This ensures that the target fluctuation score can reflect the fluctuation of driving behavior under the key evaluation indicators focused on in the key driving scenario, improving the relevance and accuracy of the target fluctuation score. In addition, the historical fluctuation scores of multiple historical trips are calculated in the same way, which also improves the relevance and accuracy of the historical fluctuation scores.

[0181] In some embodiments, S104, determining the target trip score and long-term score based on target trip features, driving habit fingerprint features, and driving scenarios includes: S1041, determining the dynamic weight of the driving scenario under multiple evaluation indicators based on the driving time of the driving scenario corresponding to the target trip; S1042, determining the target trip score based on the target trip features, the baseline features corresponding to the driving scenario, and the dynamic weights under multiple evaluation indicators; S1043, determining the long-term score based on the driving habit fingerprint features, the baseline features, and the dynamic weights under multiple evaluation indicators.

[0182] Specifically, the driving time of the target trip in each driving scenario is determined. For each driving scenario, the time proportion coefficient of that driving scenario is determined based on the ratio between the driving time of that driving scenario and the total time of the target trip.

[0183] The mapping relationship between driving scenarios and the weights of multiple evaluation indicators is pre-defined. Therefore, for each driving scenario, the basic weight of the driving scenario under multiple evaluation indicators can be determined according to the mapping relationship. Then, the duration ratio coefficient of the driving scenario is used to adjust the basic weight of the driving scenario under multiple evaluation indicators, so as to obtain the dynamic weight of the driving scenario under multiple evaluation indicators.

[0184] For example, taking driving scenario i as an example, the dynamic weight of driving scenario i under multiple evaluation indicators is determined according to formula (5).

[0185] Formula (5): ;

[0186] in, It is the dynamic weight of driving scenario i under multiple evaluation indicators. It is the basic weight of driving scenario i under multiple evaluation indicators. It is the driving duration in driving scenario i. It is the total duration of the target trip. It is the time percentage coefficient for driving scenario i.

[0187] Based on the target trip characteristics and the baseline characteristics corresponding to driving scenario i, the short-term deviation of the driving scenario under multiple evaluation indicators is determined, and the dynamic weights of the driving scenario under multiple evaluation indicators are used as the basis for this determination. We calculate the target trip score by weighted summation of the short-term deviations of driving scenario i under multiple evaluation indicators.

[0188] Based on driving habit fingerprint features and baseline features, the long-term deviation of driving scenario i under multiple evaluation indicators is determined, and the dynamic weights of driving scenario i under multiple evaluation indicators are used as the basis for further analysis. The long-term deviations of driving scenario i under multiple evaluation indicators are weighted and summed to obtain the target trip score.

[0189] In the above embodiments, the dynamic weight of the driving scenario under multiple evaluation indicators is determined based on the driving duration of the driving scenario. The weight of driving scenarios with longer driving durations is increased, while the weight of driving scenarios with shorter driving durations is decreased. The target trip score is determined based on the target trip characteristics, the baseline characteristics corresponding to the driving scenario, and the dynamic weights under multiple evaluation indicators, thereby improving the accuracy of the target trip score. The long-term score is determined based on the driving habit fingerprint characteristics, the baseline characteristics, and the dynamic weights under multiple evaluation indicators, thereby improving the accuracy of the long-term score.

[0190] In some embodiments, S1042, determining the target trip score based on the target trip features, the baseline features corresponding to the driving scenario, and the dynamic weights under multiple evaluation indicators, includes: S10421, determining the target scenario features corresponding to the driving scenario in the target trip features; S10422, determining the target deviation score of the driving scenario under multiple evaluation indicators based on the baseline features and target scenario features corresponding to the driving scenario; S10423, performing a weighted summation of the target deviation scores under multiple evaluation indicators based on the dynamic weights under multiple evaluation indicators to obtain the target trip score.

[0191] Specifically, the target trip features include target scenario features under each driving scenario. Among the baseline features and target scenario features corresponding to the driving scenarios, the baseline feature values ​​and target scenario feature values ​​corresponding to the features involved in the safety indicators are determined, the baseline feature values ​​and target scenario feature values ​​corresponding to the features involved in the energy-saving indicators are determined, and the baseline feature values ​​and target scenario feature values ​​corresponding to the features involved in the vehicle care indicators are determined.

[0192] The deviations between the baseline feature values ​​and target scenario feature values ​​corresponding to the features involved in the safety indicators are determined to obtain the target deviation score under the safety indicators; the deviations between the baseline feature values ​​and target scenario feature values ​​corresponding to the features involved in the energy-saving indicators are determined to obtain the target deviation score under the energy-saving indicators; the baseline feature values ​​and target scenario feature values ​​corresponding to the features involved in the vehicle care indicators are determined to obtain the target deviation score under the vehicle care indicators.

[0193] Based on the dynamic weights corresponding to safety indicators, energy-saving indicators, and vehicle care indicators, the target deviation scores corresponding to safety indicators, energy-saving indicators, and vehicle care indicators are weighted and summed to obtain the target travel score.

[0194] For example, the target trip involves two driving scenarios, as shown in Table 5.

[0195] Table 5

[0196]

[0197] Among them, high-speed cruise scenario Dynamic weights under multiple evaluation metrics, including: dynamic weights under safety metric j1. Dynamic weights under energy-saving index j2 Dynamic weights under vehicle care index j3 .

[0198] Taking safety indicators as an example, in a high-speed cruising scenario... Dynamic weights under safety indicators It is a high-speed cruising scenario. The duration ratio (0.4) and high-speed cruising scenario The benchmark weights corresponding to the lower safety indicators The product of.

[0199] In high-speed cruising scenarios Below, the target scene feature values ​​corresponding to the security indicators will be... and benchmark eigenvalues Substituting into the scoring function, we obtain the high-speed cruising scenario. Target deviation score under safety indicators As an example, the scoring function could be: y is the baseline feature value, and x is the target scene feature value.

[0200] In the same way, high-speed cruising scenarios can be calculated. Target deviation score under energy saving indicators High-speed cruising scenario Target deviation score under vehicle care indicators Urban congestion scenarios Target deviation score under safety indicators Urban congestion scenarios Target deviation score under energy saving indicators Urban congestion scenarios Target deviation score under vehicle care indicators .

[0201] The target trip score is obtained by weighting the target deviation score of the driving scenario under multiple evaluation indicators by using the dynamic weights of the driving scenario under multiple evaluation indicators, as shown in formula (6).

[0202] Formula (6): ;

[0203] in, It is the target itinerary score. It is the dynamic weight of driving scenario i under evaluation index j. This is the target deviation score for driving scenario i under evaluation index j; for example, in the example corresponding to Table 5, It is a high-speed cruising scenario. Dynamic weights under safety index j1 It is a high-speed cruising scenario. Target deviation score under safety indicators

[0204] Optionally, the target trip rating can be converted to a 100-point scale; for example, if the target trip rating is 154.75, after conversion to a 100-point scale, the target trip rating is 86.

[0205] In some embodiments, S1043, determining a long-term score based on driving habit fingerprint features, baseline features, and dynamic weights under multiple evaluation indicators includes: S10431, determining driving habit scenario features corresponding to the driving scenario from the driving habit fingerprint features; S10432, determining driving habit deviation scores for the driving scenario under multiple evaluation indicators based on the baseline features and driving habit scenario features corresponding to the driving scenario; S10433, weighting and summing the driving habit deviation scores under multiple evaluation indicators based on the dynamic weights under multiple evaluation indicators to obtain a long-term score.

[0206] Specifically, the driving habit fingerprint features include driving habit scenario features under various driving scenarios. Among the baseline features and driving habit scenario features corresponding to the driving scenarios, the baseline feature values ​​and driving habit scenario feature values ​​corresponding to the features involved in the safety indicators are determined, the baseline feature values ​​and driving habit scenario feature values ​​corresponding to the features involved in the energy-saving indicators are determined, and the baseline feature values ​​and driving habit scenario feature values ​​corresponding to the features involved in the vehicle care indicators are determined.

[0207] The deviations between the baseline feature values ​​and driving habit scenario feature values ​​corresponding to the features involved in the safety indicators are determined to obtain the driving habit deviation score under the safety indicators; the deviations between the baseline feature values ​​and driving habit scenario feature values ​​corresponding to the features involved in the energy-saving indicators are determined to obtain the driving habit deviation score under the energy-saving indicators; the baseline feature values ​​and driving habit scenario feature values ​​corresponding to the features involved in the vehicle care indicators are determined to obtain the driving habit deviation score under the vehicle care indicators.

[0208] Based on the dynamic weights corresponding to safety indicators, energy-saving indicators, and vehicle care indicators, the driving habit deviation scores corresponding to these three indicators are weighted and summed to obtain a long-term score.

[0209] It should be noted that the specific calculation process for the long-term score is similar to that for the target trip score, so you can refer to the above-mentioned specific calculation process for the target trip score.

[0210] In the above embodiments, the target trip score is determined based on the target trip characteristics, the baseline characteristics corresponding to the driving scenario, and the dynamic weights under multiple evaluation indicators, thereby improving the accuracy of the target trip score; the long-term score is determined based on the driving habit fingerprint characteristics, the baseline characteristics, and the dynamic weights under multiple evaluation indicators, thereby improving the accuracy of the long-term score.

[0211] In a specific example, such as Figure 6 As shown, driving behavior evaluation methods based on driving habits include:

[0212] S101 determines the driving scenario and characteristics of the target trip based on the vehicle dynamic data of the target trip; the target trip is the trip most recently completed by the driver.

[0213] S102. Based on the target trip characteristics and the historical trip characteristics corresponding to multiple historical trips of the driver, determine the driver's driving habit fingerprint characteristics;

[0214] S1031 includes:

[0215] S1031a. Based on the driving time of the driving scenario corresponding to the target trip, identify the key scenarios in the driving scenario;

[0216] S1031b, Identify the key travel features corresponding to the key scenarios in the target travel features;

[0217] S1031c, Determine the key fingerprint features corresponding to key scenarios from the driving habit fingerprint features;

[0218] S1031d includes:

[0219] S1031d1. Based on key travel features and key fingerprint features, determine the similarity under multiple evaluation indicators;

[0220] S1031d2. Determine the key weights corresponding to multiple evaluation indicators based on key scenarios;

[0221] S1031d3. Based on the key weights, the similarity corresponding to multiple evaluation indicators is weighted and summed to obtain the target fluctuation score of the target journey.

[0222] S1032. Determine the standard deviation of the score based on the target volatility score and the historical volatility scores of multiple historical journeys;

[0223] S1033. Determine the driver's driving habit entropy value based on the standard deviation of the score;

[0224] S1041. Determine the dynamic weight of the driving scenario under multiple evaluation indicators based on the driving time of the driving scenario corresponding to the target trip.

[0225] S1042 includes:

[0226] S10421. Determine the target scenario features corresponding to the driving scenario from the target trip features;

[0227] S10422. Based on the baseline features and target scene features corresponding to the driving scene, determine the target deviation score of the driving scene under multiple evaluation indicators;

[0228] S10423. Based on the dynamic weights under multiple evaluation indicators, the target deviation scores under multiple evaluation indicators are weighted and summed to obtain the target travel score.

[0229] S1043 includes:

[0230] S10431. Determine the driving habit scenario features corresponding to the driving scenario from the driving habit fingerprint features;

[0231] S10432. Based on the baseline characteristics and driving habit scenario characteristics corresponding to the driving scenario, determine the driving habit deviation score of the driving scenario under multiple evaluation indicators.

[0232] S10433. Based on the dynamic weights under multiple evaluation indicators, the driving habit deviation scores under multiple evaluation indicators are weighted and summed to obtain the long-term score.

[0233] S1051. Determine the target weight and long-term weight based on the driving habit entropy value; the driving habit entropy value is positively correlated with the target weight.

[0234] S1052. Based on the target weight and long-term weight, the target trip score and the long-term score are weighted and summed to obtain the driving behavior score.

[0235] Each step in this example has been described in the above embodiments and will not be repeated here.

[0236] In some embodiments, the method further includes: determining a dynamic rating based on driving behavior scores.

[0237] Specifically, based on driving behavior scores The score range to which the score belongs determines the dynamic rating.

[0238] For example, if driving behavior score Belongs to [90,100], with a dynamic rating of S, if the driving behavior score It belongs to [80, 90) and has a dynamic rating of A; as shown in Table 6.

[0239] Table 6: Dynamic Rating Mapping Table

[0240]

[0241] In some embodiments, such as Figure 7 As shown, the method further includes: S1061, determining a driving ability map based on the target deviation scores of the driving scenario under multiple evaluation indicators; S1062, determining the progress slope based on the target deviation scores and historical deviation scores of the driving scenario under multiple evaluation indicators, and determining the driving challenge coefficient based on the number of driving challenge tasks completed by the driver; and obtaining the driver's plasticity index by weighted summation of the target progress slope, driving challenge coefficient, and driving habit entropy value; S1063, determining the target scenario score based on the target deviation scores of the driving scenario under multiple evaluation indicators, and pushing driving challenge tasks to the driver based on the target scenario score and the plasticity index.

[0242] Specifically, the target deviation scores of the driving scenario under multiple evaluation indicators are uploaded to the cloud. The cloud then sends the target deviation scores of the driving scenario under multiple evaluation indicators to the display system. The display system determines the rating result based on the target deviation scores of the driving scenario under multiple evaluation indicators and displays the driving capability map.

[0243] For example, such as Figure 8 As shown, the driving ability map displays users' scores on scenario tags such as "highway cruising", "city congestion", "mountain winding roads" and "driving in the rain" in the form of a radar chart.

[0244] Optionally, in response to a triggered operation for any scenario label in the driving capability map, the score of the scenario label under multiple evaluation indicators is displayed, and suggested feedback is displayed. The suggested feedback is determined based on the deducted items corresponding to the scenario label, which are determined by the trip characteristics corresponding to the scenario label.

[0245] For example, the feedback suggestion is: "You scored 75 points in highway cruising, with the main reason for the low score being that the vehicle's average energy consumption was 15 kWh / 100km higher than the recommended value."

[0246] Optionally, based on the target trip characteristics, the target deviation score of the driving scenario under multiple evaluation indicators is determined. Similarly, based on the historical trip characteristics, the historical deviation score of the driving scenario under multiple evaluation indicators is determined. Based on the target deviation score and the historical deviation score of the driving scenario, the driver's progress slope in the driving scenario is determined, and the progress slope corresponding to the lowest target deviation score is taken as the target progress slope. The driving challenge coefficient is determined based on the number of driving challenge tasks completed by the driver. The driver's plasticity index is obtained by weighted summing of the target progress slope, the driving challenge coefficient, and the driving habit entropy value.

[0247] Specifically, the improvement slope of drivers in driving scenarios with low scores over a period of time is used as the target improvement slope. For example, if the average energy consumption is high during highway cruising, resulting in a low driving cruise score, but subsequent driving habits are better, the average energy consumption during highway cruising decreases, and the score in this scenario dimension rises rapidly, then the potential for improvement is high.

[0248] Driving habit entropy represents the average fluctuation of a driver's driving score. The greater the fluctuation and the more unstable the rating over a period of time, the higher the plasticity.

[0249] The effectiveness of a driver's adoption and implementation of the system's past suggestions is evaluated based on the number of challenge tasks completed and the level of enthusiasm for completing them. The more challenges a driver completes and the more actively they accept them, the higher their potential for growth.

[0250] For example, the target progress slope, driving challenge coefficient, and driving habit entropy value are weighted and summed, where the weight of the target progress slope can be 40%, the weight of the driving challenge coefficient can be 40%, and the weight of the driving habit entropy value can be 20%.

[0251] Alternatively, the plasticity grade P can be determined based on plasticity, and the plasticity grade P can be low, medium, or high.

[0252] Optionally, a target scenario score is determined based on the target deviation score of the driving scenario under multiple evaluation indicators, and a driving challenge task is determined based on the target scenario score and the driving behavior score, and the driving challenge task is pushed to the driver.

[0253] Specifically, based on driving behavior scores, several initial challenge tasks suitable for drivers are determined. Among the target deviation scores of driving scenarios under multiple evaluation indicators, the lowest score is determined. The driving scenario corresponding to the lowest score is used as the improvement scenario. Among the several initial challenge tasks, the driving challenge task corresponding to the improvement scenario is determined.

[0254] Optionally, a target scenario score is determined based on the target deviation score of the driving scenario under multiple evaluation indicators, and a driving challenge task is determined based on the target scenario score and the plasticity index, and the driving challenge task is pushed to the driver.

[0255] Specifically, the lowest score is determined among the target deviation scores of the driving scenario under multiple evaluation indicators. The driving scenario corresponding to the lowest score is taken as the improvement scenario. Among the candidate challenge tasks corresponding to the improvement scenario, the candidate challenge task that matches the plasticity index is taken as the driving challenge task.

[0256] It should be noted that driving challenge tasks are specific challenges designed for scenarios where drivers score low, aiming to improve undesirable driving habits. For example, for drivers with low scores in highway cruising scenarios, a "This Week's Challenge: Complete 3 smooth ramp entry maneuvers (longitudinal acceleration <0.25g)" can be pushed. Completing the challenge will earn virtual rewards and slightly increase the weight of relevant dimensions to encourage continuous improvement. The number and frequency of challenge pushes can be adjusted based on the plasticity index.

[0257] For example, the driving challenge task is: "Next time on a long downhill section, try to use kinetic energy recovery instead of braking. If the completion rate is >80% and a certain number of times is achieved, the task is successful."

[0258] In practical applications, completing a certain number of driving challenge tasks can provide feedback on driving habits, such as reducing driving habit entropy. This helps drivers gradually develop good habits and simultaneously improves their adaptive driving score. Suggested feedback can clearly inform drivers of the differences between their behavior and excellent habits in specific parameters, and provide suggestions. For example, suggested feedback could be: "If there are many instances of sudden braking when following closely, it is recommended to slow down to a certain distance beforehand." This can also be linked to "driving micro-challenges," offering related tasks for users to choose from, thereby achieving the goal of cultivating good driving habits and reducing driving habit entropy.

[0259] The driving behavior evaluation method based on driving habits provided in this application includes the following parts.

[0260] The first part identifies driving scenarios through multi-dimensional vehicle dynamic data.

[0261] The control system collects traditional behavioral signals from the vehicle's CAN bus (such as accelerator pedal opening, brake pressure, steering angle, vehicle speed, and driving time). By integrating these signals, it can identify the current interaction between the vehicle and the driver, such as whether the vehicle is accelerating through a corner or decelerating / braking. Additionally, it receives high-precision map data, real-time traffic flow information, weather data, and road complexity information perceived by the vehicle's vision and radar systems. This data is then integrated to identify the current driving scenario, such as urban congestion, highway cruising, mountain curves, and rainy driving. This achieves specific identification of the overall driving scenario, encompassing the interaction between the driver and the vehicle, and between the vehicle and its environment.

[0262] The second part involves extracting driving habit fingerprint features based on the driver's long-term driving habits.

[0263] By recording the characteristics of each driver's trip, and through long-term (e.g., one month) trip records, driving habit fingerprint features representing the driver's inherent driving style can be extracted. Extracting driving habit fingerprint features through long-term trip records can prevent short-term data from failing to characterize the driver's driving habits. The driving habit fingerprint feature is a multi-dimensional vector that includes deep-level features such as the driver's risk preference coefficients in different driving scenarios (e.g., average following distance, vehicle steering angular velocity, and vehicle slip ratio), operational finesse (e.g., average deceleration before stopping, average acceleration throughout the journey, and speed fluctuation standard deviation), and energy consumption patterns (e.g., vehicle energy consumption).

[0264] Part Three: Dynamic Weights and Adaptive Scoring.

[0265] By recognizing driving scenarios and combining them with the fingerprint characteristics of drivers' driving habits, driving behavior is scored. The scores are dynamically adjusted based on changes in drivers' driving habits through dynamic weighting. The weights are mainly determined by two parts:

[0266] Weighting adapted to driving scenarios: Based on the identified driving scenario, the weights of evaluation indicators such as safety and fuel efficiency are automatically adjusted. For example, in a mountainous winding road scenario, the weight of safety indicators is significantly increased, while the weight of fuel efficiency indicators can be reduced.

[0267] Driving habit entropy: This measures the volatility of a driver's behavior. New drivers with high driving habit entropy (unstable driving behavior) need to have their improvement trend emphasized, with higher weight given to recent improvement behaviors. For experienced drivers with low driving habit entropy (stable driving behavior), the focus is more on the gap between their stable and ideal behavior levels. Finally, combining the weights of driving scenario adaptation and driving habit fingerprint features, a driving behavior score is calculated, and a corresponding comprehensive rating is determined (e.g., divided into five levels: S, A, B, C, and D).

[0268] Part Four: Personalized Feedback and Driving Capability Map Generation.

[0269] The rating result is determined based on the target deviation score of the driving scenario under multiple evaluation indicators, and the driving ability map is displayed. The driving ability map shows the user's score on scenario tags such as "highway cruising", "city congestion", "mountain curves" and "driving in rainy weather" in the form of a radar chart.

[0270] By scoring the target deviation of the driving scenario under multiple evaluation indicators, the target scenario score is determined. Based on the target scenario score and the driving behavior score, the driving challenge task is determined and pushed to the driver.

[0271] This application's embodiments, by dividing driving scenarios, make driving behavior scoring more closely resemble actual driving scenarios, avoiding the unfairness caused by using the same evaluation method for different road conditions; it also achieves long-term dynamic learning through driving habit fingerprint features and driving habit entropy values, truly understanding the evolution of drivers' driving habits, making driving behavior evaluation more forward-looking and instructive; the weights, feedback, and challenge tasks vary from person to person and from time to time, possessing a high degree of personalization, forming a dedicated driving coach system, effectively improving user engagement and improvement results, and ultimately achieving multiple goals of safety, energy saving, and vehicle care.

[0272] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0273] Figure 9 This application provides a driving behavior evaluation device based on driving habits, the device comprising:

[0274] The target trip processing module 901 is used to determine the driving scenario and target trip characteristics corresponding to the target trip based on the vehicle dynamic data of the target trip; the target trip is the trip most recently completed by the driver.

[0275] The driving habit fingerprint feature determination module 902 is used to determine the driver's driving habit fingerprint features based on the target trip features and the historical trip features corresponding to multiple historical trips of the driver.

[0276] The driving habit entropy determination module 903 is used to determine the driver's driving habit entropy value based on the target trip features and driving habit fingerprint features;

[0277] The first evaluation module 904 is used to determine the target trip score and long-term score based on the target trip characteristics, driving habit fingerprint characteristics and driving scenario;

[0278] The second evaluation module 905 is used to determine the driving behavior score based on the driving habit entropy value, target trip score, and long-term score.

[0279] In some embodiments, the target trip processing module 901 is configured to classify scenarios based on vehicle dynamic data of the target trip to obtain at least one driving scenario corresponding to the target trip; determine scenario dynamic data corresponding to at least one driving scenario in the vehicle dynamic data; determine target scenario features corresponding to at least one driving scenario based on the scenario dynamic data corresponding to at least one driving scenario; and determine the driver's target trip features based on the target scenario features corresponding to at least one driving scenario.

[0280] In some embodiments, the driving habit entropy determination module 903 is used to determine the target fluctuation score of the target trip based on the target trip features and driving habit fingerprint features; determine the score standard deviation based on the target fluctuation score and the historical fluctuation scores of multiple historical trips; and determine the driver's driving habit entropy based on the score standard deviation.

[0281] In some embodiments, the driving habit entropy determination module 903 is used to determine key scenarios in the driving scenario based on the driving duration of the driving scenario corresponding to the target trip; determine key trip features corresponding to the key scenarios in the target trip features; determine key fingerprint features corresponding to the key scenarios in the driving habit fingerprint features; and determine the target fluctuation score of the target trip based on the similarity between the key trip features and the key fingerprint features.

[0282] In some embodiments, the driving habit entropy determination module 903 is used to determine the similarity under multiple evaluation indicators based on key trip features and key fingerprint features; determine the key weights corresponding to multiple evaluation indicators based on key scenarios; and perform weighted summation of the similarity corresponding to multiple evaluation indicators based on the key weights to obtain the target fluctuation score of the target trip.

[0283] In some embodiments, the first evaluation module 904 is used to determine the dynamic weight of the driving scenario under multiple evaluation indicators based on the driving time of the driving scenario corresponding to the target trip; determine the target trip score based on the target trip features, the baseline features corresponding to the driving scenario and the dynamic weight under multiple evaluation indicators; and determine the long-term score based on the driving habit fingerprint features, the baseline features and the dynamic weight under multiple evaluation indicators.

[0284] In some embodiments, the first evaluation module 904 is used to determine the target scenario features corresponding to the driving scenario in the target trip features; determine the target deviation score of the driving scenario under multiple evaluation indicators based on the benchmark features and target scenario features corresponding to the driving scenario; and perform a weighted summation of the target deviation scores under multiple evaluation indicators based on the dynamic weights under multiple evaluation indicators to obtain the target trip score.

[0285] In some embodiments, the second evaluation module 905 is used to determine the driving habit scenario features corresponding to the driving scenario in the driving habit fingerprint features; determine the driving habit deviation score of the driving scenario under multiple evaluation indicators based on the baseline features and driving habit scenario features corresponding to the driving scenario; and perform a weighted summation of the driving habit deviation scores under multiple evaluation indicators based on the dynamic weights under multiple evaluation indicators to obtain a long-term score.

[0286] In some embodiments, the second evaluation module 905 is used to determine the target weight and long-term weight based on the driving habit entropy value; the driving habit entropy value is positively correlated with the target weight; and the target trip score and the long-term score are weighted and summed based on the target weight and the long-term weight to obtain the driving behavior score.

[0287] The driving behavior evaluation device based on driving habits provided in this embodiment can execute the driving behavior evaluation method based on driving habits provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0288] Figure 10 A schematic diagram of the structure of the electronic device provided in this application. Figure 10 As shown, the electronic device 100 provided in this embodiment includes at least one processor 1001 and a memory 1002. Optionally, the device 100 further includes a communication component 1003. The processor 1001, memory 1002, and communication component 1003 are connected via a bus.

[0289] In a specific implementation, at least one processor 1001 executes computer execution instructions stored in memory 1002, causing at least one processor 1001 to perform the above-described method.

[0290] The specific implementation process of processor 1001 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0291] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0292] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0293] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0294] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0295] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0296] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0297] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0298] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0299] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0300] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0301] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0302] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0303] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A driving behavior evaluation method based on driving habits, characterized in that, include: Based on the vehicle dynamic data of the target trip, determine the driving scenario and target trip characteristics corresponding to the target trip; The target trip is the trip most recently completed by the driver; Based on the target trip characteristics and the historical trip characteristics corresponding to multiple historical trips of the driver, the driver's driving habit fingerprint characteristics are determined; Based on the target trip features and the driving habit fingerprint features, the driver's driving habit entropy value is determined; Based on the target trip characteristics, the driving habit fingerprint characteristics, and the driving scenario, a target trip score and a long-term score are determined. A driving behavior score is determined based on the driving habit entropy value, the target trip score, and the long-term score.

2. The method according to claim 1, characterized in that, The step of determining the driving scenario and target trip characteristics corresponding to the target trip based on vehicle dynamic data of the target trip includes: Based on the vehicle dynamic data of the target trip, the scenario is classified to obtain at least one driving scenario corresponding to the target trip; Determine at least one scenario dynamic data corresponding to the driving scenario from the vehicle dynamic data; Based on the scene dynamic data corresponding to at least one of the driving scenarios, determine the target scene features corresponding to at least one of the driving scenarios; The driver's target travel characteristics are determined based on at least one target scenario characteristic corresponding to the driving scenario.

3. The method according to claim 1, characterized in that, The step of determining the driver's driving habit entropy value based on the target trip features and the driving habit fingerprint features includes: Based on the target trip characteristics and the driving habit fingerprint characteristics, the target fluctuation score of the target trip is determined; The standard deviation of the score is determined based on the target volatility score and the historical volatility scores of the multiple historical journeys; The driver's driving habit entropy value is determined based on the standard deviation of the score.

4. The method according to claim 3, characterized in that, The step of determining the target fluctuation score of the target trip based on the target trip features and the driving habit fingerprint features includes: Based on the driving time of the driving scenario corresponding to the target trip, key scenarios are identified in the driving scenario; Determine the key travel features corresponding to the key scenarios from the target travel features; Determine the key fingerprint features corresponding to the key scenario from the driving habit fingerprint features; The target fluctuation score of the target journey is determined based on the similarity between the key journey features and the key fingerprint features.

5. The method according to claim 4, characterized in that, The step of determining the target fluctuation score of the target journey based on the similarity between the key journey features and the key fingerprint features includes: Based on the key travel features and the key fingerprint features, the similarity under multiple evaluation indicators is determined; Determine the key weights corresponding to the multiple evaluation indicators based on the key scenarios. Based on the key weights, the similarity scores corresponding to the multiple evaluation indicators are weighted and summed to obtain the target fluctuation score of the target journey.

6. The method according to any one of claims 1 to 5, characterized in that, The determination of the target trip score and long-term score based on the target trip features, the driving habit fingerprint features, and the driving scenario includes: Based on the driving time of the driving scenario corresponding to the target trip, determine the dynamic weight of the driving scenario under multiple evaluation indicators; The target trip score is determined based on the target trip characteristics, the baseline characteristics corresponding to the driving scenario, and the dynamic weights under multiple evaluation indicators. A long-term score is determined based on the driving habit fingerprint features, the baseline features, and the dynamic weights under multiple evaluation indicators.

7. The method according to claim 6, characterized in that, The step of determining the target trip score based on the target trip characteristics, the baseline characteristics corresponding to the driving scenario, and the dynamic weights under multiple evaluation indicators includes: Determine the target scenario features corresponding to the driving scenario from the target trip features; Based on the baseline features corresponding to the driving scenario and the target scenario features, determine the target deviation score of the driving scenario under multiple evaluation indicators; Based on the dynamic weights under multiple evaluation indicators, the target deviation scores under multiple evaluation indicators are weighted and summed to obtain the target travel score.

8. The method according to claim 6, characterized in that, The determination of a long-term score based on the driving habit fingerprint features, the baseline features, and the dynamic weights under multiple evaluation indicators includes: Determine the driving habit scenario features corresponding to the driving scenario from the driving habit fingerprint features; Based on the baseline features corresponding to the driving scenario and the driving habit scenario features, determine the driving habit deviation score of the driving scenario under multiple evaluation indicators; Based on the dynamic weights under multiple evaluation indicators, the driving habit deviation scores under multiple evaluation indicators are weighted and summed to obtain a long-term score.

9. The method according to any one of claims 1 to 5, characterized in that, The step of determining the driving behavior score based on the driving habit entropy value, the target trip score, and the long-term score includes: The target weight and long-term weight are determined based on the driving habit entropy value; the driving habit entropy value is positively correlated with the target weight. Based on the target weight and the long-term weight, the target trip score and the long-term score are weighted and summed to obtain the driving behavior score.

10. A driving behavior evaluation device based on driving habits, characterized in that, The device includes: The target trip processing module is used to determine the driving scenario and target trip characteristics corresponding to the target trip based on the vehicle dynamic data of the target trip; the target trip is the trip most recently completed by the driver; The driving habit fingerprint feature determination module is used to determine the driver's driving habit fingerprint features based on the target trip features and the historical trip features corresponding to multiple historical trips of the driver; The driving habit entropy determination module is used to determine the driver's driving habit entropy value based on the target trip features and the driving habit fingerprint features; The first evaluation module is used to determine the target trip score and long-term score based on the target trip characteristics, the driving habit fingerprint characteristics, and the driving scenario; The second evaluation module is used to determine the driving behavior score based on the driving habit entropy value, the target trip score, and the long-term score.

11. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1 to 9.

13. A computer program product, characterized in that, Includes computer execution instructions, which, when executed by a processor, implement the method as described in any one of claims 1 to 9.

Images