Driving style determination method and device, storage medium and electronic device

Abstract

This application discloses a method, apparatus, storage medium, and electronic device for determining driving style. The method includes acquiring a set of driving description information for a current driving object within a first driving cycle, wherein the driving description information set indicates multiple driving decision behaviors of the current driving object within the first driving cycle; inputting the driving description information set into a driving style classification model to obtain a first driving style label for the current driving object; acquiring second driving style labels corresponding to at least one reference driving object, wherein the at least one reference driving object is located within a road range associated with the current driving object; and determining a target driving style label for the current driving object based on the first driving style label and at least one second driving style label. This application solves the technical problem of inaccurate driving style labels for driving objects in related technologies.

Description

Technical Field

[0001] This application relates to the field of autonomous driving, and more specifically, to a method, apparatus, storage medium, and electronic device for determining driving style. Background Technology

[0002] In autonomous driving and intelligent driving assistance systems, the user profiling module plays a crucial role. Its core objective is to understand and characterize the driver's personality traits, driving habits, and preferences, thereby enabling more human-centered driving decisions that align with the driver's expectations during the decision-making and planning phase. These personality traits and preferences include not only basic driving styles (such as aggressive or mild), but may also encompass the driver's reaction time to emergencies, attitude towards risk, and tendency to avoid specific vehicles (such as trucks and motorcycles).

[0003] Currently, existing technologies typically pre-define a series of rules and standards to uniformly determine the driving style of a vehicle. While this approach is relatively intuitive, it may perform poorly when dealing with complex or unusual driving behaviors, lacking a personalized understanding of different vehicles and resulting in inaccurate determinations of driving styles. In other words, existing technologies suffer from the technical problem of inaccurately determining the driving style of a vehicle. Summary of the Invention

[0004] This application provides a method, apparatus, storage medium, and electronic device for determining driving style, so as to at least solve the technical problem of inaccurate determination of the driving style of the driving object in the prior art.

[0005] According to one aspect of the embodiments of this application, a method for determining driving style is provided, comprising: obtaining a set of driving description information of a current driving object within a first driving cycle, wherein the set of driving description information is used to indicate multiple driving decision behaviors of the current driving object within the first driving cycle; inputting the set of driving description information into a driving style classification model to obtain a first driving style label of the current driving object; obtaining a second driving style label corresponding to at least one reference driving object, wherein the at least one reference driving object is located within a road range associated with the current driving object; and determining a target driving style label of the current driving object based on the first driving style label and at least one second driving style label.

[0006] According to another aspect of the embodiments of this application, a driving style determination device is also provided, comprising: a first acquisition unit for acquiring a set of driving description information of a current driving object within a first driving cycle, wherein the set of driving description information is used to indicate multiple driving decision behaviors of the current driving object within the first driving cycle; a model output unit for inputting the set of driving description information into a driving style classification model to obtain a first driving style label of the current driving object; a second acquisition unit for acquiring a second driving style label corresponding to at least one reference driving object, wherein the at least one reference driving object is located within a road range associated with the current driving object; and a determination unit for determining a target driving style label of the current driving object based on the first driving style label and at least one second driving style label.

[0007] As an optional solution, the aforementioned determining unit further includes: a first determining module, used to determine the object type of each of at least one reference driving object; determine the corresponding label influence coefficient according to each of the at least one object type; and determine the target driving style label according to at least one second driving style label and its corresponding label influence coefficient, and the first driving style label.

[0008] As an optional solution, the aforementioned first determining module includes: a third obtaining module, used to obtain a first weight information set matching the current driving scenario, wherein the first weight information set is used to indicate the first label influence coefficient corresponding to each object type in the current driving scenario; to obtain a first driving object from at least one reference driving object; and to obtain a second label influence coefficient matching the current driving state of the first driving object from the type weight information set matching the first object type of the first driving object.

[0009] As an optional solution, the first determining module includes: a second determining module, configured to determine at least one candidate reference driving object corresponding to a second driving style label that has a standard decision relationship with the first driving style label based on the first driving style label of the current driving object; determine the label influence coefficient that matches the object type corresponding to each of the at least one candidate reference driving object; calculate the average value of the label influence coefficients corresponding to each of the at least one candidate reference driving object; and determine that the target driving style label is a label adjacent to the first driving style label if the average value is greater than a first threshold.

[0010] As an optional solution, the second determining module mentioned above includes: a model updating module, used to acquire driving data generated by the current driving object for at least one candidate reference driving object within the driving decision period, wherein the driving decision period is after the first driving period; and to update the model parameters in the driving style classification model according to the driving data of the current driving object and the corresponding target driving style label.

[0011] As an optional solution, the second acquisition unit is further configured to acquire the second driving style label sent by the reference driving object; acquire the reference driving description information set of at least one reference driving object in the first driving cycle respectively; input the at least one reference driving description information set into the driving style classification model to obtain the second driving style classification label corresponding to each of the at least one reference driving object.

[0012] As an optional solution, the aforementioned driving style determination device is also used to acquire braking activity information, driving speed information, driving position information, and following distance information; to acquire the probability coefficients corresponding to each of the multiple driving style labels output by the driving style classification model; and to determine the first driving style label from the multiple driving style labels based on the probability coefficients.

[0013] As an optional approach, the above-mentioned model output unit is also used to perform feature clustering based on the similarity between multiple historical driving description information to obtain multiple driving feature clusters; determine the standard driving style label corresponding to each of the multiple driving feature clusters; and use the multiple driving feature clusters with standard driving style labels as sample data to train a driving style classification model.

[0014] As an optional solution, the first acquisition unit is further configured to acquire a set of driving description information of the current driving object within a first driving cycle when the road type of the driving road where the current driving object is located is the target road type and the driving road is under the target traffic conditions; acquire a set of driving description information of the current driving object within a first driving cycle when the current driving object is under the target lighting environment; and acquire a set of driving description information of the current driving object within a first driving cycle when the current driving object is under the target weather conditions.

[0015] According to another aspect of the embodiments of this application, a computer program product or computer program is provided, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the driving style determination method described above.

[0016] According to another aspect of the embodiments of this application, an electronic device is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the above-described method for determining driving style through the computer program.

[0017] Through the above-described embodiments of this application, a set of driving description information of the current driving object within a first driving cycle is obtained. This set of driving description information is used to indicate multiple driving decision behaviors of the current driving object within the first driving cycle, reflecting the driving habits of the driving object. The set of driving description information is input into a driving style classification model to obtain a first driving style label of the current driving object. At least one second driving style label corresponding to each of at least one reference driving object is obtained. The at least one reference driving object is located within the road range associated with the current driving object, taking into account the influence with surrounding traffic participants. Thus, based on the first driving style label and at least one second driving style label, the target driving style label of the current driving object is determined. The label can be corrected based on actual driving behavior, improving the accuracy and dynamic adaptability of the driving style label, and solving the technical problem of inaccurate determination of the driving style of the driving object in related technologies. Attached Figure Description

[0018] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0019] Figure 1 This is a schematic diagram of the application environment of an optional method for determining driving style according to an embodiment of this application;

[0020] Figure 2 This is a flowchart of an optional method for determining a driving style according to an embodiment of this application;

[0021] Figure 3 This is a flowchart of another optional method for determining driving style according to an embodiment of this application;

[0022] Figure 4 This is a schematic diagram of an optional method for determining a driving style according to an embodiment of this application;

[0023] Figure 5 This is an optional online classification diagram according to an embodiment of this application;

[0024] Figure 6 This is a schematic diagram of a driving style determination device according to an embodiment of this application;

[0025] Figure 7 This is a schematic diagram of the structure of an optional electronic device according to an embodiment of this application. Detailed Implementation

[0026] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0027] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0028] According to one aspect of the embodiments of this application, a method for determining driving style is provided. Optionally, the method for determining driving style may be applied, but is not limited to, to applications such as... Figure 1 The hardware environment shown. Optionally, the driving style determination method provided in this application can be applied to a vehicle terminal. Figure 1 A side view of a vehicle terminal 101 is shown, which can travel on a driving surface 113. The vehicle terminal 101 includes a memory 102 storing an onboard navigation system 103 and a digital road map 104, a spatial monitoring system 117, a vehicle controller 109, a GPS (Global Positioning System) sensor 110, an HMI (Human / Machine Interface) device 111, and also includes an autonomous controller 112 and a telematics controller 114.

[0029] In one embodiment, the space monitoring system 117 includes one or more space sensors and systems for monitoring a visible area 105 in front of the vehicle terminal 101. The space monitoring system 117 also includes a space monitoring controller 118. The space sensors for monitoring the visible area 105 include a lidar sensor 106, a radar sensor 107, a camera 108, etc. The space monitoring controller 118 can be used to generate data related to the visible area 105 based on data input from the space sensors. The space monitoring controller 118 can determine the linear range, relative speed, and trajectory of the vehicle terminal 101 based on the input from the space sensors, for example, determining the vehicle's current speed and its relative speed to a vehicle in front. The space sensors of the vehicle terminal space monitoring system 117 may include object positioning sensing devices, which may include range sensors that can be used to locate objects in front, such as vehicles in front.

[0030] Camera 108 is advantageously mounted and positioned on vehicle terminal 101 in a location that allows for the capture of images of a visible area 105, wherein at least a portion of the visible area 105 includes the area in front of vehicle terminal 101 and a portion of the travel surface 113 of the trajectory of vehicle terminal 101. The visible area 105 may also include the surrounding environment. Other cameras may also be employed, for example, including a second camera positioned on the rear or side portion of vehicle terminal 101 to monitor the rear of vehicle terminal 101 and one of the right or left sides of vehicle terminal 101.

[0031] The autonomous controller 112 is configured to implement autonomous driving or advanced driver assistance system (ADAS) vehicle terminal functionality. Such functionality may include an onboard vehicle terminal control system capable of providing a certain level of driving automation. Driving automation may include a series of dynamic driving and vehicle terminal operations. Driving automation may include a certain level of automated control or intervention involving individual vehicle terminal functions (e.g., steering, acceleration, and / or braking). For example, the aforementioned autonomous controller may be used to determine the validity of a detection point trace by performing the following steps:

[0032] S102, obtain the driving description information set of the current driving object in the first driving cycle, wherein the driving description information set is used to indicate multiple driving decision behaviors of the current driving object in the first driving cycle;

[0033] S104, Input the set of driving description information into the driving style classification model to obtain the first driving style label of the current driving object;

[0034] S106 Obtain the second driving style label corresponding to at least one reference driving object, wherein at least one reference driving object is located within the road range associated with the current driving object;

[0035] S108, determine the target driving style label of the current driving object based on the first driving style label and at least one second driving style label.

[0036] HMI device 111 provides human-machine interaction for guiding the operation of infotainment systems, GPS (Global Positioning System) sensors 110, airborne navigation systems 103, and similar systems, and includes controllers. HMI device 111 monitors operator requests and provides the operator with status, service, and maintenance information about the vehicle terminal system. HMI device 111 communicates with and / or controls the operation of multiple operator interface devices. HMI device 111 may also communicate with one or more devices that monitor biometric data associated with the vehicle terminal operator. For simplicity, HMI device 111 is depicted as a single device, but in embodiments of the system described herein, it may be configured as multiple controllers and associated sensing devices.

[0037] Operator controls may be included in the passenger compartment of vehicle terminal 101 and, by way of non-limiting example, may include a steering wheel, accelerator pedal, brake pedal, and operator input device, which is an element of HMI device 111. The operator controls enable a vehicle terminal operator to interact with and instruct the operation of vehicle terminal 101 to provide passenger transport.

[0038] The airborne navigation system 103 uses a digital road map 104 for the purpose of providing navigation support and information to the vehicle terminal operator. The autonomous controller 112 uses the digital road map 104 for the purpose of controlling the operation of the autonomous vehicle terminal or the functions of the ADAS vehicle terminal.

[0039] The vehicle terminal 101 may include a telematics controller 114, which includes a wireless telematics communication system capable of communicating outside the vehicle terminal (including communicating with a communication network 115 with both wireless and wired communication capabilities). The wireless telematics communication system includes a non-airborne server 116 capable of short-range wireless communication with mobile terminals.

[0040] It should be noted that the information collected in this application (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data used for analysis, etc.) are information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of this data all comply with relevant laws, regulations, and standards, necessary confidentiality measures have been taken, and they do not violate public order and good morals. Corresponding access points are provided for users to choose to authorize or refuse. For example, interfaces are set up between this system and relevant users or organizations, providing users with corresponding access points to choose to agree to or refuse automated decision-making results; if the user chooses to refuse, the process proceeds to the expert decision-making stage.

[0041] As an optional implementation method, such as Figure 2 As shown, the method for determining the driving style can be performed by electronic devices, and the specific steps include:

[0042] S202, Obtain a set of driving description information of the current driving object in the first driving cycle, wherein the set of driving description information is used to indicate multiple driving decision behaviors of the current driving object in the first driving cycle;

[0043] S204, Input the set of driving description information into the driving style classification model to obtain the first driving style label of the current driving object;

[0044] S206, obtain the second driving style label corresponding to at least one reference driving object, wherein at least one reference driving object is located within the road range associated with the current driving object;

[0045] S208, determine the target driving style label of the current driving object based on the first driving style label and at least one second driving style label.

[0046] In S202 of the above embodiment, a set of driving description information of the current driving object within the first driving cycle is obtained. The set of driving description information is used to indicate multiple driving decision behaviors of the current driving object within the first driving cycle. The set of driving description information includes parameters and data reflecting the driving behavior of the current driving object, such as lane change frequency, braking / acceleration frequency, average absolute acceleration / deceleration, average value / standard deviation of acceleration, standard deviation of lane centering error, following distance, and TTC longitudinal speed with the preceding vehicle. The driving data of the current driving object can be collected in real time by vehicle sensors (such as cameras, radar, ABS system, etc.).

[0047] In step S204 above, the driving description information set is input into the driving style classification model to obtain the first driving style label of the current driving object. It can be understood that the driving style includes, but is not limited to, the following types: aggressive type, characterized by rapid lane changes, frequent acceleration and deceleration, high average speed, small following distance, and a tendency to take aggressive measures when facing risks; attempting to overtake quickly in traffic jams and being more aggressive in judging green and yellow lights; moderate type, characterized by smooth driving, fewer lane changes, gentle acceleration and deceleration, and a larger following distance; and economical type, characterized by low-speed driving, avoiding unnecessary acceleration and deceleration, preferring to use cruise control, minimizing energy consumption, paying more attention to fuel efficiency or electricity consumption while driving, tending to choose the economic mode, and reducing unnecessary operations to reduce driving costs.

[0048] As an optional implementation, key features are extracted from the driving description information set, standardized, and input into the driving style classification model for prediction to obtain the first driving style label of the current driving object.

[0049] It should be noted that the above driving style classification model can be implemented through a process as follows. Figure 3 The following method was used for training:

[0050] S302, Data Acquisition and Data Analysis; Analyze the acquired data, select the region of interest for sliding window slicing, and extract features that reflect the driver profile, such as: lane change frequency, braking / acceleration frequency, mean absolute acceleration / deceleration, average / standard deviation of acceleration, standard deviation of lane centering error, following distance, and TTC longitudinal speed with the vehicle in front.

[0051] S304, Data cleaning, data slicing, feature extraction; S306, Standardization, feature selection, dimensionality reduction, model training; S308, Adding labels; After extracting features, standardize the features, and then use unsupervised learning K-Means algorithm or Gaussian mixture model GMM for clustering. Add labels (aggressive, general, mild) to the clustering results. GMM is a soft clustering algorithm that can give the probability that a sample belongs to each cluster.

[0052] Specifically, after clustering, classification algorithms such as Support Vector Machine (SVM), KNN, or classification trees can be selected for classification. The classification model is then trained iteratively to obtain the weight values ​​of the classification network.

[0053]

[0054]

[0055]

[0056] In the formula, exp is an exponential function with the natural constant e as its base. , , These represent the probability values ​​for the driver's aggressive, average, and mild driving profiles, respectively. , , The values ​​represent the linear classifications of the driver's aggressive, normal, and mild driving profiles, respectively. These values ​​are the results of clustering, and the regression equation coefficients are then trained based on the following equation:

[0057]

[0058]

[0059]

[0060] In the formula, , , The regression weights are set for each feature of the driver's aggressive, normal, and mild driving profiles. , , These represent the intercepts of the linear classification equations for the driver's aggressive, average, and mild driving profiles, respectively.

[0061] S310, Model Training, which involves retraining the model to obtain the weight values ​​for each feature; the above process is illustrated in the diagram. Figure 4 The stages shown are feature extraction, cluster analysis, and classification algorithm.

[0062] Therefore, the online classification first obtains the model's input values, then performs feature preprocessing, and finally uses a classification algorithm to classify the data, outputting probability values ​​for three driver profile labels. The specific equation is the inverse process of the above formula, as shown in the flowchart. Figure 5 As shown, where, Figure 5 The inputs on the left, such as AvgALgt, VarALgt, AvgALat, VarALat, and TimeGap, are feature values ​​derived from scene analysis.

[0063] In step S206 above, a second driving style label corresponding to at least one reference driving object is obtained, wherein at least one reference driving object is located within the road range associated with the current driving object; the reference driving object is other vehicles or traffic participants around the current driving object, whose driving style may affect the driving decision of the current driving object; the road range includes a certain range of adjacent or closely spaced lanes.

[0064] In step S208 above, the target driving style label of the current driving object is determined based on the first driving style label and at least one second driving style label. Optionally, for example, if the nearest reference driving object to the current driving object has an "aggressive" driving style, it may be assigned a higher influence coefficient. If the first driving style label of the vehicle is "mild," then ideally, the vehicle should make a decision to avoid the reference vehicle when it encounters it. However, in actual decision-making, the vehicle makes multiple overtaking maneuvers, and the influence coefficient of the reference driving object is relatively large. In this case, it is necessary to redetermine the corresponding target driving style label for the vehicle's individualization. For example, the redetermined target driving style label may be "mild" for urban roads and "aggressive" for highway roads, or the user's profile label may be adjusted as "mildly aggressive" overall.

[0065] Through the above-described embodiments of this application, a set of driving description information of the current driving object within a first driving cycle is obtained. This set of driving description information is used to indicate multiple driving decision behaviors of the current driving object within the first driving cycle, reflecting the driving habits of the driving object. The set of driving description information is input into a driving style classification model to obtain a first driving style label of the current driving object. At least one second driving style label corresponding to each of at least one reference driving object is obtained. The at least one reference driving object is located within the road range associated with the current driving object, taking into account the influence with surrounding traffic participants. Thus, based on the first driving style label and at least one second driving style label, the target driving style label of the current driving object is determined. The label can be corrected based on actual driving behavior, improving the accuracy and dynamic adaptability of the driving style label, and solving the technical problem of inaccurate determination of the driving style of the driving object in related technologies.

[0066] In one optional implementation, determining the target driving style label of the current driving object based on a first driving style label and at least one second driving style label includes:

[0067] S1, determine the object type of at least one reference driving object;

[0068] S2, determine the corresponding label influence coefficient based on at least one object type;

[0069] S3. Determine the target driving style label based on at least one second driving style label and their respective label influence coefficients, and the first driving style label.

[0070] As an optional implementation, in step S1 above, the object type of the reference driving object includes, but is not limited to, types with different safety levels and vehicle sizes. For example, school buses, ambulances, and fire trucks have higher safety levels and therefore greater impact. Compared with cars and tricycles, passenger cars are larger in size, have more blind spots, and longer braking distances, and therefore also have a greater impact.

[0071] In step S2 above, different types of driving objects have different degrees of influence on the label of the current driving object, and have different label influence coefficients. For example, the label influence coefficient of a school bus is greater than that of a tricycle.

[0072] In step S3 above, as an optional implementation, when the vehicle's first driving style label is "mild" and it encounters a reference driving object with an "aggressive" style, the standard should be biased towards taking avoidance measures. However, if the influence coefficient of the reference driving object is extremely high (greater than the preset threshold) and the vehicle actually exhibits overtaking behavior, it is necessary to personalize the adjustment of the current driving object so that the target driving style label is more biased towards "aggressive".

[0073] Understandably, this application allows users to personalize their driving style labels by providing feedback on their preferences or based on specific user actions (such as frequent manual takeover). Even if users A and B have similar driving data at the beginning, their profile labels will be personalized based on the general label classification if they provide feedback or make adjustments through the mechanisms provided by the system.

[0074] Through the above-described embodiments of this application, at least one reference driving object's object type is determined; a corresponding label influence coefficient is determined based on at least one object type, quantifying the influence of different types of driving objects on the accuracy of the current driving style label, and considering the correlation between vehicle type and driving style; a target driving style label is determined based on at least one second driving style label and its corresponding label influence coefficient, and the first driving style label, which not only considers the current driver's style but also incorporates the style influence of surrounding vehicles and environmental characteristics, thereby improving the accuracy of the driving style label.

[0075] In one optional implementation, the label influence coefficient is determined according to at least one object type, including one of the following:

[0076] Method 1: Obtain the first weight information set that matches the current driving scenario. The first weight information set is used to indicate the first label influence coefficient of each object type in the current driving scenario.

[0077] Method 2: Obtain a first driving object from at least one reference driving object; obtain the second label influence coefficient that matches the current driving state of the first driving object from the first object type matching type weight information set of the first driving object.

[0078] In the above method one, the current driving scenario includes different road scenarios, lighting scenarios, weather scenarios, etc. For different scenarios, the label influence coefficient of the reference driving object is different. Taking the road scenario as an example, the label influence coefficient of large vehicles is greater in the highway scenario, and the label influence coefficient of sedans that are more flexible in urban scenarios is greater.

[0079] By taking into account the specific needs of different scenarios, the first weight information set contains the label influence coefficients of different object types in specific scenarios. This allows us to determine the degree of influence of different vehicle types on style labels based on different scenarios, thereby improving the accuracy of determining the driving style label of the current driving object.

[0080] In the above method two, the first driving object can be the driving object that is closest to the current driving object and is judged to be likely to interact. For example, the first driving object is merging into the lane where the current driving object is located from a fork, or the first driving object is the driving object with the fastest driving speed behind the current driving object. The label influence coefficient is different depending on the driving state of the reference driving object. For example, the label influence coefficient of a high-speed vehicle is greater than that of a low-speed vehicle. If the overtaking object is a reference driving object with a larger label influence coefficient, the current driving object is biased to be labeled "aggressive".

[0081] By taking into account the different degrees of influence of different driving states of the reference driving object on the driving style of the driving object, it is possible to determine the degree of influence on the style label of the current driving object based on different driving states, thereby improving the accuracy of determining the driving style label of the current driving object.

[0082] In one optional implementation, determining a target driving style label based on at least one second driving style label and its corresponding label influence coefficient, and a first driving style label, includes:

[0083] S1, based on the first driving style label of the current driving object, determine at least one candidate reference driving object corresponding to the second driving style label that has a standard decision relationship with the first driving style label;

[0084] S2, determine the label influence coefficient that matches the object type corresponding to at least one candidate reference driving object;

[0085] S3, calculate the average value of the label influence coefficients corresponding to at least one candidate reference driving object;

[0086] S4, if the average value is greater than the first threshold, determine the target driving style label as the label adjacent to the first driving style label.

[0087] Optionally, the above-mentioned standard decision-making relationship can be understood as a set of driving style labels with unified decision-making standards. If the current driving object is labeled as "mild" and vehicle B is labeled as "aggressive", then the decision when they meet should be for the current driving object to give way to vehicle B. Conversely, if the current driving object is labeled as "aggressive" and vehicle B is labeled as "mild", then the decision when they meet should be for the current driving object to overtake vehicle B. If the current driving object is labeled as "mild" and vehicle B is labeled as "normal" (between "mild" and "aggressive"), then the decision when they meet should be for the current driving object to maintain its driving state, etc., without specific restrictions.

[0088] The following describes steps S1-S4 using a complete implementation method:

[0089] Assuming the model outputs driving style labels including: mild, normal, and aggressive, the first driving style label of the current driving object is determined to be "mild" in the first driving cycle, and the second driving style label with standard decision relationship is determined to be "aggressive". This is because these two label styles can be generally regarded as driving objects with the "mild" label avoiding driving objects with the "aggressive" label when they are close to each other.

[0090] For example, if the reference driving object A adjacent to the current driving object in the left and right lanes is identified as a truck, the label influence coefficient corresponding to the truck is 4; reference driving object B is a tricycle, the corresponding label influence coefficient is 2; reference driving object C is a school bus, the corresponding label influence coefficient is 6; and reference driving object D is a car, the corresponding label influence coefficient is 3.

[0091] Based on the collected data, it was determined that the current driving object overtook the truck, school bus, and sedan. The average value calculated based on the corresponding label influence coefficient is (4+6+3) / 3=4.3. Assuming the first threshold is 3.5, it was determined that the value exceeded the first threshold, and the label of the current driving object needs to be personalized. In this scenario, it can be determined that the current driving object usually engages in aggressive behavior, and the user's driving style label needs to be updated to the "normal" label, which is one level adjacent to the "mild" label. That is, if driving data similar to the first driving cycle is collected in the future, it is determined that the driving style label of the current driving object is actually "normal".

[0092] By introducing a label influence coefficient and threshold mechanism, it is possible to identify the driving behavior patterns of the current driving object when facing different types of reference driving objects based on the actual driving behavior of the current driving object, and to dynamically adjust the personalized driving style label, thereby improving the accuracy and real-time performance of the label.

[0093] In one optional implementation, after determining that the target driving style label is adjacent to the first driving style label when the average value is greater than a first threshold, the process includes:

[0094] S1, Obtain driving data generated by the current driving object for at least one candidate reference driving object within the driving decision cycle, wherein the driving decision cycle is after the first driving cycle;

[0095] S2, based on the driving data of the current driving object and the corresponding target driving style label, update the model parameters in the driving style classification model.

[0096] In step S1 above, driving data generated by the current driving object to at least one candidate reference driving object within the driving decision period is obtained, wherein the driving decision period is after the first driving period; optionally, the driving decision period is the time period during which interactive behavior is generated after the style label of the driving object is determined, and the driving data includes interactive behavior data between the current driving object and the reference driving object, such as relative speed, acceleration, and distance.

[0097] In step S2 above, the model parameters in the driving style classification model are updated based on the driving data of the current driving object and the corresponding target driving style label. The target driving style label is a style label based on the personalized analysis of the actual interaction behavior of the current driving object. Based on the corrected target driving style label and the corresponding driving data, which may also include the driving data before the decision, the model parameters in the driving style classification model are updated to train a driving style classification model that is more in line with the personalization of the current driving object. Through deep learning of individual driver behavior, the autonomous driving system can better understand and predict the driver's reaction in specific situations, thereby making decisions that are closer to human driving behavior. With user authorization, the profile information can be contributed to a larger data pool through anonymization, which can obtain more accurate driving style labels of surrounding vehicles in the driving scenario and make better driving decisions.

[0098] By retraining and updating the model parameters based on the personalized target driving style labels, the driving style classification model can better adapt to changes in the driving style of the current driving object, thereby improving the accuracy of the style labels.

[0099] In one optional implementation, obtaining a second driving style label corresponding to at least one reference driving object includes at least one of the following:

[0100] Method 1: Obtain the second driving style tag sent by the reference driving object;

[0101] Method 2: Obtain at least one set of reference driving description information for each reference driving object within the first driving cycle, input the at least one set of reference driving description information into the driving style classification model, and obtain the second driving style classification label corresponding to each of the at least one reference driving object.

[0102] Optionally, in the above method one, with user authorization, the reference driving object can establish a real-time data exchange channel through V2V (Vehicle-to-Vehicle) or V2X (Vehicle-to-Everything) communication technology. Then, the reference driving object sends its second driving style label, which has been processed by its internal driving style classification model, to the vehicle through the communication protocol. The vehicle's communication module can receive and decode these label information and make decision-making actions based on the driving style labels.

[0103] Optionally, in the above method two, the driving description information (such as driving speed, acceleration, following distance, etc.) of the surrounding reference driving objects can be collected and measured by reading the perception system (such as radar, camera, GPS) of the driving object over a period of time. The collected data can then be analyzed using the driving style classification model of the current driving object to obtain the second driving style label of the reference driving object.

[0104] Through these two methods, intelligent driving systems can not only directly acquire information about the driving styles of other road users using real-time communication technology, but also autonomously analyze collected driving data to predict the driving styles of surrounding vehicles. This not only improves the accuracy of driving decisions but also enhances the system's adaptability and safety.

[0105] In one alternative implementation,

[0106] S1, Obtain a set of driving description information of the currently driving object within the first driving cycle, including at least one of the following:

[0107] S1-1, Braking activity information, driving speed information, driving position information, following distance information;

[0108] Understandably, the aforementioned braking activity information includes braking / acceleration frequency, reflecting the frequency of vehicle braking and acceleration operations. A high braking / acceleration frequency may indicate that the driver tends to have an aggressive driving style, while a lower frequency may indicate a more moderate driving habit. The aforementioned driving speed information includes average absolute acceleration / deceleration. The aforementioned driving position information includes the average / standard deviation of acceleration and the standard deviation of lane centering error. The statistics of acceleration (i.e., the change in the rate of change of speed, Jerk) reflect the smoothness and comfort of the vehicle's driving path. The standard deviation of lane centering error measures the statistical value of the degree of deviation of the vehicle from the center line of the lane when driving in the lane. The larger the standard deviation, the worse the stability of lane keeping, which may reflect a more aggressive or unstable driving style. The aforementioned following distance information includes following distance and the longitudinal speed (TTC) of the vehicle in front. Following distance describes the distance between the vehicle and the vehicle in front. TTC is an important safety indicator, calculating the time from the current time to the possible collision with the vehicle in front. Its relationship with longitudinal speed directly reflects the driver's assessment of the potential collision risk.

[0109] S2, input the driving description information set into the driving style classification model to obtain the first driving style label of the current driving object, including:

[0110] S2-1, Obtain the probability coefficients corresponding to each of the multiple driving style labels output by the driving style classification model;

[0111] S2-2, determine the first driving style label from multiple driving style labels based on probability coefficients.

[0112] The above steps S2-1 to S2-2 are described in one optional implementation: The preprocessed driving description information set is input into the model, including braking activity, driving speed, driving position and following distance information; the model outputs the probability coefficients of three style labels according to the input information, for example: "aggressive" probability is 0.2, "normal" probability is 0.3 and "mild" probability is 0.5; and then the "mild" with the highest probability is selected from the prediction results as the first driving style label of the current driving object.

[0113] In one optional implementation, before inputting the driving description information set into the driving style classification model to obtain the first driving style label of the current driving object, the method further includes:

[0114] S1. Based on the similarity between multiple historical driving description information, feature clustering is performed to obtain multiple driving feature clusters;

[0115] S2, determine the standard driving style label corresponding to each of the multiple driving feature clusters;

[0116] S3 uses multiple driving feature clusters with standard driving style labels as sample data to train a driving style classification model.

[0117] In step S1 above, similarity is used to measure the degree of similarity between different historical driving description information. It is usually calculated based on a distance metric (such as Euclidean distance, Mahalanobis distance, etc.) or a similarity coefficient (such as cosine similarity). Feature clustering is to group historical driving description information with similar features into different driving feature clusters under an unsupervised learning framework. The clustering algorithm can be K-means, DBSCAN, hierarchical clustering, etc. Multiple driving feature clusters are the results of clustering. Each cluster represents a set of historical driving description information with similar driving behaviors or styles.

[0118] In step S2 above, driving behavior experts determine a standard driving style label for each driving feature cluster based on the analysis results. This can be done by combining historical driving description information with the analysis of driving decision-making behaviors in each driving feature cluster. For example, the aggressive driving style cluster may include behaviors such as frequent lane changes and sudden acceleration. Based on the behavior patterns, a standard driving style label is determined for each cluster. The determined standard driving style label is then associated with the corresponding driving feature cluster to establish a mapping relationship between the feature cluster and the driving style label.

[0119] In step S3 above, driving feature cluster data with standard driving style labels are organized to form a training dataset. A neural network is used as a driving style classification model. The model is trained by inputting the training dataset. Through multiple iterations, the model parameters are adjusted to improve prediction accuracy. At the same time, the model performance is verified using a test dataset, and necessary parameter tuning is performed.

[0120] Through the above steps, the intelligent driving system can train a classification model that can accurately predict driving styles based on historical driving description information, through feature clustering and expert evaluation, thereby achieving personalized recognition of driver styles and improving the accuracy of determining the driving style label of the driving object.

[0121] In one optional implementation, obtaining the set of driving description information of the currently driving object within the first driving cycle further includes:

[0122] S1, if the road type of the current driving object is the target road type and the driving road is under the target traffic conditions, obtain the set of driving description information of the current driving object in the first driving cycle;

[0123] S2, when the current driving object is in the target lighting environment, obtain the set of driving description information of the current driving object in the first driving cycle;

[0124] S3, under the condition that the current driving object is in the target weather condition, obtain the set of driving description information of the current driving object in the first driving cycle.

[0125] It should be noted that, in some optional implementations, the driving data collected may be representative data selected from certain specific scenarios.

[0126] Optionally, in step S1 above, the target road type is a predefined road type or a road type required by a specific task, such as highway, urban road, rural road, etc.; the target traffic condition is the traffic condition required by a specific task or scenario, including but not limited to congestion, smooth traffic, construction zone, etc.; for example, the collected data is a set of driving description information of the driving object under highway congestion conditions.

[0127] By selectively acquiring driving description information under specific road types and traffic conditions, the relevance and accuracy of the collected data are increased. This is because drivers' driving habits and styles vary significantly under different road types and traffic conditions. By focusing on data under target environmental conditions, driver characteristics can be captured more accurately, thereby improving the accuracy of driving style classification.

[0128] Optionally, in step S2 above, the target lighting environment refers to the lighting conditions under specific conditions or as required by the task, such as daytime, nighttime, low light, or direct sunlight. For example, the collected data is a set of driving description information of a vehicle under dim lighting conditions.

[0129] Lighting conditions (such as daytime, nighttime, bright light, or shadow) directly affect a driver's visibility and reaction ability. By collecting driving description information under target lighting conditions, intelligent driving systems can learn typical driver behaviors under these special lighting conditions, thereby improving the accuracy of driving style recognition in different lighting environments and optimizing driving assistance strategies.

[0130] Optionally, in step S3 above, the target weather conditions, such as sunny, rainy, snowy, or foggy weather, are confirmed by the vehicle's weather sensor showing that the current weather is rainy, the road surface is slippery, and the visibility is reduced. The vehicle is then identified as being in the target weather condition of "rainy". The vehicle's driving description information set under this weather condition is then collected.

[0131] The above steps focus on driving description information under specific weather conditions (such as rain, snow, fog, etc.). When driving in the rain, drivers may use the brakes more frequently, maintain a greater safe distance, and reduce lane changes. By acquiring data under these specific weather conditions, we can learn the characteristics of the driving style of the driving object in the rain, improve the accuracy of model training, and thus improve the accuracy of the model in recognizing the driving style characteristics of the driving object.

[0132] Understandably, the size and weight of large vehicles like trucks make rapid lane changes or accelerations physically more difficult, while cars are more agile. Therefore, the same speed change or lane-changing behavior may be perceived differently by drivers of different vehicle types. For example, a smooth lane change might be considered aggressive by a truck driver, while the same behavior might be seen as typical driving style by a car driver. Therefore, driving style classification models can be trained to produce different results for different types of vehicles, and the scheme described above in this application can be used in combination with these models.

[0133] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0134] According to another aspect of the embodiments of this application, a driving style determination apparatus for implementing the above-described driving style determination method is also provided. For example... Figure 6 As shown, the device includes:

[0135] The first acquisition unit 602 acquires a set of driving description information of the current driving object within the first driving cycle, wherein the set of driving description information is used to indicate multiple driving decision behaviors of the current driving object within the first driving cycle;

[0136] The model output unit 604 inputs the driving description information set into the driving style classification model to obtain the first driving style label of the current driving object;

[0137] The second acquisition unit 606 acquires a second driving style label corresponding to at least one reference driving object, wherein at least one reference driving object is located within the road range associated with the current driving object;

[0138] The determining unit 608 determines the target driving style label of the current driving object based on the first driving style label and at least one second driving style label.

[0139] Optionally, the determining unit 608 further includes: a first determining module, configured to determine the object type of each of at least one reference driving object; determine the corresponding label influence coefficient according to each of the at least one object type; and determine the target driving style label according to at least one second driving style label and its corresponding label influence coefficient, and the first driving style label.

[0140] Optionally, the first determining module includes: a third obtaining module, used to obtain a first weight information set matching the current driving scenario, wherein the first weight information set is used to indicate the first label influence coefficient corresponding to each object type in the current driving scenario; to obtain a first driving object from at least one reference driving object; and to obtain a second label influence coefficient matching the current driving state of the first driving object from the type weight information set matching the first object type of the first driving object.

[0141] Optionally, the first determining module includes: a second determining module, configured to determine at least one candidate reference driving object corresponding to a second driving style label that has a standard decision relationship with the first driving style label based on the first driving style label of the current driving object; determine the label influence coefficient that matches the object type corresponding to each of the at least one candidate reference driving object; calculate the average value of the label influence coefficients corresponding to each of the at least one candidate reference driving object; and determine that the target driving style label is a label adjacent to the first driving style label if the average value is greater than a first threshold.

[0142] Optionally, the second determining module mentioned above includes: a model updating module, used to acquire driving data generated by the current driving object for at least one candidate reference driving object within the driving decision period, wherein the driving decision period is after the first driving period; and to update the model parameters in the driving style classification model according to the driving data of the current driving object and the corresponding target driving style label.

[0143] Optionally, the second acquisition unit 606 is further configured to acquire the second driving style label sent by the reference driving object; acquire the reference driving description information set of at least one reference driving object in the first driving cycle respectively; input the at least one reference driving description information set into the driving style classification model to obtain the second driving style classification label corresponding to each of the at least one reference driving object.

[0144] Optionally, the aforementioned driving style determination device is also used to acquire braking activity information, driving speed information, driving position information, and following distance information; to acquire the probability coefficients corresponding to each of the multiple driving style labels output by the driving style classification model; and to determine the first driving style label from the multiple driving style labels based on the probability coefficients.

[0145] Optionally, the above-mentioned model output unit 604 is also used to perform feature clustering based on the similarity between multiple historical driving description information to obtain multiple driving feature clusters; determine the standard driving style label corresponding to each of the multiple driving feature clusters; and use the multiple driving feature clusters with standard driving style labels as sample data to train a driving style classification model.

[0146] Optionally, the first acquisition unit 602 is further configured to acquire a set of driving description information of the current driving object within a first driving cycle when the road type of the driving road where the current driving object is located is the target road type and the driving road is under the target traffic conditions; acquire a set of driving description information of the current driving object within a first driving cycle when the current driving object is under the target lighting environment; and acquire a set of driving description information of the current driving object within a first driving cycle when the current driving object is under the target weather conditions.

[0147] For specific implementation examples, please refer to the examples shown in the above method for determining driving style, which will not be repeated here.

[0148] The memory 702 can be used to store software programs and modules, such as the program instructions / modules corresponding to the driving style determination method and device in this embodiment of the invention. The processor 704 executes various functional applications and data processing by running the software programs and modules stored in the memory 702, thereby realizing the aforementioned driving style determination method. The memory 702 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 702 may further include memory remotely located relative to the processor 704, and these remote memories can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof. Specifically, the memory 702 may be used, but is not limited to, for storing file information such as target files. As an example, such as Figure 7 As shown, the memory 702 may include, but is not limited to, the first acquisition unit 602, the model output unit 604, the second acquisition unit 606, and the determination unit 608 from the driving style determination device. Furthermore, it may include, but is not limited to, other module units from the driving style determination device, which will not be elaborated upon in this example.

[0149] Optionally, the transmission device 706 described above is used to receive or send data via a network. Specific examples of the network described above may include wired networks and wireless networks. In one example, the transmission device 706 includes a Network Interface Controller (NIC), which can be connected to other network devices and a router via a network cable to communicate with the Internet or a local area network. In another example, the transmission device 706 is a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0150] In addition, the aforementioned electronic device also includes a display 708 and a connection bus 710 for connecting the various module components in the aforementioned electronic device.

[0151] According to one aspect of this application, a computer program product is provided, comprising a computer program / instructions containing program code for performing the methods shown in the flowchart. In such embodiments, the computer program can be downloaded and installed from a network via a communication component, and / or installed from a removable medium. When the computer program is executed by a central processing unit, it performs various functions provided in embodiments of this application.

[0152] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0153] It should be noted that the computer system of the electronic device is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0154] Specifically, according to embodiments of this application, the processes described in the various method flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication component, and / or installed from a removable medium. When the computer program is executed by a central processing unit, it performs various functions defined in the system of this application.

[0155] According to one aspect of this application, a computer-readable storage medium is provided, wherein a processor of a computer device reads computer instructions from the computer-readable storage medium, and executes the computer instructions to cause the computer device to perform the methods provided in the various alternative implementations described above.

[0156] Optionally, in this embodiment, the computer-readable storage medium described above may be configured to store a computer program for performing the following steps:

[0157] S1, obtain the set of driving description information of the current driving object in the first driving cycle, wherein the set of driving description information is used to indicate multiple driving decision behaviors of the current driving object in the first driving cycle;

[0158] S2, input the set of driving description information into the driving style classification model to obtain the first driving style label of the current driving object;

[0159] S3, obtain the second driving style label corresponding to at least one reference driving object, wherein at least one reference driving object is located within the road range associated with the current driving object;

[0160] S4, determine the target driving style label of the current driving object based on the first driving style label and at least one second driving style label.

[0161] Optionally, in this embodiment, those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware of an electronic device. The program can be stored in a computer-readable storage medium, which may include: flash drive, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0162] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0163] If the integrated units in the above embodiments are implemented as software functional units and sold or used as independent products, they can be stored in the aforementioned computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or 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 one or more computer devices (which may be personal computers, servers, or network devices, etc.) to execute all or part of the steps of the methods of the various embodiments of this application.

[0164] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0165] In the several embodiments provided in this application, it should be understood that the disclosed user equipment can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and 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 through some interfaces; the indirect coupling or communication connection of units or modules may be electrical or other forms.

[0166] 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.

[0167] Furthermore, the functional units in the various embodiments of this application 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. The integrated unit can be implemented in hardware or as a software functional unit.

[0168] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method of determining a driving style, characterized by include: Obtain a set of driving description information of the current driving object within a first driving cycle, wherein the set of driving description information is used to indicate multiple driving decision behaviors of the current driving object within the first driving cycle; Input the set of driving description information into the driving style classification model to obtain the first driving style label of the current driving object; Obtain a second driving style label corresponding to at least one reference driving object, wherein at least one of the reference driving objects is located within the road range associated with the current driving object; Determine the object type of at least one of the reference driving objects; Determine the corresponding tag influence coefficient for each of at least one of the object types; Based on the first driving style label of the current driving object, determine at least one candidate reference driving object corresponding to the second driving style label that has a standard decision relationship with the first driving style label; Determine the label influence coefficient that matches the object type corresponding to at least one of the candidate reference driving objects; Calculate the average value of the label influence coefficients corresponding to at least one of the candidate reference driving objects; If the average value is greater than the first threshold, the target driving style label is determined to be the label adjacent to the first driving style label.

2. The method according to claim 1, characterized in that, Determining the corresponding tag influence coefficient based on at least one of the object types includes one of the following: Obtain a first set of weight information that matches the current driving scenario, wherein the first set of weight information is used to indicate the first label influence coefficient corresponding to each of the object types in the current driving scenario; Obtain a first driving object from at least one of the reference driving objects; and obtain a second label influence coefficient that matches the current driving state of the first driving object from the first object type matching type weight information set of the first driving object.

3. The method according to claim 1, characterized in that, After determining that the target driving style label is an adjacent label to the first driving style label when the average value is greater than the first threshold, the process includes: Acquire driving data generated by the current driving object for at least one of the candidate reference driving objects within a driving decision cycle, wherein the driving decision cycle is after the first driving cycle; Based on the driving data of the current driving object and the corresponding target driving style label, update the model parameters in the driving style classification model.

4. The method according to claim 1, characterized in that, The step of obtaining the second driving style label corresponding to at least one of the reference driving objects includes at least one of the following: Obtain the second driving style tag sent by the reference driving object; Obtain at least one set of reference driving description information for each of the reference driving objects within the first driving cycle, and input the at least one set of reference driving description information into the driving style classification model to obtain the second driving style label corresponding to each of the at least one reference driving object.

5. The method according to claim 1, characterized in that, The process of obtaining the set of driving description information of the currently driving object within the first driving cycle includes at least one of the following: Braking activity information, driving speed information, driving position information, following distance information; The step of inputting the driving description information set into the driving style classification model to obtain the first driving style label of the current driving object includes: Obtain the probability coefficients corresponding to each of the multiple driving style labels output by the driving style classification model; The first driving style label is determined from the plurality of driving style labels based on the probability coefficient.

6. The method according to claim 1, characterized in that, Before inputting the driving description information set into the driving style classification model to obtain the first driving style label of the current driving object, the method further includes: Multiple driving feature clusters are obtained by clustering features based on the similarity between multiple historical driving description information. Determine the standard driving style label corresponding to each of the multiple driving feature clusters; The driving style classification model is trained by using multiple driving feature clusters with the standard driving style label as sample data.

7. The method according to claim 1, characterized in that, The step of obtaining the set of driving description information of the currently driving object within the first driving cycle also includes: If the road type of the current driving object is the target road type and the driving road is under the target traffic conditions, obtain the set of driving description information of the current driving object in the first driving cycle; When the current driving object is in the target lighting environment, obtain the set of driving description information of the current driving object within the first driving cycle; When the current driving object is in the target weather condition, obtain the set of driving description information of the current driving object within the first driving cycle.

8. A device for determining driving style, characterized in that, include: The first acquisition unit acquires a set of driving description information of the current driving object within a first driving cycle, wherein the set of driving description information is used to indicate multiple driving decision behaviors of the current driving object within the first driving cycle; The model output unit inputs the driving description information set into the driving style classification model to obtain the first driving style label of the current driving object; The second acquisition unit acquires a second driving style label corresponding to at least one reference driving object, wherein at least one of the reference driving objects is located within the road range associated with the current driving object; The determining unit determines the object type of at least one of the reference driving objects; determines the corresponding label influence coefficient for each of the at least one object type; determines at least one candidate reference driving object corresponding to a second driving style label that has a standard decision relationship with the first driving style label of the current driving object; determines the label influence coefficient that matches the object type corresponding to each of the at least one candidate reference driving object; calculates the average value of the label influence coefficients corresponding to each of the at least one candidate reference driving object; and if the average value is greater than a first threshold, determines the target driving style label as the label adjacent to the first driving style label.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein the program is executed by an electronic device to perform the method according to any one of claims 1 to 7.

10. An electronic device comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor is configured to execute the method described in any one of claims 1 to 7 through the computer program.

11. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method described in any one of claims 1 to 7.

Images