Driving style configuration method, model training method, data processing method, device and equipment

Abstract

The invention provides a driving style configuration method, a model training method, a data processing method, devices and equipment. Relates to the technical field of intelligent driving. The method comprises: acquiring style configuration information of the autonomous mobile device (S201), acquiring a target driving style based on the style configuration information (S202), and configuring the driving style of the autonomous mobile device based on the target driving style (S203). According to the method, the preference and demand of the user can be identified through the style configuration information, so that the driving style meeting the expectation of the user is configured, and the user experience is improved.

Description

Technical Field

[0001] This application relates to the field of intelligent driving technology, and in particular, to a method for configuring driving styles, a method for model training, a method for data processing, a device, and a device. Background Art

[0002] Autonomous mobile devices have intelligent driving systems that can provide configuration options for driving styles when driving autonomously or assisting in driving. For example, if the autonomous mobile device is an intelligent driving vehicle, the intelligent driving system can provide configuration options for driving styles such as selecting the gear for following distance and the percentage of the maximum speed limit, so that the performance of the intelligent driving system better meets the expectations or needs of users.

[0003] In the prior art, the driving styles of autonomous mobile devices are usually preset by the autonomous mobile device manufacturer during the production stage. These preset driving style schemes are general and fixed after being determined. However, in the above method, when ordinary users want to customize their driving styles, it is not well applicable to ordinary users, resulting in poor user experience. Summary of the Invention

[0004] This application provides a method for configuring driving styles, a method for model training, a method for data processing, a device, and a device to solve the problem that the preset driving styles in the prior art cannot well meet the personalized needs of ordinary users, resulting in poor user experience.

[0005] In a first aspect, this application provides a method for configuring a driving style, the method including:

[0006] Obtain the style configuration information of the autonomous mobile device;

[0007] Based on the style configuration information, obtain the target driving style;

[0008] Based on the target driving style, configure the driving style of the autonomous mobile device.

[0009] In a possible implementation manner, the style configuration information includes any one of the following:

[0010] The driving control method selected by the user;

[0011] The image data and / or audio data collected by the autonomous mobile device;

[0012] The position information selected by the user in the style palette, where the style palette includes color adjustment areas corresponding to a plurality of pre-obtained driving styles.

[0013] In a possible implementation, if the style configuration information is the driving control mode selected by the user, then obtaining the target driving style based on the style configuration information includes:

[0014] Based on the driving control mode and the pre-configured mapping relationship, determining the target driving style corresponding to the driving control mode; wherein, the mapping relationship includes the driving styles corresponding to different driving control modes.

[0015] In a possible implementation, the driving control mode is determined based on at least one driving scenario among a congestion scenario, a starting time at an intersection, a lane change power selection, a turning comfort level, a cut-in handling method, and a safety distance maintained from an autonomous mobile device traveling normally.

[0016] In a possible implementation, if the style configuration information is the image data and / or audio data collected by the autonomous mobile device, then obtaining the target driving style based on the style configuration information includes:

[0017] Processing the image data and / or audio data based on a pre-obtained data processing model to obtain a style description vector;

[0018] Selecting the target driving style corresponding to the style description vector from a preset driving style library; wherein, the driving style library stores style description vectors corresponding to different driving styles.

[0019] In a possible implementation, when the style configuration information is image data, the data processing model is a visual processing model;

[0020] When the style configuration information is audio data, the data processing model is a language processing model;

[0021] When the style configuration information includes image data and audio data, the data processing model is a visual language processing model.

[0022] In a possible implementation, the style description vector includes a scene description vector and / or a preference instruction vector;

[0023] Wherein, the scene description vector is used to indicate the driving scene where the autonomous mobile device is located, and the preference instruction vector is used to indicate the user's preference for the driving style.

[0024] In a possible implementation, the selecting the target driving style corresponding to the style description vector from a preset driving style library includes:

[0025] Select the driving style corresponding to the target description vector with the highest similarity to the style description vector from the driving style library as the target driving style.

[0026] In a possible implementation manner, the method further includes:

[0027] Receive the data processing model sent by the server and deploy the data processing model.

[0028] In a possible implementation manner, if the style configuration information is the position information selected by the user in the style palette, then obtaining the target driving style based on the style configuration information includes:

[0029] If the position information is within the color palette area corresponding to any driving style in the style palette, determine the driving style as the target driving style;

[0030] If the position information is not within the color palette area corresponding to any driving style, determine at least one driving style within the preset range of the position information from the style palette;

[0031] Determine the target driving style according to the at least one driving style.

[0032] In a possible implementation manner, determining the target driving style according to the at least one driving style includes:

[0033] Determine the target driving style according to the distance between the color palette area corresponding to each driving style in the at least one driving style and the position information.

[0034] In a possible implementation manner, the method further includes:

[0035] Receive the style palette data sent by the server;

[0036] Display the style palette on the user interface of the autonomous mobile device.

[0037] In a second aspect, the present application provides a model training method for driving styles, including:

[0038] Obtain a plurality of training samples, each training sample includes data to be processed and a style description vector corresponding to the data to be processed, and the data to be processed includes image data and / or audio data collected by the autonomous mobile device;

[0039] Train a deep learning model according to the plurality of training samples to obtain a data processing model, and the data processing model is used to analyze the input data to obtain the style description vector of the driving style.

[0040] In a possible implementation, the data processing model is any one of a visual processing model, a language processing model, or a vision-language processing model;

[0041] Among them, the visual processing model is used to analyze image data to obtain a style description vector of the driving style;

[0042] The language processing model is used to analyze audio data to obtain a style description vector of the driving style;

[0043] The vision-language processing model is used to analyze image data and audio data to obtain a style description vector of the driving style.

[0044] In a possible implementation, the style description vector includes a scene description vector and / or a preference instruction vector;

[0045] Among them, the scene description vector is used to indicate the driving scene where the autonomous mobile device is located, and the preference instruction vector is used to indicate the user's preference for the driving style.

[0046] In a third aspect, the present application provides a data processing method for driving style, including:

[0047] Obtain driving data of multiple users;

[0048] For each user, according to the driving data of the user, obtain the driving style vector corresponding to the user;

[0049] Perform dimensionality reduction clustering analysis on the driving style vectors corresponding to the multiple users to obtain a style palette.

[0050] In a possible implementation, the step of, for each user, obtaining the driving style vector corresponding to the user according to the driving data of the user includes:

[0051] Analyze the driving data of the user according to a preset style parsing model to obtain the driving style vector corresponding to the user.

[0052] In a fourth aspect, the present application provides a model training method for driving style, including:

[0053] Obtain multiple training samples, each training sample including the driving data of a user and the driving style vector corresponding to the driving data;

[0054] Train a deep learning model according to the multiple training samples to obtain a style parsing model.

[0055] In a fifth aspect, the present application provides a configuration device for driving style, the device includes:

[0056] The first acquisition module is configured to acquire the style configuration information of the autonomous mobile device;

[0057] The first processing module is configured to acquire the target driving style based on the style configuration information;

[0058] The configuration module is configured to configure the driving style of the autonomous mobile device based on the target driving style.

[0059] In a sixth aspect, the present application provides a model training device for driving styles, including:

[0060] The second acquisition module is configured to acquire a plurality of training samples, each training sample including data to be processed and a style description vector corresponding to the data to be processed, where the data to be processed includes image data and / or audio data collected by the autonomous mobile device;

[0061] The second processing module is configured to train a deep learning model according to the plurality of training samples to obtain a data processing model, and the data processing model is configured to analyze the input data to obtain a style description vector of the driving style.

[0062] In a seventh aspect, the present application provides a data processing device for driving styles, including:

[0063] The third acquisition module is configured to acquire the driving data of a plurality of users;

[0064] The third processing module is configured to, for each user, acquire a driving style vector corresponding to the user according to the driving data of the user;

[0065] The analysis module is configured to perform dimensionality reduction clustering analysis on the driving style vectors corresponding to the plurality of users to obtain a style palette.

[0066] In an eighth aspect, the present application provides a model training device for driving styles, including:

[0067] The fourth acquisition module is configured to acquire a plurality of training samples, each training sample including the driving data of a user and a driving style vector corresponding to the driving data;

[0068] The fourth processing module is configured to train a deep learning model according to the plurality of training samples to obtain a style parsing model.

[0069] In a ninth aspect, the present application provides an electronic device, including: a processor, a memory, and a communication interface;

[0070] The memory stores computer execution instructions;

[0071] The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of the first aspect.

[0072] In a tenth aspect, the present application provides a server, including: a processor, a memory, and a communication interface;

[0073] The memory stores computer-executable instructions;

[0074] The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of the second aspect to the fourth aspect.

[0075] In an eleventh aspect, the present application provides a computer-readable storage medium, in which computer-executable instructions are stored, and when the computer-executable instructions are executed by a processor, they are used to implement the method according to any one of the first aspect to the fourth aspect above.

[0076] In a twelfth aspect, the present application provides a program product, including: a computer program, when the program product runs on a computer, enabling the computer to execute the method according to any one of the first aspect to the fourth aspect above.

[0077] In a thirteenth aspect, the present application provides a computer program, when the computer program is executed by a processor, it is used to execute the method according to any one of the first aspect to the fourth aspect above.

[0078] In a fourteenth aspect, the present application provides an autonomous mobile device, including the electronic device according to the ninth aspect above.

[0079] The driving style configuration method, model training method, data processing method, device and equipment provided by the present application include obtaining the style configuration information of an autonomous device, and based on the style configuration information, obtaining a target driving style, and configuring the driving style of the autonomous mobile device based on the target driving style. In the above process, the style configuration information can be analyzed and utilized to identify the preferences and needs of the user, so as to configure a target driving style that meets the user's expectations and improve the user experience. Description of the Drawings

[0080] The drawings here are incorporated into the specification and constitute a part of this specification, showing the embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application.

[0081] Figure 1 It is a schematic diagram of the application scenario provided by the embodiment of the present application;

[0082] Figure 2 It is a schematic flowchart of the first embodiment of the driving style processing method provided by the present application;

[0083] Figure 3 Schematic flowchart of the second embodiment of the driving style processing method provided by this application;

[0084] Figure 4 Schematic flowchart of the third embodiment of the driving style processing method provided by this application;

[0085] Figure 5 Schematic flowchart of the fourth embodiment of the driving style processing method provided by this application;

[0086] Figure 6 Schematic flowchart of the fifth embodiment of the driving style processing method provided by this application;

[0087] Figure 7 Schematic diagram of the style palette provided by the embodiment of this application;

[0088] Figure 8 Schematic flowchart of the sixth embodiment of the driving style processing method provided by this application;

[0089] Figure 9 Schematic flowchart of the first embodiment of the model training method for driving style provided by this application;

[0090] Figure 10 Schematic diagram of the principle for obtaining the vector database provided by the embodiment of this application;

[0091] Figure 11 Schematic diagram of the principle for applying the vector database provided by the embodiment of this application;

[0092] Figure 12 Schematic flowchart of the data processing method for driving style provided by the embodiment of this application;

[0093] Figure 13 Schematic flowchart of the second embodiment of the model training method for driving style provided by this application;

[0094] Figure 14 Schematic diagram of the principle for obtaining the style palette provided by the embodiment of this application;

[0095] Figure 15 Schematic diagram of the principle for applying the style palette provided by the embodiment of this application;

[0096] Figure 16 Schematic diagram of the structure of the driving style configuration device provided by the embodiment of this application;

[0097] Figure 17 Schematic diagram of the structure of the first embodiment of the model training device for driving style provided by this application;

[0098] Figure 18A data processing device for driving styles provided in an embodiment of the present application;

[0099] Figure 19 A schematic structural diagram of the second embodiment of a model training device for driving styles provided in the present application;

[0100] Figure 20 A schematic structural diagram of an electronic device provided in an embodiment of the present application. Detailed implementation manners

[0101] In the embodiments of the present application, the term "and / or" describes the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally represents an "or" relationship between the front and rear associated objects.

[0102] In the embodiments of the present application, the term "plural" means two or more, and other quantifiers are similar thereto.

[0103] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Apparently, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0104] Figure 1 A schematic diagram of an application scenario provided in an embodiment of the present application. Please refer to Figure 1 , in an autonomous mobile device, an intelligent driving system can be set up, and the intelligent driving system can perform autonomous driving or assisted driving according to the driving styles configured therein.

[0105] Optionally, the autonomous mobile device can be a flying car, an intelligent driving vehicle, a mobile robot, an intelligent ship, or other devices that can move and are configured with an intelligent driving system.

[0106] For example, in an intelligent driving vehicle, an assisted driving system can be set up. The assisted driving system can perform assisted driving according to driving style 1 and driving style 2 configured therein. One manifestation of driving style 1 can be that the following distance is greater than 3 meters (m), and one manifestation of driving style 2 can be to maintain the maximum speed limit. During the slow driving of the vehicle, the following distance greater than 3 m can be maintained according to driving style 1; when driving on a highway, the maximum speed limit can be maintained on the highway according to driving style 2.

[0107] In related technologies, the driving style of autonomous mobile devices is usually preset by manufacturers during the production stage according to requirements such as following distance or percentage of the maximum speed limit. These preset driving style schemes are often general and aim to meet the basic needs of most users. However, once these driving styles are determined, they are usually fixed. When ordinary users want to customize the driving style, they are not well-suited to ordinary users, resulting in a poor user experience.

[0108] To address the above problems, the inventors found during the research on personalized driving style schemes that each user has different acceptance levels and preferences for the dynamic characteristics of vehicle driving. For example, some users may prefer a conservative driving style, focusing on safety and smoothness; some users may prefer an aggressive driving style, pursuing speed and excitement; and some users may be more inclined to a comfortable driving experience, emphasizing comfort and smoothness. Therefore, the fixed driving styles preset by manufacturers during the production stage often cannot meet these diverse user needs. Based on this, the inventors found through multiple experiments that personalized driving style customization can be achieved by obtaining the style configuration information of autonomous mobile devices. Specifically, the style configuration information of autonomous mobile devices can be obtained, and through the style configuration information, the target driving style can be obtained, and then the driving style of the autonomous mobile device can be configured to enable dynamic adjustment according to the user's preferences. Based on this, the present application proposes a method for configuring driving styles to meet the diverse needs of ordinary users for driving styles and improve the user experience.

[0109] The following uses specific embodiments to elaborate in detail on the technical solutions of the present application and how the technical solutions of the present application solve the above technical problems. These specific embodiments below can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below with reference to the accompanying drawings.

[0110] Figure 2 It is a schematic flowchart of the first embodiment of the method for processing driving styles provided by the present application. Please refer to Figure 2 , the method includes:

[0111] S201. Obtain the style configuration information of the autonomous mobile device.

[0112] In this step, the autonomous mobile device can obtain the style configuration information related to the driving style.

[0113] The style configuration information may include any one of the following:

[0114] The driving control method selected by the user;

[0115] The image data and / or audio data collected by the autonomous mobile device;

[0116] The position information selected by the user in the style palette, where the style palette includes multiple pre-acquired color palette areas corresponding to driving styles.

[0117] Regarding the driving control method selected by the user included in the style configuration information, it can be a global configuration style different from conventional following distances and speed limits. It can subdivide driving scenarios and allow the user to adjust the decision-making tendency of the intelligent driving system in the subdivided scenarios. Among them, the subdivided scenarios can be some predefined style options.

[0118] For ease of understanding, the following will use an intelligent driving vehicle as an example of an autonomous mobile device for illustration.

[0119] For example, the driving control method selected by the user can be determined based on a congestion scenario, and the vehicle can follow the preceding vehicle 2 seconds (s) after the preceding vehicle starts.

[0120] Regarding the image data and / or audio data collected by the autonomous mobile device included in the style configuration information, it can be the image data of the driving process collected by the image acquisition device of the autonomous mobile device, and the audio data input by the user collected by the audio acquisition device. Among them, the image acquisition device can be a camera, and the audio acquisition device can be an in-vehicle microphone.

[0121] For example, the image data of the autonomous mobile device during driving can be collected by the camera of the autonomous mobile device. The voice input by the user can be collected by the microphone in the vehicle cabin.

[0122] Regarding the position information selected by the user in the style palette included in the style configuration information, it can be the position of a specific style area selected by the user in the style palette. The style palette is a visual interface that displays multiple predefined driving styles, and each style corresponds to a specific color palette area. The user determines their preferred style configuration information by selecting these areas.

[0123] For example, there are color areas for a sporty and aggressive style, a sporty and neutral style, and a smooth and efficient style in the style palette. The user can select the position of the color area for the sporty and aggressive style in the style palette to determine that their preferred style configuration information includes the sporty and aggressive style.

[0124] S202. Obtain the target driving style based on the style configuration information.

[0125] In this step, the target driving style can be obtained based on the style configuration information. The target driving style is a driving style that matches the style configuration information. Different driving styles have different performances. The driving style can be configured through the style configuration information, so that the driving performance of the autonomous vehicle conforms to the corresponding driving style. The driving performance includes, for example, at least one of the following: 1. The aggressiveness of overtaking, which can be reflected by changing the cooling time for overtaking and lane changing, the sensitivity to the traffic efficiency comparison for triggering overtaking and lane changing, the minimum distance for overtaking and lane changing in the opposite direction of the traffic direction, etc.; 2. The aggressiveness of lane changing, which can be reflected by changing the magnitude of acceleration and deceleration during overtaking, the reaction delay of the vehicle behind in the target lane, the maximum deceleration of the vehicle behind in the target lane, whether it is allowed to change lanes by crossing the solid line, etc.; 3. The aggressiveness of bypassing, which can be reflected by the minimum bypassing distance from the intersection, the maximum lateral distance of bypassing, whether to bypass by crossing the line, etc.; 4. The aggressiveness in the longitudinal direction, which can be reflected by the maximum acceleration and deceleration of the vehicle itself, the following distance of the vehicle itself, the deviation between the vehicle speed of the vehicle itself on the road with a map speed limit and the map speed limit, etc.

[0126] When the style configuration information is the driving control method selected by the user, the target driving style corresponding to the driving control method can be determined based on the driving control method and the pre-configured mapping relationship.

[0127] For example, based on the driving control method of the vehicle following the leading vehicle 2s after the leading vehicle starts, and the pre-configured mapping relationship, the target driving style corresponding to the driving control method can be determined as a conservative style.

[0128] When the style configuration information is the image data and / or audio data collected by the autonomous mobile device, the target driving style can be obtained by processing the image data and / or audio data based on the pre-acquired data processing model. Among them, the data processing model can be a trained deep learning model. For example, the deep learning model can be a convolutional neural network model.

[0129] For example, the target driving style can be obtained by processing the image data and / or audio data collected by the autonomous mobile device during driving based on the pre-acquired convolutional neural network model. The performance of the target driving style can include: automatically reducing the speed when identifying a waterlogged section to safely pass through the area.

[0130] When the style configuration information is the position information selected by the user in the style palette, the target driving style can be obtained according to the position information selected by the user.

[0131] For example, according to the position of the user selecting the color area of the sporty aggressive style in the style palette, the target driving style can be determined as the sporty aggressive style.

[0132] S203. Configure the driving style of the autonomous mobile device based on the target driving style.

[0133] In this step, the driving style of the autonomous mobile device can be configured according to the determined target driving style based on the style configuration information.

[0134] For example, the driving style of the autonomous mobile device can be configured according to the conservative style determined by the driving control method selected by the user, or the target driving style obtained from the data processing model, "when identifying a waterlogged section, automatically reduce the speed to safely pass through the area", or the sports aggressive style selected through the style palette.

[0135] In an alternative embodiment, the driving style of the autonomous mobile device can be configured in the following manner. An end-to-end driving model can be configured in the autonomous mobile device. This driving model can perform inference on a deep learning model by directly inputting data from various sensors, such as cameras, radars, and lidars, and finally output the running trajectory of the vehicle. Different driving styles can be achieved by adjusting the parameters of the driving model. The changes in these parameters can directly affect the dynamic performance and response characteristics of the vehicle, enabling the vehicle to output an expected running trajectory according to specific driving style requirements.

[0136] Optionally, to achieve a personalized and dynamic driving experience, the three schemes included in the style configuration information can be used flexibly in combination.

[0137] For example, first, based on the driving control method of following the vehicle in front after 2 seconds of starting, the corresponding target driving style 1 can be determined. The target driving style 1 can be a conservative style. Then, using the image data collected by the intelligent driving vehicle, the target driving style 2 can be obtained. The performance of the target driving style 2 can include: when identifying a waterlogged section, automatically reducing the speed to safely pass through the area. Furthermore, the driving style of the intelligent driving vehicle can be configured to include the target driving style 1 and the target driving style 2.

[0138] Another example is that the target driving style 1 can be determined first based on the driving control method. Then, the target driving style 2 can be obtained through the image data collected by the intelligent driving vehicle. Next, the target driving style 3 can be determined according to the position information selected by the user in the style palette. Finally, the driving style of the intelligent driving vehicle can be configured to include the target driving style 1, the target driving style 2, and the target driving style 3.

[0139] In the embodiments of the present application, the style configuration information of the autonomous device can be obtained to determine the target driving style. Then, according to the target driving style, the driving style of the autonomous mobile device is configured. In the above process, by obtaining the style configuration information of the autonomous device to determine and configure the target driving style, the personalization and dynamic adjustment of the driving experience of the autonomous mobile device are realized to better meet the diverse needs of users.

[0140] Based on the Figure 2 embodiments shown above, below, in combination with Figure 3 , how to configure the driving style of the autonomous mobile device through the driving control method selected by the user will be described in detail.

[0141] Figure 3 FIG. is a schematic flowchart of Embodiment 2 of the method for processing the driving style provided by the present application. Please refer to Figure 3 , the method may include:

[0142] S301. Obtain the style configuration information of the autonomous mobile device, where the style configuration information is the driving control method selected by the user.

[0143] In this step, the autonomous mobile device can determine the style configuration information according to the driving control method selected by the user.

[0144] Optionally, an interactive interface can be set in the autonomous mobile device, and the interactive interface can provide various driving control method options in different scenarios. The user can select a suitable driving control method in different driving scenarios according to their own needs and preferences.

[0145] Optionally, the driving control method can be determined based on at least one driving scenario among a congestion scenario, a starting time at an intersection, a lane-changing power selection, a turning comfort, a cut-in handling method, and a safety distance maintained from a normally moving autonomous mobile device.

[0146] Based on the congestion scenario, in the case of traffic congestion, the intelligent driving vehicle can detect the distance and speed of the vehicle ahead through radar and camera, and automatically adjust its own vehicle speed to maintain a safe distance, reducing frequent acceleration and braking operations, so as to achieve smooth driving.

[0147] For example, based on the congestion scenario, driving control method 1 can be determined. Among them, driving control method 1 can be that when it is detected that the current traffic is congested and the distance from the vehicle ahead is 5 meters, when the vehicle ahead decelerates, the intelligent driving vehicle automatically reduces its vehicle speed to maintain a safe distance; when the vehicle ahead accelerates or the lane becomes unobstructed, that is, when it is detected that the current is not a traffic congestion scenario, the system gradually returns to the preset cruise speed and vehicle distance.

[0148] Based on the starting time at intersections, such as at intersections waiting for traffic lights, intelligent driving vehicles can automatically determine the optimal starting time according to the changes of the traffic lights, so as to improve traffic efficiency and reduce waiting time.

[0149] For example, the driving control mode 2 can be determined based on the starting time at intersections. Among them, the driving control mode 2 can be to start within 1 second after the green light comes on, that is, when the traffic light is about to change from red to green, the intelligent driving vehicle prepares in advance, and at the moment when the green light comes on, automatically controls the vehicle's accelerator pedal to achieve a rapid start.

[0150] Based on the lane-changing power selection, when a lane-changing opportunity is recognized, appropriate power output can be selected according to the current traffic conditions and vehicle performance to achieve safe and efficient lane-changing.

[0151] For example, the driving control mode 3 and the driving control mode 4 can be determined based on the lane-changing power selection. Among them, the driving control mode 3 can be on urban roads where the traffic is relatively dense. The intelligent driving vehicle only changes lanes when the vehicle in the front lane is severely slow and the adjacent lane is significantly unobstructed, and selects a lower power output to complete the lane-changing smoothly; the driving control mode 4 can be on highways where the traffic flow is large but the vehicle speed is fast. The intelligent driving vehicle frequently takes advantage of lane-changing opportunities to maintain a high driving speed. When a higher speed opportunity in the adjacent lane is detected, a higher power output is selected to quickly change lanes.

[0152] Exemplarily, for the driving control mode 4, on a highway, an intelligent driving vehicle is driving in a lane with a speed limit of 120 km / h. It is detected that the vehicle speed in the front of the current lane is slow, while the left adjacent lane is unobstructed and there are no vehicles blocking. After recognizing this lane-changing opportunity, the intelligent driving vehicle can select a higher power output, quickly accelerate and complete the lane-changing to reach a speed close to 120 km / h.

[0153] Based on the steering comfort, it can be how fast an intelligent driving vehicle turns when turning. The faster the turn, the higher the traffic efficiency, but the comfort of the intelligent driving vehicle will be reduced.

[0154] For example, the driving control mode 5 and the driving control mode 6 can be determined based on the steering comfort. Among them, the driving control mode 5 can be to select a higher turning speed to improve traffic efficiency under a large turning radius and good road surface conditions. The driving control mode 6 can be to select a lower turning speed to ensure passenger comfort and vehicle stability under a small turning radius or poor road surface conditions.

[0155] Exemplarily, for the driving control mode 6, when the turning radius of the intelligent driving vehicle is less than 5 m, it turns at a vehicle speed of 10 km / h.

[0156] Based on the cut-in handling method, when an intelligent driving vehicle faces the cut-in intention of other vehicles, it can either be conservatively courteous or strive for the right of way of its own vehicle while ensuring safety. In the conservative and courteous way, it can achieve giving consumers additional sense of security by sacrificing traffic efficiency.

[0157] For example, based on the cut-in handling method, driving control methods 7 and 8 can be determined. Among them, driving control method 7 can be that when an intelligent driving vehicle detects the cut-in intention of other vehicles, it actively decelerates or maintains the current speed to give way to the cut-in vehicle. Driving control method 8 can be that when an intelligent driving vehicle detects the cut-in intention of other vehicles, after evaluating safety, it selects appropriate acceleration or maintains speed to reduce the possibility of being cut in.

[0158] Exemplarily, for driving control method 8, on a highway, an intelligent driving vehicle is driving in the middle lane. When it detects that a vehicle in the right lane starts to shift left, showing a cut-in intention. After a quick evaluation, the intelligent driving vehicle confirms that there is sufficient safety distance between it and the cut-in vehicle and the road conditions are good. So, it can choose to moderately accelerate and maintain the leading position in the current lane, thus avoiding being cut in.

[0159] Based on the safety distance maintained from a normally moving autonomous device, it can be the safety distance maintained between an intelligent driving vehicle and the vehicle in the front or adjacent lane to ensure driving safety and smoothness.

[0160] For example, if the normally moving autonomous device is a large truck, based on the safety distance maintained from the large truck, driving control methods 9 and 10 can be determined. Among them, driving control method 9 can be to normally maintain a distance from the large truck while ensuring safety. Driving control method 10 can be to seek opportunities to quickly overtake and stay away from the large truck.

[0161] Exemplarily, for driving control method 10, when an intelligent driving vehicle is driving on a highway, if there is a large truck ahead, the intelligent driving vehicle can evaluate the current traffic conditions and road conditions and look for a safe overtaking opportunity.

[0162] It should be noted that the driving scenarios for determining the driving control mode described above can be further divided in more detail. For example, the driving control mode can also include driving scenarios under different weather conditions. In clear weather, an intelligent driving vehicle may choose a more aggressive overtaking strategy, taking advantage of the good visibility and road surface conditions to quickly overtake a large truck. In rainy or foggy weather, the strategy will be adjusted to increase the safety distance from large trucks and other vehicles to ensure safe driving under slippery or low visibility conditions. In addition, on snowy or icy roads, an intelligent driving vehicle may choose a more conservative driving style, avoiding unnecessary overtaking maneuvers to reduce the risk of skidding or losing control.

[0163] For example, the style configuration information of the intelligent driving vehicle can be obtained, and the style configuration information is the driving control mode 10 selected by the user.

[0164] S302. Based on the driving control mode and the pre-configured mapping relationship, determine the target driving style corresponding to the driving control mode.

[0165] In this step, according to the driving control mode selected by the user and in combination with the pre-configured mapping relationship, the target driving style corresponding to this driving control mode can be determined. Among them, the mapping relationship includes the driving styles corresponding to different driving control modes.

[0166] For example, in the pre-configured mapping relationship, it can include the driving control mode shown in Table 1 and the driving style corresponding to each driving control mode.

[0167] Table 1

[0168] Driving control mode Driving style Driving control mode Driving style 1 Congested slow driving 6 Comfortable turning 2 Signal start 7 Yielding to cut-ins 3 Smooth lane change 8 Right-of-way contention 4 Quick lane change 8 Safe following 5 Efficient turning 10 Overtaking and keeping distance

[0169] It should be noted that the style names corresponding to the driving styles, such as slow driving in congestion, starting at signals, smooth lane change, etc., only provide a brief description of the style and do not constitute any limitation to this application.

[0170] For example, according to the driving control mode 10 selected by the user, through the mapping relationship shown in Table 1, the target driving style corresponding to the driving control mode 10 can be determined as the overtaking and keeping away style.

[0171] In an optional implementation manner, if the driving control mode selected by the user includes multiple ones, then according to the pre-configured mapping relationship, the target driving style corresponding to each driving control mode can be determined. For example, if the driving control modes selected by the user include driving control mode 1 and driving control mode 10, then through the mapping relationship shown in Table 1, the target driving style corresponding to driving control mode 1 can be determined as the slow driving in congestion style, and the target driving style corresponding to driving control mode 10 can be determined as the overtaking and keeping away style.

[0172] S303. Configure the driving style of the autonomous mobile device based on the target driving style.

[0173] In this step, the driving style of the autonomous mobile device can be configured based on the target driving style determined according to the mapping relationship.

[0174] For example, based on the overtaking-away style determined according to the mapping relationship, the driving style of the intelligent driving vehicle can be configured as the overtaking-away style.

[0175] For another example, based on the mapping relationship, the congestion-slowing style and the overtaking-away style determined can be used to configure the driving style of the intelligent driving vehicle to include the congestion-slowing style and the overtaking-away style.

[0176] In the embodiments of the present application, the driving control method selected by the user can be obtained as the style configuration information of the autonomous mobile device, and then the target driving style corresponding to the driving control method can be determined through the pre-configured mapping relationship. Finally, according to the target driving style, the corresponding driving style configuration is performed on the autonomous mobile device. In the above process, the user can select the driving control method according to personal preferences to configure the driving style of the autonomous mobile device to meet the specific needs of the user, thereby providing a personalized driving experience.

[0177] In Figure 2 Based on the above-described embodiments, below, in combination with Figure 4 , it is described in detail how to configure the driving style of the autonomous mobile device through the image data and / or audio data collected by the autonomous mobile device.

[0178] Figure 4 This is a schematic flowchart of the third embodiment of the driving style processing method provided by the present application. Please refer to Figure 4 , the method may include:

[0179] S401. Obtain the style configuration information of the autonomous mobile device, where the style configuration information is the image data and / or audio data collected by the autonomous mobile device.

[0180] In this step, the image data during the movement of the autonomous mobile device can be collected by an image acquisition device, and the audio data of the user can be collected by an audio acquisition device. The style configuration information of the autonomous mobile device is determined according to the image data and / or audio data.

[0181] Obtain the image data during the movement of the autonomous mobile device through the image acquisition device. For example, the image data 1 in front of the intelligent driving vehicle during driving can be collected through a camera. The image data 1 may include an image of a water accumulation section in front of the intelligent driving vehicle, and the style configuration information of the intelligent driving vehicle is determined according to the image data 1.

[0182] Collect the audio data of the user through an audio acquisition device. For example, the output speech of the user can be collected through a microphone to obtain Audio Data 1. Audio Data 1 can be "When encountering a waterlogged section, the intelligent driving vehicle decelerates and drives through the waterlogged section to prevent skidding", and determine the style configuration information of the intelligent driving vehicle according to Audio Data 1.

[0183] For example, the style configuration information of the intelligent driving vehicle can be obtained. The style configuration information is Image Data 1 collected by the intelligent driving vehicle.

[0184] S402. Process the image data and / or audio data based on a pre-obtained data processing model to obtain a style description vector.

[0185] In this step, the image data and / or audio data can be input into the pre-obtained data processing model for processing to obtain a style description vector.

[0186] Optionally, when the style configuration information is image data, the data processing model is a visual processing model; when the style configuration information is audio data, the data processing model is a language processing model; when the style configuration information includes image data and audio data, the data processing model is a visual language processing model.

[0187] In a specific implementation, the style description vector can be represented by a high-dimensional feature space. In this high-dimensional feature space, each dimension can correspond to a specific driving style feature or parameter. By quantifying and combining these features or parameters in the high-dimensional space, a comprehensive style description vector can be formed to accurately describe the driving style of the autonomous mobile device.

[0188] For example, the style description vector can be a 64-dimensional vector used to describe the relevant driving style. Exemplarily, the 64-dimensional vector can be [1, 2,..., 64], and each numerical value represents the quantization index of the corresponding feature or parameter.

[0189] For example, the image data 1 can be processed through a visual processing model to obtain Style Description Vector 1.

[0190] S403. Select the target driving style corresponding to the style description vector from a preset driving style library.

[0191] In this step, the target driving style corresponding to the style description vector can be selected from the preset driving style library according to the obtained style description vector.

[0192] In a specific embodiment, the driving style corresponding to the target description vector with the highest similarity to the style description vector can be selected from the driving style library as the target driving style. Optionally, the style description vectors included in the driving style library and the driving styles corresponding to the style description vectors can be obtained through user calibration or through inference of driving data by a pre-trained style acquisition model.

[0193] For example, by analyzing the style description vector, driving behaviors can be classified into specific driving styles, such as adventurous style, cautious style, steady style, etc.

[0194] For example, in the driving style library, there can be style description vectors as shown in Table 2 and the driving style corresponding to each style description vector.

[0195] Table 2

[0196] Style description vector Driving style A Adventurous style B Cautious style C Steady style D Sporty style E Balanced style

[0197] For example, the driving style corresponding to the target description vector with the highest similarity to style description vector 1 can be selected from the driving style library as the target driving style. Among them, the target description vector can be style description vector A, and the driving style corresponding to style description vector A can be an adventurous style, that is, the target driving style is an adventurous style.

[0198] S404. Configure the driving style of the autonomous mobile device based on the target driving style.

[0199] For example, based on the adventurous style determined by style description vector 1, the driving style of the intelligent driving vehicle can be configured as an adventurous style.

[0200] In the embodiments of the present application, the image data and / or audio data collected by the autonomous mobile device can be obtained as the style configuration information of the autonomous mobile device. By using a pre-obtained data processing model, the image data and / or audio data are processed to obtain a style description vector. Furthermore, the target driving style corresponding to the style description vector can be selected from a preset driving style library to configure the driving style of the autonomous mobile device. In the above process, through an advanced data processing model and a driving style library, by using the image data and / or audio data collected by the autonomous mobile device, intelligent driving style configuration can be realized, thereby improving the user experience.

[0201] Figure 5 This is a schematic flowchart of the fourth embodiment of the method for processing the driving style provided by the present application. Please refer to Figure 5 , on the basis of the embodiment shown in Figure 4 , this method may include:

[0202] S501. Receive the data processing model sent by the server and deploy the data processing model.

[0203] In this step, the autonomous mobile device can receive the data processing model trained by the server and deploy it in the local system of the autonomous mobile device. Among them, the data processing model can be a visual processing model, a language processing model, or a visual language processing model.

[0204] For example, an intelligent driving vehicle can receive and deploy the visual processing model sent by the server.

[0205] S502. Obtain the style configuration information of the autonomous mobile device, where the style configuration information is the image data and / or audio data collected by the autonomous mobile device.

[0206] For example, an intelligent driving vehicle can obtain the style configuration information, where the style configuration information is the image data 1 collected by the intelligent driving vehicle.

[0207] S503. Process the image data and / or audio data based on the pre-obtained data processing model to obtain a style description vector.

[0208] Optionally, the style description vector includes a scene description vector and / or a preference instruction vector; among them, the scene description vector is used to indicate the driving scene where the autonomous mobile device is located, and the preference instruction vector is used to indicate the user's preference for the driving style.

[0209] For example, if the style description vector is a scene description vector, the image data 1 can be processed based on the deployed visual processing model to obtain the scene description vector 1.

[0210] For another example, if the style description vector is a preference instruction vector, the image data 1 can be processed based on the deployed visual processing model to obtain the preference instruction vector 1.

[0211] For yet another example, if the style description vector includes a scene description vector and a preference instruction vector, the image data 1 can be processed based on the deployed visual processing model to obtain the scene description vector 2 and the preference instruction vector 2.

[0212] S504. Select the driving style corresponding to the target description vector with the highest similarity to the style description vector from the driving style library as the target driving style.

[0213] In this step, the driving style corresponding to the target description vector with the highest similarity to the style description vector obtained according to the data processing model can be selected from the driving style library as the target driving style. In specific implementation, the driving style corresponding to the target description vector with the highest similarity to the style description vector can be selected from the driving style library as the target driving style.

[0214] Optionally, any one of the similarity measurement methods such as cosine similarity, Euclidean distance, and Manhattan distance can be used to select the target description vector with the highest similarity to the style description vector from the driving style library.

[0215] In an alternative embodiment, if the style description vector is a scene description vector or a preference instruction vector, the style description vector may include all-dimensional environmental features and driving style features in the scene description vector and the preference instruction vector. Through a single scene description vector or preference instruction vector, the target driving style can be determined.

[0216] For example, through the Euclidean distance measurement method, the scene description vector A with the highest similarity to the scene description vector 1 can be selected from the driving style library as the target description vector. Furthermore, the driving style corresponding to the target description vector can be obtained as the target driving style. Exemplarily, the scene description vector A may be the same as the style description vector A, and the target driving style may be an adventurous style.

[0217] It should be noted that if the style description vector is a scene description vector and at least two different driving scene description vectors are included in the description vector selected from the driving style library with the highest similarity to the scene description vector, the driving styles corresponding to the at least two different driving scene description vectors can be recommended to the user through the interaction interface, and the target driving style can be determined according to the user's selection. Optionally, in addition to the user selecting the target driving style on the interaction interface, the driving style that is the latest / most frequently used in the driving scene description vector can also be determined as the target driving style to better conform to the user's habits.

[0218] Another example is that through the cosine similarity measurement method, the preference instruction vector A with the highest similarity to the preference instruction vector 1 can be selected from the driving style library as the target description vector. Furthermore, the driving style corresponding to the target description vector can be obtained as the target driving style. Exemplarily, the preference instruction vector A may be the same as the style description vector A, and the target driving style may be an adventurous style.

[0219] In another alternative embodiment, the style description vector includes a scene description vector and a preference instruction vector, and the characteristics of the scene description vector and the preference instruction vector can be comprehensively considered to determine the target driving style. In this case, the style description vector is not only a reflection of a single scene or preference instruction, but a fusion of the two, which can better adapt to complex driving environments and personalized needs.

[0220] For example, through the Euclidean distance metric method, the scene description vector B with the highest similarity to the scene description vector 2 can be selected from the driving style library, and the preference instruction vector B with the highest similarity to the preference instruction vector 2 can be selected from the driving style library. According to the scene description vector B, driving style 1 can be determined. According to the preference instruction vector 2, driving style 2 can be determined. Among them, driving style 1 can be the most suitable driving method based on the current environmental conditions, such as choosing a cautious style in congested traffic; driving style 2 can reflect the driver's personal preference, such as being more inclined to a sports style.

[0221] Furthermore, through a style fusion strategy, driving style 1 and driving style 2 can be combined to form a comprehensive target driving style. The fusion process can adjust the weights according to specific application scenarios and user requirements. For example, if the current driving environment has high requirements for safety, then a greater weight can be given to driving style 1; if the driver's preference is more important, then the weight of driving style 2 can be increased.

[0222] Finally, the obtained target driving style can be a balanced style, which can ensure safety while satisfying a certain degree of driving pleasure.

[0223] S505. Configure the driving style of the autonomous mobile device based on the target driving style.

[0224] For example, the driving style of the intelligent driving vehicle can be configured based on the obtained adventurous style or balanced style.

[0225] In the embodiments of the present application, the autonomous mobile device can receive the data processing model sent by the server and deploy it. The obtained style configuration information is processed through the data processing model to generate a style description vector, and the driving style corresponding to the target description vector with the highest similarity to the style description vector is selected from the driving style library as the target driving style. According to the target driving style, the driving style of the autonomous mobile device is configured. In the above process, through the deployed data processing model, the image data and / or audio data included in the style configuration information can be processed to identify the user's preference and / or the current driving environment, so as to select a suitable driving style and provide a personalized driving experience.

[0226] In Figure 2 Based on the shown embodiments, below, in combination with Figure 6 , it is detailed how to configure the driving style of the autonomous mobile device through the position information selected by the user in the style palette.

[0227] Figure 6 This is a schematic flowchart of the fifth embodiment of the driving style processing method provided by the present application. Please refer to Figure 6 , the method may include:

[0228] S601. Obtain the style configuration information of the autonomous mobile device, where the style configuration information is the position information selected by the user in the style palette.

[0229] In this step, the autonomous mobile device can obtain the position information selected by the user in the style palette as the style configuration information.

[0230] Figure 7 This is a schematic diagram of the style palette provided by the embodiment of the present application. Please refer to Figure 7 , and in the style palette, there can be multiple driving styles such as a sporty aggressive style, a sporty neutral style, and a smooth and efficient style.

[0231] S602. If the position information is within the color palette area corresponding to any driving style in the style palette, then determine the driving style as the target driving style.

[0232] In this step, it can be determined that the driving style is the target driving style according to the position information falling within the area corresponding to any driving style in the style palette.

[0233] For example, it can be determined that the sporty aggressive style is the target driving style according to the position information falling within the area corresponding to the sporty aggressive style in the style palette.

[0234] S603. If the position information is not within the color palette area corresponding to any driving style, then determine at least one driving style within the preset range of the position information from the style palette.

[0235] In this step, the autonomous mobile device can search for at least one driving style within the preset range of the position information in the style palette according to the position information not being within the color palette area corresponding to any driving style.

[0236] Optionally, the preset range can be set according to user requirements. For example, the horizontal direction of the style palette is the abscissa, the vertical direction is the ordinate, and the numerical value is represented by the number of pixels. If the preset range is 5 pixels, then a circular search range centered on the position information point and expanding 5 pixels on the style palette can be constructed. Within this circular range with a radius of 5 pixels, the driving style closest to the current position information can be searched for.

[0237] Optionally, the preset range can also be a rectangular area, a square area, or an area of other shapes.

[0238] It should be noted that there can be multiple preset ranges. For example, if the driving style closest to the current position information is not found within a circular preset range of 5 pixels, the preset range can be expanded, such as to a circular range of 10 pixels, and the driving style closest to the current position information is searched for within this range.

[0239] S604. Determine a target driving style according to at least one driving style.

[0240] In this step, the target driving style can be determined according to at least one driving style found within the preset range of the position information. In a specific implementation, the target driving style can be determined according to the distance between the color palette area corresponding to each driving style in at least one driving style and the position information.

[0241] For example, the smooth and efficient style, the sporty neutral style, and the neutral style all fall within a circular preset range of 5 pixels. The center coordinates of the smooth and efficient style are coordinate A(14, 4), the center coordinates of the sporty neutral style are coordinate B(12, 6), the center coordinates of the neutral style are coordinate C(14, 9), and the center coordinates of the user's current position information are coordinate O(15, 8). Through calculation, it is found that the distance between the neutral style and the user's position information is the smallest. Therefore, the neutral style can be selected as the target driving style.

[0242] In an alternative implementation, each position on the style color palette (this position can be the color palette area corresponding to each driving style) can correspond to a parameter group (or parameter vector) for driving style adjustment. The user can determine the parameter group (or parameter vector) corresponding to the driving style by selecting the position information on the style color palette for configuring the corresponding driving style (target driving style). Among them, the corresponding parameter group (or parameter vector) can be determined through the mapping relationship between the preselected stored position and the parameter group; or it can be obtained by adjusting the parameters of the parameter group (or parameter vector) corresponding to the nearest driving style according to the distance between the position information selected by the user and the nearest driving style, where the smaller the distance, the smaller the adjustment amplitude, and the larger the distance, the larger the adjustment amplitude. The specific adjustment amplitude can be determined according to experience or a preset model.

[0243] S605. Configure the driving style of the autonomous mobile device based on the target driving style.

[0244] For example, the driving style of the intelligent driving vehicle can be configured based on the sporty aggressive style or the neutral style obtained from the style color palette.

[0245] In an embodiment of the present application, the position information selected by the user in the style palette can be obtained as the style configuration information. If the position information is within the color palette area corresponding to the driving style, the driving style is determined as the target driving style; if the position information is not within any color palette area, at least one driving style is selected within a preset range, and the target driving style is determined accordingly, and finally the driving style of the autonomous mobile device is configured. In the above process, the user can customize their driving style by selecting the position information on the color palette to better adapt to their personal preferences, so that the autonomous mobile device provides a driving experience that better meets the user's needs.

[0246] Figure 8 It is a schematic flowchart of Embodiment 6 of the method for processing the driving style provided by the present application. Please refer to Figure 8 , on the basis of the embodiment shown in Figure 6 , the method may include:

[0247] S801. Receive the style palette data sent by the server.

[0248] In this step, the style palette can be generated by the server and sent to the autonomous mobile device.

[0249] In an alternative embodiment, it can also be that the server sends the style palette data (such as a high-dimensional style vector) to the autonomous mobile device, and the autonomous mobile device generates the style palette.

[0250] For example, the server can send the high-dimensional style vectors corresponding to various driving styles such as the aggressive style, the sporty neutral style, and the smooth and efficient style to the intelligent driving vehicle, and the intelligent driving vehicle generates the Figure 7 style palette shown.

[0251] S802. Display the style palette on the user interface of the autonomous mobile device.

[0252] In this step, after receiving the style palette sent by the server, the autonomous mobile device can display it on the user interface of the autonomous mobile device, so as to generate style configuration information according to the position information selected by the user in the style palette.

[0253] S803. Obtain the style configuration information of the autonomous mobile device, where the style configuration information is the position information selected by the user in the style palette.

[0254] S804. If the position information is within the color palette area corresponding to any driving style in the style palette, determine the driving style as the target driving style.

[0255] S805. If the position information is not within the color palette area corresponding to any driving style, determine at least one driving style within the preset range of the position information from the style color palette.

[0256] It should be noted that for the implementation of step S803, reference can be made to step S601; for the implementation of step S804, reference can be made to step S602; for the implementation of step S805, reference can be made to step S603, which will not be repeated here.

[0257] S806. Determine the target driving style according to the distance between the color palette area corresponding to each driving style in the at least one driving style and the position information.

[0258] Optionally, the target driving style can be determined in the following two ways.

[0259] Method 1: Distance minimization. In the specific implementation, the autonomous mobile device can calculate the distance between the position information selected by the user and the center of the color palette area corresponding to each driving style. The style with the minimum distance from the user's position information is determined as the target driving style.

[0260] Method 2: Driving style fusion. In the specific implementation, the autonomous mobile device can determine the candidate driving styles within the preset range according to the position information selected by the user. According to the distance between the position information selected by the user and the center of the color palette area corresponding to each driving style, a weight coefficient is assigned to each candidate driving style based on these distances, and different driving styles are fused according to the weight coefficients to generate a comprehensive driving style. Generally, the closer the driving style, the higher the weight coefficient it can obtain.

[0261] For example, the central coordinates of the user's current position information are coordinate O(15, 8), and the distances from the smooth and efficient style, the sporty neutral style, and the neutral style are 4, 3.5, and 2.5 respectively. Among them, the central coordinates of the smooth and efficient style are coordinate A(14, 4), the central coordinates of the sporty neutral style are coordinate B(12, 6), and the central coordinates of the neutral style are coordinate C(14, 9). By calculating the reciprocals of these distances and normalizing them, the weight coefficient of the smooth and efficient style can be obtained as 0.25, the weight coefficient of the sporty neutral style is 0.3, and the weight coefficient of the neutral style is 0.45. Therefore, the autonomous mobile device can fuse different driving styles according to these weight coefficients to generate a comprehensive driving style. Exemplarily, this comprehensive driving style may be a new style, close to the neutral style, and can be named "smooth neutral style".

[0262] Optionally, in addition to the above two ways to determine the target driving style, the target driving style can also be determined by means of a parameter group (or parameter vector).

[0263] For example, a style palette contains multiple color regions, and each color region represents a specific driving style. Each color region is pre-mapped to a specific parameter group. For example, the red region may be mapped to a parameter group that sets the vehicle to accelerate faster and steer more sensitively, while the parameter group for the blue region may set the vehicle to accelerate more smoothly and steer more stably. Users can select their desired driving style by choosing a certain color region on the style palette. According to the color region selected by the user, the pre-stored mapping relationship can be looked up in the style palette to directly obtain the corresponding parameter group to configure the driving style of the intelligent driving vehicle.

[0264] Optionally, if the position selected by the user is between two or more color regions, the parameter group can be dynamically adjusted according to the distance between the selected position and the nearest preset style. For example, if the position information selected by the user is between the red region and the blue region, the parameters between the parameter groups of the red region and the blue region can be adjusted according to the distance ratio between the position information selected by the user and the red region and the blue region, generating a new parameter group that is between the red and blue styles in characteristics, thus providing the user with a more personalized driving style option.

[0265] S807. Configure the driving style of the autonomous mobile device based on the target driving style.

[0266] For example, the driving style of the intelligent driving vehicle can be configured based on the sporty aggressive style or the smooth neutral style obtained from the style palette.

[0267] In the embodiments of the present application, the autonomous mobile device can receive and display the style palette sent by the server, obtain the position information selected by the user in the palette as the style configuration information; if the position information is in the color region of a certain driving style, determine that driving style as the target driving style; if the position information is not in any color region, select at least one driving style within a preset range and determine the target driving style according to its distance from the position information, and finally configure the driving style of the autonomous mobile device based on the target driving style. In the above process, by allowing the user to select position information in the style palette, the autonomous mobile device can adjust the driving style according to the user's preferences, thus providing a more personalized user experience.

[0268] In the above Figure 3 or Figure 4 Based on the embodiments shown above, below, in combination with Figure 9 , the training process of the data processing model will be described in detail. The training process of the data processing model can be performed on the server.

[0269] Figure 9The flowchart of the first embodiment of the model training method for driving style provided by this application. Please refer to Figure 9 , the method may include:

[0270] S901. Obtain a plurality of training samples.

[0271] In this step, a plurality of training samples may be obtained. Each training sample includes data to be processed and a style description vector corresponding to the data to be processed. The data to be processed includes image data and / or audio data collected by the autonomous mobile device.

[0272] Optionally, the server may obtain image data and / or audio data from the autonomous mobile device through various methods such as mobile application upload, Application Programming Interface (API), cloud storage service, Internet of Things platform, data synchronization mechanism, or third-party service integration.

[0273] For the style description vector of the data to be processed, the user may customize the style description vector when uploading the image and / or audio data, or select the style description vector from a predefined list.

[0274] S902. Train the deep learning model according to the plurality of training samples to obtain a data processing model.

[0275] In this step, the deep learning model may be trained according to the obtained plurality of training samples to obtain a data processing model. The data processing model is used to analyze the input data to obtain the style description vector of the driving style.

[0276] Optionally, the data processing model is any one of a visual processing model, a language processing model, or a visual language processing model;

[0277] Among them, the visual processing model is used to analyze the image data to obtain the style description vector of the driving style;

[0278] The language processing model is used to analyze the audio data to obtain the style description vector of the driving style;

[0279] The visual language processing model is used to analyze the image data and the audio data to obtain the style description vector of the driving style.

[0280] If the data processing model is a visual processing model, each training sample for training the deep model may include image data and the style description vector corresponding to the image data;

[0281] Optionally, the deep learning model corresponding to the visual processing model can also be a visual model based on the Transformer architecture, a Graph Neural Network (GNN) model, or a self-supervised learning model, in addition to the convolutional neural network model.

[0282] If the data processing model is a language processing model, each training sample for training the deep model can include audio data and a style description vector corresponding to the audio data.

[0283] Optionally, the deep learning model corresponding to the language processing model can also be a language model based on a Recurrent Neural Network (RNN), a Long Short-Term Memory (LSTM), a Gated Recurrent Unit (GRU), and the Transformer architecture, in addition to the convolutional neural network model.

[0284] If the data processing model is a vision-language processing model, each training sample for training the deep model can include image data, audio data, and style description vectors corresponding to the image data and the audio data. Among them, the audio data is used to describe the image data.

[0285] Optionally, the deep learning model corresponding to the vision-language processing model can also be a multimodal model based on the Transformer architecture, a graph neural network model, a self-supervised learning model, and other models capable of processing multimodal data, in addition to the convolutional neural network model.

[0286] It should be noted that the selection and application of the above deep learning models are determined according to different data processing requirements. The various models illustrated are only examples, and the types of deep learning models in this application are not limited.

[0287] In an optional implementation manner, the style description vector includes a scene description vector and / or a preference instruction vector;

[0288] Among them, the scene description vector is used to indicate the driving scene where the autonomous mobile device is located, and the preference instruction vector is used to indicate the user's preference for the driving style.

[0289] In the embodiments of the present application, multiple training samples can be obtained, and the deep learning model can be trained according to the training samples to obtain a data processing model. This model can analyze the image data and / or audio data collected by the autonomous mobile device, generate a style description vector, and configure the driving style of the autonomous mobile device accordingly to meet the personalized needs and preferences of the user.

[0290] In a possible design, the driving style of the autonomous mobile device can also be configured by means of the cooperation of multiple data processing models.

[0291] Figure 10 The following is a schematic diagram of the principle for obtaining the vector database provided by the embodiments of the present application. Please refer to Figure 10 , the visual language large model in the server can be used to perform recognition processing on the image data collected by the image acquisition device and the audio data collected by the audio acquisition device, obtain the scene description vector and the preference instruction vector, and store the obtained scene description vector and preference instruction vector in the vector database. Among them, the visual language large model can be the visual language processing model in the above embodiments.

[0292] Figure 11 The following is a schematic diagram of the principle for applying the vector database provided by the embodiments of the present application. Please refer to Figure 11 , the autonomous mobile device can use the visual large model to perform recognition processing on the image data collected by the image acquisition device, obtain the scene description vector, search and match the database according to the scene description vector, retrieve the preference instruction vector from the vector database, and configure the driving style of the autonomous mobile device according to the scene description vector and the preference instruction vector. Among them, the visual large model can be the visual processing model in the above embodiments, and configuring the driving style of the autonomous mobile device can include configuring the driving style of the automatic driving system or the assisted driving system.

[0293] It should be noted that Figure 10 the visual language large model in Figure 11 and the visual large model in

[0294] can also be the same model, that is, it can be a visual processing model, a language processing model, or a visual language processing model. It is used to obtain the historical configuration information of the autonomous mobile device, obtain its corresponding style description vector, and store it. During the subsequent use of the model, the corresponding style description vector can be directly called according to the obtained image data and / or audio data to configure the driving style of the autonomous mobile device, realizing user-defined style adjustment.

[0295] Based on the above Figure 6 or Figure 8 shown embodiments, below, in combination with Figure 12, the process of obtaining the style palette is described in detail. The process of obtaining the style palette can be performed on the server.

[0296] Figure 12 It is a schematic flowchart of the data processing method for driving style provided by the embodiments of the present application. Please refer to Figure 12 , the method may include:

[0297] S1201. Obtain the driving data of multiple users.

[0298] In this step, the driving data of multiple users can be obtained. The driving data of each user may include at least one of driving data, environmental data, driving habits, and status data. Among them, the driving data may include data such as vehicle speed and acceleration, the environmental data may include data such as road conditions and weather conditions, the driving habits may include data such as turning habits and speeding behaviors, and the status data may include data such as engine speed and temperature.

[0299] Optionally, the sources of the driving data of multiple users can be diversified to provide a richer and more personalized driving style.

[0300] For example, the system service provider of intelligent driving vehicles can upload the professional data of different driving styles (such as the driving data of star drivers or professional racers) collected to the server. In addition, users can also choose to upload their own driving data to the server.

[0301] S1202. For each user, obtain the driving style vector corresponding to the user according to the driving data of the user.

[0302] In this step, the driving style vector corresponding to each user can be obtained according to the driving data of each user.

[0303] Optionally, the driving data of the user can be analyzed according to a preset style parsing model to obtain the driving style vector corresponding to the user. Among them, the driving style vector can be a set of high-dimensional numerical values, which can reflect various behavioral characteristics of the user during driving, such as acceleration habits, braking frequency, turning methods, speed preferences, etc.

[0304] Optionally, the style parsing model can be a pre-trained deep learning model or an artificial intelligence (AI) model.

[0305] S1203. Perform dimensionality reduction clustering analysis on the driving style vectors corresponding to multiple users to obtain the style palette.

[0306] In this step, dimensionality reduction clustering analysis can be performed on the driving style vectors in the vector space to obtain K clusters, where K≥2. The driving style corresponding to each cluster can be used to construct a style palette.

[0307] Optionally, dimensionality reduction techniques such as Principal Component Analysis (PCA) or t-Distributed Stochastic Neighbor Embedding (t-SNE) can be used to transform the high-dimensional driving style vectors into a lower-dimensional space. For example, the 64-dimensional driving style vectors can be transformed into 2 dimensions for visualization and further analysis.

[0308] Furthermore, clustering analysis can be performed on the dimensionality-reduced driving style vectors to identify different driving style categories. The clustering algorithm can include any one of the K-means clustering algorithm, hierarchical clustering, and DBSCAN clustering algorithm. After the clustering analysis is completed, each cluster can represent a typical driving style. By visualizing the clustering results as a "style palette", each color or marker represents a specific driving style. For example, the style palette is a two-dimensional canvas that shows different styles in the high-dimensional space and the style differences between users. Styles with a high degree of similarity are closer, while styles with a large difference are farther apart. In other embodiments, the style palette can also be a three-dimensional canvas.

[0309] In the embodiments of this application, driving data of multiple users is obtained, and corresponding driving style vectors are generated for each user based on this data. Dimensionality reduction and clustering analysis are performed on these driving style vectors to construct a style palette. Users can customize their driving styles by selecting position information in the style palette. This personalized setting can better meet the preferences and needs of users and improve the user experience.

[0310] In Figure 12 Based on the embodiments shown below, in combination with Figure 13 The training process of the style parsing model will be described in detail. The training process of the style parsing model can be performed on the server.

[0311] Figure 13 It is a schematic flowchart of the second embodiment of the model training method for driving styles provided by this application. Please refer to Figure 13 This method may include:

[0312] S1301. Obtain a plurality of training samples.

[0313] In this step, the server can obtain multiple training samples, each of which includes a user's driving data and a driving style vector corresponding to the driving data.

[0314] The server can obtain the user's driving data through various methods such as mobile application upload, API interface, cloud storage service, Internet of Things platform, data synchronization mechanism, or third-party service integration. For example, the server can interact with other applications or devices (such as cameras, sensors) through an open API interface to obtain the user's driving data.

[0315] The driving style vector can be customized by the user according to the driving data, which includes at least one of driving data, environmental data, driving habits, and status data. By analyzing these data, the user can identify their own driving patterns and habits, and thus customize the driving style vector corresponding to the driving data.

[0316] S1302. Train the deep learning model according to multiple training samples to obtain a style parsing model.

[0317] In this step, the deep learning model can be trained according to the obtained multiple training samples to obtain a style parsing model for analyzing the user's driving data and obtaining the driving style vector.

[0318] Optionally, the deep learning model can be a convolutional neural network model.

[0319] In the embodiments of the present application, the deep learning model can be trained according to the obtained multiple training samples to obtain a style parsing model. Through the style parsing model, the user's driving data can be analyzed to identify the user's driving style vectors. These driving style vectors not only help the user understand their own driving habits and patterns, but also provide important basic data support for the establishment of the style palette. This process can help the user achieve personalized driving style customization and optimization.

[0320] Figure 14 This is a schematic diagram of the principle for obtaining the style palette provided by the embodiments of the present application. Please refer to Figure 14 , the deep learning model can be trained with the driving data of driver 1, the driving data of driver 2, ……, and the driving data of driver N to obtain an autonomous driving model or an assisted driving model, the driving style vector of driver 1, the driving style vector of driver 2, ……, and the driving style vector of driver N. Among them, the driving style vector of each driver can be represented by a high-dimensional vector.

[0321] By performing dimensionality reduction clustering analysis on the driving style vectors of N drivers in the vector space, a style palette can be obtained.

[0322] Among them, each driver can correspond to a user, and the automatic driving model or the assisted driving model can be the style parsing model in the above embodiments.

[0323] Figure 15 It is a schematic diagram of the principle of applying the style color palette provided by the embodiments of the present application. Please refer to Figure 15 , the user can make selections and fine-tuning in the style color palette to determine the configured driving style. According to the driving style configured by the user, configure the automatic driving system or the assisted driving system of the autonomous mobile device to obtain a style preference system configured by the user. The automatic driving system or the assisted driving system can control the autonomous mobile device according to the driving style configured by the user to match the driving style preference of the user.

[0324] Figure 16 It is a schematic diagram of the structure of the driving style configuration device provided by the embodiments of the present application. Please refer to Figure 16 , the driving style configuration device 10 includes:

[0325] The first acquisition module 11 is used to acquire the style configuration information of the autonomous mobile device;

[0326] The first processing module 12 is used to obtain the target driving style based on the style configuration information;

[0327] The configuration module 13 is used to configure the driving style of the autonomous mobile device based on the target driving style.

[0328] The driving style configuration device provided by the embodiments of the present application can execute the technical solutions shown in the above method embodiments, and its implementation principles and beneficial effects are similar, so details are not described herein again.

[0329] In a possible implementation manner, the style configuration information includes any one of the following:

[0330] The driving control method selected by the user;

[0331] The image data and / or audio data collected by the autonomous mobile device;

[0332] The position information selected by the user in the style color palette, and the style color palette includes a plurality of color adjustment areas corresponding to pre-acquired driving styles.

[0333] In a possible implementation manner, if the style configuration information is the driving control method selected by the user, the first processing module 12 is specifically used for:

[0334] Based on the driving control mode and a pre-configured mapping relationship, determine the target driving style corresponding to the driving control mode; wherein, the mapping relationship includes driving styles corresponding to different driving control modes.

[0335] In a possible implementation manner, the driving control mode is determined based on at least one driving scenario among a congestion scenario, a starting time at an intersection, a lane change power selection, a turning comfort, a cut-in handling method, and a safety distance maintained from an autonomous mobile device traveling normally.

[0336] In a possible implementation manner, if the style configuration information is image data and / or audio data collected by the autonomous mobile device, the first processing module 12 is specifically configured to:

[0337] Process the image data and / or audio data based on a pre-acquired data processing model to obtain a style description vector;

[0338] Select the target driving style corresponding to the style description vector from a preset driving style library; wherein, the driving style library stores style description vectors corresponding to different driving styles.

[0339] In a possible implementation manner, when the style configuration information is image data, the data processing model is a visual processing model;

[0340] When the style configuration information is audio data, the data processing model is a language processing model;

[0341] When the style configuration information includes image data and audio data, the data processing model is a visual language processing model.

[0342] In a possible implementation manner, the style description vector includes a scene description vector and / or a preference instruction vector;

[0343] Wherein, the scene description vector is used to indicate the driving scene where the autonomous mobile device is located, and the preference instruction vector is used to indicate the user's preference for the driving style.

[0344] In a possible implementation manner, the first processing module 12 is specifically configured to:

[0345] Select the driving style corresponding to the target description vector with the highest similarity to the style description vector from the driving style library as the target driving style.

[0346] In a possible implementation manner, the first processing module 12 is further configured to:

[0347] Receive a data processing model sent by the server and deploy the data processing model.

[0348] In a possible implementation manner, if the style configuration information is the position information selected by the user in the style palette, the first processing module 12 is specifically configured to:

[0349] If the position information is within the color palette area corresponding to any driving style in the style palette, determine the driving style as the target driving style;

[0350] If the position information is not within the color palette area corresponding to any driving style, determine at least one driving style within the preset range of the position information from the style palette;

[0351] Determine the target driving style according to the at least one driving style.

[0352] In a possible implementation manner, the first processing module 12 is specifically configured to:

[0353] Determine the target driving style according to the distance between the color palette area corresponding to each driving style in the at least one driving style and the position information.

[0354] In a possible implementation manner, the first processing module 12 is further configured to:

[0355] Receive the style palette data sent by the server;

[0356] Display the style palette on the user interface of the autonomous mobile device.

[0357] The driving style configuration device provided in the embodiments of the present application can execute the technical solutions shown in the above method embodiments, and the implementation principles and beneficial effects are similar, which will not be elaborated here.

[0358] Figure 17 This is a schematic structural diagram of the first embodiment of the model training device for driving styles provided by the present application. Please refer to Figure 17 , for the model training device 20 for driving styles, including:

[0359] A second acquisition module 21, configured to acquire a plurality of training samples, each training sample including data to be processed and a style description vector corresponding to the data to be processed, where the data to be processed includes image data and / or audio data collected by the autonomous mobile device;

[0360] A second processing module 22, configured to train a deep learning model according to the plurality of training samples to obtain a data processing model, where the data processing model is used to analyze the input data to obtain a style description vector of the driving style.

[0361] In a possible implementation, the data processing model is any one of a visual processing model, a language processing model, or a visual language processing model;

[0362] wherein, the visual processing model is used to analyze image data to obtain a style description vector of the driving style;

[0363] the language processing model is used to analyze audio data to obtain a style description vector of the driving style;

[0364] the visual language processing model is used to analyze image data and audio data to obtain a style description vector of the driving style.

[0365] In a possible implementation, the style description vector includes a scene description vector and / or a preference instruction vector;

[0366] wherein, the scene description vector is used to indicate the driving scene where the autonomous mobile device is located, and the preference instruction vector is used to indicate the user's preference for the driving style.

[0367] The model training device for the driving style provided by the embodiments of the present application can execute the technical solutions shown in the above method embodiments, and its implementation principle and beneficial effects are similar, and will not be elaborated here.

[0368] Figure 18 This is a data processing device for the driving style provided by the embodiments of the present application. Please refer to Figure 18 , the data processing device 30 for the driving style includes:

[0369] A third acquisition module 31, configured to acquire driving data of multiple users;

[0370] A third processing module 32, configured to, for each user, obtain a driving style vector corresponding to the user according to the driving data of the user;

[0371] An analysis module 33, configured to perform dimensionality reduction clustering analysis on the driving style vectors corresponding to the multiple users to obtain a style palette.

[0372] In a possible implementation, the third processing module 32 is specifically configured to:

[0373] Analyze the driving data of the user according to a preset style parsing model to obtain a driving style vector corresponding to the user.

[0374] The data processing device for the driving style provided by the embodiments of the present application can execute the technical solutions shown in the above method embodiments, and its implementation principle and beneficial effects are similar, and will not be elaborated here.

[0375] Figure 19Schematic diagram of the second embodiment of the model training device for driving style provided by this application. Please refer to Figure 19 , the model training device 40 for driving style includes:

[0376] A fourth acquisition module 41, configured to acquire a plurality of training samples, each training sample including driving data of a user and a driving style vector corresponding to the driving data;

[0377] A fourth processing module 42, configured to train a deep learning model according to the plurality of training samples to obtain a style parsing model.

[0378] The model training device for driving style provided by the embodiment of this application can execute the technical solutions shown in the above method embodiments, and its implementation principles and beneficial effects are similar, and will not be elaborated here.

[0379] Figure 20 Schematic diagram of the electronic device provided by the embodiment of this application. Please refer to Figure 20 , the electronic device 50 may include a processor 51, a memory 52, and a communication interface 54. Exemplarily, the processor 51, the memory 52, and the communication interface 54 are interconnected through a bus 53.

[0380] The memory 52 stores computer execution instructions;

[0381] The processor 51 executes the computer execution instructions stored in the memory 52, so that the processor 51 executes the driving style processing method provided in the above method embodiment.

[0382] The electronic device provided by the embodiment of this application may be a control device of an autonomous mobile device (such as a domain controller of a vehicle), or an autonomous mobile device or a server. When it is a control device of an autonomous mobile device or an autonomous mobile device, it can be used to execute the method provided in any method embodiment related to the above autonomous mobile device. When it is a server, it can be used to execute the method provided in any method embodiment related to the above server, and its implementation principles and technical effects are similar, and will not be elaborated here.

[0383] The embodiment of this application provides an autonomous mobile device, including the above electronic device. The autonomous mobile device may be an autonomous mobile device such as a vehicle, a ship, a drone, or a mobile robot.

[0384] The embodiment of this application provides a computer-readable storage medium, in which computer execution instructions are stored, and when the computer execution instructions are executed by a processor, they are used to implement the method described in the above method embodiment.

[0385] An embodiment of the present application provides a program product, including: a computer program, when the program product runs on a computer, enabling the computer to execute the method described in the above method embodiment.

[0386] An embodiment of the present application provides a computer program, when the computer program is executed by a processor, for executing the method described in the above method embodiment.

[0387] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0388] The present application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the specified functions in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0389] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means, and the instruction means implement the specified functions in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0390] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the specified functions in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0391] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0392] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0393] Computer-readable media include permanent and non-permanent, removable and non-removable media that can be used to store information by any method or technology. Information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic tape, disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include temporary computer-readable media (transitory media), such as modulated data signals and carrier waves.

[0394] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0395] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A configuration method for driving style, characterized in that The method includes: Obtaining the style configuration information of the autonomous mobile device; Based on the style configuration information, obtaining the target driving style; Based on the target driving style, configuring the driving style of the autonomous mobile device.

2. The method according to claim 1, characterized in that The style configuration information includes any one of the following: The driving control method selected by the user; The image data and / or audio data collected by the autonomous mobile device; The position information selected by the user in the style palette, where the style palette includes color palette areas corresponding to multiple pre-obtained driving styles.

3. The method according to claim 1 or 2, characterized in that, If the style configuration information is the driving control method selected by the user, then the obtaining the target driving style based on the style configuration information includes: Based on the driving control method and a pre-configured mapping relationship, determining the target driving style corresponding to the driving control method; wherein, the mapping relationship includes driving styles corresponding to different driving control methods.

4. The method according to claim 2 or 3, characterized in that The driving control method is determined based on at least one driving scenario among a congestion scenario, a starting time at an intersection, a lane change power selection, a turning comfort level, a cut-in handling method, and a safety distance maintained from a normally driving autonomous mobile device.

5. The method according to claim 1 or 2, characterized in that, If the style configuration information is the image data and / or audio data collected by the autonomous mobile device, then the obtaining the target driving style based on the style configuration information includes: Processing the image data and / or audio data based on a pre-obtained data processing model to obtain a style description vector; Selecting the target driving style corresponding to the style description vector from a preset driving style library; wherein, the driving style library stores style description vectors corresponding to different driving styles.

6. The method according to claim 5, wherein When the style configuration information is image data, the data processing model is a visual processing model; When the style configuration information is audio data, the data processing model is a language processing model; When the style configuration information includes image data and audio data, the data processing model is a visual language processing model.

7. The method according to claim 5 or 6, characterized in that, The style description vector includes a scene description vector and / or a preference instruction vector; Wherein, the scene description vector is used to indicate the driving scene where the autonomous mobile device is located, and the preference instruction vector is used to indicate the user's preference for the driving style.

8. The method according to any one of claims 5 to 7, characterized in that, The selecting the target driving style corresponding to the style description vector from a preset driving style library includes: Selecting the driving style corresponding to the target description vector with the highest similarity to the style description vector from the driving style library as the target driving style.

9. The method according to any one of claims 5 to 8, characterized in that, The method further includes: Receiving a data processing model sent by a server and deploying the data processing model.

10. The method according to claim 1 or 2, characterized in that, If the style configuration information is the position information selected by the user in the style palette, then the obtaining the target driving style based on the style configuration information includes: If the position information is within the color palette area corresponding to any driving style in the style palette, then determining the driving style as the target driving style; If the position information is not within the color palette area corresponding to any driving style, at least one driving style within the preset range of the position information is determined from the style color palette; Based on the at least one driving style, the target driving style is determined.

11. The method according to claim 10, wherein The determining the target driving style based on the at least one driving style includes: Based on the distance between the color palette area corresponding to each driving style in the at least one driving style and the position information, the target driving style is determined.

12. The method according to claim 10 or 11, characterized in that The method further includes: Receiving style color palette data sent by the server; Displaying the style color palette on the user interface of the autonomous mobile device.

13. A method for training a model for driving style, characterized in that, Includes: Obtaining a plurality of training samples, each training sample including data to be processed and a style description vector corresponding to the data to be processed, where the data to be processed includes image data and / or audio data collected by the autonomous mobile device; Based on the plurality of training samples, training a deep learning model to obtain a data processing model, where the data processing model is used to analyze input data to obtain a style description vector of the driving style.

14. The method according to claim 13, wherein The data processing model is any one of a visual processing model, a language processing model, or a visual language processing model; Among them, the visual processing model is used to analyze image data to obtain a style description vector of the driving style; The language processing model is used to analyze audio data to obtain a style description vector of the driving style; The visual language processing model is used to analyze image data and audio data to obtain a style description vector of the driving style.

15. The method according to claim 13 or 14, characterized in that The style description vector includes a scene description vector and / or a preference instruction vector; Among them, the scene description vector is used to indicate the driving scene where the autonomous mobile device is located, and the preference instruction vector is used to indicate the user's preference for the driving style.

16. A data processing method for driving style, characterized in that Includes: Obtaining driving data of a plurality of users; For each user, based on the driving data of the user, obtaining a driving style vector corresponding to the user; Performing dimensionality reduction clustering analysis on the driving style vectors corresponding to the plurality of users to obtain a style color palette.

17. The method according to claim 16, wherein The obtaining, for each user, a driving style vector corresponding to the user based on the driving data of the user includes: Analyzing the driving data of the user according to a preset style parsing model to obtain a driving style vector corresponding to the user.

18. A model training method for driving style, characterized in that, Includes: Obtaining a plurality of training samples, each training sample including the driving data of a user and a driving style vector corresponding to the driving data; Based on the plurality of training samples, training a deep learning model to obtain a style parsing model.

19. An apparatus for configuring a driving style, characterized in that The device includes: A first obtaining module, configured to obtain style configuration information of the autonomous mobile device; A first processing module, configured to obtain a target driving style based on the style configuration information; A configuration module, configured to configure the driving style of the autonomous mobile device based on the target driving style.

20. A model training device for driving style, characterized in that Includes: A second obtaining module, configured to obtain a plurality of training samples, each training sample including data to be processed and a style description vector corresponding to the data to be processed, where the data to be processed includes image data and / or audio data collected by the autonomous mobile device; A second processing module, configured to train a deep learning model according to the multiple training samples to obtain a data processing model, where the data processing model is used to analyze the input data to obtain a style description vector of the driving style.

21. A data processing device for driving styles, characterized in that, Comprising: A third acquisition module, configured to acquire driving data of multiple users; A third processing module, configured to, for each user, acquire a driving style vector corresponding to the user according to the driving data of the user; An analysis module, configured to perform dimensionality reduction clustering analysis on the driving style vectors corresponding to the multiple users to obtain a style palette.

22. A model training device for driving style, characterized in that, Comprising: A fourth acquisition module, configured to acquire multiple training samples, where each training sample includes the driving data of a user and the driving style vector corresponding to the driving data; A fourth processing module, configured to train a deep learning model according to the multiple training samples to obtain a style parsing model.

23. An electronic device, characterized in that, Comprising: A processor, a memory, and a communication interface; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 12.

24. A server, characterized in that, Comprising: A processor, a memory, and a communication interface; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 13 to 18.

25. A computer-readable storage medium, characterized in that, Computer-executable instructions are stored in the computer-readable storage medium, and when the computer-executable instructions are executed by a processor, they are used to implement the method according to any one of claims 1 to 18.

26. A program product, characterized in that, Comprising: A computer program, when the program product runs on a computer, causes the computer to execute the method according to any one of claims 1 to 18 above.

27. A computer program, characterized in that, When the computer program is executed by a processor, it is used to execute the method according to any one of claims 1 to 18 above.

28. An autonomous mobile device, characterized in that, Comprising: The electronic device according to claim 23.

Images