A vehicle instrument device control method and control system supporting online processing

Abstract

The present application provides a vehicle instrument device control method and control system supporting online processing, which relates to the technical field of vehicle instrument display, and includes: determining a screen display layout template library and a real-time riding scenario; smoothly replacing the standardized screen display configuration of the vehicle instrument device after scheduling the real-time screen display layout in the screen display layout template library; receiving the real-time ambient brightness and outputting a brightness adjustment strategy; performing screen display layout analysis according to the real-time weather data and real-time terrain data; refreshing the real-time instrument screen display by using the brightness adjustment strategy and the layout optimization strategy; presetting a screen display refresh window, and taking the screen display refresh window as a constraint to perform adaptive periodic online refresh control on the real-time instrument screen display. Through the present application, the technical problem in the prior art that the adaptability of the instrument device control is low due to the lack of perception of real-time index changes and scene recognition functions can be solved, and the adaptability of the instrument device control is improved by real-time scheduling and adjustment of the screen display dynamics.

Description

Technical Field

[0001] This application relates to the technical field of vehicle instrument displays, and particularly to a control method and control system for vehicle instrument devices that support online processing. Background Art

[0002] With the rapid development of intelligent transportation technology, cycling vehicles (such as electric bicycles, shared bicycles, etc.) have gradually become an important part of modern urban transportation. As an important part of cycling vehicles, vehicle instrument devices provide basic cycling data (such as speed, mileage, battery level, etc.) while also undertaking the key task of information interaction between users and vehicles. To meet the increasingly diverse needs of users, the existing control methods for cycling vehicle instrument devices mainly adopt fixed screen display layouts and static indicator display methods, usually only providing the display of basic data such as speed, mileage, and battery level, resulting in the difficulty of meeting the information needs of users in different cycling scenarios (such as commuting, mountain biking, racing, etc.), thus affecting the cycling efficiency and safety.

[0003] In summary, there is a technical problem in the prior art that due to the lack of perception of real-time indicator changes and scene recognition functions, the adaptability of instrument device control is low, further affecting cycling efficiency and safety. Summary of the Invention

[0004] The purpose of this application is to provide a control method and control system for vehicle instrument devices that support online processing, so as to solve the technical problem in the prior art that due to the lack of perception of real-time indicator changes and scene recognition functions, the adaptability of instrument device control is low, further affecting cycling efficiency and safety.

[0005] In view of the above problems, this application provides a control method and control system for vehicle instrument devices that support online processing.

[0006] In a first aspect, the present application provides a vehicle instrument device control method supporting online processing. The vehicle instrument device control method supporting online processing is implemented through a vehicle instrument device control system supporting online processing. Among them, the vehicle instrument device control method supporting online processing includes: receiving historical riding data, and performing retrospective attention analysis on the historical riding data to obtain a screen display layout template library, wherein a plurality of sample riding scenarios and a plurality of sample screen display layouts are associated and stored in the screen display layout template library; fusing and analyzing instrument input data and real-time terrain data to output a real-time riding scenario, wherein the real-time terrain data is obtained in real time through a cloud navigation service; after scheduling a real-time screen display layout in the screen display layout template library according to the real-time riding scenario, smoothly replacing the standardized screen display configuration of the vehicle instrument device with the real-time screen display layout to generate a real-time instrument screen display; after receiving the real-time ambient brightness detected and transmitted back by a light sensor, inputting the real-time ambient brightness into an ambient brightness adjustment network for brightness adjustment analysis to output a brightness adjustment strategy; obtaining real-time weather data through an API, and performing screen display layout analysis according to the real-time weather data and the real-time terrain data to output a layout optimization strategy; refreshing the real-time instrument screen display by using the brightness adjustment strategy and the layout optimization strategy; presetting a screen display refresh window, and performing adaptive periodic online refresh control on the real-time instrument screen display with the screen display refresh window as a constraint.

[0007] In a second aspect, the present application further provides a vehicle instrument device control system supporting online processing, which is used to execute a vehicle instrument device control method supporting online processing as described in the first aspect. Among them, the vehicle instrument device control system supporting online processing includes: a historical data analysis module, which is used to receive historical riding data and perform retrospective attention analysis on the historical riding data to obtain a screen display layout template library. Among them, multiple sample riding scenarios and multiple sample screen display layouts are associated and stored in the screen display layout template library; a riding scenario analysis module, which is used to fuse and analyze instrument input data and real-time terrain data and output a real-time riding scenario. Among them, the real-time terrain data is obtained in real time through a cloud navigation service; a screen display layout scheduling module, which is used to schedule a real-time screen display layout in the screen display layout template library according to the real-time riding scenario, and then use the real-time screen display layout to smoothly replace the standardized screen display configuration of the vehicle instrument device to generate a real-time instrument screen display; an ambient brightness adjustment module, which is used to receive the real-time ambient brightness detected and transmitted back by a light sensor, input the real-time ambient brightness into an ambient brightness adjustment network for brightness adjustment analysis, and output a brightness adjustment strategy; a screen display layout optimization module, which is used to obtain real-time weather data through an API and perform screen display layout analysis according to the real-time weather data and the real-time terrain data, and output a layout optimization strategy; a screen display update module, which is used to refresh the real-time instrument screen display by using the brightness adjustment strategy and the layout optimization strategy; an online refresh control module, which is used to preset a screen display refresh window and perform adaptive periodic online refresh control on the real-time instrument screen display with the screen display refresh window as a constraint.

[0008] One or more technical solutions provided in the present application have at least the following technical effects or advantages:

[0009] By receiving historical riding data and performing retrospective attention analysis on the historical riding data, a screen display layout template library is obtained. Among them, multiple sample riding scenarios and multiple sample screen display layouts are associated and stored in the screen display layout template library; by fusing and analyzing instrument input data and real-time terrain data, a real-time riding scenario is output, where the real-time terrain data is obtained in real time through a cloud navigation service; after scheduling a real-time screen display layout in the screen display layout template library according to the real-time riding scenario, the real-time screen display layout is used to smoothly replace the standardized screen display configuration of the vehicle instrument device to generate a real-time instrument screen display; after receiving the real-time ambient brightness detected and transmitted back by a light sensor, the real-time ambient brightness is input into an ambient brightness adjustment network for brightness adjustment analysis to output a brightness adjustment strategy; real-time weather data is obtained through an API, and screen display layout analysis is performed according to the real-time weather data and the real-time terrain data to output a layout optimization strategy; the real-time instrument screen display is refreshed using the brightness adjustment strategy and the layout optimization strategy; a screen display refresh window is preset, and adaptive periodic online refresh control is performed on the real-time instrument screen display with the screen display refresh window as a constraint. That is to say, by determining the screen display layout template library according to historical riding data, combining real-time terrain data and instrument data to determine the current riding scenario, dynamically scheduling the real-time screen display layout, intelligently adjusting the display brightness and layout according to light conditions and weather changes, and presetting the screen display refresh window to automatically update the screen display information according to real-time changes, the adaptability, intelligent level of the vehicle instrument device, and the safety and efficiency of the user's riding are improved.

[0010] The above description is only an overview of the technical solution of this application. In order to be able to understand the technical means of this application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of this application more obvious and understandable, the following specifically gives the specific implementation manners of this application. It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of this application, nor is it used to limit the scope of this application. Other features of this application will become easily understood through the following description. Brief Description of the Drawings

[0011] In order to more clearly illustrate the technical solutions in this application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings described below are only exemplary, and for those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0012] Figure 1 It is a schematic flowchart of a control method for a vehicle instrument device supporting online processing in this application;

[0013] Figure 2This is a schematic structural diagram of a vehicle instrument device control system that supports online processing in this application.

[0014] Explanation of reference numerals in the drawings: Historical data analysis module 11, riding scenario analysis module 12, screen display layout scheduling module 13, ambient brightness adjustment module 14, screen display layout optimization module 15, screen display update module 16, online refresh control module 17. Detailed implementation manners

[0015] This application provides a vehicle instrument device control method and a control system that support online processing, and solves the technical problem in the prior art that due to the lack of perception of real-time index changes and scenario recognition functions, the adaptability of instrument device control is low, further affecting riding efficiency and safety. By determining a screen display layout template library based on historical riding data, combining real-time terrain data and instrument data to determine the current riding scenario, dynamically scheduling the real-time screen display layout, intelligently adjusting the display brightness and layout according to light conditions and weather changes, and presetting a screen display refresh window to automatically update the screen display information according to real-time changes, the adaptability, intelligent level of the vehicle instrument device, and the safety and efficiency of user riding are improved.

[0016] Next, the technical solutions in this application will be described clearly and completely with reference to the drawings. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments of this application. It should be understood that this application is not limited by the example embodiments described here. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of this application. Additionally, it should be noted that for the sake of description, only the parts related to this application are shown in the drawings rather than all of them.

[0017] Embodiment 1, please refer to the attached Figure 1 , this application provides a vehicle instrument device control method that supports online processing. Among them, the vehicle instrument device control method that supports online processing is applied to a vehicle instrument device control system that supports online processing. The vehicle instrument device control method that supports online processing specifically includes the following steps:

[0018] S100: Receive historical riding data, and perform retrospective attention analysis on the historical riding data to obtain a screen display layout template library, where multiple sample riding scenarios and multiple sample screen display layouts are associated and stored in the screen display layout template library.

[0019] Specifically, receive historical cycling data, including cycling records in different scenarios over a past period of time and the user's usage and query behaviors regarding various metrics. Conduct retrospective attention analysis on the historical cycling data to count the attention frequencies of the user for various cycling metrics in different cycling scenarios. According to the results of the retrospective attention analysis, that is, the attention frequencies of various cycling metrics in each scenario, design corresponding screen display layout templates for each cycling scenario. The retrospective attention analysis refers to analyzing the historical cycling data, counting the attention degrees of the user for various metrics (such as heart rate, power, etc.) in different scenarios, and designing the screen display layout based on the attention degrees.

[0020] The screen display layout template library is a database used to store predefined screen display layouts. Each template is associated with a certain cycling scenario (such as commuting mode, competitive mode, etc.) and is used to dynamically adjust the screen information display to optimize the user experience. For example, the historical cycling data shows that the user has 100 query records in the commuting scenario, among which, navigation is 50 times, speed is 25 times, and time is 25 times. Then the screen display layout for the commuting mode is: navigation occupies 50% of the screen, as the main content of attention, located in the center of the screen; speed occupies 25% of the screen, located on the right side of the screen; time occupies 25% of the screen, located on the left side of the screen. The historical cycling data shows that the user has 40 query records in the competitive mode, among which, speed is 16 times, power output is 12 times, and heart rate is 12 times. Then the screen display layout for the competitive mode is: speed occupies 40% of the screen, as the core content of attention, displayed at the top of the screen; power output occupies 30% of the screen, displayed on the left side of the screen; heart rate occupies 30% of the screen, displayed on the right side of the screen.

[0021] Through this allocation, the screen display layout can meet the actual needs of the user in different scenarios and improve the usage experience. Associate the generated screen display layout templates with the corresponding cycling scenarios and store them in the screen display layout template library to complete the construction of the screen display layout template library. The role of the template library is to standardize the screen display layout and provide a reference for the real-time layout in different scenarios. Provide optimized screen display layouts according to different cycling scenarios (such as commuting, competition) to meet the core attention needs of the user in specific scenarios.

[0022] S200: Integrate and analyze the instrument input data and the real-time terrain data, and output the real-time cycling scenario, where the real-time terrain data is obtained in real time through the cloud navigation service.

[0023] Specifically, obtain the current instrument input data, that is, the data collected from the instrument devices of the cycling vehicle (such as speedometers, power meters, etc.), including the current cycling state, user physiological state, and vehicle state, such as cycling speed, mileage, remaining battery power, user physiological state (such as heart rate), etc. At the same time, obtain the current terrain data, that is, the geographical and terrain information of the cycling environment, including slope, road surface type (such as asphalt, dirt, gravel, etc.), road conditions (such as flat road, uphill, downhill, etc.), etc., which helps to judge the complexity and difficulty of the current cycling. Usually, it is obtained in real time through GPS, sensors, and cloud navigation services. The cloud navigation service is a navigation service provided by using the cloud computing platform, which can obtain and process geographical information data in real time, provide detailed terrain information, real-time traffic conditions, road conditions, etc., and can be dynamically updated according to the position and state of the cyclist.

[0024] By combining the instrument data and terrain information, identify the specific situation of the current cycling. Input the instrument input data and real-time terrain data into the trained cycling scenario recognition network to identify the current cycling scenario, including flat road, uphill, downhill, turning, bumpy road, etc. The cycling scenario recognition network is a model constructed based on the CNN network and is locally trained through historical data (such as instrument input records, terrain records, and cycling scenarios) to complete the construction of the cycling scenario recognition network. Through the cycling scenario recognition network, obtain multiple cycling scenario probability parameters. Conduct probability transfer analysis on the multiple cycling scenario probability parameters, and combine the most likely cycling scenario after scenario transfer to determine the current cycling scenario. By considering both the state of the cyclist and environmental conditions, accurately identify the current cycling scenario (such as uphill, flat ground, downhill, etc.) to ensure the accuracy and real-time nature of the information.

[0025] S300: After scheduling the real-time screen display layout in the screen display layout template library according to the real-time cycling scenario, smoothly replace the standardized screen display configuration of the vehicle instrument device with the real-time screen display layout to generate a real-time instrument screen display.

[0026] Specifically, traverse the screen display layout template library according to the determined real-time riding scenario (such as commuting mode, competitive mode, mountain biking, etc.) to determine the current real-time screen display layout. That is to say, according to the current riding scenario, the screen display configuration retrieved and applied from the screen display layout template library determines the display method and distribution of various indicators on the screen. When the scenario changes, smoothly replace the currently displayed content (standardized screen display configuration) with the new screen display layout (real-time screen display layout). The smooth replacement technology is used to gradually change the screen content in the form of an animation to avoid the visual abruptness caused by direct jumping. Smooth replacement is a visual transition effect, usually achieved through a transition animation, making the switching of the screen display layout more fluent without producing an abrupt or unnatural transition. The change of the displayed content can be smoothed through gradient, sliding or other animation effects.

[0027] The standardized screen display configuration refers to the initial or default display mode of the device, which may include the layout of fixed content, such as the default display of information such as speed, battery level, and time. According to the current riding needs of the user corresponding to the real-time riding scenario, determine which data is the most important and display it preferentially. According to the selected layout configuration, render and display various indicators (such as speed, battery level, heart rate, etc.) at the preset positions. For example, the information on the screen gradually disappears and is replaced by new content, with a gradual transition in color or content, presenting a smooth change; different screen display areas can be switched through a sliding effect, with new information sliding into the old information area and the old information sliding out of the area. When the user starts entering the competitive mode, the navigation information on the screen will first gradually disappear, while data such as speed, power output, and heart rate will gradually appear. The content in the navigation area gradually slides out of the screen, while the speed, power, and heart rate data slide in from the edge of the screen. By introducing animation or transition effects, the visual abruptness is reduced, and the user can smoothly transition to the new screen display layout, enhancing the fluency and comfort of operation and ensuring that the device can display the indicators that the user is most concerned about according to different riding scenarios (such as commuting, competition, etc.).

[0028] S400: After receiving the real-time ambient brightness detected and transmitted back by the light sensor, input the real-time ambient brightness into the ambient brightness adjustment network for brightness adjustment analysis, and output a brightness adjustment strategy.

[0029] Specifically, the ambient brightness is monitored in real time by a light sensor, and the detected real-time ambient brightness data is transmitted to the control unit of the vehicle instrument wirelessly or wiredly. The control unit inputs the real-time ambient brightness data detected by the light sensor into the ambient brightness adjustment network for brightness adjustment analysis. A light sensor is a sensor that can sense the light intensity (such as natural light or artificial light), and is usually used to measure the brightness in the environment.

[0030] An ambient light adjustment network is a system or network used to process and analyze real-time ambient light data. It usually employs machine learning or adaptive algorithms to generate light adjustment strategies. Based on the changing trend of ambient light data, it determines appropriate adjustment strategies, such as increasing or decreasing the screen brightness, controlling the lights, etc. After receiving the real-time ambient light data, the ambient light adjustment network evaluates the brightness condition of the current environment through built-in analysis algorithms (such as rule-based logic, machine learning, deep learning, etc.) and determines whether to adjust the display screen brightness or indoor lighting. According to the analysis results, the system outputs a set of light adjustment strategies, including increasing brightness, decreasing brightness, optimizing energy efficiency, etc.

[0031] Obtain a large number of ambient light samples and related parameters for training the ambient light adjustment network, that is, the ambient light sets and corresponding light adjustment strategy sets in different environments and different time periods over a past period. Select a suitable type of neural network, such as a convolutional neural network (CNN) or a recurrent neural network (RNN). Design the structure of the network, including the number of layers, the number of neurons in each layer, activation functions, etc. Use the collected data to adjust the parameters in the neural network so that it can accurately predict the light adjustment strategy. Perform data preprocessing on the ambient light sets and corresponding light adjustment strategy sets, and divide them into a training set and a validation set. Train the neural network using the training set, and use an appropriate optimization algorithm (such as gradient descent) to adjust the network parameters to minimize the difference between the predicted value and the actual value. Verify the accuracy of the model through the validation set, and adjust the network structure or parameters according to the verification results to improve its generalization ability. Stop training until the model reaches the convergence condition, and integrate the trained ambient light adjustment network into the control unit of the vehicle instrument so that it can receive real-time ambient light data for light adjustment analysis and output light adjustment strategies.

[0032] Determine corresponding light adjustment strategies according to the change of real-time ambient light. For example, if the ambient light is lower than a certain threshold (such as 300 Lux), automatically increase the screen brightness or turn on the lights; if the ambient light is higher than a certain threshold (such as 7000 Lux), automatically decrease the screen brightness and even turn off unnecessary lights. By automatically adjusting the brightness of the display device and lights according to the ambient light, avoid the impact of too bright or too dark environment on the user's visual comfort. When the ambient light is high, automatically lower the device brightness to save energy and extend the battery life.

[0033] S500: Obtain real-time weather data through the API, and perform screen display layout analysis based on the real-time weather data and real-time terrain data, and output a layout optimization strategy.

[0034] Specifically, by obtaining real-time weather data, weather information including temperature, humidity, wind speed, precipitation probability, ultraviolet intensity, etc. is obtained from a weather service platform through an established interface. An API (Application Programming Interface) is an interface that allows different software systems to communicate with each other, providing a standardized method to access specific functions or data. For example, public weather APIs such as OpenWeatherMap, WeatherAPI, etc. are used to obtain current weather data. Real-time weather data is information about environmental conditions provided in real-time by weather service providers, such as temperature, humidity, wind speed, precipitation amount, etc.

[0035] Through a Geographic Information System (GIS) or a navigation service, real-time data about the terrain is obtained, including road slope, curvature, road surface type, driving path, etc. The real-time weather data and terrain data are input into the screen display layout analysis module for comprehensive analysis. For example, the screen display layout is adjusted according to different weather and terrain conditions. When the temperature is relatively high, the area for displaying temperature and humidity is increased; when the wind speed is relatively high, since the wind speed has a greater impact on cyclists, the area for displaying wind speed data is enlarged; when it is detected that the cyclist enters an uphill section, the display priority of power output or driving speed data is increased to help the cyclist better adjust the strategy. According to the results of the screen display layout analysis, a layout optimization strategy is output, which determines the display priority and display content of each item of information on the display screen. According to the output layout optimization strategy, the content of the display screen is adjusted. For example, if the likelihood of rain is detected to increase, a weather warning is preferentially displayed on the screen, and even the brightness is adjusted to improve visibility. By dynamically adjusting the display layout, users can clearly and conveniently obtain the most important information such as temperature, wind speed, slope, etc. during cycling, optimizing the cycling experience.

[0036] S600: Refresh the real-time instrument screen display by adopting the brightness adjustment strategy and the layout optimization strategy.

[0037] Specifically, the brightness adjustment strategy is to dynamically adjust the brightness level of the instrument screen according to the brightness adjustment strategy output by the ambient brightness adjustment network, optimizing the clarity and energy consumption of the screen display, ensuring that the screen content is visible in strong light, not dazzling and energy-saving in low-light environments. The layout optimization strategy is to dynamically adjust the display area size, position, and priority of each functional module (such as heart rate, speed, temperature, power, etc.) on the display screen according to the optimization results output by the screen display layout analysis module, enhancing the readability of important information while reducing the interference of unnecessary information.

[0038] Combine the brightness adjustment strategy and the layout optimization strategy to update the real-time instrument display. Apply the brightness parameter in the brightness adjustment strategy to the screen hardware of the instrument device, and adjust the backlight intensity or screen contrast to achieve brightness adjustment. Reallocate the space size of the display module according to the priority, and smoothly replace the real-time instrument display of the vehicle instrument device. When the brightness or layout changes, use transition animations (such as gradual brightness, smooth scaling of modules) to improve the user experience. After completing the brightness and layout refresh, load the real-time user physiological data, vehicle status data, and environmental data into the new layout. By refreshing the real-time instrument display using the brightness adjustment strategy and the layout optimization strategy, it is possible to achieve adaptive adjustment of brightness and layout, improve the intelligence level of the riding device, provide the best display effect of important data in real time, meet the usage requirements of users under various environmental conditions, and ensure the efficient and energy-saving operation of the device.

[0039] S700: Preset a display refresh window, and use the display refresh window as a constraint to perform adaptive periodic online refresh control on the real-time instrument display.

[0040] Specifically, preset a display refresh window, that is, a time range or constraint condition for controlling the screen refresh frequency and update cycle. For example, each 1 minute is a refresh cycle to ensure that the display data is updated within this time interval, so as to achieve the purpose of balancing real-time performance and system resource utilization. Set a display refresh time range with strong adaptability according to the hardware performance of the riding device and user requirements. Use the display refresh window as a constraint to perform online refresh control on the real-time instrument display through adaptive periodicity.

[0041] Adaptive periodicity means dynamically adjusting the refresh cycle of the display. According to the real-time data change rate or the frequency of priority changes, set a reasonable refresh frequency. For example, when the user's heart rate or riding speed changes violently, shorten the refresh cycle. If the data change amplitude exceeds a preset threshold (such as the heart rate change exceeds 10 bpm), shorten the refresh cycle. If the data change amplitude is small or stable for a long time, extend the refresh cycle to save energy. According to the refresh window and priority, update the display content in real time, and frequently refresh high-priority modules (such as speed, heart rate); refresh low-priority modules (such as body temperature, battery power) according to a longer cycle. During each refresh cycle, monitor the current data change situation and dynamically adjust the refresh strategy for the next cycle. By presetting the display refresh window and implementing adaptive periodic online refresh control, it is possible to effectively improve the intelligence and user experience of the riding instrument device, provide fast feedback of real-time data, and optimize the refresh frequency according to actual needs, taking into account both performance and energy efficiency.

[0042] Furthermore, S100 of this application includes:

[0043] Preset a time series division scale, and divide the historical cycling data with the time series division scale as a constraint to obtain M historical cycling subsets, where the historical cycling subsets include historical instrument input records and historical terrain records; after inputting the M historical cycling subsets into the cycling scenario recognition network for scenario probability recognition, obtain M historical cycling scenarios through probability transition analysis; perform scenario consistency aggregation on the M historical cycling scenarios to obtain multiple historical cycling scenario sets of the multiple sample cycling scenarios; aggregate the M cycling index query records of the M historical cycling subsets according to the multiple historical cycling scenario sets to obtain multiple groups of cycling index query records; perform retrospective attention analysis on the multiple groups of cycling index query records to obtain the screen display layout template library.

[0044] Specifically, preset a time series division scale, which is the time interval used for dividing time series data, such as dividing data by dimensions such as minutes, hours, days, cycling stages, etc. Obtain historical cycling data, and divide the historical cycling data through the time series division scale to obtain M historical cycling subsets, and each subset contains historical instrument input records and terrain records within a specific time period.

[0045] Input the M historical cycling subsets obtained by the division into the cycling scenario recognition network to identify the cycling scenario probability corresponding to each subset. By analyzing the change of the scenario probability, that is, analyzing the scenario probability distribution through the scenario state transition matrix, determine the scenario classification of each historical cycling subset to obtain M historical cycling scenarios. The specific process is similar to the process of performing probability transition analysis on multiple cycling scenario probability parameters and outputting the real-time cycling scenario, which is described in detail in the corresponding dependent claim steps. Briefly, input the M historical cycling subsets into the cycling scenario recognition network to obtain multiple cycling scenario probability parameters corresponding to each subset respectively. Input the multiple cycling scenario probability parameters corresponding to each subset into the scenario state transition matrix to obtain multiple updated scenario probability parameters, add the multiple cycling scenario probability parameters and the multiple updated scenario probability parameters to obtain the scenario probability sequence corresponding to this subset, and perform a maximum value call to obtain the historical cycling scenario corresponding to this subset. Perform this step for all M historical cycling subsets to obtain M historical cycling scenarios.

[0046] Aggregate the identified M historical cycling scenarios for scene consistency, classify multiple similar scenarios (such as uphill slopes of different intensities) into one category to reduce redundancy and improve efficiency. Scene consistency aggregation refers to the process of integrating multiple historical cycling scenarios to find consistent scene features, such as merging similar scenarios into one through a clustering algorithm. Through consistency aggregation, multiple sets of historical cycling scenarios of multiple sample cycling scenarios are obtained. Query historical cycling metrics for the M historical cycling subsets to obtain M cycling metric query records, including average speed, power consumption, terrain type, etc. Aggregate the M cycling metric query records according to the multiple sets of historical cycling scenarios to obtain multiple sets of cycling metric query records, and the multiple sets of historical cycling scenarios and multiple sets of cycling metric query records are corresponding.

[0047] Statistically analyze the backtracking query frequency of multiple sets of cycling metric query records, that is, count the frequency of each cycling metric appearing in the historical cycling data, and analyze which metrics are the key concerns of users in different scenarios. According to the query frequencies of multiple sets of metrics, determine multiple user attention sequences, including the sequences of user attention in different scenarios. Analyze the metric display layout based on the multiple user attention sequences to determine multiple sample display layouts, that is, how the display layouts should be in different scenarios. For example, in the commuting scenario, the attention to navigation is 50%, the attention to the current speed is 25%, and the attention to time is 25%. Then navigation occupies 50% of the screen, and speed and time each occupy 25% of the screen.

[0048] Associate and store multiple sample display layouts and multiple sample cycling scenarios to obtain a display layout template library. That is, associate each sample cycling scenario with its corresponding display layout and store it in the display layout template library. Through time series division scale and scene aggregation, convert the huge historical cycling data into meaningful scene sets. The generated display layout template library can automatically switch the display content according to real-time data, improve the information display effect during cycling, and make it more in line with the needs of cyclists.

[0049] Furthermore, the present application further includes the following steps:

[0050] Statistically analyze the backtracking query frequency of the multiple sets of cycling metric query records to obtain multiple sets of metric query frequencies, where the cycling metrics include heart rate metrics, blood oxygen saturation metrics, body temperature metrics, driving pace metrics, battery metrics, power output metrics, exercise load metrics, fatigue index metrics, and energy efficiency metrics; construct multiple user attention sequences according to the multiple sets of metric query frequencies; analyze the metric display layout based on the metric attention of the multiple user attention sequences to generate the multiple sample display layouts; associate and store the multiple sample cycling scenarios and multiple sample display layouts to complete the construction of the display layout template library.

[0051] Specifically, multiple groups of cycling index query records record various physiological, environmental, or vehicle status indexes related to the cycling process, such as heart rate, blood oxygen saturation, body temperature, battery level, power, etc. The backtracking query frequency of each cycling index in the historical cycling data is statistically analyzed, and the query frequency of each index in different scenarios (such as uphill, flat ground, etc.) is counted to obtain the query frequency of each index. Cycling indexes include heart rate, blood oxygen saturation, body temperature, driving pace, battery level, power output, exercise load, fatigue index, and energy efficiency, etc., which reflect the states of the driving user and the vehicle.

[0052] Based on the query frequencies of multiple groups of indexes, a attention sequence is generated for each cycling index. The indexes are sorted according to their query frequencies (or percentages) to obtain multiple user attention sequences. Each attention index has an attention percentage, which determines the display size of the index on the index screen layout. A higher percentage indicates that the user pays more attention. For example, when the user is cycling uphill, they often check their heart rate 80 times, accounting for 40% of the total query frequency, and the attention percentage is 40%; the frequency of checking the battery level is 40 times, accounting for 20% of the total query frequency, and the attention percentage is 20%.

[0053] Based on the index attention of multiple user attention sequences, the index screen layout analysis is carried out to determine the display size and position of each index in the screen layout, and multiple sample screen layouts are generated. Each layout corresponds to a sample cycling scenario. The user attention sequence is a generated attention sequence based on the query frequency of the cycling index and its display size in the screen layout (determined by the attention percentage). Each index concerned by the user will be assigned a weight (percentage) according to the frequency, which determines the display size or priority of the index on the screen. By analyzing the attention of each cycling index, it is determined which indexes need to be displayed in a large area and which can be displayed in a small area, so as to optimize the space utilization and information display of the screen.

[0054] Based on multiple sample riding scenarios, determine the sample screen display layout corresponding to each sample riding scenario, associate each sample riding scenario with its corresponding screen display layout, and store them in the screen display layout template library. Each riding scenario (such as flat road, uphill, competition, etc.) has a corresponding sample screen display layout, and these layouts can be dynamically switched according to real-time data. For example, in the commuting scenario, navigation, current speed, and average speed occupy a relatively large display area, while heart rate and fatigue index are displayed in a smaller area; in the competition scenario, the user pays more attention to the real-time speed, followed by power output, then heart rate, and battery level and navigation occupy a smaller display area; in the fitness scenario, the user has a relatively high concern for calorie consumption and heart rate, so they occupy a relatively large display area. By analyzing the query frequency of riding metrics, automatically adjust the screen display content according to the user's attention needs, ensure that the metrics the user cares most about are displayed first, allocate appropriate display ratios for each metric, effectively utilize the screen space, avoid excessive information stacking, and ensure that the screen display is concise and information-rich.

[0055] Further, step S200 of the present application includes:

[0056] Interactively obtain multiple sample instrument input records, multiple sample terrain records, and multiple sample scenario probability records of the multiple sample riding scenarios, where the sample instrument input records include multiple historical instrument inputs, and the historical instrument inputs include historical riding state parameters, historical user physiological data, and historical vehicle state data; construct a riding scenario recognition network based on a CNN network; use the multiple sample instrument input records, multiple sample terrain records, and multiple sample riding scenarios as training data to adjust and optimize the parameters of the riding scenario recognition network to complete the localization of the riding scenario recognition network; load the instrument input data and real-time terrain data into the riding scenario recognition network, and output multiple riding scenario probability parameters of the multiple sample riding scenarios; perform probability transfer analysis on the multiple riding scenario probability parameters to output the real-time riding scenario.

[0057] Specifically, multiple sample cycling scenarios are obtained by interacting with the dashboard terminal or the Internet of Things platform. That is, in different cycling environments, based on the specific cycling situations recorded in historical data, including various possible cycling situations such as commuting, racing, mountain biking, fitness, etc. Historical data related to multiple sample cycling scenarios is collected, including multiple sample instrument input records, multiple sample terrain records, and multiple sample scenario probability records. Multiple sample instrument input records refer to the input data recorded by historical instrument devices, including historical cycling state parameters (such as speed, power output, mileage, etc.), user physiological data (such as heart rate, body temperature, etc.), and vehicle state data (such as battery level, tire pressure, etc.). Multiple sample terrain records refer to the historical terrain data related to the cycling environment, including road slope, curvature, road surface type, etc. Multiple sample scenario probability records refer to recording different cycling scenarios that occurred during historical cycling and their occurrence probabilities.

[0058] A cycling scenario recognition network is constructed using a Convolutional Neural Network (CNN), which includes multiple convolutional layers, pooling layers, and fully connected layers. The convolutional layers are used to extract local features of the image, the pooling layers are used to reduce the feature dimension and computational amount, and the fully connected layers are used to map the extracted features to the final classification labels. The CNN network is a deep learning method that is particularly good at processing data with a grid structure, such as images and videos. Data preprocessing is performed on multiple sample instrument input records, multiple sample terrain records, and multiple sample cycling scenarios, including data cleaning, normalization, etc. Pair the historical cycling data (such as user heart rate, cycling speed, etc.), terrain data (such as slope, curvature, etc.), and labels of cycling scenarios (such as mountain, flat, urban, etc.) with the actual scenario labels to form a complete training dataset.

[0059] Input multiple sample instrument input records, multiple sample terrain records, and multiple sample cycling scenarios into the cycling scenario recognition network as the training set, with the aim of enabling the network to learn how to identify cycling scenarios based on the input instrument data and terrain information. By adjusting the hyperparameters of the model, the performance of the network can be improved, including the learning rate, network structure, etc. The learning rate controls the update step size of the model in each iteration. By using a learning rate decay strategy or trying different learning rate values, it can help the model converge quickly and avoid getting stuck in local optima. Network structure optimization includes selecting different numbers of convolutional layers, convolutional kernel sizes, pooling layers, etc. Adjusting these structural parameters can enable the network to extract features more effectively.

[0060] By using the optimized parameters above, local training of the network is carried out. Localization refers to adapting the trained general model to specific application scenarios, such as different geographical locations, user behaviors, hardware characteristics, etc. The terrain in different geographical locations may vary, so the model needs to be locally trained according to local terrain features (such as slope, road surface type, etc.). Different users have different riding habits. For example, in some places, speed is more emphasized during riding, while in other places, endurance may be more emphasized. Therefore, adjustments need to be made according to different riding habits. Different riding devices (such as electric bicycles, shared bicycles, etc.) may have different sensor data, so the model needs to be customized according to the device characteristics.

[0061] After the local training is completed, the network will be trained using new sample data (including instrument input, terrain record, and scene labels), and continuously optimize the parameters to finally achieve the best recognition effect. At this time, the model has been customized for a specific riding scenario and can more accurately identify the current riding environment. The riding scenario recognition network refers to a model that identifies the current riding environment or scenario through deep learning (such as a CNN network), and is trained through the input instrument data, terrain data, and historical riding scenario data to identify and classify the probability parameters of real-time riding scenarios.

[0062] The instrument input data and real-time terrain data are passed through the network input layer to the riding scenario recognition network. The instrument input data extracts key features through a processing network (for example, a feedforward neural network) to describe the user's physiological state, riding speed, etc. The real-time terrain data extracts geographical features, such as slope, road surface type, etc., through a convolutional neural network (CNN). The instrument input data refers to the data obtained from the instrument system of the riding vehicle, including but not limited to riding state parameters, user physiological data, and vehicle state data, such as parameters related to the riding state, such as vehicle speed, heart rate, battery power, power output, etc. The real-time terrain data refers to the data of the terrain features of the current riding environment, usually including information such as road slope, curvature, and road surface type.

[0063] In the riding scenario recognition network, the input instrument data and terrain data will be fused to form a comprehensive feature vector. The instrument data is feature-extracted through a fully connected layer; the terrain data features are extracted through a convolutional layer network structure, and these two parts of features are combined into a final feature vector for subsequent scenario recognition. The fused feature vector will be calculated through a classification layer (usually a softmax layer) to output the probability of each riding scenario. For example, the model may output: mountain riding 40%, commuting riding 35%, flat riding 15%, competitive riding 10%.

[0064] Obtain the scene switching records in historical rides to form historical scene alternation records. Analyze the transition probabilities between each pair of scenes, aggregate multiple sets of sample scene transition probabilities together, and construct a scene state transition matrix. Each row of the matrix represents the current riding scene, and each column represents the target riding scene. Input multiple riding scene probability parameters into the scene state transition matrix and calculate to obtain multiple updated scene probability parameters. That is to say, through matrix operations, calculate the probability distribution of the next most likely riding scene.

[0065] Sum up the probability parameters of multiple riding scenes and the updated scene probability parameters to obtain a comprehensive sample scene probability sequence. These summed results represent the total probability of each scene, helping to further confirm which scene is most likely to be the current riding scene. By making a maximum value call on the summed sample scene probability sequence, select the riding scene with the largest probability value as the current real-time riding scene. That is to say, through the probability distribution of the current scene and the probability distribution of the next most likely riding scene, determine the most suitable current riding scene. Through the training and optimization of the CNN network, the device can identify the current riding scene in real time and accurately. According to the recognition results of different riding scenes, the instrument device can dynamically adjust the display content, timely adjust the display according to the riding state and environmental changes, enhance the user's perception of the current riding state, and avoid affecting riding safety due to insufficient information or inaccurate display.

[0066] Furthermore, this application also includes the following steps:

[0067] Interactively obtain historical scene alternation records; calculate multiple sets of sample scene transition probabilities of the multiple sample riding scenes according to the historical scene alternation records; construct a scene state transition matrix using the multiple sets of sample scene transition probabilities; input the multiple riding scene probability parameters into the scene state transition matrix and calculate to obtain multiple updated scene probability parameters; sum up the multiple riding scene probability parameters and the multiple updated scene probability parameters, and serialize the multiple sample riding scenes according to the summation result to obtain a sample scene probability sequence; make a maximum value call on the sample scene probability sequence and output the real-time riding scene.

[0068] Specifically, interact with the Internet of Things platform to obtain historical scene alternation records, that is, during the cycling process, the conversion records between different scenes, including the changes from one cycling scene to another, such as the transition from urban cycling to mountain cycling. For example, the user may transition from flat road cycling to mountain cycling and then to urban cycling, etc. The historical data will help judge the switching frequency and method between such scenes. By analyzing the historical scene alternation records, calculate the transition probability between each pair of scenes, that is, the probability of transitioning from one cycling scene to another. The historical scene alternation records contain the scene switching information during the cycling process. By counting these switching times, the transition probability can be calculated. Through calculation, the probability of each scene transition can be obtained by the ratio of the number of transitions to the total number of transitions. For example, if the number of times of switching from flat road to uphill is 300 times and the total number of switchings is 1000 times, then the transition probability from flat road to uphill is 0.3.

[0069] Construct a scene state transition matrix based on the calculated multiple groups of sample scene transition probabilities, which is used to represent between different scenes. Each row of the matrix represents the current cycling scene, each column represents the target scene, and each element represents the probability of transitioning from the current cycling scene to the target cycling scene. Input multiple cycling scene probability parameters into the scene state transition matrix. Usually, multiply multiple cycling scene probability parameters by the scene state transition matrix to obtain multiple updated scene probability parameters, that is, the probability values of the updated cycling scenes, which reflect the possibility of the current scene after the scene transition.

[0070] Sum up multiple cycling scene probability parameters and multiple updated scene probability parameters to form a comprehensive sample scene probability sequence. Summing up means performing a superposition operation on multiple probability parameters. Usually, obtain a more accurate scene prediction value by weighted summing the probabilities of multiple scenes. Serialize multiple sample cycling scenes according to the summing result, arrange the summing results at each time point in a certain order (such as the summing result) to form a sample scene probability sequence, which represents the scene probability evolution process during the entire cycling process and indicates the possible cycling scenes and their corresponding probabilities at each time point during the entire cycling process.

[0071] Perform a maximum value call on the sample scenario probability sequence to find the cycling scenario with the highest probability value from multiple sample scenario probability sequences, thereby outputting the most likely real-time cycling scenario. The maximum value call is performed by comparing the values of each data point and selecting the maximum value. From the entire sample scenario probability sequence, select the cycling scenario with the highest probability at the current moment to obtain the real-time cycling scenario. The real-time cycling scenario is not static and will be continuously updated as time goes by and the input data changes. For example, as the cyclist enters different terrains (such as from the city to the mountains), new probability values are recalculated based on the new real-time terrain data and historical data, and a maximum value call is performed to ensure that the output scenario is always the most suitable for the current environment and state.

[0072] Accurately predict the current cycling scenario through historical scenario alternation records and scenario transition probabilities, and dynamically adjust the prediction of the current scenario according to the changes during the cycling process through the scenario state transition matrix, determine the most likely real-time cycling scenario, effectively predict and identify the current cycling environment, thereby providing more accurate instrument information display and cycling suggestions, and improving the safety and user experience of cycling.

[0073] Furthermore, step S300 of this application includes:

[0074] Collect physiological data of the cycling user through a wearable device to obtain real-time user physiological data. Among them, the real-time user physiological data includes real-time heart rate, real-time blood oxygen saturation, and real-time body temperature; obtain real-time vehicle state data by calling from a vehicle state sensing array. Among them, the real-time vehicle state data includes real-time driving speed, real-time power information, and real-time power output; perform fusion analysis on the real-time user physiological data and real-time vehicle state data to obtain real-time cycling state parameters. Among them, the real-time cycling state parameters include real-time exercise load, real-time fatigue index, and real-time energy efficiency; perform a screen display priority judgment based on the real-time cycling state parameters and real-time cycling scenario to obtain a real-time screen display priority; after dynamically rearranging the real-time instrument screen display according to the real-time screen display priority, load the real-time cycling state parameters, real-time user physiological data, and real-time vehicle state data onto the real-time instrument screen display.

[0075] Specifically, collect the physiological data of the cycling user through a wearable device. For example, through a heart rate belt, smart watch or other sensor devices worn by the user, collect the heart rate, blood oxygen saturation, and body temperature data of the user in real time. The real-time user physiological data includes real-time heart rate, real-time blood oxygen saturation, and real-time body temperature, that is, the current heart beating frequency of the cycling user (used to evaluate exercise intensity and cardiovascular burden), the saturation of oxygen in the blood (used to judge the body's oxygen supply situation and breathing frequency), and the body surface or core body temperature.

[0076] Through a sensor array installed on the vehicle, the speed, acceleration, power status of the vehicle, and the power output data of the rider are collected. The real-time vehicle status data includes real-time riding speed, real-time power information, and real-time power output. Among them, the real-time riding speed refers to the current riding speed of the vehicle (unit: km / h or m / s) and acceleration (the rate of speed change, unit: m / s²), reflecting the riding dynamics; the real-time power information refers to the remaining power of the vehicle battery (usually expressed as a percentage), used to measure the endurance; the real-time power output refers to the physical output of the rider, usually related to the pedal output power monitored by the vehicle sensor, reflecting the physical energy consumed by the user.

[0077] The real-time user physiological data and real-time vehicle status data are fused and analyzed to obtain real-time riding status parameters, including real-time exercise load, real-time fatigue index, and real-time energy efficiency. Among them, the real-time exercise load is an index to measure the current exercise intensity, obtained by the joint analysis of heart rate and power output; the real-time fatigue index is to evaluate the user's fatigue degree according to the trends of long-term heart rate and blood oxygen saturation; the real-time energy efficiency is calculated dynamically based on the battery endurance and the vehicle speed.

[0078] The change in the user's heart rate reflects the heart burden, and the power output reflects the physical consumption of the user. The combination of the two can comprehensively evaluate the real-time exercise load. That is to say, the real-time exercise load = the ratio of the real-time heart rate to the maximum heart rate + the ratio of the real-time power output to the maximum power output. The change trends of long-term heart rate and blood oxygen level are key indicators of the user's fatigue state. A high heart rate and a decrease in blood oxygen level indicate a gradual increase in fatigue. For example, if the heart rate rises from 140 bpm to 165 bpm and the blood oxygen drops from 97% to 92% in the past 30 minutes, the fatigue index is 85 (value range 0 - 100). The energy efficiency measures the user's unit energy consumption performance, that is, the battery endurance time at the current speed. The real-time energy efficiency can be obtained by the ratio of the battery endurance (such as mileage) to the current vehicle speed, indicating the driving efficiency that the user can maintain at the current speed.

[0079] Determine multiple sample riding state thresholds corresponding to multiple sample riding scenarios, that is, the reference range or upper and lower limits of the state parameters of each riding scenario, used to judge the deviation degree of the real-time state. Determine the corresponding real-time riding state threshold according to the real-time riding scenario, compare the real-time riding state parameters and the real-time riding state threshold item by item, calculate the deviation, and obtain the real-time riding state deviation. Sort according to the magnitude of the absolute value of the deviation, and give priority to displaying the indicators with serious deviations in prominent positions to obtain the real-time screen display priority, judge which data is more urgent or important, and ensure that the most serious deviation indicator is displayed on the screen first. For example, when the rider's fatigue index is too high, display this indicator first to prompt the rider to rest; if the energy efficiency is low (such as 1.2), then remind the user to reduce the speed to extend the endurance time.

[0080] According to the screen display priority judgment, the layout of the instrument screen display content is adjusted in real time. The content with higher priority is displayed in the center of the screen or occupies a larger area, and the secondary content is placed in the corner position and displayed in a smaller font. For example, when the fatigue index is high, this item of information in the screen display will be displayed in a large font on the screen, while reducing the proportion of other information. Dynamic rearrangement is to adjust the display content and layout order on the screen according to the real-time screen display priority. The real-time cycling state parameters (such as exercise load, fatigue index, energy efficiency), real-time user physiological data, and real-time vehicle state data are loaded into the vehicle instrument screen display according to the adjusted layout. By dynamically adjusting the screen display layout, the cyclist can always see the data that needs to be focused on the most. According to the real-time monitoring of energy efficiency, it helps the cyclist reasonably allocate physical strength and battery energy, improve the endurance during cycling, timely prompt the cyclist to pay attention to rest or adjust the cycling intensity, prevent accidents caused by excessive fatigue, and thus improve cycling efficiency and safety.

[0081] Furthermore, the present application further includes the following steps:

[0082] Interact to obtain multiple sample cycling state thresholds for the multiple sample cycling scenarios; call the real-time cycling state threshold from the multiple sample cycling state thresholds according to the real-time cycling scenario; traverse the real-time cycling state threshold with the real-time cycling state parameters to obtain the real-time cycling state deviation, where the real-time cycling state deviation includes exercise load deviation, fatigue index deviation, and energy efficiency deviation; perform screen display priority judgment according to the real-time cycling state deviation to obtain the real-time screen display priority.

[0083] Specifically, interact with the Internet of Things platform or the historical cycling database to obtain multiple sample cycling state thresholds for multiple cycling scenarios, that is, the state parameter thresholds preset for multiple cycling scenarios, including the reference range or upper and lower limits of exercise load, fatigue index, and energy efficiency, which are used to judge the degree of deviation of the real-time state. According to the real-time cycling scenario (such as commuting mode or competitive mode), traverse among the multiple sample cycling state thresholds to determine the real-time cycling state threshold corresponding to the real-time cycling scenario.

[0084] Compare the real-time cycling status parameters (exercise load, fatigue index, energy efficiency) item by item with the real-time cycling status thresholds, calculate the deviations, and obtain the real-time cycling status deviations, including exercise load deviation, fatigue index deviation, and energy efficiency deviation. The larger the absolute value of the deviation, the more serious the deviation from the status; a positive deviation indicates exceeding the upper limit, and a negative deviation indicates falling below the lower limit. Sort the display priorities of each indicator according to the severity of the deviation. The more serious the deviation, the higher the priority. Sort according to the size of the absolute value of the deviation, and preferentially display the indicators with serious deviations in prominent positions. For example, if the fatigue index deviation is the largest, the information related to the fatigue index occupies a more prominent position on the screen. Based on the deviation analysis, accurately locate the abnormal degree of the status parameters, improve the sensitivity of the cyclist to their own and the vehicle's status, and dynamically rearrange the screen display layout according to the severity of the deviation to ensure that the most critical information is always prominently displayed.

[0085] In summary, the vehicle instrument device control method supporting online processing provided by this application has the following technical effects:

[0086] By receiving historical cycling data and performing retrospective attention analysis on the historical cycling data, a screen display layout template library is obtained. Among them, multiple sample cycling scenarios and multiple sample screen display layouts are associated and stored in the screen display layout template library; fuse and analyze the instrument input data and real-time terrain data, and output the real-time cycling scenario. Among them, the real-time terrain data is obtained in real time through the cloud navigation service; after scheduling the real-time screen display layout in the screen display layout template library according to the real-time cycling scenario, use the real-time screen display layout to smoothly replace the standardized screen display configuration of the vehicle instrument device to generate a real-time instrument screen display; after receiving the real-time ambient brightness detected and transmitted back by the light sensor, input the real-time ambient brightness into the ambient brightness adjustment network for brightness adjustment analysis, and output a brightness adjustment strategy; obtain real-time weather data through the API, and perform screen display layout analysis according to the real-time weather data and real-time terrain data, and output a layout optimization strategy; use the brightness adjustment strategy and layout optimization strategy to refresh the real-time instrument screen display; preset a screen display refresh window, and use the screen display refresh window as a constraint to perform adaptive periodic online refresh control on the real-time instrument screen display. That is to say, by determining the screen display layout template library according to historical cycling data, combining real-time terrain data and instrument data to determine the current cycling scenario, dynamically scheduling the real-time screen display layout, intelligently adjusting the display brightness and layout according to light conditions and weather changes, and presetting the screen display refresh window to automatically update the screen display information according to real-time changes, the adaptability, intelligent level of the vehicle instrument device, and the safety and efficiency of the user's cycling are improved.

[0087] Embodiment 2. Based on the same inventive concept as the vehicle instrument device control method supporting online processing in the foregoing Embodiment 1, this application also provides a vehicle instrument device control system supporting online processing. Please refer to the appendixFigure 2 , the vehicle instrument device control system supporting online processing includes:

[0088] A historical data analysis module 11, which is used to receive historical riding data and perform retrospective attention analysis on the historical riding data to obtain a screen display layout template library. Among them, multiple sample riding scenarios and multiple sample screen display layouts are associated and stored in the screen display layout template library; a riding scenario analysis module 12, which is used to fuse and analyze instrument input data and real-time terrain data and output a real-time riding scenario. Among them, the real-time terrain data is obtained in real time through a cloud navigation service; a screen display layout scheduling module 13, which is used to schedule a real-time screen display layout in the screen display layout template library according to the real-time riding scenario, and then use the real-time screen display layout to smoothly replace the standardized screen display configuration of the vehicle instrument device to generate a real-time instrument screen display; an ambient brightness adjustment module 14, which is used to receive the real-time ambient brightness detected and transmitted back by a light sensor, input the real-time ambient brightness into an ambient brightness adjustment network for brightness adjustment analysis, and output a brightness adjustment strategy; a screen display layout optimization module 15, which is used to obtain real-time weather data through an API and perform screen display layout analysis according to the real-time weather data and real-time terrain data, and output a layout optimization strategy; a screen display update module 16, which is used to refresh the real-time instrument screen display by using the brightness adjustment strategy and the layout optimization strategy; an online refresh control module 17, which is used to preset a screen display refresh window and perform adaptive periodic online refresh control on the real-time instrument screen display with the screen display refresh window as a constraint.

[0089] Further, the historical data analysis module 11 in the vehicle instrument device control system supporting online processing is further used for:

[0090] Preset a time series division scale, and divide the historical riding data with the time series division scale as a constraint to obtain M historical riding subsets. Among them, the historical riding subsets include historical instrument input records and historical terrain records; input the M historical riding subsets into a riding scenario recognition network for scenario probability recognition, and then obtain M historical riding scenarios through probability transfer analysis; perform scenario consistency aggregation on the M historical riding scenarios to obtain multiple historical riding scenario sets of the multiple sample riding scenarios; aggregate the M riding index query records of the M historical riding subsets according to the multiple historical riding scenario sets to obtain multiple groups of riding index query records; perform retrospective attention analysis on the multiple groups of riding index query records to obtain the screen display layout template library.

[0091] Further, the historical data analysis module 11 in the vehicle instrument device control system supporting online processing is further configured to:

[0092] Statistically analyze the index backtracking query frequency of the multiple groups of cycling index query records to obtain multiple groups of index query frequencies. Among them, the cycling indexes include heart rate index, blood oxygen saturation index, body temperature index, driving pace index, power consumption index, power output index, exercise load index, fatigue index, and energy efficiency index; construct multiple user attention sequences according to the multiple groups of index query frequencies; perform index display layout analysis based on the index attention of the multiple user attention sequences to generate the multiple sample display layouts; and associatively store the multiple sample cycling scenarios and the multiple sample display layouts to complete the construction of the display layout template library.

[0093] Further, the cycling scenario analysis module 12 in the vehicle instrument device control system supporting online processing is further configured to:

[0094] Interactively obtain multiple sample instrument input records, multiple sample terrain records, and multiple sample scenario probability records of the multiple sample cycling scenarios. Among them, the sample instrument input records include multiple historical instrument inputs, and the historical instrument inputs include historical cycling state parameters, historical user physiological data, and historical vehicle state data; construct a cycling scenario recognition network based on a CNN network; use the multiple sample instrument input records, multiple sample terrain records, and multiple sample cycling scenarios as training data to optimize the parameters of the cycling scenario recognition network to complete the localization of the cycling scenario recognition network; load the instrument input data and real-time terrain data into the cycling scenario recognition network and output multiple cycling scenario probability parameters of the multiple sample cycling scenarios; perform probability transfer analysis on the multiple cycling scenario probability parameters and output the real-time cycling scenario.

[0095] Further, the cycling scenario analysis module 12 in the vehicle instrument device control system supporting online processing is further configured to:

[0096] Interactively obtain historical scenario alternation records; calculate multiple groups of sample scenario transition probabilities of the multiple sample cycling scenarios according to the historical scenario alternation records; construct a scenario state transition matrix using the multiple groups of sample scenario transition probabilities; input the multiple cycling scenario probability parameters into the scenario state transition matrix and calculate multiple updated scenario probability parameters; sum the multiple cycling scenario probability parameters and the multiple updated scenario probability parameters, and serialize the multiple sample cycling scenarios according to the summation result to obtain a sample scenario probability sequence; perform a maximum value call on the sample scenario probability sequence and output the real-time cycling scenario.

[0097] Furthermore, the screen display layout scheduling module 13 in the vehicle instrument device control system supporting online processing is further configured to:

[0098] Collect physiological data of cycling users through wearable devices to obtain real-time user physiological data, where the real-time user physiological data includes real-time heart rate, real-time blood oxygen saturation, and real-time body temperature; call real-time vehicle state data from the vehicle state sensing array, where the real-time vehicle state data includes real-time riding speed, real-time power information, and real-time power output; perform fusion analysis on the real-time user physiological data and real-time vehicle state data to obtain real-time riding state parameters, where the real-time riding state parameters include real-time exercise load, real-time fatigue index, and real-time energy efficiency; determine the screen display priority according to the real-time riding state parameters and real-time riding scenarios to obtain the real-time screen display priority; after dynamically rearranging the real-time instrument screen display according to the real-time screen display priority, load the real-time riding state parameters, real-time user physiological data, and real-time vehicle state data to the real-time instrument screen display.

[0099] Furthermore, the screen display layout scheduling module 13 in the vehicle instrument device control system supporting online processing is further configured to:

[0100] Interactively obtain multiple sample riding state thresholds for the multiple sample riding scenarios; call the real-time riding state threshold from the multiple sample riding state thresholds according to the real-time riding scenario; traverse the real-time riding state threshold with the real-time riding state parameters to obtain real-time riding state deviations, where the real-time riding state deviations include exercise load deviation, fatigue index deviation, and energy efficiency deviation; determine the screen display priority according to the real-time riding state deviations to obtain the real-time screen display priority.

[0101] Each embodiment in this specification is described in a progressive manner. The key point of each embodiment is to illustrate the differences from other embodiments. The Figure 1 A vehicle instrument device control method and specific examples in the first embodiment are equally applicable to the vehicle instrument device control system in this embodiment. Through the detailed description of the vehicle instrument device control method supporting online processing above, those skilled in the art can clearly know the vehicle instrument device control system in this embodiment. Therefore, for the sake of simplicity of the specification, it will not be elaborated here. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple. For related parts, please refer to the description in the method part.

[0102] The foregoing description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0103] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is also intended to include these changes and modifications.

Claims

1. A vehicle instrument device control method supporting online processing, characterized in that Including: Receiving historical riding data, and performing retrospective attention analysis on the historical riding data to obtain a screen display layout template library, wherein a plurality of sample riding scenarios and a plurality of sample screen display layouts are associated and stored in the screen display layout template library; Fusion-analyzing instrument input data and real-time terrain data, and outputting a real-time riding scenario, wherein the real-time terrain data is obtained in real time through a cloud navigation service; After scheduling a real-time screen display layout in the screen display layout template library according to the real-time riding scenario, smoothly replacing the standardized screen display configuration of the vehicle instrument device with the real-time screen display layout to generate a real-time instrument screen display; After receiving the real-time ambient brightness detected and transmitted back by a light sensor, inputting the real-time ambient brightness into an ambient brightness adjustment network for brightness adjustment analysis, and outputting a brightness adjustment strategy; Obtaining real-time weather data through an API, and performing screen display layout analysis according to the real-time weather data and the real-time terrain data, and outputting a layout optimization strategy; Refreshing the real-time instrument screen display by using the brightness adjustment strategy and the layout optimization strategy; Presetting a screen display refresh window, and performing adaptive periodic online refresh control on the real-time instrument screen display with the screen display refresh window as a constraint; Receiving historical riding data, and performing retrospective attention analysis on the historical riding data to obtain a screen display layout template library, including: Presetting a time series division scale, and dividing the historical riding data with the time series division scale as a constraint to obtain M historical riding subsets, wherein the historical riding subsets include historical instrument input records and historical terrain records; Inputting the M historical riding subsets into a riding scenario recognition network for scenario probability recognition, and obtaining M historical riding scenarios through probability transfer analysis; Performing scenario consistency aggregation on the M historical riding scenarios to obtain multiple historical riding scenario sets of the multiple sample riding scenarios; Aggregating M riding index query records of the M historical riding subsets according to the multiple historical riding scenario sets to obtain multiple groups of riding index query records; Performing retrospective attention analysis on the multiple groups of riding index query records to obtain the screen display layout template library.

2. The vehicle instrument device control method supporting online processing according to claim 1, characterized in that Fusion-analyzing instrument input data and real-time terrain data, and outputting a real-time riding scenario, including: Interactively obtaining multiple sample instrument input records, multiple sample terrain records, and multiple sample scenario probability records of the multiple sample riding scenarios, wherein the sample instrument input records include multiple historical instrument inputs, and the historical instrument inputs include historical riding state parameters, historical user physiological data, and historical vehicle state data; Constructing a riding scenario recognition network based on a CNN network; Using the multiple sample instrument input records, the multiple sample terrain records, and the multiple sample riding scenarios as training data to perform parameter tuning and optimization of the riding scenario recognition network, and completing the localization of the riding scenario recognition network; Loading the instrument input data and the real-time terrain data into the riding scenario recognition network, and outputting multiple riding scenario probability parameters of the multiple sample riding scenarios; Performing probability transfer analysis on the multiple riding scenario probability parameters, and outputting the real-time riding scenario.

3. A vehicle instrument device control method supporting online processing as described in claim 1, characterized in that, Perform retrospective attention analysis on the multiple groups of cycling index query records to obtain the screen display layout template library, including: Perform statistical analysis on the index retrospective query frequency of the multiple groups of cycling index query records to obtain multiple groups of index query frequencies. Among them, the cycling indexes include heart rate index, blood oxygen saturation index, body temperature index, driving pace index, battery level index, power output index, exercise load index, fatigue index index, and energy efficiency index; Construct multiple user attention sequences according to the multiple groups of index query frequencies; Conduct index screen display layout analysis based on the index attention degrees of the multiple user attention sequences to generate the multiple sample screen display layouts; Associate and store the multiple sample cycling scenarios and the multiple sample screen display layouts to complete the construction of the screen display layout template library.

4. A method for controlling a vehicle instrument device supporting online processing according to claim 1, characterized in that Use the real-time screen display layout to smoothly replace the standardized screen display configuration of the vehicle instrument device to generate a real-time instrument screen display. After that, it includes: Collect physiological data of cycling users through wearable devices to obtain real-time user physiological data. Among them, the real-time user physiological data includes real-time heart rate, real-time blood oxygen saturation, and real-time body temperature; Call real-time vehicle state data from the vehicle state sensing array. Among them, the real-time vehicle state data includes real-time driving pace, real-time battery level information, and real-time power output; Conduct fusion analysis on the real-time user physiological data and the real-time vehicle state data to obtain real-time cycling state parameters. Among them, the real-time cycling state parameters include real-time exercise load, real-time fatigue index, and real-time energy efficiency; Judge the screen display priority according to the real-time cycling state parameters and the real-time cycling scenario to obtain the real-time screen display priority; After dynamically rearranging the real-time instrument screen display according to the real-time screen display priority, load the real-time cycling state parameters, real-time user physiological data, and real-time vehicle state data into the real-time instrument screen display.

5. The vehicle instrument device control method supporting online processing according to claim 2, characterized in that, Conduct probability transfer analysis on the multiple cycling scenario probability parameters and output the real-time cycling scenario, including: Interactively obtain historical scenario alternation records; Calculate multiple groups of sample scenario transition probabilities of the multiple sample cycling scenarios according to the historical scenario alternation records; Construct a scenario state transition matrix using the multiple groups of sample scenario transition probabilities; Input the multiple cycling scenario probability parameters into the scenario state transition matrix and calculate to obtain multiple updated scenario probability parameters; Sum the multiple cycling scenario probability parameters and the multiple updated scenario probability parameters, and serialize the multiple sample cycling scenarios according to the summation result to obtain a sample scenario probability sequence; Call the maximum value of the sample scenario probability sequence and output the real-time cycling scenario.

6. The vehicle instrument device control method supporting online processing according to claim 4, characterized in that, Judge the screen display priority according to the real-time cycling state parameters and the real-time cycling scenario to obtain the real-time screen display priority, including: Interactively obtain multiple sample cycling state thresholds of the multiple sample cycling scenarios. The cycling state threshold is the reference range or upper and lower limits of the state parameters of each cycling scenario; Call the real-time cycling state threshold from the multiple sample cycling state thresholds according to the real-time cycling scenario; Traverse the real-time riding state threshold with the real-time riding state parameters to obtain the real-time riding state deviation, where the real-time riding state deviation includes a motion load deviation, a fatigue index deviation, and an energy efficiency deviation; Judge the screen display priority according to the real-time riding state deviation to obtain the real-time screen display priority.

7. A vehicle instrument device control system that supports online processing, characterized in that, For implementing the steps of the method for controlling a vehicle instrument device supporting online processing according to any one of claims 1 to 6, the system for controlling a vehicle instrument device supporting online processing includes: A historical data analysis module, which is configured to receive historical riding data and perform retrospective attention analysis on the historical riding data to obtain a screen display layout template library, where a plurality of sample riding scenarios and a plurality of sample screen display layouts are associated and stored in the screen display layout template library; A riding scenario analysis module, which is configured to perform fusion analysis on instrument input data and real-time terrain data and output a real-time riding scenario, where the real-time terrain data is obtained in real time through a cloud navigation service; A screen display layout scheduling module, which is configured to schedule a real-time screen display layout in the screen display layout template library according to the real-time riding scenario, and then smoothly replace the standardized screen display configuration of the vehicle instrument device with the real-time screen display layout to generate a real-time instrument screen display; An ambient brightness adjustment module, which is configured to receive the real-time ambient brightness detected and transmitted back by a light sensor, input the real-time ambient brightness into an ambient brightness adjustment network for brightness adjustment analysis, and output a brightness adjustment strategy; A screen display layout optimization module, which is configured to obtain real-time weather data through an API and perform screen display layout analysis according to the real-time weather data and the real-time terrain data, and output a layout optimization strategy; A screen display update module, which is configured to refresh the real-time instrument screen display by using the brightness adjustment strategy and the layout optimization strategy; An online refresh control module, which is configured to preset a screen display refresh window and perform adaptive periodic online refresh control on the real-time instrument screen display with the screen display refresh window as a constraint.

Images