A user portrait generation method for driving habit tracking

Abstract

The present invention relates to the technical field of driving habit tracking, and discloses a method for generating a user portrait for driving habit tracking, the method comprising: performing time-series alignment on a user driving behavior data sequence and a neighboring vehicle behavior data sequence; using a neighboring vehicle impact assessment model to assess the degree of driving impact of the user driving behavior data; eliminating user driving behavior data with a driving impact degree higher than a preset impact threshold, forming a user driving behavior habit data sequence, and constructing a driving habit index system; based on the driving habit index system, generating a user driving habit portrait corresponding to the user driving behavior habit data sequence. The present invention is based on the spatiotemporal distance between a neighboring vehicle and a user-driven vehicle, and uses a spatiotemporal clustering method to obtain the degree of driving impact of abnormal behavior of neighboring vehicles on user driving behavior data, and screens a data sequence representing the user's driving habits to calculate a driving index value, and obtains a user driving habit portrait based on driving habit tracking.

Description

Technical Field

[0001] The present invention relates to the technical field of driving habit tracking, and in particular to a method for generating a user portrait for driving habit tracking. Background Art

[0002] In recent years, the popularity of IoT technology, vehicle sensors, smart phones and vehicle-to-everything (V2X) has enabled real-time collection of multi-dimensional data (such as vehicle speed, acceleration, braking intensity, steering angle, etc.) during vehicle operation. These data can not only reflect the driver's behavior patterns, but also explore the driver's potential risk habits, behavioral preferences, and safety hazards such as fatigue driving. At the same time, the advancement of artificial intelligence and big data analysis technology has provided strong technical support for the generation of driving habit portraits. By building an accurate driver portrait, personalized driving assistance systems, driving training, and traffic accident prediction can be realized. However, current user driving habit analysis is often based on the user's own data, ignoring the impact of other vehicles on the user's driving habits, resulting in confusion in the evaluation of user driving habits. Summary of the invention

[0003] In view of this, the present invention proposes a user portrait generation method for driving habit tracking, which tracks the user's driving habits through comprehensive analysis of the user's driving behavior data and the behavior data of neighboring vehicles, thereby achieving the purpose of generating an accurate user driving habit portrait.

[0004] To achieve the above object, the present invention provides a method for generating a user portrait for driving habit tracking, comprising the following steps:

[0005] S1: collecting a user driving behavior data sequence and a neighboring vehicle behavior data sequence of a neighboring vehicle during the user's driving process, and performing time series alignment on the user driving behavior data sequence and the neighboring vehicle behavior data sequence;

[0006] S2: using a neighboring vehicle impact assessment model to assess the degree of driving impact of neighboring vehicles on the user's driving behavior data, wherein the neighboring vehicle impact assessment model includes an input layer, a spatiotemporal clustering module, and an impact assessment module;

[0007] S3: removing user driving behavior data whose driving impact degree is higher than a preset impact threshold from the user driving behavior data sequence, taking the retained user driving behavior data as user driving behavior habit data, and forming a user driving behavior habit data sequence, and constructing a driving habit index system, wherein the driving habit index system includes a safe driving index, an energy utilization efficiency index, and a driving proficiency index;

[0008] S4: Calculate the index value of the user's driving behavior habit data sequence based on the driving habit index system to obtain the user's index value vector under the driving habit index system as the user's driving habit profile.

[0009] As a further improvement method of the present invention:

[0010] Optionally, the user driving behavior data sequence is:

[0011] ;

[0012] ;

[0013] in:

[0014] represents the user's driving behavior data sequence, represents the driving behavior data of the nth group of users, , N represents the total number of user driving behavior data, Represents user driving behavior data The i-th type of data in the user driving behavior data, wherein the 1st to 7th types of data in the user driving behavior data are the data collection time, the vehicle position, speed, acceleration, braking intensity, steering wheel steering angle and throttle angle of the driven vehicle;

[0015] The vehicle driven by the user collects the driving behavior data of the user and collects the behavior data of neighboring vehicles. The neighboring vehicles are vehicles whose distance from the vehicle driven by the user is less than a preset distance threshold. The neighboring vehicle behavior data sequence is :

[0016] ;

[0017] ;

[0018] ;

[0019] in:

[0020] represents the mth group of neighboring vehicle behavior data, , M represents the total number of neighboring vehicle behavior data, Represents the behavior data of neighboring vehicles The behavior data of the sth neighboring vehicle in Represents the behavior data of neighboring vehicles The number of neighboring vehicles;

[0021] Representing behavioral data The jth type of data in the neighboring vehicle behavior data, wherein the first to sixth types of data in the neighboring vehicle behavior data are the data collection time, the vehicle position, speed, acceleration, heading angle and behavior event of the neighboring vehicle, respectively; the data collection time of all behavior data in the neighboring vehicle behavior data is consistent;

[0022] The behavior events of the neighboring vehicles include sudden acceleration behavior, sudden deceleration behavior and lane change behavior. The behavior events are identified based on the acceleration change and position change of the neighboring vehicles under the behavior data of the neighboring vehicles. If the behavior event is identified, the data corresponding to the behavior event is 1, and if the behavior event is not identified, the data corresponding to the behavior event is 0;

[0023] The user driving behavior data sequence and the neighboring vehicle behavior data sequence are time-series aligned.

[0024] Optionally, the timing alignment process is:

[0025] Extract the data collection time from the neighboring vehicle behavior data and the user's driving behavior data, take the data collection time in the user's driving behavior data as the standard time, and construct an objective function that describes the standard deviation between the data collection time of the neighboring vehicle behavior data and the standard time. :

[0026] ;

[0027] ;

[0028] ;

[0029] in:

[0030] represents the data collection time of the mth group of neighboring vehicle behavior data, represents the temporal alignment parameter of the mth group of neighboring vehicle behavior data, represents the time series alignment result of the data collection time of the mth group of neighboring vehicle behavior data, Representation and The closest standard time, Indicates that the vehicle driven by the user is at standard time speed, represents the time series alignment parameter vector to be solved, with the goal of minimizing the standard deviation of the data collection time between the neighboring vehicle behavior data sequence and the user's driving behavior data sequence;

[0031] represents the average speed of all neighboring vehicles in the mth group of neighboring vehicle behavior data;

[0032] Indicates timing alignment parameters The time series information weight for the mth group of neighboring vehicle behavior data;

[0033] Construct the objective function The constraints are:

[0034] ;

[0035] in:

[0036] Indicates the preset parameter threshold;

[0037] Introducing the Lagrange multiplier, the objective function The constraints are constructed as a Lagrangian function, and the Lagrangian function is solved to obtain the timing alignment parameters;

[0038] The data collection time of each group of neighboring vehicle behavior data is aligned using the timing alignment parameters, and the timing alignment result is used as the data collection time of all behavior data under the neighboring vehicle behavior data to form a user driving behavior data sequence and a neighboring vehicle behavior data sequence after timing alignment.

[0039] Optionally, the input layer is used to receive the time-aligned user driving behavior data sequence and the neighboring vehicle behavior data sequence, the spatiotemporal clustering module is used to perform spatiotemporal clustering on the neighboring vehicle behavior data based on the aligned data collection time to obtain the spatiotemporal clustering clusters corresponding to each user driving behavior data, and the impact assessment module maps and transforms the spatiotemporal clustering clusters to obtain the driving impact degree of the user driving behavior data.

[0040] The process of using the input layer, spatiotemporal clustering module and impact assessment module to assess the driving impact of neighboring vehicles on the user's driving behavior data is as follows:

[0041] The input layer is used to receive the user driving behavior data sequence and the neighboring vehicle behavior data sequence after the time sequence alignment;

[0042] The spatiotemporal clustering module is used to perform spatiotemporal clustering on the neighboring vehicle behavior data based on the aligned data collection time. The spatiotemporal clustering method is:

[0043] The temporal and spatial distance between each group of behavior data and the user's driving behavior data is calculated, where the behavior data of the sth neighboring vehicle in the mth group of neighboring vehicle behavior data and the user's driving behavior data are The space-time distance between :

[0044] ;

[0045] in:

[0046] Represents user driving behavior data The data collection time in represents the data collection time after time-series alignment in the behavior data of the sth neighboring vehicle in the mth group of neighboring vehicle behavior data, Represents user driving behavior data The vehicle position in represents the vehicle position in the behavior data of the sth neighboring vehicle in the mth group of neighboring vehicle behavior data;

[0047] represents the L2 norm;

[0048] The behavior data of each neighboring vehicle is attributed to the driving behavior data of the user with the closest spatiotemporal distance, forming a spatiotemporal cluster of each user's driving behavior data, and filtering the behavior data in the spatiotemporal cluster whose spatiotemporal distance with the user's driving behavior data is higher than a preset distance threshold;

[0049] The impact assessment module maps and transforms the filtered spatiotemporal clusters as the driving impact degree of the user's driving behavior data.

[0050] The user driving behavior data The corresponding spatiotemporal cluster is , spatiotemporal clustering contains multiple groups of neighboring vehicle behavior data, the spatiotemporal clustering cluster The mapping conversion method is:

[0051] ;

[0052] in:

[0053] Represents spatiotemporal clusters The mapping conversion result, that is, the user driving behavior data The degree of driving influence;

[0054] Represents user driving behavior data The steering wheel angle in Represents user driving behavior data The speed of Represents user driving behavior data The acceleration in

[0055] Represents spatiotemporal clusters The mean heading angle of all behavioral data in , Represents spatiotemporal clusters The mean speed of all behavioral data in , Represents spatiotemporal clusters The mean acceleration of all behavioral data in ;

[0056] Represents the preset cluster number threshold, Represents spatiotemporal clusters The sum of all behavioral events in the behavioral data.

[0057] Optionally, the safe driving index is used to measure whether the user's driving behavior is safe, the energy utilization efficiency index is used to measure whether the user efficiently utilizes the vehicle's energy during driving, and the driving proficiency index is used to evaluate the user's driving control ability of the vehicle and the proficiency of driving skills.

[0058] Optionally, the index value vector of the user under the driving habit index system is composed of the index values ​​of the user in the driving habit index system, and the index values ​​of the user in the driving habit index system include the index values ​​of the user in the safe driving index, the energy utilization efficiency index and the driving proficiency index, wherein the higher the index values ​​of the three indicators are, the safer the user's driving habits are, the more energy-efficient they are, and the more proficient their driving skills are, and the highest index value is 100:

[0059] The index value of the safe driving index is calculated as follows:

[0060] ;

[0061] in:

[0062] Indicates the index value of the safe driving index, Indicates the number of rapid accelerations in the user's driving behavior habit data sequence, Indicates the number of emergency brakes. Indicates the number of sharp turns;

[0063] The indicator weights are the number of sudden accelerations, the number of sudden brakes, and the number of sharp turns;

[0064] The energy efficiency index is calculated as follows:

[0065] ;

[0066] in:

[0067] Indicates the index value of energy utilization efficiency index, Represents the ratio between the sum of the braking intensities and the total number of brakes in the user's driving behavior habit data sequence, and counts the number of user driving behavior habit data in the user's driving behavior habit data sequence whose braking intensities exceed a preset braking threshold as the total number of brakes in the user's driving behavior habit data sequence;

[0068] The weights are energy waste index and smooth driving index respectively;

[0069] It is the ratio between the mean speed and the mean braking intensity in the user's driving behavior habit data sequence;

[0070] The indicator value of the driving proficiency indicator is calculated as follows:

[0071] ;

[0072] in:

[0073] Indicates the index value of the driving proficiency index, Indicates the ratio of the total number of brakes in the user's driving behavior habit data sequence to the total number of the user's driving behavior habit data. It represents the ratio of the speed standard deviation to the speed mean. It represents the ratio of the standard deviation of acceleration to the mean acceleration. , It describes the coordination between the accelerator and the brake, the stability of the speed and acceleration during the user's driving process. are the indicator weights of the three.

[0074] In order to solve the above problem, the present invention provides an electronic device, the electronic device comprising:

[0075] A memory storing at least one instruction;

[0076] Communication interface, enabling electronic equipment to communicate; and

[0077] The processor executes the instructions stored in the memory to implement the user portrait generation method for driving habit tracking described above.

[0078] In order to solve the above problems, the present invention also provides a computer-readable storage medium, in which at least one instruction is stored. The at least one instruction is executed by a processor in an electronic device to implement the user portrait generation method for driving habit tracking described above.

[0079] Compared with the prior art, the present invention proposes a method for generating a user portrait for tracking driving habits, which has the following advantages:

[0080] First, this scheme proposes a timing alignment method, which collects the user driving behavior data sequence during the user's driving process and the neighboring vehicle behavior data sequence of the neighboring vehicles respectively. Since there is data transmission in the process of collecting the neighboring vehicle behavior data, there is a transmission delay. Then, the objective function is constructed based on the standard deviation of the two data collection times, and the timing alignment parameters of the neighboring vehicle behavior data are generated with the goal of minimizing the standard deviation. In the solution process, the timing information weight of each group of neighboring vehicle behavior data is generated in combination with the vehicle's speed information. The closer the speed between the neighboring vehicle and the user's driving vehicle, the higher the timing information weight, and the closer the timing alignment result of this group of neighboring vehicle behavior data is in terms of speed information. Timing alignment is performed based on the data collection time and speed information.

[0081] At the same time, this scheme proposes a user portrait generation method, which combines the aligned data collection time and position of the neighboring vehicles to calculate the spatiotemporal distance between the neighboring vehicles and the user's driving vehicle, and uses the spatiotemporal clustering method to obtain the spatiotemporal clustering clusters of the user's driving behavior data. The spatiotemporal clustering clusters are mapped and transformed to obtain the driving influence degree of the user's driving behavior data. The more frequent the behavior events of the neighboring vehicles, the greater the difference in vehicle routes, and the greater the speed and acceleration differences, the more abnormal the behavior of the neighboring vehicles, and the higher the degree of influence on the user's driving behavior data. The user's driving behavior data is invalid data under abnormal neighboring vehicles, and the user's driving behavior habit data sequence that represents the user's normal driving habits is extracted. A variety of driving habit indicators are used to calculate the indicator values ​​of the user's driving behavior habit data sequence. The higher the indicator values ​​of the three indicators, the safer the user's driving habits, the more energy-saving, and the more skilled the driving skills, and the user's driving habit portrait based on driving habit tracking is obtained. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Figure 1 A flowchart of a method for generating a user portrait for tracking driving habits provided by an embodiment of the present invention.

[0083] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0084] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0085] The embodiment of the present application provides a method for generating a user portrait for tracking driving habits. The execution subject of the method for generating a user portrait for tracking driving habits includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided by the embodiment of the present application. In other words, the method for generating a user portrait for tracking driving habits can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0086] Reference Figure 1 , Embodiment 1 of the present invention is:

[0087] A method for generating a user portrait for tracking driving habits comprises the following steps:

[0088] S1: Collecting a user driving behavior data sequence and a neighboring vehicle behavior data sequence of a neighboring vehicle during the user's driving process, and performing time alignment on the user driving behavior data sequence and the neighboring vehicle behavior data sequence.

[0089] The user driving behavior data sequence is composed of multiple sets of user driving behavior data, and the neighboring vehicle behavior data sequence is composed of multiple sets of neighboring vehicle behavior data;

[0090] The user driving behavior data includes the data collection time, the vehicle position, speed, acceleration, braking intensity, steering wheel steering angle and throttle angle of the driven vehicle, and the neighboring vehicle behavior data includes the data collection time, the vehicle position, speed, acceleration, heading angle and behavior events of the neighboring vehicles;

[0091] The user driving behavior data sequence is:

[0092] ;

[0093] ;

[0094] in:

[0095] represents the user's driving behavior data sequence, represents the driving behavior data of the nth group of users, , N represents the total number of user driving behavior data, Represents user driving behavior data The i-th type of data in the user driving behavior data, wherein the 1st to 7th types of data in the user driving behavior data are the data collection time, the vehicle position, speed, acceleration, braking intensity, steering wheel steering angle and throttle angle of the driven vehicle;

[0096] The vehicle driven by the user automatically collects the user's driving behavior data, wherein the vehicle position is collected by using a GPS module, the speed is collected by deploying a speed sensor on the wheel, the acceleration is collected by deploying an inertial measurement unit on the wheel, the braking intensity is collected by using a brake pressure sensor to measure the hydraulic pressure in the braking system, and the hydraulic pressure is used as the braking intensity. The higher the hydraulic pressure, the higher the braking intensity. The steering wheel steering angle is collected by reading the steering wheel steering angle in the CAN bus connected to the steering system, and the throttle angle is collected by reading the throttle angle in the CAN bus connected to the engine control unit;

[0097] The vehicle driven by the user collects the driving behavior data of the user and collects the behavior data of neighboring vehicles. The neighboring vehicles are vehicles whose distance from the vehicle driven by the user is less than a preset distance threshold. The neighboring vehicle behavior data sequence is :

[0098] ;

[0099] ;

[0100] ;

[0101] in:

[0102] represents the mth group of neighboring vehicle behavior data, , M represents the total number of neighboring vehicle behavior data, Represents the behavior data of neighboring vehicles The behavior data of the sth neighboring vehicle in Represents the behavior data of neighboring vehicles The number of neighboring vehicles;

[0103] Representing behavioral data The jth type of data in the neighboring vehicle behavior data, wherein the first to sixth types of data in the neighboring vehicle behavior data are the data collection time, the vehicle position, speed, acceleration, heading angle and behavior event of the neighboring vehicle, respectively; the data collection time of all behavior data in the neighboring vehicle behavior data is consistent;

[0104] The behavior events of the neighboring vehicles include sudden acceleration behavior, sudden deceleration behavior and lane change behavior. The behavior events are identified based on the acceleration change and position change of the neighboring vehicles under the behavior data of the neighboring vehicles. If the behavior event is identified, the data corresponding to the behavior event is 1, and if the behavior event is not identified, the data corresponding to the behavior event is 0;

[0105] The neighboring vehicle behavior data is collected by sharing the neighboring vehicles using the Internet of Vehicles;

[0106] The user driving behavior data sequence and the neighboring vehicle behavior data sequence are time-series aligned.

[0107] The timing alignment process is as follows:

[0108] Extract the data collection time from the neighboring vehicle behavior data and the user's driving behavior data, take the data collection time in the user's driving behavior data as the standard time, and construct an objective function that describes the standard deviation between the data collection time of the neighboring vehicle behavior data and the standard time. :

[0109] ;

[0110] ;

[0111] ;

[0112] in:

[0113] represents the data collection time of the mth group of neighboring vehicle behavior data, represents the temporal alignment parameter of the mth group of neighboring vehicle behavior data, represents the time series alignment result of the data collection time of the mth group of neighboring vehicle behavior data, Representation and The closest standard time, Indicates that the vehicle driven by the user is at standard time speed, represents the time series alignment parameter vector to be solved, with the goal of minimizing the standard deviation of the data collection time between the neighboring vehicle behavior data sequence and the user's driving behavior data sequence;

[0114] represents the average speed of all neighboring vehicles in the mth group of neighboring vehicle behavior data;

[0115] Indicates timing alignment parameters The time series information weight for the mth group of neighboring vehicle behavior data;

[0116] Construct the objective function The constraints are:

[0117] ;

[0118] in:

[0119] Indicates the preset parameter threshold;

[0120] Introducing the Lagrange multiplier, the objective function The constraints are constructed as a Lagrangian function, and the Lagrangian function is solved to obtain the timing alignment parameters;

[0121] The data collection time of each group of neighboring vehicle behavior data is aligned using the timing alignment parameters, and the timing alignment result is used as the data collection time of all behavior data under the neighboring vehicle behavior data to form a user driving behavior data sequence and a neighboring vehicle behavior data sequence after timing alignment.

[0122] S2: Use the neighboring vehicle impact assessment model to assess the degree of driving influence of neighboring vehicles on the user's driving behavior data.

[0123] The neighbor vehicle impact assessment model includes an input layer, a spatiotemporal clustering module and an impact assessment module; the spatiotemporal clustering module is used to perform spatiotemporal clustering on the neighbor vehicle behavior data based on the aligned data collection time to obtain the spatiotemporal clustering clusters corresponding to each user's driving behavior data, and the impact assessment module maps and transforms the spatiotemporal clustering clusters to obtain the driving impact degree of the user's driving behavior data.

[0124] The process of using the input layer, spatiotemporal clustering module and impact assessment module to assess the driving impact of neighboring vehicles on the user's driving behavior data is as follows:

[0125] The input layer is used to receive the user driving behavior data sequence and the neighboring vehicle behavior data sequence after the time sequence alignment;

[0126] The spatiotemporal clustering module is used to perform spatiotemporal clustering on the neighboring vehicle behavior data based on the aligned data collection time. The spatiotemporal clustering method is:

[0127] The temporal and spatial distance between each group of behavior data and the user's driving behavior data is calculated, where the behavior data of the sth neighboring vehicle in the mth group of neighboring vehicle behavior data and the user's driving behavior data are The space-time distance between :

[0128] ;

[0129] in:

[0130] Represents user driving behavior data The data collection time in represents the data collection time after time-series alignment in the behavior data of the sth neighboring vehicle in the mth group of neighboring vehicle behavior data, Represents user driving behavior data The vehicle position in represents the vehicle position in the behavior data of the sth neighboring vehicle in the mth group of neighboring vehicle behavior data;

[0131] represents the L2 norm;

[0132] The behavior data of each neighboring vehicle is attributed to the driving behavior data of the user with the closest spatiotemporal distance, forming a spatiotemporal cluster of each user's driving behavior data, and filtering the behavior data in the spatiotemporal cluster whose spatiotemporal distance with the user's driving behavior data is higher than a preset distance threshold;

[0133] The impact assessment module maps and transforms the filtered spatiotemporal clusters as the driving impact degree of the user's driving behavior data.

[0134] The user driving behavior data The corresponding spatiotemporal cluster is , spatiotemporal clustering contains multiple groups of neighboring vehicle behavior data, the spatiotemporal clustering cluster The mapping conversion method is:

[0135] ;

[0136] in:

[0137] Represents spatiotemporal clusters The mapping conversion result, that is, the user driving behavior data The degree of driving influence;

[0138] Represents user driving behavior data The steering wheel angle in Represents user driving behavior data The speed of Represents user driving behavior data The acceleration in

[0139] Represents spatiotemporal clusters The mean heading angle of all behavioral data in , Represents spatiotemporal clusters The mean speed of all behavioral data in , Represents spatiotemporal clusters The mean acceleration of all behavioral data in ;

[0140] Represents the preset cluster number threshold, Represents spatiotemporal clusters The sum of all behavioral events in the behavioral data.

[0141] S3: Eliminate user driving behavior data whose driving impact degree is higher than a preset impact threshold in the user driving behavior data sequence, use the retained user driving behavior data as user driving behavior habit data, and form a user driving behavior habit data sequence to construct a driving habit index system.

[0142] The safe driving index is used to measure whether the user's driving behavior is safe, and the safe driving index is quantified based on acceleration and steering wheel steering angle. The energy utilization efficiency index is used to measure whether the user efficiently utilizes the vehicle's energy during driving, and the energy utilization efficiency index is quantified based on speed and braking intensity. The driving proficiency index is used to evaluate the user's driving control ability of the vehicle and the proficiency of driving skills, and the driving proficiency index is quantified based on speed, braking intensity and acceleration.

[0143] S4: Calculate the index value of the user's driving behavior habit data sequence based on the driving habit index system to obtain the user's index value vector under the driving habit index system as the user's driving habit profile.

[0144] The index value vector of the user under the driving habit index system is composed of the index values ​​of the user in the driving habit index system. The index values ​​of the user in the driving habit index system include the index values ​​of the user in the safe driving index, the energy efficiency index and the driving proficiency index. The higher the index values ​​of the three indicators, the safer the user's driving habits, the more energy-efficient they are and the more proficient their driving skills are. The highest index value is 100:

[0145] The index value of the safe driving index is calculated as follows:

[0146] ;

[0147] in:

[0148] Indicates the index value of the safe driving index, Indicates the number of rapid accelerations in the user's driving behavior habit data sequence, Indicates the number of emergency brakes. Indicates the number of sharp turns. If the acceleration in the user's driving behavior habit data is a positive value and exceeds the preset acceleration threshold, then a sharp acceleration is recorded. If the acceleration in the user's driving behavior habit data is a negative value and exceeds the preset deceleration threshold, then a sudden brake is recorded. If the lateral acceleration in the user's driving behavior habit data is a positive value and exceeds the preset turning threshold, then a sharp turn is recorded. The lateral acceleration is calculated as follows:

[0149] ;

[0150] in:

[0151] represents the lateral acceleration, Indicates the speed in the user's driving behavior data. Indicates the change rate of the steering wheel angle in the user's driving behavior habit data, where the change rate is calculated as the ratio of the difference between the current steering wheel angle and the steering wheel angle in the previous set of user driving behavior habit data to the steering wheel angle in the previous set of user driving behavior habit data. Represents the difference in data collection time between two groups of user driving behavior habit data;

[0152] The indicator weights are the number of sudden accelerations, the number of sudden brakes, and the number of sharp turns;

[0153] The energy efficiency index is calculated as follows:

[0154] ;

[0155] in:

[0156] Indicates the index value of energy utilization efficiency index, Represents the ratio between the sum of the braking intensities and the total number of brakes in the user's driving behavior habit data sequence, and counts the number of user driving behavior habit data in the user's driving behavior habit data sequence whose braking intensities exceed a preset braking threshold as the total number of brakes in the user's driving behavior habit data sequence;

[0157] The weights are energy waste index and smooth driving index respectively;

[0158] It is the ratio between the mean speed and the mean braking intensity in the user's driving behavior habit data sequence;

[0159] The indicator value of the driving proficiency indicator is calculated as follows:

[0160] ;

[0161] in:

[0162] Indicator value representing the driving proficiency indicator, Indicates the ratio of the total number of brakes in the user's driving behavior habit data sequence to the total number of the user's driving behavior habit data. It represents the ratio of the speed standard deviation to the speed mean. It represents the ratio of the standard deviation of acceleration to the mean acceleration. , It describes the coordination between the accelerator and the brake, the stability of the speed and acceleration during the user's driving process. are the indicator weights of the three.

[0163] It should be understood that the embodiment is for illustration only and the scope of the patent application is not limited to this structure.

[0164] It should be noted that the serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments. And the terms "including", "comprising" or any other variants thereof in this article are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "including a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0165] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0166] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for generating a user profile for tracking driving habits, characterized in that: The method comprises: S1: collecting a user driving behavior data sequence and a neighboring vehicle behavior data sequence of a neighboring vehicle during the user's driving process, and performing time series alignment on the user driving behavior data sequence and the neighboring vehicle behavior data sequence; The user driving behavior data sequence is composed of multiple sets of user driving behavior data, and the neighboring vehicle behavior data sequence is composed of multiple sets of neighboring vehicle behavior data; The user driving behavior data includes the data collection time, the vehicle position, speed, acceleration, braking intensity, steering wheel steering angle and throttle angle of the driven vehicle, and the neighboring vehicle behavior data includes the data collection time, the vehicle position, speed, acceleration, heading angle and behavior events of the neighboring vehicles; S2: using a neighboring vehicle impact assessment model to assess the degree of driving impact of neighboring vehicles on the user's driving behavior data, wherein the neighboring vehicle impact assessment model includes an input layer, a spatiotemporal clustering module, and an impact assessment module; The input layer is used to receive the user driving behavior data sequence and the neighboring vehicle behavior data sequence after the time sequence alignment, the spatiotemporal clustering module is used to perform spatiotemporal clustering on the neighboring vehicle behavior data based on the aligned data collection time to obtain the spatiotemporal clustering cluster corresponding to each user driving behavior data, and the impact assessment module performs mapping conversion on the spatiotemporal clustering cluster to obtain the driving impact degree of the user driving behavior data; The process of using the input layer, spatiotemporal clustering module and impact assessment module to assess the driving impact of neighboring vehicles on the user's driving behavior data is as follows: The input layer is used to receive the user driving behavior data sequence and the neighboring vehicle behavior data sequence after the time sequence alignment; The spatiotemporal clustering module is used to perform spatiotemporal clustering on the behavior data of neighboring vehicles based on the aligned data collection time, and attribute the behavior data of each neighboring vehicle to the driving behavior data of the user with the closest spatiotemporal distance to form a spatiotemporal cluster of each user's driving behavior data, and filter the behavior data in the spatiotemporal clustering cluster whose spatiotemporal distance with the user's driving behavior data is higher than a preset distance threshold; The impact assessment module maps and transforms the filtered spatiotemporal clusters as the driving impact degree of the user's driving behavior data; The user driving behavior data The corresponding spatiotemporal cluster is , spatiotemporal clustering contains multiple groups of neighboring vehicle behavior data, the spatiotemporal clustering cluster The mapping conversion method is: ； in: represents the nth group of user driving behavior data in the user driving behavior data sequence, , N represents the total number of user driving behavior data in the user driving behavior data sequence; Represents spatiotemporal clusters The mapping conversion result, that is, the user driving behavior data The degree of driving influence; Represents user driving behavior data The steering wheel angle in Represents user driving behavior data The speed of Represents user driving behavior data The acceleration in Represents spatiotemporal clusters The mean heading angle of all behavioral data in , Represents spatiotemporal clusters The mean speed of all behavioral data in , Represents spatiotemporal clusters The mean acceleration of all behavioral data in ; Represents the preset cluster number threshold, Represents spatiotemporal clusters The sum of all behavioral events in the behavioral data; S3: removing user driving behavior data whose driving impact degree is higher than a preset impact threshold from the user driving behavior data sequence, taking the retained user driving behavior data as user driving behavior habit data, and forming a user driving behavior habit data sequence, and constructing a driving habit index system, wherein the driving habit index system includes a safe driving index, an energy utilization efficiency index, and a driving proficiency index; S4: Calculate the index value of the user's driving behavior habit data sequence based on the driving habit index system to obtain the user's index value vector under the driving habit index system as the user's driving habit profile.

2. The method for generating a user portrait for tracking driving habits according to claim 1, characterized in that: The user driving behavior data sequence is: ； ； in: represents the user's driving behavior data sequence, represents the driving behavior data of the nth group of users, , N represents the total number of user driving behavior data, Represents user driving behavior data The i-th type of data in the user driving behavior data, wherein the 1st to 7th types of data in the user driving behavior data are the data collection time, the vehicle position, speed, acceleration, braking intensity, steering wheel steering angle and throttle angle of the driven vehicle; The vehicle driven by the user collects the driving behavior data of the user and collects the behavior data of neighboring vehicles. The neighboring vehicles are vehicles whose distance from the vehicle driven by the user is less than a preset distance threshold. The neighboring vehicle behavior data sequence is : ； ； ； in: represents the mth group of neighboring vehicle behavior data, , M represents the total number of neighboring vehicle behavior data, Represents the behavior data of neighboring vehicles The behavior data of the sth neighboring vehicle in Represents the behavior data of neighboring vehicles The number of neighboring vehicles; Representing behavioral data The jth data in the neighboring vehicle behavior data, wherein the first to sixth data in the neighboring vehicle behavior data are the data collection time, the vehicle position, speed, acceleration, heading angle and behavior event of the neighboring vehicle, respectively; the data collection time of all behavior data in the neighboring vehicle behavior data is consistent; The user driving behavior data sequence and the neighboring vehicle behavior data sequence are time-series aligned.

3. The method for generating a user portrait for tracking driving habits according to claim 2, characterized in that: The timing alignment process is as follows: Extract the data collection time from the neighboring vehicle behavior data and the user's driving behavior data, take the data collection time in the user's driving behavior data as the standard time, and construct an objective function that describes the standard deviation between the data collection time of the neighboring vehicle behavior data and the standard time. : ； ； ； in: represents the data collection time of the mth group of neighboring vehicle behavior data, represents the temporal alignment parameter of the mth group of neighboring vehicle behavior data, represents the time series alignment result of the data collection time of the mth group of neighboring vehicle behavior data, Representation and The closest standard time, Indicates that the vehicle driven by the user is at standard time speed, represents the time series alignment parameter vector to be solved, with the goal of minimizing the standard deviation of the data collection time between the neighboring vehicle behavior data sequence and the user's driving behavior data sequence; represents the average speed of all neighboring vehicles in the mth group of neighboring vehicle behavior data; Indicates timing alignment parameters The time series information weight for the mth group of neighboring vehicle behavior data; Construct the objective function The constraints are: ； in: Indicates the preset parameter threshold; Introducing the Lagrange multiplier, the objective function The constraints are constructed as a Lagrangian function, and the Lagrangian function is solved to obtain the timing alignment parameters; The data collection time of each group of neighboring vehicle behavior data is aligned using the timing alignment parameters, and the timing alignment result is used as the data collection time of all behavior data under the neighboring vehicle behavior data to form a user driving behavior data sequence and a neighboring vehicle behavior data sequence after timing alignment.

4. The method for generating a user portrait for tracking driving habits according to claim 1, characterized in that: The safe driving index is used to measure whether the user's driving behavior is safe, the energy utilization efficiency index is used to measure whether the user efficiently utilizes the vehicle's energy during driving, and the driving proficiency index is used to evaluate the user's driving control ability of the vehicle and the proficiency of driving skills.

5. The method for generating a user portrait for tracking driving habits according to claim 4, characterized in that: The index value vector of the user under the driving habit index system is composed of the index values ​​of the user in the driving habit index system. The index values ​​of the user in the driving habit index system include the index values ​​of the user in the safe driving index, the energy efficiency index and the driving proficiency index. The higher the index values ​​of the three indicators, the safer the user's driving habits, the more energy-efficient they are and the more proficient their driving skills are. The highest index value is 100: The index value of the safe driving index is calculated as follows: ； in: Indicates the index value of the safe driving index, Indicates the number of rapid accelerations in the user's driving behavior habit data sequence, Indicates the number of emergency brakes. Indicates the number of sharp turns; The indicator weights are the number of sudden accelerations, the number of sudden brakes, and the number of sharp turns; The energy efficiency index is calculated as follows: ； in: Indicates the index value of energy utilization efficiency index, Represents the ratio between the sum of the braking intensities and the total number of brakes in the user's driving behavior habit data sequence, and counts the number of user driving behavior habit data in the user's driving behavior habit data sequence whose braking intensities exceed a preset braking threshold as the total number of brakes in the user's driving behavior habit data sequence; The weights are energy waste index and smooth driving index respectively; It is the ratio between the mean speed and the mean braking intensity in the user's driving behavior habit data sequence; The indicator value of the driving proficiency indicator is calculated as follows: ； in: Indicator value representing the driving proficiency indicator, Indicates the ratio of the total number of brakes in the user's driving behavior habit data sequence to the total number of the user's driving behavior habit data. It represents the ratio of the speed standard deviation to the speed mean. It represents the ratio of the standard deviation of acceleration to the mean acceleration. , It describes the coordination between the accelerator and the brake, the stability of the speed and acceleration during the user's driving process. are the indicator weights of the three.

Images